Series comparison

-[PULL 00/34] Block patches
+[Qemu-devel] [PULL v2 00/24] Block patches
-The following changes since commit f17783e706ab9c7b3a2b69cf48e4f0ba40664f54:
+The following changes since commit 56f9e46b841c7be478ca038d8d4085d776ab4b0d:
-  Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20191220' into staging (2020-01-03 18:50:33 +0000)
+  Merge remote-tracking branch 'remotes/armbru/tags/pull-qapi-2017-02-20' into staging (2017-02-20 17:42:47 +0000)
-are available in the Git repository at:
+are available in the git repository at:
-  https://github.com/XanClic/qemu.git tags/pull-block-2020-01-06
+  git://github.com/stefanha/qemu.git tags/block-pull-request
-for you to fetch changes up to 503ca1262bab2c11c533a4816d1ff4297d4f58a6:
+for you to fetch changes up to a7b91d35bab97a2d3e779d0c64c9b837b52a6cf7:
-  backup-top: Begin drain earlier (2020-01-06 14:26:23 +0100)
+  coroutine-lock: make CoRwlock thread-safe and fair (2017-02-21 11:39:40 +0000)
 ----------------------------------------------------------------
-Block patches:
+Pull request
-- Minor fixes and tests from the freeze period (too minor to be included
-  in 4.2)
+v2:
-- Allow many bash iotests to test qcow2's external data file feature
+ * Rebased to resolve scsi conflicts
 - Add compress filter driver
 - Fix Python iotests after 6f6e1698a6
 - Fix for the backup job
 ----------------------------------------------------------------
-Andrey Shinkevich (3):
-  block: introduce compress filter driver
-  qcow2: Allow writing compressed data of multiple clusters
-  tests/qemu-iotests: add case to write compressed data of multiple
-    clusters
-Max Reitz (28):
+Paolo Bonzini (24):
-  block: Add bdrv_qapi_perm_to_blk_perm()
+  block: move AioContext, QEMUTimer, main-loop to libqemuutil
-  block: Use bdrv_qapi_perm_to_blk_perm()
+  aio: introduce aio_co_schedule and aio_co_wake
-  blkdebug: Allow taking/unsharing permissions
+  block-backend: allow blk_prw from coroutine context
-  iotests: Add @error to wait_until_completed
+  test-thread-pool: use generic AioContext infrastructure
-  iotests: Add test for failing mirror complete
+  io: add methods to set I/O handlers on AioContext
-  iotests: s/qocw2/qcow2/
+  io: make qio_channel_yield aware of AioContexts
-  iotests/qcow2.py: Add dump-header-exts
+  nbd: convert to use qio_channel_yield
-  iotests/qcow2.py: Split feature fields into bits
+  coroutine-lock: reschedule coroutine on the AioContext it was running
-  iotests: Add _filter_json_filename
+    on
-  iotests: Filter refcount_order in 036
+  blkdebug: reschedule coroutine on the AioContext it is running on
-  iotests: Replace IMGOPTS by _unsupported_imgopts
+  qed: introduce qed_aio_start_io and qed_aio_next_io_cb
-  iotests: Drop compat=1.1 in 050
+  aio: push aio_context_acquire/release down to dispatching
-  iotests: Let _make_test_img parse its parameters
+  block: explicitly acquire aiocontext in timers that need it
-  iotests: Add -o and --no-opts to _make_test_img
+  block: explicitly acquire aiocontext in callbacks that need it
-  iotests: Inject space into -ocompat=0.10 in 051
+  block: explicitly acquire aiocontext in bottom halves that need it
-  iotests: Replace IMGOPTS= by -o
+  block: explicitly acquire aiocontext in aio callbacks that need it
-  iotests: Replace IMGOPTS='' by --no-opts
+  aio-posix: partially inline aio_dispatch into aio_poll
-  iotests: Drop IMGOPTS use in 267
+  async: remove unnecessary inc/dec pairs
-  iotests: Avoid qemu-img create
+  block: document fields protected by AioContext lock
-  iotests: Use _rm_test_img for deleting test images
+  coroutine-lock: make CoMutex thread-safe
-  iotests: Avoid cp/mv of test images
+  coroutine-lock: add limited spinning to CoMutex
-  iotests: Make 091 work with data_file
+  test-aio-multithread: add performance comparison with thread-based
-  iotests: Make 110 work with data_file
+    mutexes
-  iotests: Make 137 work with data_file
+  coroutine-lock: place CoMutex before CoQueue in header
-  iotests: Make 198 work with data_file
+  coroutine-lock: add mutex argument to CoQueue APIs
-  iotests: Disable data_file where it cannot be used
+  coroutine-lock: make CoRwlock thread-safe and fair
   iotests: Allow check -o data_file
   backup-top: Begin drain earlier
-PanNengyuan (1):
+ Makefile.objs                       |   4 -
-  throttle-groups: fix memory leak in throttle_group_set_limit:
+ stubs/Makefile.objs                 |   1 +
+ tests/Makefile.include              |  19 +-
-Philippe Mathieu-Daudé (1):
+ util/Makefile.objs                  |   6 +-
-  tests/qemu-iotests: Update tests to recent desugarized -accel option
+ block/nbd-client.h                  |   2 +-
+ block/qed.h                         |   3 +
-Vladimir Sementsov-Ogievskiy (1):
+ include/block/aio.h                 |  38 ++-
-  qcow2-bitmaps: fix qcow2_can_store_new_dirty_bitmap
+ include/block/block_int.h           |  64 +++--
+ include/io/channel.h                |  72 +++++-
- block.c                          |  47 +++++----
+ include/qemu/coroutine.h            |  84 ++++---
- block/Makefile.objs              |   1 +
+ include/qemu/coroutine_int.h        |  11 +-
- block/backup-top.c               |   4 +-
+ include/sysemu/block-backend.h      |  14 +-
- block/blkdebug.c                 |  93 ++++++++++++++++-
+ tests/iothread.h                    |  25 ++
- block/filter-compress.c          | 168 +++++++++++++++++++++++++++++++
+ block/backup.c                      |   2 +-
- block/qcow2-bitmap.c             |  41 ++++----
+ block/blkdebug.c                    |   9 +-
- block/qcow2.c                    | 102 ++++++++++++++-----
+ block/blkreplay.c                   |   2 +-
- block/throttle-groups.c          |   4 +-
+ block/block-backend.c               |  13 +-
- include/block/block.h            |   1 +
+ block/curl.c                        |  44 +++-
- qapi/block-core.json             |  24 ++++-
+ block/gluster.c                     |   9 +-
- tests/qemu-iotests/007           |   5 +-
+ block/io.c                          |  42 +---
- tests/qemu-iotests/014           |   2 +
+ block/iscsi.c                       |  15 +-
- tests/qemu-iotests/015           |   5 +-
+ block/linux-aio.c                   |  10 +-
- tests/qemu-iotests/019           |   6 +-
+ block/mirror.c                      |  12 +-
- tests/qemu-iotests/020           |   6 +-
+ block/nbd-client.c                  | 119 +++++----
- tests/qemu-iotests/024           |  10 +-
+ block/nfs.c                         |   9 +-
- tests/qemu-iotests/026           |   5 +-
+ block/qcow2-cluster.c               |   4 +-
- tests/qemu-iotests/028           |   2 +-
+ block/qed-cluster.c                 |   2 +
- tests/qemu-iotests/029           |   7 +-
+ block/qed-table.c                   |  12 +-
- tests/qemu-iotests/031           |   9 +-
+ block/qed.c                         |  58 +++--
- tests/qemu-iotests/031.out       |  36 +++----
+ block/sheepdog.c                    |  31 +--
- tests/qemu-iotests/036           |  15 +--
+ block/ssh.c                         |  29 +--
- tests/qemu-iotests/036.out       |  66 ++----------
+ block/throttle-groups.c             |   4 +-
- tests/qemu-iotests/039           |  27 ++---
+ block/win32-aio.c                   |   9 +-
- tests/qemu-iotests/039.out       |  22 ++--
+ dma-helpers.c                       |   2 +
- tests/qemu-iotests/041           |  44 ++++++++
+ hw/9pfs/9p.c                        |   2 +-
- tests/qemu-iotests/041.out       |   4 +-
+ hw/block/virtio-blk.c               |  19 +-
- tests/qemu-iotests/043           |   4 +-
+ hw/scsi/scsi-bus.c                  |   2 +
- tests/qemu-iotests/046           |   2 +
+ hw/scsi/scsi-disk.c                 |  15 ++
- tests/qemu-iotests/048           |   4 +-
+ hw/scsi/scsi-generic.c              |  20 +-
- tests/qemu-iotests/050           |   8 +-
+ hw/scsi/virtio-scsi.c               |   7 +
- tests/qemu-iotests/051           |   7 +-
+ io/channel-command.c                |  13 +
- tests/qemu-iotests/053           |   4 +-
+ io/channel-file.c                   |  11 +
- tests/qemu-iotests/058           |   7 +-
+ io/channel-socket.c                 |  16 +-
- tests/qemu-iotests/059           |  20 ++--
+ io/channel-tls.c                    |  12 +
- tests/qemu-iotests/060           |  14 +--
+ io/channel-watch.c                  |   6 +
- tests/qemu-iotests/060.out       |  20 ++--
+ io/channel.c                        |  97 ++++++--
- tests/qemu-iotests/061           |  63 ++++++------
+ nbd/client.c                        |   2 +-
- tests/qemu-iotests/061.out       |  72 ++++++-------
+ nbd/common.c                        |   9 +-
- tests/qemu-iotests/062           |   5 +-
+ nbd/server.c                        |  94 +++-----
- tests/qemu-iotests/063           |  18 ++--
+ stubs/linux-aio.c                   |  32 +++
- tests/qemu-iotests/063.out       |   3 +-
+ stubs/set-fd-handler.c              |  11 -
- tests/qemu-iotests/066           |   7 +-
+ tests/iothread.c                    |  91 +++++++
- tests/qemu-iotests/067           |   6 +-
+ tests/test-aio-multithread.c        | 463 ++++++++++++++++++++++++++++++++++++
- tests/qemu-iotests/068           |   6 +-
+ tests/test-thread-pool.c            |  12 +-
- tests/qemu-iotests/069           |   2 +-
+ aio-posix.c => util/aio-posix.c     |  62 ++---
- tests/qemu-iotests/071           |   7 +-
+ aio-win32.c => util/aio-win32.c     |  30 +--
- tests/qemu-iotests/073           |   4 +
+ util/aiocb.c                        |  55 +++++
- tests/qemu-iotests/074           |   4 +-
+ async.c => util/async.c             |  84 ++++++-
- tests/qemu-iotests/079           |   3 +-
+ iohandler.c => util/iohandler.c     |   0
- tests/qemu-iotests/080           |   7 +-
+ main-loop.c => util/main-loop.c     |   0
- tests/qemu-iotests/081           |   6 +-
+ util/qemu-coroutine-lock.c          | 254 ++++++++++++++++++--
- tests/qemu-iotests/085           |  18 ++--
+ util/qemu-coroutine-sleep.c         |   2 +-
- tests/qemu-iotests/085.out       |   8 +-
+ util/qemu-coroutine.c               |   8 +
- tests/qemu-iotests/088           |   2 +-
+ qemu-timer.c => util/qemu-timer.c   |   0
- tests/qemu-iotests/090           |   2 +
+ thread-pool.c => util/thread-pool.c |   8 +-
- tests/qemu-iotests/091           |   2 +-
+ trace-events                        |  11 -
- tests/qemu-iotests/091.out       |   2 -
+ util/trace-events                   |  17 +-
- tests/qemu-iotests/092           |   2 +-
+files changed, 1712 insertions(+), 533 deletions(-)
- tests/qemu-iotests/094           |   4 +-
+ create mode 100644 tests/iothread.h
- tests/qemu-iotests/095           |   5 +-
+ create mode 100644 stubs/linux-aio.c
- tests/qemu-iotests/098           |   6 +-
+ create mode 100644 tests/iothread.c
- tests/qemu-iotests/099           |  10 +-
+ create mode 100644 tests/test-aio-multithread.c
- tests/qemu-iotests/103           |   5 +-
+ rename aio-posix.c => util/aio-posix.c (94%)
- tests/qemu-iotests/106           |   2 +-
+ rename aio-win32.c => util/aio-win32.c (95%)
- tests/qemu-iotests/108           |  10 +-
+ create mode 100644 util/aiocb.c
- tests/qemu-iotests/109           |   4 +-
+ rename async.c => util/async.c (82%)
- tests/qemu-iotests/110           |  11 +-
+ rename iohandler.c => util/iohandler.c (100%)
- tests/qemu-iotests/110.out       |   4 +-
+ rename main-loop.c => util/main-loop.c (100%)
- tests/qemu-iotests/111           |   3 +-
+ rename qemu-timer.c => util/qemu-timer.c (100%)
- tests/qemu-iotests/112           |  37 +++----
+ rename thread-pool.c => util/thread-pool.c (97%)
  tests/qemu-iotests/114           |   2 +
  tests/qemu-iotests/115           |   3 +-
  tests/qemu-iotests/121           |   9 +-
  tests/qemu-iotests/122           |   6 +-
  tests/qemu-iotests/123           |   4 +-
  tests/qemu-iotests/125           |   2 +-
  tests/qemu-iotests/137           |  17 +++-
  tests/qemu-iotests/137.out       |   6 +-
  tests/qemu-iotests/138           |   8 +-
  tests/qemu-iotests/141           |   4 +-
  tests/qemu-iotests/142           |   2 +-
  tests/qemu-iotests/144           |   4 +-
  tests/qemu-iotests/153           |  12 +--
  tests/qemu-iotests/156           |  10 +-
  tests/qemu-iotests/159           |   2 +-
  tests/qemu-iotests/160           |   3 +-
  tests/qemu-iotests/161           |   4 +-
  tests/qemu-iotests/170           |   2 +-
  tests/qemu-iotests/172           |   6 +-
  tests/qemu-iotests/173           |   3 +-
  tests/qemu-iotests/174           |   2 +-
  tests/qemu-iotests/175           |   2 +-
  tests/qemu-iotests/176           |   7 +-
  tests/qemu-iotests/178           |   6 +-
  tests/qemu-iotests/182           |   2 +-
  tests/qemu-iotests/183           |   2 +-
  tests/qemu-iotests/185           |   4 +-
  tests/qemu-iotests/187           |   6 +-
  tests/qemu-iotests/190           |   4 +-
  tests/qemu-iotests/191           |  11 +-
  tests/qemu-iotests/195           |   2 +-
  tests/qemu-iotests/197           |   6 +-
  tests/qemu-iotests/198           |   6 +-
  tests/qemu-iotests/198.out       |   4 +-
  tests/qemu-iotests/200           |   7 +-
  tests/qemu-iotests/201           |   6 +-
  tests/qemu-iotests/214           |  46 ++++++++-
  tests/qemu-iotests/214.out       |  14 +++
  tests/qemu-iotests/215           |   6 +-
  tests/qemu-iotests/217           |   3 +-
  tests/qemu-iotests/220           |   5 +-
  tests/qemu-iotests/225           |   2 +-
  tests/qemu-iotests/229           |   3 +-
  tests/qemu-iotests/232           |   4 +-
  tests/qemu-iotests/235           |   2 +-
  tests/qemu-iotests/243           |  10 +-
  tests/qemu-iotests/244           |  15 +--
  tests/qemu-iotests/247           |   4 +-
  tests/qemu-iotests/249           |   4 +-
  tests/qemu-iotests/250           |   5 +-
  tests/qemu-iotests/252           |   2 +-
  tests/qemu-iotests/261           |   5 +-
  tests/qemu-iotests/265           |   2 +-
  tests/qemu-iotests/267           |  17 ++--
  tests/qemu-iotests/273           |   3 +
  tests/qemu-iotests/check         |   6 +-
  tests/qemu-iotests/common.filter |  47 ++++++++-
  tests/qemu-iotests/common.rc     |  63 ++++++++++--
  tests/qemu-iotests/iotests.py    |  18 ++--
  tests/qemu-iotests/qcow2.py      |  23 ++++-
 files changed, 1139 insertions(+), 552 deletions(-)
  create mode 100644 block/filter-compress.c
 --
-.24.1
+.9.3

-[PULL 19/34] iotests: Replace IMGOPTS='' by --no-opts
+[Qemu-devel] [PULL v2 01/24] block: move AioContext, QEMUTimer, main-loop to libqemuutil
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-13-mreitz@redhat.com
+AioContext is fairly self contained, the only dependency is QEMUTimer but
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+that in turn doesn't need anything else.  So move them out of block-obj-y
 to avoid introducing a dependency from io/ to block-obj-y.
 main-loop and its dependency iohandler also need to be moved, because
 later in this series io/ will call iohandler_get_aio_context.
 [Changed copyright "the QEMU team" to "other QEMU contributors" as
 suggested by Daniel Berrange and agreed by Paolo.
 --Stefan]
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Message-id: 20170213135235.12274-2-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/071 | 4 ++--
+ Makefile.objs                       |  4 ---
- tests/qemu-iotests/174 | 2 +-
+ stubs/Makefile.objs                 |  1 +
- tests/qemu-iotests/178 | 4 ++--
+ tests/Makefile.include              | 11 ++++----
- tests/qemu-iotests/197 | 4 ++--
+ util/Makefile.objs                  |  6 +++-
- tests/qemu-iotests/215 | 4 ++--
+ block/io.c                          | 29 -------------------
-files changed, 9 insertions(+), 9 deletions(-)
+ stubs/linux-aio.c                   | 32 +++++++++++++++++++++
+ stubs/set-fd-handler.c              | 11 --------
-diff --git a/tests/qemu-iotests/071 b/tests/qemu-iotests/071
+ aio-posix.c => util/aio-posix.c     |  2 +-
-index XXXXXXX..XXXXXXX 100755
+ aio-win32.c => util/aio-win32.c     |  0
---- a/tests/qemu-iotests/071
+ util/aiocb.c                        | 55 +++++++++++++++++++++++++++++++++++++
-+++ b/tests/qemu-iotests/071
+ async.c => util/async.c             |  3 +-
-@@ -XXX,XX +XXX,XX @@ echo
+ iohandler.c => util/iohandler.c     |  0
- echo "=== Testing blkverify through filename ==="
+ main-loop.c => util/main-loop.c     |  0
- echo
+ qemu-timer.c => util/qemu-timer.c   |  0
+ thread-pool.c => util/thread-pool.c |  2 +-
--TEST_IMG="$TEST_IMG.base" IMGOPTS="" IMGFMT="raw" _make_test_img $IMG_SIZE |\
+ trace-events                        | 11 --------
-+TEST_IMG="$TEST_IMG.base" IMGFMT="raw" _make_test_img --no-opts $IMG_SIZE |\
+ util/trace-events                   | 11 ++++++++
-     _filter_imgfmt
+files changed, 114 insertions(+), 64 deletions(-)
- _make_test_img $IMG_SIZE
+ create mode 100644 stubs/linux-aio.c
- $QEMU_IO -c "open -o driver=raw,file.driver=blkverify,file.raw.filename=$TEST_IMG.base $TEST_IMG" \
+ rename aio-posix.c => util/aio-posix.c (99%)
-@@ -XXX,XX +XXX,XX @@ echo
+ rename aio-win32.c => util/aio-win32.c (100%)
- echo "=== Testing blkverify through file blockref ==="
+ create mode 100644 util/aiocb.c
- echo
+ rename async.c => util/async.c (99%)
+ rename iohandler.c => util/iohandler.c (100%)
--TEST_IMG="$TEST_IMG.base" IMGOPTS="" IMGFMT="raw" _make_test_img $IMG_SIZE |\
+ rename main-loop.c => util/main-loop.c (100%)
-+TEST_IMG="$TEST_IMG.base" IMGFMT="raw" _make_test_img --no-opts $IMG_SIZE |\
+ rename qemu-timer.c => util/qemu-timer.c (100%)
-     _filter_imgfmt
+ rename thread-pool.c => util/thread-pool.c (99%)
- _make_test_img $IMG_SIZE
- $QEMU_IO -c "open -o driver=raw,file.driver=blkverify,file.raw.filename=$TEST_IMG.base,file.test.driver=$IMGFMT,file.test.file.filename=$TEST_IMG" \
+diff --git a/Makefile.objs b/Makefile.objs
-diff --git a/tests/qemu-iotests/174 b/tests/qemu-iotests/174
+index XXXXXXX..XXXXXXX 100644
-index XXXXXXX..XXXXXXX 100755
+--- a/Makefile.objs
---- a/tests/qemu-iotests/174
++++ b/Makefile.objs
-+++ b/tests/qemu-iotests/174
+@@ -XXX,XX +XXX,XX @@ chardev-obj-y = chardev/
-@@ -XXX,XX +XXX,XX @@ _unsupported_fmt raw
+ #######################################################################
+ # block-obj-y is code used by both qemu system emulation and qemu-img
- size=256K
+-block-obj-y = async.o thread-pool.o
--IMGFMT=raw IMGKEYSECRET= IMGOPTS= _make_test_img $size | _filter_imgfmt
+ block-obj-y += nbd/
-+IMGFMT=raw IMGKEYSECRET= _make_test_img --no-opts $size | _filter_imgfmt
+ block-obj-y += block.o blockjob.o
+-block-obj-y += main-loop.o iohandler.o qemu-timer.o
- echo
+-block-obj-$(CONFIG_POSIX) += aio-posix.o
- echo "== reading wrong format should fail =="
+-block-obj-$(CONFIG_WIN32) += aio-win32.o
-diff --git a/tests/qemu-iotests/178 b/tests/qemu-iotests/178
+ block-obj-y += block/
-index XXXXXXX..XXXXXXX 100755
+ block-obj-y += qemu-io-cmds.o
---- a/tests/qemu-iotests/178
+ block-obj-$(CONFIG_REPLICATION) += replication.o
-+++ b/tests/qemu-iotests/178
+diff --git a/stubs/Makefile.objs b/stubs/Makefile.objs
-@@ -XXX,XX +XXX,XX @@ $QEMU_IMG measure -O foo "$TEST_IMG" # unknown image file format
+index XXXXXXX..XXXXXXX 100644
+--- a/stubs/Makefile.objs
- make_test_img_with_fmt() {
++++ b/stubs/Makefile.objs
-     # Shadow global variables within this function
+@@ -XXX,XX +XXX,XX @@ stub-obj-y += get-vm-name.o
--    local IMGFMT="$1" IMGOPTS=""
+ stub-obj-y += iothread.o
--    _make_test_img "$2"
+ stub-obj-y += iothread-lock.o
-+    local IMGFMT="$1"
+ stub-obj-y += is-daemonized.o
-+    _make_test_img --no-opts "$2"
++stub-obj-$(CONFIG_LINUX_AIO) += linux-aio.o
  stub-obj-y += machine-init-done.o
  stub-obj-y += migr-blocker.o
  stub-obj-y += monitor.o
 diff --git a/tests/Makefile.include b/tests/Makefile.include
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/Makefile.include
 +++ b/tests/Makefile.include
@@ -XXX,XX +XXX,XX @@ check-unit-y += tests/test-visitor-serialization$(EXESUF)
  check-unit-y += tests/test-iov$(EXESUF)
  gcov-files-test-iov-y = util/iov.c
  check-unit-y += tests/test-aio$(EXESUF)
 +gcov-files-test-aio-y = util/async.c util/qemu-timer.o
 +gcov-files-test-aio-$(CONFIG_WIN32) += util/aio-win32.c
 +gcov-files-test-aio-$(CONFIG_POSIX) += util/aio-posix.c
  check-unit-y += tests/test-throttle$(EXESUF)
  gcov-files-test-aio-$(CONFIG_WIN32) = aio-win32.c
  gcov-files-test-aio-$(CONFIG_POSIX) = aio-posix.c
@@ -XXX,XX +XXX,XX @@ tests/check-qjson$(EXESUF): tests/check-qjson.o $(test-util-obj-y)
  tests/check-qom-interface$(EXESUF): tests/check-qom-interface.o $(test-qom-obj-y)
  tests/check-qom-proplist$(EXESUF): tests/check-qom-proplist.o $(test-qom-obj-y)
 -tests/test-char$(EXESUF): tests/test-char.o qemu-timer.o \
 -    $(test-util-obj-y) $(qtest-obj-y) $(test-block-obj-y) $(chardev-obj-y)
 +tests/test-char$(EXESUF): tests/test-char.o $(test-util-obj-y) $(qtest-obj-y) $(test-io-obj-y) $(chardev-obj-y)
  tests/test-coroutine$(EXESUF): tests/test-coroutine.o $(test-block-obj-y)
  tests/test-aio$(EXESUF): tests/test-aio.o $(test-block-obj-y)
  tests/test-throttle$(EXESUF): tests/test-throttle.o $(test-block-obj-y)
@@ -XXX,XX +XXX,XX @@ tests/test-vmstate$(EXESUF): tests/test-vmstate.o \
      migration/vmstate.o migration/qemu-file.o \
          migration/qemu-file-channel.o migration/qjson.o \
      $(test-io-obj-y)
 -tests/test-timed-average$(EXESUF): tests/test-timed-average.o qemu-timer.o \
 -    $(test-util-obj-y)
 +tests/test-timed-average$(EXESUF): tests/test-timed-average.o $(test-util-obj-y)
  tests/test-base64$(EXESUF): tests/test-base64.o \
      libqemuutil.a libqemustub.a
  tests/ptimer-test$(EXESUF): tests/ptimer-test.o tests/ptimer-test-stubs.o hw/core/ptimer.o libqemustub.a
@@ -XXX,XX +XXX,XX @@ tests/usb-hcd-ehci-test$(EXESUF): tests/usb-hcd-ehci-test.o $(libqos-usb-obj-y)
  tests/usb-hcd-xhci-test$(EXESUF): tests/usb-hcd-xhci-test.o $(libqos-usb-obj-y)
  tests/pc-cpu-test$(EXESUF): tests/pc-cpu-test.o
  tests/postcopy-test$(EXESUF): tests/postcopy-test.o
 -tests/vhost-user-test$(EXESUF): tests/vhost-user-test.o qemu-timer.o \
 +tests/vhost-user-test$(EXESUF): tests/vhost-user-test.o $(test-util-obj-y) \
      $(qtest-obj-y) $(test-io-obj-y) $(libqos-virtio-obj-y) $(libqos-pc-obj-y) \
      $(chardev-obj-y)
  tests/qemu-iotests/socket_scm_helper$(EXESUF): tests/qemu-iotests/socket_scm_helper.o
 diff --git a/util/Makefile.objs b/util/Makefile.objs
 index XXXXXXX..XXXXXXX 100644
 --- a/util/Makefile.objs
 +++ b/util/Makefile.objs
@@ -XXX,XX +XXX,XX @@
  util-obj-y = osdep.o cutils.o unicode.o qemu-timer-common.o
  util-obj-y += bufferiszero.o
  util-obj-y += lockcnt.o
 +util-obj-y += aiocb.o async.o thread-pool.o qemu-timer.o
 +util-obj-y += main-loop.o iohandler.o
 +util-obj-$(CONFIG_POSIX) += aio-posix.o
  util-obj-$(CONFIG_POSIX) += compatfd.o
  util-obj-$(CONFIG_POSIX) += event_notifier-posix.o
  util-obj-$(CONFIG_POSIX) += mmap-alloc.o
  util-obj-$(CONFIG_POSIX) += oslib-posix.o
  util-obj-$(CONFIG_POSIX) += qemu-openpty.o
  util-obj-$(CONFIG_POSIX) += qemu-thread-posix.o
 -util-obj-$(CONFIG_WIN32) += event_notifier-win32.o
  util-obj-$(CONFIG_POSIX) += memfd.o
 +util-obj-$(CONFIG_WIN32) += aio-win32.o
 +util-obj-$(CONFIG_WIN32) += event_notifier-win32.o
  util-obj-$(CONFIG_WIN32) += oslib-win32.o
  util-obj-$(CONFIG_WIN32) += qemu-thread-win32.o
  util-obj-y += envlist.o path.o module.o
 diff --git a/block/io.c b/block/io.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/io.c
 +++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ BlockAIOCB *bdrv_aio_flush(BlockDriverState *bs,
      return &acb->common;
  }
- qemu_io_with_fmt() {
+-void *qemu_aio_get(const AIOCBInfo *aiocb_info, BlockDriverState *bs,
-diff --git a/tests/qemu-iotests/197 b/tests/qemu-iotests/197
+-                   BlockCompletionFunc *cb, void *opaque)
-index XXXXXXX..XXXXXXX 100755
+-{
---- a/tests/qemu-iotests/197
+-    BlockAIOCB *acb;
-+++ b/tests/qemu-iotests/197
+-
-@@ -XXX,XX +XXX,XX @@ if [ "$IMGFMT" = "vpc" ]; then
+-    acb = g_malloc(aiocb_info->aiocb_size);
- fi
+-    acb->aiocb_info = aiocb_info;
- _make_test_img 4G
+-    acb->bs = bs;
- $QEMU_IO -c "write -P 55 3G 1k" "$TEST_IMG" | _filter_qemu_io
+-    acb->cb = cb;
--IMGPROTO=file IMGFMT=qcow2 IMGOPTS= TEST_IMG_FILE="$TEST_WRAP" \
+-    acb->opaque = opaque;
--    _make_test_img -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create
+-    acb->refcnt = 1;
-+IMGPROTO=file IMGFMT=qcow2 TEST_IMG_FILE="$TEST_WRAP" \
+-    return acb;
-+    _make_test_img --no-opts -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create
+-}
- $QEMU_IO -f qcow2 -c "write -z -u 1M 64k" "$TEST_WRAP" | _filter_qemu_io
+-
+-void qemu_aio_ref(void *p)
- # Ensure that a read of two clusters, but where one is already allocated,
+-{
-diff --git a/tests/qemu-iotests/215 b/tests/qemu-iotests/215
+-    BlockAIOCB *acb = p;
-index XXXXXXX..XXXXXXX 100755
+-    acb->refcnt++;
---- a/tests/qemu-iotests/215
+-}
-+++ b/tests/qemu-iotests/215
+-
-@@ -XXX,XX +XXX,XX @@ if [ "$IMGFMT" = "vpc" ]; then
+-void qemu_aio_unref(void *p)
- fi
+-{
- _make_test_img 4G
+-    BlockAIOCB *acb = p;
- $QEMU_IO -c "write -P 55 3G 1k" "$TEST_IMG" | _filter_qemu_io
+-    assert(acb->refcnt > 0);
--IMGPROTO=file IMGFMT=qcow2 IMGOPTS= TEST_IMG_FILE="$TEST_WRAP" \
+-    if (--acb->refcnt == 0) {
--    _make_test_img -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create
+-        g_free(acb);
-+IMGPROTO=file IMGFMT=qcow2 TEST_IMG_FILE="$TEST_WRAP" \
+-    }
-+    _make_test_img --no-opts -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create
+-}
- $QEMU_IO -f qcow2 -c "write -z -u 1M 64k" "$TEST_WRAP" | _filter_qemu_io
+-
+ /**************************************************************/
- # Ensure that a read of two clusters, but where one is already allocated,
+ /* Coroutine block device emulation */
 diff --git a/stubs/linux-aio.c b/stubs/linux-aio.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/stubs/linux-aio.c
@@ -XXX,XX +XXX,XX @@
 +/*
 + * Linux native AIO support.
 + *
 + * Copyright (C) 2009 IBM, Corp.
 + * Copyright (C) 2009 Red Hat, Inc.
 + *
 + * This work is licensed under the terms of the GNU GPL, version 2 or later.
 + * See the COPYING file in the top-level directory.
 + */
 +#include "qemu/osdep.h"
 +#include "block/aio.h"
 +#include "block/raw-aio.h"
 +
 +void laio_detach_aio_context(LinuxAioState *s, AioContext *old_context)
 +{
 +    abort();
 +}
 +
 +void laio_attach_aio_context(LinuxAioState *s, AioContext *new_context)
 +{
 +    abort();
 +}
 +
 +LinuxAioState *laio_init(void)
 +{
 +    abort();
 +}
 +
 +void laio_cleanup(LinuxAioState *s)
 +{
 +    abort();
 +}
 diff --git a/stubs/set-fd-handler.c b/stubs/set-fd-handler.c
 index XXXXXXX..XXXXXXX 100644
 --- a/stubs/set-fd-handler.c
 +++ b/stubs/set-fd-handler.c
@@ -XXX,XX +XXX,XX @@ void qemu_set_fd_handler(int fd,
  {
      abort();
  }
 -
 -void aio_set_fd_handler(AioContext *ctx,
 -                        int fd,
 -                        bool is_external,
 -                        IOHandler *io_read,
 -                        IOHandler *io_write,
 -                        AioPollFn *io_poll,
 -                        void *opaque)
 -{
 -    abort();
 -}
 diff --git a/aio-posix.c b/util/aio-posix.c
 similarity index 99%
 rename from aio-posix.c
 rename to util/aio-posix.c
 index XXXXXXX..XXXXXXX 100644
 --- a/aio-posix.c
 +++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@
  #include "qemu/rcu_queue.h"
  #include "qemu/sockets.h"
  #include "qemu/cutils.h"
 -#include "trace-root.h"
 +#include "trace.h"
  #ifdef CONFIG_EPOLL_CREATE1
  #include <sys/epoll.h>
  #endif
 diff --git a/aio-win32.c b/util/aio-win32.c
 similarity index 100%
 rename from aio-win32.c
 rename to util/aio-win32.c
 diff --git a/util/aiocb.c b/util/aiocb.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/util/aiocb.c
@@ -XXX,XX +XXX,XX @@
 +/*
 + * BlockAIOCB allocation
 + *
 + * Copyright (c) 2003-2017 Fabrice Bellard and other QEMU contributors
 + *
 + * Permission is hereby granted, free of charge, to any person obtaining a copy
 + * of this software and associated documentation files (the "Software"), to deal
 + * in the Software without restriction, including without limitation the rights
 + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 + * copies of the Software, and to permit persons to whom the Software is
 + * furnished to do so, subject to the following conditions:
 + *
 + * The above copyright notice and this permission notice shall be included in
 + * all copies or substantial portions of the Software.
 + *
 + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
 + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 + * THE SOFTWARE.
 + */
 +
 +#include "qemu/osdep.h"
 +#include "block/aio.h"
 +
 +void *qemu_aio_get(const AIOCBInfo *aiocb_info, BlockDriverState *bs,
 +                   BlockCompletionFunc *cb, void *opaque)
 +{
 +    BlockAIOCB *acb;
 +
 +    acb = g_malloc(aiocb_info->aiocb_size);
 +    acb->aiocb_info = aiocb_info;
 +    acb->bs = bs;
 +    acb->cb = cb;
 +    acb->opaque = opaque;
 +    acb->refcnt = 1;
 +    return acb;
 +}
 +
 +void qemu_aio_ref(void *p)
 +{
 +    BlockAIOCB *acb = p;
 +    acb->refcnt++;
 +}
 +
 +void qemu_aio_unref(void *p)
 +{
 +    BlockAIOCB *acb = p;
 +    assert(acb->refcnt > 0);
 +    if (--acb->refcnt == 0) {
 +        g_free(acb);
 +    }
 +}
 diff --git a/async.c b/util/async.c
 similarity index 99%
 rename from async.c
 rename to util/async.c
 index XXXXXXX..XXXXXXX 100644
 --- a/async.c
 +++ b/util/async.c
@@ -XXX,XX +XXX,XX @@
  /*
 - * QEMU System Emulator
 + * Data plane event loop
   *
   * Copyright (c) 2003-2008 Fabrice Bellard
 + * Copyright (c) 2009-2017 QEMU contributors
   *
   * Permission is hereby granted, free of charge, to any person obtaining a copy
   * of this software and associated documentation files (the "Software"), to deal
 diff --git a/iohandler.c b/util/iohandler.c
 similarity index 100%
 rename from iohandler.c
 rename to util/iohandler.c
 diff --git a/main-loop.c b/util/main-loop.c
 similarity index 100%
 rename from main-loop.c
 rename to util/main-loop.c
 diff --git a/qemu-timer.c b/util/qemu-timer.c
 similarity index 100%
 rename from qemu-timer.c
 rename to util/qemu-timer.c
 diff --git a/thread-pool.c b/util/thread-pool.c
 similarity index 99%
 rename from thread-pool.c
 rename to util/thread-pool.c
 index XXXXXXX..XXXXXXX 100644
 --- a/thread-pool.c
 +++ b/util/thread-pool.c
@@ -XXX,XX +XXX,XX @@
  #include "qemu/queue.h"
  #include "qemu/thread.h"
  #include "qemu/coroutine.h"
 -#include "trace-root.h"
 +#include "trace.h"
  #include "block/thread-pool.h"
  #include "qemu/main-loop.h"
 diff --git a/trace-events b/trace-events
 index XXXXXXX..XXXXXXX 100644
 --- a/trace-events
 +++ b/trace-events
@@ -XXX,XX +XXX,XX @@
  #
  # The <format-string> should be a sprintf()-compatible format string.
 -# aio-posix.c
 -run_poll_handlers_begin(void *ctx, int64_t max_ns) "ctx %p max_ns %"PRId64
 -run_poll_handlers_end(void *ctx, bool progress) "ctx %p progress %d"
 -poll_shrink(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
 -poll_grow(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
 -
 -# thread-pool.c
 -thread_pool_submit(void *pool, void *req, void *opaque) "pool %p req %p opaque %p"
 -thread_pool_complete(void *pool, void *req, void *opaque, int ret) "pool %p req %p opaque %p ret %d"
 -thread_pool_cancel(void *req, void *opaque) "req %p opaque %p"
 -
  # ioport.c
  cpu_in(unsigned int addr, char size, unsigned int val) "addr %#x(%c) value %u"
  cpu_out(unsigned int addr, char size, unsigned int val) "addr %#x(%c) value %u"
 diff --git a/util/trace-events b/util/trace-events
 index XXXXXXX..XXXXXXX 100644
 --- a/util/trace-events
 +++ b/util/trace-events
@@ -XXX,XX +XXX,XX @@
  # See docs/tracing.txt for syntax documentation.
 +# util/aio-posix.c
 +run_poll_handlers_begin(void *ctx, int64_t max_ns) "ctx %p max_ns %"PRId64
 +run_poll_handlers_end(void *ctx, bool progress) "ctx %p progress %d"
 +poll_shrink(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
 +poll_grow(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
 +
 +# util/thread-pool.c
 +thread_pool_submit(void *pool, void *req, void *opaque) "pool %p req %p opaque %p"
 +thread_pool_complete(void *pool, void *req, void *opaque, int ret) "pool %p req %p opaque %p ret %d"
 +thread_pool_cancel(void *req, void *opaque) "req %p opaque %p"
 +
  # util/buffer.c
  buffer_resize(const char *buf, size_t olen, size_t len) "%s: old %zd, new %zd"
  buffer_move_empty(const char *buf, size_t len, const char *from) "%s: %zd bytes from %s"
 --
-.24.1
+.9.3

-[PULL 30/34] block: introduce compress filter driver
+[Qemu-devel] [PULL v2 02/24] aio: introduce aio_co_schedule and aio_co_wake
-From: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-Allow writing all the data compressed through the filter driver.
+aio_co_wake provides the infrastructure to start a coroutine on a "home"
-The written data will be aligned by the cluster size.
+AioContext.  It will be used by CoMutex and CoQueue, so that coroutines
-Based on the QEMU current implementation, that data can be written to
+don't jump from one context to another when they go to sleep on a
-unallocated clusters only. May be used for a backup job.
+mutex or waitqueue.  However, it can also be used as a more efficient
+alternative to one-shot bottom halves, and saves the effort of tracking
-Suggested-by: Max Reitz <mreitz@redhat.com>
+which AioContext a coroutine is running on.
-Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
+aio_co_schedule is the part of aio_co_wake that starts a coroutine
-Message-id: 1575288906-551879-2-git-send-email-andrey.shinkevich@virtuozzo.com
+on a remove AioContext, but it is also useful to implement e.g.
-[mreitz: Replace NULL bdrv_get_format_name() by "(no format)"]
+bdrv_set_aio_context callbacks.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
 The implementation of aio_co_schedule is based on a lock-free
 multiple-producer, single-consumer queue.  The multiple producers use
 cmpxchg to add to a LIFO stack.  The consumer (a per-AioContext bottom
 half) grabs all items added so far, inverts the list to make it FIFO,
 and goes through it one item at a time until it's empty.  The data
 structure was inspired by OSv, which uses it in the very code we'll
 "port" to QEMU for the thread-safe CoMutex.
 Most of the new code is really tests.
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Message-id: 20170213135235.12274-3-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- block/Makefile.objs     |   1 +
+ tests/Makefile.include       |   8 +-
- block/filter-compress.c | 168 ++++++++++++++++++++++++++++++++++++++++
+ include/block/aio.h          |  32 +++++++
- qapi/block-core.json    |  10 ++-
+ include/qemu/coroutine_int.h |  11 ++-
-files changed, 175 insertions(+), 4 deletions(-)
+ tests/iothread.h             |  25 +++++
- create mode 100644 block/filter-compress.c
+ tests/iothread.c             |  91 ++++++++++++++++++
+ tests/test-aio-multithread.c | 213 +++++++++++++++++++++++++++++++++++++++++++
-diff --git a/block/Makefile.objs b/block/Makefile.objs
+ util/async.c                 |  65 +++++++++++++
  util/qemu-coroutine.c        |   8 ++
  util/trace-events            |   4 +
 files changed, 453 insertions(+), 4 deletions(-)
  create mode 100644 tests/iothread.h
  create mode 100644 tests/iothread.c
  create mode 100644 tests/test-aio-multithread.c
 diff --git a/tests/Makefile.include b/tests/Makefile.include
 index XXXXXXX..XXXXXXX 100644
---- a/block/Makefile.objs
+--- a/tests/Makefile.include
-+++ b/block/Makefile.objs
++++ b/tests/Makefile.include
-@@ -XXX,XX +XXX,XX @@ block-obj-y += crypto.o
+@@ -XXX,XX +XXX,XX @@ check-unit-y += tests/test-aio$(EXESUF)
+ gcov-files-test-aio-y = util/async.c util/qemu-timer.o
- block-obj-y += aio_task.o
+ gcov-files-test-aio-$(CONFIG_WIN32) += util/aio-win32.c
- block-obj-y += backup-top.o
+ gcov-files-test-aio-$(CONFIG_POSIX) += util/aio-posix.c
-+block-obj-y += filter-compress.o
++check-unit-y += tests/test-aio-multithread$(EXESUF)
++gcov-files-test-aio-multithread-y = $(gcov-files-test-aio-y)
- common-obj-y += stream.o
++gcov-files-test-aio-multithread-y += util/qemu-coroutine.c tests/iothread.c
+ check-unit-y += tests/test-throttle$(EXESUF)
-diff --git a/block/filter-compress.c b/block/filter-compress.c
+-gcov-files-test-aio-$(CONFIG_WIN32) = aio-win32.c
 -gcov-files-test-aio-$(CONFIG_POSIX) = aio-posix.c
  check-unit-y += tests/test-thread-pool$(EXESUF)
  gcov-files-test-thread-pool-y = thread-pool.c
  gcov-files-test-hbitmap-y = util/hbitmap.c
@@ -XXX,XX +XXX,XX @@ test-qapi-obj-y = tests/test-qapi-visit.o tests/test-qapi-types.o \
      $(test-qom-obj-y)
  test-crypto-obj-y = $(crypto-obj-y) $(test-qom-obj-y)
  test-io-obj-y = $(io-obj-y) $(test-crypto-obj-y)
 -test-block-obj-y = $(block-obj-y) $(test-io-obj-y)
 +test-block-obj-y = $(block-obj-y) $(test-io-obj-y) tests/iothread.o
  tests/check-qint$(EXESUF): tests/check-qint.o $(test-util-obj-y)
  tests/check-qstring$(EXESUF): tests/check-qstring.o $(test-util-obj-y)
@@ -XXX,XX +XXX,XX @@ tests/check-qom-proplist$(EXESUF): tests/check-qom-proplist.o $(test-qom-obj-y)
  tests/test-char$(EXESUF): tests/test-char.o $(test-util-obj-y) $(qtest-obj-y) $(test-io-obj-y) $(chardev-obj-y)
  tests/test-coroutine$(EXESUF): tests/test-coroutine.o $(test-block-obj-y)
  tests/test-aio$(EXESUF): tests/test-aio.o $(test-block-obj-y)
 +tests/test-aio-multithread$(EXESUF): tests/test-aio-multithread.o $(test-block-obj-y)
  tests/test-throttle$(EXESUF): tests/test-throttle.o $(test-block-obj-y)
  tests/test-blockjob$(EXESUF): tests/test-blockjob.o $(test-block-obj-y) $(test-util-obj-y)
  tests/test-blockjob-txn$(EXESUF): tests/test-blockjob-txn.o $(test-block-obj-y) $(test-util-obj-y)
 diff --git a/include/block/aio.h b/include/block/aio.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/block/aio.h
 +++ b/include/block/aio.h
@@ -XXX,XX +XXX,XX @@ typedef void QEMUBHFunc(void *opaque);
  typedef bool AioPollFn(void *opaque);
  typedef void IOHandler(void *opaque);
 +struct Coroutine;
  struct ThreadPool;
  struct LinuxAioState;
@@ -XXX,XX +XXX,XX @@ struct AioContext {
      bool notified;
      EventNotifier notifier;
 +    QSLIST_HEAD(, Coroutine) scheduled_coroutines;
 +    QEMUBH *co_schedule_bh;
 +
      /* Thread pool for performing work and receiving completion callbacks.
       * Has its own locking.
       */
@@ -XXX,XX +XXX,XX @@ static inline bool aio_node_check(AioContext *ctx, bool is_external)
  }
  /**
 + * aio_co_schedule:
 + * @ctx: the aio context
 + * @co: the coroutine
 + *
 + * Start a coroutine on a remote AioContext.
 + *
 + * The coroutine must not be entered by anyone else while aio_co_schedule()
 + * is active.  In addition the coroutine must have yielded unless ctx
 + * is the context in which the coroutine is running (i.e. the value of
 + * qemu_get_current_aio_context() from the coroutine itself).
 + */
 +void aio_co_schedule(AioContext *ctx, struct Coroutine *co);
 +
 +/**
 + * aio_co_wake:
 + * @co: the coroutine
 + *
 + * Restart a coroutine on the AioContext where it was running last, thus
 + * preventing coroutines from jumping from one context to another when they
 + * go to sleep.
 + *
 + * aio_co_wake may be executed either in coroutine or non-coroutine
 + * context.  The coroutine must not be entered by anyone else while
 + * aio_co_wake() is active.
 + */
 +void aio_co_wake(struct Coroutine *co);
 +
 +/**
   * Return the AioContext whose event loop runs in the current thread.
   *
   * If called from an IOThread this will be the IOThread's AioContext.  If
 diff --git a/include/qemu/coroutine_int.h b/include/qemu/coroutine_int.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/qemu/coroutine_int.h
 +++ b/include/qemu/coroutine_int.h
@@ -XXX,XX +XXX,XX @@ struct Coroutine {
      CoroutineEntry *entry;
      void *entry_arg;
      Coroutine *caller;
 +
 +    /* Only used when the coroutine has terminated.  */
      QSLIST_ENTRY(Coroutine) pool_next;
 +
      size_t locks_held;
 -    /* Coroutines that should be woken up when we yield or terminate */
 +    /* Coroutines that should be woken up when we yield or terminate.
 +     * Only used when the coroutine is running.
 +     */
      QSIMPLEQ_HEAD(, Coroutine) co_queue_wakeup;
 +
 +    /* Only used when the coroutine has yielded.  */
 +    AioContext *ctx;
      QSIMPLEQ_ENTRY(Coroutine) co_queue_next;
 +    QSLIST_ENTRY(Coroutine) co_scheduled_next;
  };
  Coroutine *qemu_coroutine_new(void);
 diff --git a/tests/iothread.h b/tests/iothread.h
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
-+++ b/block/filter-compress.c
++++ b/tests/iothread.h
 @@ -XXX,XX +XXX,XX @@
 +/*
-+ * Compress filter block driver
++ * Event loop thread implementation for unit tests
 + *
-+ * Copyright (c) 2019 Virtuozzo International GmbH
++ * Copyright Red Hat Inc., 2013, 2016
 + *
-+ * Author:
++ * Authors:
-+ *   Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
++ *  Stefan Hajnoczi   <stefanha@redhat.com>
-+ *   (based on block/copy-on-read.c by Max Reitz)
++ *  Paolo Bonzini     <pbonzini@redhat.com>
 + *
-+ * This program is free software; you can redistribute it and/or
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * modify it under the terms of the GNU General Public License as
++ * See the COPYING file in the top-level directory.
 + * published by the Free Software Foundation; either version 2 or
 + * (at your option) any later version of the License.
 + *
 + * This program is distributed in the hope that it will be useful,
 + * but WITHOUT ANY WARRANTY; without even the implied warranty of
 + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 + * GNU General Public License for more details.
 + *
 + * You should have received a copy of the GNU General Public License
 + * along with this program; if not, see <http://www.gnu.org/licenses/>.
 + */
++#ifndef TEST_IOTHREAD_H
++#define TEST_IOTHREAD_H
++
++#include "block/aio.h"
++#include "qemu/thread.h"
++
++typedef struct IOThread IOThread;
++
++IOThread *iothread_new(void);
++void iothread_join(IOThread *iothread);
++AioContext *iothread_get_aio_context(IOThread *iothread);
++
++#endif
+diff --git a/tests/iothread.c b/tests/iothread.c
+new file mode 100644
+index XXXXXXX..XXXXXXX
+--- /dev/null
++++ b/tests/iothread.c
+@@ -XXX,XX +XXX,XX @@
++/*
++ * Event loop thread implementation for unit tests
++ *
++ * Copyright Red Hat Inc., 2013, 2016
++ *
++ * Authors:
++ *  Stefan Hajnoczi   <stefanha@redhat.com>
++ *  Paolo Bonzini     <pbonzini@redhat.com>
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
 +
 +#include "qemu/osdep.h"
-+#include "block/block_int.h"
-+#include "qemu/module.h"
 +#include "qapi/error.h"
-+
++#include "block/aio.h"
-+
++#include "qemu/main-loop.h"
-+static int compress_open(BlockDriverState *bs, QDict *options, int flags,
++#include "qemu/rcu.h"
-+                         Error **errp)
++#include "iothread.h"
-+{
++
-+    bs->file = bdrv_open_child(NULL, options, "file", bs, &child_file, false,
++struct IOThread {
-+                               errp);
++    AioContext *ctx;
-+    if (!bs->file) {
++
-+        return -EINVAL;
++    QemuThread thread;
-+    }
++    QemuMutex init_done_lock;
-+
++    QemuCond init_done_cond;    /* is thread initialization done? */
-+    if (!bs->file->bs->drv || !block_driver_can_compress(bs->file->bs->drv)) {
++    bool stopping;
-+        error_setg(errp,
++};
-+                   "Compression is not supported for underlying format: %s",
++
-+                   bdrv_get_format_name(bs->file->bs) ?: "(no format)");
++static __thread IOThread *my_iothread;
 +
-+        return -ENOTSUP;
++AioContext *qemu_get_current_aio_context(void)
-+    }
++{
-+
++    return my_iothread ? my_iothread->ctx : qemu_get_aio_context();
-+    bs->supported_write_flags = BDRV_REQ_WRITE_UNCHANGED |
++}
-+        (BDRV_REQ_FUA & bs->file->bs->supported_write_flags);
++
-+
++static void *iothread_run(void *opaque)
-+    bs->supported_zero_flags = BDRV_REQ_WRITE_UNCHANGED |
++{
-+        ((BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK) &
++    IOThread *iothread = opaque;
-+            bs->file->bs->supported_zero_flags);
++
-+
++    rcu_register_thread();
-+    return 0;
++
-+}
++    my_iothread = iothread;
-+
++    qemu_mutex_lock(&iothread->init_done_lock);
-+
++    iothread->ctx = aio_context_new(&error_abort);
-+static int64_t compress_getlength(BlockDriverState *bs)
++    qemu_cond_signal(&iothread->init_done_cond);
-+{
++    qemu_mutex_unlock(&iothread->init_done_lock);
-+    return bdrv_getlength(bs->file->bs);
++
-+}
++    while (!atomic_read(&iothread->stopping)) {
-+
++        aio_poll(iothread->ctx, true);
-+
++    }
-+static int coroutine_fn compress_co_preadv_part(BlockDriverState *bs,
++
-+                                                uint64_t offset, uint64_t bytes,
++    rcu_unregister_thread();
-+                                                QEMUIOVector *qiov,
++    return NULL;
-+                                                size_t qiov_offset,
++}
-+                                                int flags)
++
-+{
++void iothread_join(IOThread *iothread)
-+    return bdrv_co_preadv_part(bs->file, offset, bytes, qiov, qiov_offset,
++{
-+                               flags);
++    iothread->stopping = true;
-+}
++    aio_notify(iothread->ctx);
-+
++    qemu_thread_join(&iothread->thread);
-+
++    qemu_cond_destroy(&iothread->init_done_cond);
-+static int coroutine_fn compress_co_pwritev_part(BlockDriverState *bs,
++    qemu_mutex_destroy(&iothread->init_done_lock);
-+                                                 uint64_t offset,
++    aio_context_unref(iothread->ctx);
-+                                                 uint64_t bytes,
++    g_free(iothread);
-+                                                 QEMUIOVector *qiov,
++}
-+                                                 size_t qiov_offset, int flags)
++
-+{
++IOThread *iothread_new(void)
-+    return bdrv_co_pwritev_part(bs->file, offset, bytes, qiov, qiov_offset,
++{
-+                                flags | BDRV_REQ_WRITE_COMPRESSED);
++    IOThread *iothread = g_new0(IOThread, 1);
-+}
++
-+
++    qemu_mutex_init(&iothread->init_done_lock);
-+
++    qemu_cond_init(&iothread->init_done_cond);
-+static int coroutine_fn compress_co_pwrite_zeroes(BlockDriverState *bs,
++    qemu_thread_create(&iothread->thread, NULL, iothread_run,
-+                                                  int64_t offset, int bytes,
++                       iothread, QEMU_THREAD_JOINABLE);
-+                                                  BdrvRequestFlags flags)
++
-+{
++    /* Wait for initialization to complete */
-+    return bdrv_co_pwrite_zeroes(bs->file, offset, bytes, flags);
++    qemu_mutex_lock(&iothread->init_done_lock);
-+}
++    while (iothread->ctx == NULL) {
-+
++        qemu_cond_wait(&iothread->init_done_cond,
-+
++                       &iothread->init_done_lock);
-+static int coroutine_fn compress_co_pdiscard(BlockDriverState *bs,
++    }
-+                                             int64_t offset, int bytes)
++    qemu_mutex_unlock(&iothread->init_done_lock);
-+{
++    return iothread;
-+    return bdrv_co_pdiscard(bs->file, offset, bytes);
++}
-+}
++
-+
++AioContext *iothread_get_aio_context(IOThread *iothread)
-+
++{
-+static void compress_refresh_limits(BlockDriverState *bs, Error **errp)
++    return iothread->ctx;
-+{
++}
-+    BlockDriverInfo bdi;
+diff --git a/tests/test-aio-multithread.c b/tests/test-aio-multithread.c
-+    int ret;
+new file mode 100644
-+
+index XXXXXXX..XXXXXXX
-+    if (!bs->file) {
+--- /dev/null
 +++ b/tests/test-aio-multithread.c
@@ -XXX,XX +XXX,XX @@
 +/*
 + * AioContext multithreading tests
 + *
 + * Copyright Red Hat, Inc. 2016
 + *
 + * Authors:
 + *  Paolo Bonzini    <pbonzini@redhat.com>
 + *
 + * This work is licensed under the terms of the GNU LGPL, version 2 or later.
 + * See the COPYING.LIB file in the top-level directory.
 + */
 +
 +#include "qemu/osdep.h"
 +#include <glib.h>
 +#include "block/aio.h"
 +#include "qapi/error.h"
 +#include "qemu/coroutine.h"
 +#include "qemu/thread.h"
 +#include "qemu/error-report.h"
 +#include "iothread.h"
 +
 +/* AioContext management */
 +
 +#define NUM_CONTEXTS 5
 +
 +static IOThread *threads[NUM_CONTEXTS];
 +static AioContext *ctx[NUM_CONTEXTS];
 +static __thread int id = -1;
 +
 +static QemuEvent done_event;
 +
 +/* Run a function synchronously on a remote iothread. */
 +
 +typedef struct CtxRunData {
 +    QEMUBHFunc *cb;
 +    void *arg;
 +} CtxRunData;
 +
 +static void ctx_run_bh_cb(void *opaque)
 +{
 +    CtxRunData *data = opaque;
 +
 +    data->cb(data->arg);
 +    qemu_event_set(&done_event);
 +}
 +
 +static void ctx_run(int i, QEMUBHFunc *cb, void *opaque)
 +{
 +    CtxRunData data = {
 +        .cb = cb,
 +        .arg = opaque
 +    };
 +
 +    qemu_event_reset(&done_event);
 +    aio_bh_schedule_oneshot(ctx[i], ctx_run_bh_cb, &data);
 +    qemu_event_wait(&done_event);
 +}
 +
 +/* Starting the iothreads. */
 +
 +static void set_id_cb(void *opaque)
 +{
 +    int *i = opaque;
 +
 +    id = *i;
 +}
 +
 +static void create_aio_contexts(void)
 +{
 +    int i;
 +
 +    for (i = 0; i < NUM_CONTEXTS; i++) {
 +        threads[i] = iothread_new();
 +        ctx[i] = iothread_get_aio_context(threads[i]);
 +    }
 +
 +    qemu_event_init(&done_event, false);
 +    for (i = 0; i < NUM_CONTEXTS; i++) {
 +        ctx_run(i, set_id_cb, &i);
 +    }
 +}
 +
 +/* Stopping the iothreads. */
 +
 +static void join_aio_contexts(void)
 +{
 +    int i;
 +
 +    for (i = 0; i < NUM_CONTEXTS; i++) {
 +        aio_context_ref(ctx[i]);
 +    }
 +    for (i = 0; i < NUM_CONTEXTS; i++) {
 +        iothread_join(threads[i]);
 +    }
 +    for (i = 0; i < NUM_CONTEXTS; i++) {
 +        aio_context_unref(ctx[i]);
 +    }
 +    qemu_event_destroy(&done_event);
 +}
 +
 +/* Basic test for the stuff above. */
 +
 +static void test_lifecycle(void)
 +{
 +    create_aio_contexts();
 +    join_aio_contexts();
 +}
 +
 +/* aio_co_schedule test.  */
 +
 +static Coroutine *to_schedule[NUM_CONTEXTS];
 +
 +static bool now_stopping;
 +
 +static int count_retry;
 +static int count_here;
 +static int count_other;
 +
 +static bool schedule_next(int n)
 +{
 +    Coroutine *co;
 +
 +    co = atomic_xchg(&to_schedule[n], NULL);
 +    if (!co) {
 +        atomic_inc(&count_retry);
 +        return false;
 +    }
 +
 +    if (n == id) {
 +        atomic_inc(&count_here);
 +    } else {
 +        atomic_inc(&count_other);
 +    }
 +
 +    aio_co_schedule(ctx[n], co);
 +    return true;
 +}
 +
 +static void finish_cb(void *opaque)
 +{
 +    schedule_next(id);
 +}
 +
 +static coroutine_fn void test_multi_co_schedule_entry(void *opaque)
 +{
 +    g_assert(to_schedule[id] == NULL);
 +    atomic_mb_set(&to_schedule[id], qemu_coroutine_self());
 +
 +    while (!atomic_mb_read(&now_stopping)) {
 +        int n;
 +
 +        n = g_test_rand_int_range(0, NUM_CONTEXTS);
 +        schedule_next(n);
 +        qemu_coroutine_yield();
 +
 +        g_assert(to_schedule[id] == NULL);
 +        atomic_mb_set(&to_schedule[id], qemu_coroutine_self());
 +    }
 +}
 +
 +
 +static void test_multi_co_schedule(int seconds)
 +{
 +    int i;
 +
 +    count_here = count_other = count_retry = 0;
 +    now_stopping = false;
 +
 +    create_aio_contexts();
 +    for (i = 0; i < NUM_CONTEXTS; i++) {
 +        Coroutine *co1 = qemu_coroutine_create(test_multi_co_schedule_entry, NULL);
 +        aio_co_schedule(ctx[i], co1);
 +    }
 +
 +    g_usleep(seconds * 1000000);
 +
 +    atomic_mb_set(&now_stopping, true);
 +    for (i = 0; i < NUM_CONTEXTS; i++) {
 +        ctx_run(i, finish_cb, NULL);
 +        to_schedule[i] = NULL;
 +    }
 +
 +    join_aio_contexts();
 +    g_test_message("scheduled %d, queued %d, retry %d, total %d\n",
 +                  count_other, count_here, count_retry,
 +                  count_here + count_other + count_retry);
 +}
 +
 +static void test_multi_co_schedule_1(void)
 +{
 +    test_multi_co_schedule(1);
 +}
 +
 +static void test_multi_co_schedule_10(void)
 +{
 +    test_multi_co_schedule(10);
 +}
 +
 +/* End of tests.  */
 +
 +int main(int argc, char **argv)
 +{
 +    init_clocks();
 +
 +    g_test_init(&argc, &argv, NULL);
 +    g_test_add_func("/aio/multi/lifecycle", test_lifecycle);
 +    if (g_test_quick()) {
 +        g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_1);
 +    } else {
 +        g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_10);
 +    }
 +    return g_test_run();
 +}
 diff --git a/util/async.c b/util/async.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/async.c
 +++ b/util/async.c
@@ -XXX,XX +XXX,XX @@
  #include "qemu/main-loop.h"
  #include "qemu/atomic.h"
  #include "block/raw-aio.h"
 +#include "qemu/coroutine_int.h"
 +#include "trace.h"
  /***********************************************************/
  /* bottom halves (can be seen as timers which expire ASAP) */
@@ -XXX,XX +XXX,XX @@ aio_ctx_finalize(GSource     *source)
      }
  #endif
 +    assert(QSLIST_EMPTY(&ctx->scheduled_coroutines));
 +    qemu_bh_delete(ctx->co_schedule_bh);
 +
      qemu_lockcnt_lock(&ctx->list_lock);
      assert(!qemu_lockcnt_count(&ctx->list_lock));
      while (ctx->first_bh) {
@@ -XXX,XX +XXX,XX @@ static bool event_notifier_poll(void *opaque)
      return atomic_read(&ctx->notified);
  }
 +static void co_schedule_bh_cb(void *opaque)
 +{
 +    AioContext *ctx = opaque;
 +    QSLIST_HEAD(, Coroutine) straight, reversed;
 +
 +    QSLIST_MOVE_ATOMIC(&reversed, &ctx->scheduled_coroutines);
 +    QSLIST_INIT(&straight);
 +
 +    while (!QSLIST_EMPTY(&reversed)) {
 +        Coroutine *co = QSLIST_FIRST(&reversed);
 +        QSLIST_REMOVE_HEAD(&reversed, co_scheduled_next);
 +        QSLIST_INSERT_HEAD(&straight, co, co_scheduled_next);
 +    }
 +
 +    while (!QSLIST_EMPTY(&straight)) {
 +        Coroutine *co = QSLIST_FIRST(&straight);
 +        QSLIST_REMOVE_HEAD(&straight, co_scheduled_next);
 +        trace_aio_co_schedule_bh_cb(ctx, co);
 +        qemu_coroutine_enter(co);
 +    }
 +}
 +
  AioContext *aio_context_new(Error **errp)
  {
      int ret;
@@ -XXX,XX +XXX,XX @@ AioContext *aio_context_new(Error **errp)
      }
      g_source_set_can_recurse(&ctx->source, true);
      qemu_lockcnt_init(&ctx->list_lock);
 +
 +    ctx->co_schedule_bh = aio_bh_new(ctx, co_schedule_bh_cb, ctx);
 +    QSLIST_INIT(&ctx->scheduled_coroutines);
 +
      aio_set_event_notifier(ctx, &ctx->notifier,
                             false,
                             (EventNotifierHandler *)
@@ -XXX,XX +XXX,XX @@ fail:
      return NULL;
  }
 +void aio_co_schedule(AioContext *ctx, Coroutine *co)
 +{
 +    trace_aio_co_schedule(ctx, co);
 +    QSLIST_INSERT_HEAD_ATOMIC(&ctx->scheduled_coroutines,
 +                              co, co_scheduled_next);
 +    qemu_bh_schedule(ctx->co_schedule_bh);
 +}
 +
 +void aio_co_wake(struct Coroutine *co)
 +{
 +    AioContext *ctx;
 +
 +    /* Read coroutine before co->ctx.  Matches smp_wmb in
 +     * qemu_coroutine_enter.
 +     */
 +    smp_read_barrier_depends();
 +    ctx = atomic_read(&co->ctx);
 +
 +    if (ctx != qemu_get_current_aio_context()) {
 +        aio_co_schedule(ctx, co);
 +        return;
 +    }
 +
-+    ret = bdrv_get_info(bs->file->bs, &bdi);
++    if (qemu_in_coroutine()) {
-+    if (ret < 0 || bdi.cluster_size == 0) {
++        Coroutine *self = qemu_coroutine_self();
-+        return;
++        assert(self != co);
-+    }
++        QSIMPLEQ_INSERT_TAIL(&self->co_queue_wakeup, co, co_queue_next);
-+
++    } else {
-+    bs->bl.request_alignment = bdi.cluster_size;
++        aio_context_acquire(ctx);
-+}
++        qemu_coroutine_enter(co);
-+
++        aio_context_release(ctx);
-+
++    }
-+static void compress_eject(BlockDriverState *bs, bool eject_flag)
++}
-+{
++
-+    bdrv_eject(bs->file->bs, eject_flag);
+ void aio_context_ref(AioContext *ctx)
-+}
+ {
-+
+     g_source_ref(&ctx->source);
-+
+diff --git a/util/qemu-coroutine.c b/util/qemu-coroutine.c
 +static void compress_lock_medium(BlockDriverState *bs, bool locked)
 +{
 +    bdrv_lock_medium(bs->file->bs, locked);
 +}
 +
 +
 +static bool compress_recurse_is_first_non_filter(BlockDriverState *bs,
 +                                                 BlockDriverState *candidate)
 +{
 +    return bdrv_recurse_is_first_non_filter(bs->file->bs, candidate);
 +}
 +
 +
 +static BlockDriver bdrv_compress = {
 +    .format_name                        = "compress",
 +
 +    .bdrv_open                          = compress_open,
 +    .bdrv_child_perm                    = bdrv_filter_default_perms,
 +
 +    .bdrv_getlength                     = compress_getlength,
 +
 +    .bdrv_co_preadv_part                = compress_co_preadv_part,
 +    .bdrv_co_pwritev_part               = compress_co_pwritev_part,
 +    .bdrv_co_pwrite_zeroes              = compress_co_pwrite_zeroes,
 +    .bdrv_co_pdiscard                   = compress_co_pdiscard,
 +    .bdrv_refresh_limits                = compress_refresh_limits,
 +
 +    .bdrv_eject                         = compress_eject,
 +    .bdrv_lock_medium                   = compress_lock_medium,
 +
 +    .bdrv_co_block_status               = bdrv_co_block_status_from_file,
 +
 +    .bdrv_recurse_is_first_non_filter   = compress_recurse_is_first_non_filter,
 +
 +    .has_variable_length                = true,
 +    .is_filter                          = true,
 +};
 +
 +static void bdrv_compress_init(void)
 +{
 +    bdrv_register(&bdrv_compress);
 +}
 +
 +block_init(bdrv_compress_init);
 diff --git a/qapi/block-core.json b/qapi/block-core.json
 index XXXXXXX..XXXXXXX 100644
---- a/qapi/block-core.json
+--- a/util/qemu-coroutine.c
-+++ b/qapi/block-core.json
++++ b/util/qemu-coroutine.c
 @@ -XXX,XX +XXX,XX @@
- # @copy-on-read: Since 3.0
+ #include "qemu/atomic.h"
- # @blklogwrites: Since 3.0
+ #include "qemu/coroutine.h"
- # @blkreplay: Since 4.2
+ #include "qemu/coroutine_int.h"
-+# @compress: Since 5.0
++#include "block/aio.h"
- #
- # Since: 2.9
+ enum {
- ##
+     POOL_BATCH_SIZE = 64,
- { 'enum': 'BlockdevDriver',
+@@ -XXX,XX +XXX,XX @@ void qemu_coroutine_enter(Coroutine *co)
-   'data': [ 'blkdebug', 'blklogwrites', 'blkreplay', 'blkverify', 'bochs',
+     }
--            'cloop', 'copy-on-read', 'dmg', 'file', 'ftp', 'ftps', 'gluster',
--            'host_cdrom', 'host_device', 'http', 'https', 'iscsi', 'luks',
+     co->caller = self;
--            'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels', 'qcow',
++    co->ctx = qemu_get_current_aio_context();
--            'qcow2', 'qed', 'quorum', 'raw', 'rbd',
++
-+            'cloop', 'compress', 'copy-on-read', 'dmg', 'file', 'ftp', 'ftps',
++    /* Store co->ctx before anything that stores co.  Matches
-+            'gluster', 'host_cdrom', 'host_device', 'http', 'https', 'iscsi',
++     * barrier in aio_co_wake.
-+            'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
++     */
-+            'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
++    smp_wmb();
-             { 'name': 'replication', 'if': 'defined(CONFIG_REPLICATION)' },
++
-             'sheepdog',
+     ret = qemu_coroutine_switch(self, co, COROUTINE_ENTER);
-             'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat', 'vxhs' ] }
-@@ -XXX,XX +XXX,XX @@
+     qemu_co_queue_run_restart(co);
-       'blkreplay':  'BlockdevOptionsBlkreplay',
+diff --git a/util/trace-events b/util/trace-events
-       'bochs':      'BlockdevOptionsGenericFormat',
+index XXXXXXX..XXXXXXX 100644
-       'cloop':      'BlockdevOptionsGenericFormat',
+--- a/util/trace-events
-+      'compress':   'BlockdevOptionsGenericFormat',
++++ b/util/trace-events
-       'copy-on-read':'BlockdevOptionsGenericFormat',
+@@ -XXX,XX +XXX,XX @@ run_poll_handlers_end(void *ctx, bool progress) "ctx %p progress %d"
-       'dmg':        'BlockdevOptionsGenericFormat',
+ poll_shrink(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
-       'file':       'BlockdevOptionsFile',
+ poll_grow(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
 +# util/async.c
 +aio_co_schedule(void *ctx, void *co) "ctx %p co %p"
 +aio_co_schedule_bh_cb(void *ctx, void *co) "ctx %p co %p"
 +
  # util/thread-pool.c
  thread_pool_submit(void *pool, void *req, void *opaque) "pool %p req %p opaque %p"
  thread_pool_complete(void *pool, void *req, void *opaque, int ret) "pool %p req %p opaque %p ret %d"
 --
-.24.1
+.9.3

-[PULL 20/34] iotests: Drop IMGOPTS use in 267
+[Qemu-devel] [PULL v2 03/24] block-backend: allow blk_prw from coroutine context
-Overwriting IMGOPTS means ignoring all user-supplied options, which is
+From: Paolo Bonzini <pbonzini@redhat.com>
 not what we want.  Replace the current IMGOPTS use by a new BACKING_FILE
 variable.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+qcow2_create2 calls this.  Do not run a nested event loop, as that
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+breaks when aio_co_wake tries to queue the coroutine on the co_queue_wakeup
-Message-id: 20191107163708.833192-14-mreitz@redhat.com
+list of the currently running one.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
 Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Message-id: 20170213135235.12274-4-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/267 | 12 ++++++++----
+ block/block-backend.c | 12 ++++++++----
 file changed, 8 insertions(+), 4 deletions(-)
-diff --git a/tests/qemu-iotests/267 b/tests/qemu-iotests/267
+diff --git a/block/block-backend.c b/block/block-backend.c
-index XXXXXXX..XXXXXXX 100755
+index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/267
+--- a/block/block-backend.c
-+++ b/tests/qemu-iotests/267
++++ b/block/block-backend.c
-@@ -XXX,XX +XXX,XX @@ size=128M
+@@ -XXX,XX +XXX,XX @@ static int blk_prw(BlockBackend *blk, int64_t offset, uint8_t *buf,
  run_test()
  {
--    _make_test_img $size
+     QEMUIOVector qiov;
-+    if [ -n "$BACKING_FILE" ]; then
+     struct iovec iov;
-+        _make_test_img -b "$BACKING_FILE" $size
+-    Coroutine *co;
-+    else
+     BlkRwCo rwco;
-+        _make_test_img $size
-+    fi
+     iov = (struct iovec) {
-     printf "savevm snap0\ninfo snapshots\nloadvm snap0\n" | run_qemu "$@" | _filter_date
+@@ -XXX,XX +XXX,XX @@ static int blk_prw(BlockBackend *blk, int64_t offset, uint8_t *buf,
          .ret    = NOT_DONE,
      };
 -    co = qemu_coroutine_create(co_entry, &rwco);
 -    qemu_coroutine_enter(co);
 -    BDRV_POLL_WHILE(blk_bs(blk), rwco.ret == NOT_DONE);
 +    if (qemu_in_coroutine()) {
 +        /* Fast-path if already in coroutine context */
 +        co_entry(&rwco);
 +    } else {
 +        Coroutine *co = qemu_coroutine_create(co_entry, &rwco);
 +        qemu_coroutine_enter(co);
 +        BDRV_POLL_WHILE(blk_bs(blk), rwco.ret == NOT_DONE);
 +    }
      return rwco.ret;
  }
-@@ -XXX,XX +XXX,XX @@ echo
- TEST_IMG="$TEST_IMG.base" _make_test_img $size
--IMGOPTS="backing_file=$TEST_IMG.base" \
-+BACKING_FILE="$TEST_IMG.base" \
- run_test -blockdev driver=file,filename="$TEST_IMG.base",node-name=backing-file \
-          -blockdev driver=file,filename="$TEST_IMG",node-name=file \
-          -blockdev driver=$IMGFMT,file=file,backing=backing-file,node-name=fmt
--IMGOPTS="backing_file=$TEST_IMG.base" \
-+BACKING_FILE="$TEST_IMG.base" \
- run_test -blockdev driver=file,filename="$TEST_IMG.base",node-name=backing-file \
-          -blockdev driver=$IMGFMT,file=backing-file,node-name=backing-fmt \
-          -blockdev driver=file,filename="$TEST_IMG",node-name=file \
-@@ -XXX,XX +XXX,XX @@ echo
- echo "=== -blockdev with NBD server on the backing file ==="
- echo
--IMGOPTS="backing_file=$TEST_IMG.base" _make_test_img $size
-+_make_test_img -b "$TEST_IMG.base" $size
- cat <<EOF |
- nbd_server_start unix:$SOCK_DIR/nbd
- nbd_server_add -w backing-fmt
 --
-.24.1
+.9.3

-[PULL 34/34] backup-top: Begin drain earlier
+[Qemu-devel] [PULL v2 04/24] test-thread-pool: use generic AioContext infrastructure
-When dropping backup-top, we need to drain the node before freeing the
+From: Paolo Bonzini <pbonzini@redhat.com>
 BlockCopyState.  Otherwise, requests may still be in flight and then the
 assertion in shres_destroy() will fail.
-(This becomes visible in intermittent failure of 056.)
+Once the thread pool starts using aio_co_wake, it will also need
 qemu_get_current_aio_context().  Make test-thread-pool create
 an AioContext with qemu_init_main_loop, so that stubs/iothread.c
 and tests/iothread.c can provide the rest.
-Cc: qemu-stable@nongnu.org
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-Message-id: 20191219182638.104621-1-mreitz@redhat.com
+Reviewed-by: Fam Zheng <famz@redhat.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Message-id: 20170213135235.12274-5-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- block/backup-top.c | 4 ++--
+ tests/test-thread-pool.c | 12 +++---------
-file changed, 2 insertions(+), 2 deletions(-)
+file changed, 3 insertions(+), 9 deletions(-)
-diff --git a/block/backup-top.c b/block/backup-top.c
+diff --git a/tests/test-thread-pool.c b/tests/test-thread-pool.c
 index XXXXXXX..XXXXXXX 100644
---- a/block/backup-top.c
+--- a/tests/test-thread-pool.c
-+++ b/block/backup-top.c
++++ b/tests/test-thread-pool.c
-@@ -XXX,XX +XXX,XX @@ void bdrv_backup_top_drop(BlockDriverState *bs)
+@@ -XXX,XX +XXX,XX @@
-     BDRVBackupTopState *s = bs->opaque;
+ #include "qapi/error.h"
-     AioContext *aio_context = bdrv_get_aio_context(bs);
+ #include "qemu/timer.h"
+ #include "qemu/error-report.h"
--    block_copy_state_free(s->bcs);
++#include "qemu/main-loop.h"
  static AioContext *ctx;
  static ThreadPool *pool;
@@ -XXX,XX +XXX,XX @@ static void test_cancel_async(void)
  int main(int argc, char **argv)
  {
      int ret;
 -    Error *local_error = NULL;
 -    init_clocks();
 -
-     aio_context_acquire(aio_context);
+-    ctx = aio_context_new(&local_error);
+-    if (!ctx) {
-     bdrv_drained_begin(bs);
+-        error_reportf_err(local_error, "Failed to create AIO Context: ");
+-        exit(1);
-+    block_copy_state_free(s->bcs);
+-    }
-+
++    qemu_init_main_loop(&error_abort);
-     s->active = false;
++    ctx = qemu_get_current_aio_context();
-     bdrv_child_refresh_perms(bs, bs->backing, &error_abort);
+     pool = aio_get_thread_pool(ctx);
-     bdrv_replace_node(bs, backing_bs(bs), &error_abort);
      g_test_init(&argc, &argv, NULL);
@@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv)
      ret = g_test_run();
 -    aio_context_unref(ctx);
      return ret;
  }
 --
-.24.1
+.9.3

-[PULL 33/34] tests/qemu-iotests: Update tests to recent desugarized -accel option
+[Qemu-devel] [PULL v2 05/24] io: add methods to set I/O handlers on AioContext
-From: Philippe Mathieu-Daudé <philmd@redhat.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-Commit 6f6e1698a6 desugarized "-machine accel=" to a list
+This is in preparation for making qio_channel_yield work on
-of "-accel" options. Since now "-machine accel" and "-accel"
+AioContexts other than the main one.
-became incompatible, update the iotests to the new format.
+Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
-Error reported here:
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-https://gitlab.com/qemu-project/qemu/-/jobs/385801004#L3400
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+Reviewed-by: Fam Zheng <famz@redhat.com>
-Reported-by: GitLab CI
+Message-id: 20170213135235.12274-6-pbonzini@redhat.com
-Fixes: 6f6e1698a6 (vl: configure accelerators from -accel options)
+Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
 Message-id: 20200106130951.29873-1-philmd@redhat.com
 Signed-off-by: Max Reitz <mreitz@redhat.com>
 ---
- tests/qemu-iotests/235   | 2 +-
+ include/io/channel.h | 25 +++++++++++++++++++++++++
- tests/qemu-iotests/check | 6 +++---
+ io/channel-command.c | 13 +++++++++++++
-files changed, 4 insertions(+), 4 deletions(-)
+ io/channel-file.c    | 11 +++++++++++
+ io/channel-socket.c  | 16 +++++++++++-----
-diff --git a/tests/qemu-iotests/235 b/tests/qemu-iotests/235
+ io/channel-tls.c     | 12 ++++++++++++
-index XXXXXXX..XXXXXXX 100755
+ io/channel-watch.c   |  6 ++++++
---- a/tests/qemu-iotests/235
+ io/channel.c         | 11 +++++++++++
-+++ b/tests/qemu-iotests/235
+files changed, 89 insertions(+), 5 deletions(-)
-@@ -XXX,XX +XXX,XX @@ qemu_img_create('-f', iotests.imgfmt, '-o', 'preallocation=metadata', disk,
-                 str(size))
+diff --git a/include/io/channel.h b/include/io/channel.h
+index XXXXXXX..XXXXXXX 100644
- vm = QEMUMachine(iotests.qemu_prog)
+--- a/include/io/channel.h
--vm.add_args('-machine', 'accel=kvm:tcg')
++++ b/include/io/channel.h
-+vm.add_args('-accel', 'kvm', '-accel', 'tcg')
+@@ -XXX,XX +XXX,XX @@
- if iotests.qemu_default_machine == 's390-ccw-virtio':
-         vm.add_args('-no-shutdown')
+ #include "qemu-common.h"
- vm.add_args('-drive', 'id=src,file=' + disk)
+ #include "qom/object.h"
-diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check
++#include "block/aio.h"
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/check
+ #define TYPE_QIO_CHANNEL "qio-channel"
-+++ b/tests/qemu-iotests/check
+ #define QIO_CHANNEL(obj)                                    \
-@@ -XXX,XX +XXX,XX @@ export QEMU_PROG="$(type -p "$QEMU_PROG")"
+@@ -XXX,XX +XXX,XX @@ struct QIOChannelClass {
+                      off_t offset,
- case "$QEMU_PROG" in
+                      int whence,
-     *qemu-system-arm|*qemu-system-aarch64)
+                      Error **errp);
--        export QEMU_OPTIONS="-nodefaults -display none -machine virt,accel=qtest"
++    void (*io_set_aio_fd_handler)(QIOChannel *ioc,
-+        export QEMU_OPTIONS="-nodefaults -display none -machine virt -accel qtest"
++                                  AioContext *ctx,
-         ;;
++                                  IOHandler *io_read,
-     *qemu-system-tricore)
++                                  IOHandler *io_write,
--        export QEMU_OPTIONS="-nodefaults -display none -machine tricore_testboard,accel=qtest"
++                                  void *opaque);
-+        export QEMU_OPTIONS="-nodefaults -display none -machine tricore_testboard -accel qtest"
+ };
-         ;;
-     *)
+ /* General I/O handling functions */
--        export QEMU_OPTIONS="-nodefaults -display none -machine accel=qtest"
+@@ -XXX,XX +XXX,XX @@ void qio_channel_yield(QIOChannel *ioc,
-+        export QEMU_OPTIONS="-nodefaults -display none -accel qtest"
+ void qio_channel_wait(QIOChannel *ioc,
-         ;;
+                       GIOCondition condition);
- esac
++/**
 + * qio_channel_set_aio_fd_handler:
 + * @ioc: the channel object
 + * @ctx: the AioContext to set the handlers on
 + * @io_read: the read handler
 + * @io_write: the write handler
 + * @opaque: the opaque value passed to the handler
 + *
 + * This is used internally by qio_channel_yield().  It can
 + * be used by channel implementations to forward the handlers
 + * to another channel (e.g. from #QIOChannelTLS to the
 + * underlying socket).
 + */
 +void qio_channel_set_aio_fd_handler(QIOChannel *ioc,
 +                                    AioContext *ctx,
 +                                    IOHandler *io_read,
 +                                    IOHandler *io_write,
 +                                    void *opaque);
 +
  #endif /* QIO_CHANNEL_H */
 diff --git a/io/channel-command.c b/io/channel-command.c
 index XXXXXXX..XXXXXXX 100644
 --- a/io/channel-command.c
 +++ b/io/channel-command.c
@@ -XXX,XX +XXX,XX @@ static int qio_channel_command_close(QIOChannel *ioc,
  }
 +static void qio_channel_command_set_aio_fd_handler(QIOChannel *ioc,
 +                                                   AioContext *ctx,
 +                                                   IOHandler *io_read,
 +                                                   IOHandler *io_write,
 +                                                   void *opaque)
 +{
 +    QIOChannelCommand *cioc = QIO_CHANNEL_COMMAND(ioc);
 +    aio_set_fd_handler(ctx, cioc->readfd, false, io_read, NULL, NULL, opaque);
 +    aio_set_fd_handler(ctx, cioc->writefd, false, NULL, io_write, NULL, opaque);
 +}
 +
 +
  static GSource *qio_channel_command_create_watch(QIOChannel *ioc,
                                                   GIOCondition condition)
  {
@@ -XXX,XX +XXX,XX @@ static void qio_channel_command_class_init(ObjectClass *klass,
      ioc_klass->io_set_blocking = qio_channel_command_set_blocking;
      ioc_klass->io_close = qio_channel_command_close;
      ioc_klass->io_create_watch = qio_channel_command_create_watch;
 +    ioc_klass->io_set_aio_fd_handler = qio_channel_command_set_aio_fd_handler;
  }
  static const TypeInfo qio_channel_command_info = {
 diff --git a/io/channel-file.c b/io/channel-file.c
 index XXXXXXX..XXXXXXX 100644
 --- a/io/channel-file.c
 +++ b/io/channel-file.c
@@ -XXX,XX +XXX,XX @@ static int qio_channel_file_close(QIOChannel *ioc,
  }
 +static void qio_channel_file_set_aio_fd_handler(QIOChannel *ioc,
 +                                                AioContext *ctx,
 +                                                IOHandler *io_read,
 +                                                IOHandler *io_write,
 +                                                void *opaque)
 +{
 +    QIOChannelFile *fioc = QIO_CHANNEL_FILE(ioc);
 +    aio_set_fd_handler(ctx, fioc->fd, false, io_read, io_write, NULL, opaque);
 +}
 +
  static GSource *qio_channel_file_create_watch(QIOChannel *ioc,
                                                GIOCondition condition)
  {
@@ -XXX,XX +XXX,XX @@ static void qio_channel_file_class_init(ObjectClass *klass,
      ioc_klass->io_seek = qio_channel_file_seek;
      ioc_klass->io_close = qio_channel_file_close;
      ioc_klass->io_create_watch = qio_channel_file_create_watch;
 +    ioc_klass->io_set_aio_fd_handler = qio_channel_file_set_aio_fd_handler;
  }
  static const TypeInfo qio_channel_file_info = {
 diff --git a/io/channel-socket.c b/io/channel-socket.c
 index XXXXXXX..XXXXXXX 100644
 --- a/io/channel-socket.c
 +++ b/io/channel-socket.c
@@ -XXX,XX +XXX,XX @@ qio_channel_socket_set_blocking(QIOChannel *ioc,
          qemu_set_block(sioc->fd);
      } else {
          qemu_set_nonblock(sioc->fd);
 -#ifdef WIN32
 -        WSAEventSelect(sioc->fd, ioc->event,
 -                       FD_READ | FD_ACCEPT | FD_CLOSE |
 -                       FD_CONNECT | FD_WRITE | FD_OOB);
 -#endif
      }
      return 0;
  }
@@ -XXX,XX +XXX,XX @@ qio_channel_socket_shutdown(QIOChannel *ioc,
      return 0;
  }
 +static void qio_channel_socket_set_aio_fd_handler(QIOChannel *ioc,
 +                                                  AioContext *ctx,
 +                                                  IOHandler *io_read,
 +                                                  IOHandler *io_write,
 +                                                  void *opaque)
 +{
 +    QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
 +    aio_set_fd_handler(ctx, sioc->fd, false, io_read, io_write, NULL, opaque);
 +}
 +
  static GSource *qio_channel_socket_create_watch(QIOChannel *ioc,
                                                  GIOCondition condition)
  {
@@ -XXX,XX +XXX,XX @@ static void qio_channel_socket_class_init(ObjectClass *klass,
      ioc_klass->io_set_cork = qio_channel_socket_set_cork;
      ioc_klass->io_set_delay = qio_channel_socket_set_delay;
      ioc_klass->io_create_watch = qio_channel_socket_create_watch;
 +    ioc_klass->io_set_aio_fd_handler = qio_channel_socket_set_aio_fd_handler;
  }
  static const TypeInfo qio_channel_socket_info = {
 diff --git a/io/channel-tls.c b/io/channel-tls.c
 index XXXXXXX..XXXXXXX 100644
 --- a/io/channel-tls.c
 +++ b/io/channel-tls.c
@@ -XXX,XX +XXX,XX @@ static int qio_channel_tls_close(QIOChannel *ioc,
      return qio_channel_close(tioc->master, errp);
  }
 +static void qio_channel_tls_set_aio_fd_handler(QIOChannel *ioc,
 +                                               AioContext *ctx,
 +                                               IOHandler *io_read,
 +                                               IOHandler *io_write,
 +                                               void *opaque)
 +{
 +    QIOChannelTLS *tioc = QIO_CHANNEL_TLS(ioc);
 +
 +    qio_channel_set_aio_fd_handler(tioc->master, ctx, io_read, io_write, opaque);
 +}
 +
  static GSource *qio_channel_tls_create_watch(QIOChannel *ioc,
                                               GIOCondition condition)
  {
@@ -XXX,XX +XXX,XX @@ static void qio_channel_tls_class_init(ObjectClass *klass,
      ioc_klass->io_close = qio_channel_tls_close;
      ioc_klass->io_shutdown = qio_channel_tls_shutdown;
      ioc_klass->io_create_watch = qio_channel_tls_create_watch;
 +    ioc_klass->io_set_aio_fd_handler = qio_channel_tls_set_aio_fd_handler;
  }
  static const TypeInfo qio_channel_tls_info = {
 diff --git a/io/channel-watch.c b/io/channel-watch.c
 index XXXXXXX..XXXXXXX 100644
 --- a/io/channel-watch.c
 +++ b/io/channel-watch.c
@@ -XXX,XX +XXX,XX @@ GSource *qio_channel_create_socket_watch(QIOChannel *ioc,
      GSource *source;
      QIOChannelSocketSource *ssource;
 +#ifdef WIN32
 +    WSAEventSelect(socket, ioc->event,
 +                   FD_READ | FD_ACCEPT | FD_CLOSE |
 +                   FD_CONNECT | FD_WRITE | FD_OOB);
 +#endif
 +
      source = g_source_new(&qio_channel_socket_source_funcs,
                            sizeof(QIOChannelSocketSource));
      ssource = (QIOChannelSocketSource *)source;
 diff --git a/io/channel.c b/io/channel.c
 index XXXXXXX..XXXXXXX 100644
 --- a/io/channel.c
 +++ b/io/channel.c
@@ -XXX,XX +XXX,XX @@ GSource *qio_channel_create_watch(QIOChannel *ioc,
  }
 +void qio_channel_set_aio_fd_handler(QIOChannel *ioc,
 +                                    AioContext *ctx,
 +                                    IOHandler *io_read,
 +                                    IOHandler *io_write,
 +                                    void *opaque)
 +{
 +    QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
 +
 +    klass->io_set_aio_fd_handler(ioc, ctx, io_read, io_write, opaque);
 +}
 +
  guint qio_channel_add_watch(QIOChannel *ioc,
                              GIOCondition condition,
                              QIOChannelFunc func,
 --
-.24.1
+.9.3

-[PULL 21/34] iotests: Avoid qemu-img create
+[Qemu-devel] [PULL v2 06/24] io: make qio_channel_yield aware of AioContexts
-Use _make_test_img whenever possible.  This way, we will not ignore
+From: Paolo Bonzini <pbonzini@redhat.com>
-user-specified image options.
+Support separate coroutines for reading and writing, and place the
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+read/write handlers on the AioContext that the QIOChannel is registered
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+with.
-Message-id: 20191107163708.833192-15-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Message-id: 20170213135235.12274-7-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/094 | 2 +-
+ include/io/channel.h | 47 ++++++++++++++++++++++++++--
- tests/qemu-iotests/111 | 3 +--
+ io/channel.c         | 86 +++++++++++++++++++++++++++++++++++++++-------------
- tests/qemu-iotests/123 | 2 +-
+files changed, 109 insertions(+), 24 deletions(-)
- tests/qemu-iotests/153 | 2 +-
- tests/qemu-iotests/200 | 4 ++--
+diff --git a/include/io/channel.h b/include/io/channel.h
-files changed, 6 insertions(+), 7 deletions(-)
+index XXXXXXX..XXXXXXX 100644
+--- a/include/io/channel.h
-diff --git a/tests/qemu-iotests/094 b/tests/qemu-iotests/094
++++ b/include/io/channel.h
-index XXXXXXX..XXXXXXX 100755
+@@ -XXX,XX +XXX,XX @@
---- a/tests/qemu-iotests/094
-+++ b/tests/qemu-iotests/094
+ #include "qemu-common.h"
-@@ -XXX,XX +XXX,XX @@ _supported_proto nbd
+ #include "qom/object.h"
- _unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat"
++#include "qemu/coroutine.h"
+ #include "block/aio.h"
- _make_test_img 64M
--$QEMU_IMG create -f $IMGFMT "$TEST_DIR/source.$IMGFMT" 64M | _filter_img_create
+ #define TYPE_QIO_CHANNEL "qio-channel"
-+TEST_IMG_FILE="$TEST_DIR/source.$IMGFMT" IMGPROTO=file _make_test_img 64M
+@@ -XXX,XX +XXX,XX @@ struct QIOChannel {
+     Object parent;
- _launch_qemu -drive if=none,id=src,file="$TEST_DIR/source.$IMGFMT",format=raw \
+     unsigned int features; /* bitmask of QIOChannelFeatures */
-              -nodefaults
+     char *name;
-diff --git a/tests/qemu-iotests/111 b/tests/qemu-iotests/111
++    AioContext *ctx;
-index XXXXXXX..XXXXXXX 100755
++    Coroutine *read_coroutine;
---- a/tests/qemu-iotests/111
++    Coroutine *write_coroutine;
-+++ b/tests/qemu-iotests/111
+ #ifdef _WIN32
-@@ -XXX,XX +XXX,XX @@ _supported_fmt qed qcow qcow2 vmdk
+     HANDLE event; /* For use with GSource on Win32 */
- _supported_proto file
+ #endif
- _unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat"
+@@ -XXX,XX +XXX,XX @@ guint qio_channel_add_watch(QIOChannel *ioc,
--$QEMU_IMG create -f $IMGFMT -b "$TEST_IMG.inexistent" "$TEST_IMG" 2>&1 \
--    | _filter_testdir | _filter_imgfmt
+ /**
-+_make_test_img -b "$TEST_IMG.inexistent"
++ * qio_channel_attach_aio_context:
++ * @ioc: the channel object
- # success, all done
++ * @ctx: the #AioContext to set the handlers on
- echo '*** done'
++ *
-diff --git a/tests/qemu-iotests/123 b/tests/qemu-iotests/123
++ * Request that qio_channel_yield() sets I/O handlers on
-index XXXXXXX..XXXXXXX 100755
++ * the given #AioContext.  If @ctx is %NULL, qio_channel_yield()
---- a/tests/qemu-iotests/123
++ * uses QEMU's main thread event loop.
-+++ b/tests/qemu-iotests/123
++ *
-@@ -XXX,XX +XXX,XX @@ _supported_os Linux
++ * You can move a #QIOChannel from one #AioContext to another even if
- SRC_IMG="$TEST_DIR/source.$IMGFMT"
++ * I/O handlers are set for a coroutine.  However, #QIOChannel provides
++ * no synchronization between the calls to qio_channel_yield() and
- _make_test_img 1M
++ * qio_channel_attach_aio_context().
--$QEMU_IMG create -f $IMGFMT "$SRC_IMG" 1M | _filter_img_create
++ *
-+TEST_IMG_FILE=$SRC_IMG IMGPROTO=file _make_test_img 1M
++ * Therefore you should first call qio_channel_detach_aio_context()
++ * to ensure that the coroutine is not entered concurrently.  Then,
- $QEMU_IO -c 'write -P 42 0 1M' "$SRC_IMG" | _filter_qemu_io
++ * while the coroutine has yielded, call qio_channel_attach_aio_context(),
++ * and then aio_co_schedule() to place the coroutine on the new
-diff --git a/tests/qemu-iotests/153 b/tests/qemu-iotests/153
++ * #AioContext.  The calls to qio_channel_detach_aio_context()
-index XXXXXXX..XXXXXXX 100755
++ * and qio_channel_attach_aio_context() should be protected with
---- a/tests/qemu-iotests/153
++ * aio_context_acquire() and aio_context_release().
-+++ b/tests/qemu-iotests/153
++ */
-@@ -XXX,XX +XXX,XX @@ for opts1 in "" "read-only=on" "read-only=on,force-share=on"; do
++void qio_channel_attach_aio_context(QIOChannel *ioc,
++                                    AioContext *ctx);
-     echo
++
-     echo "== Creating test image =="
++/**
--    $QEMU_IMG create -f $IMGFMT "${TEST_IMG}" -b ${TEST_IMG}.base | _filter_img_create
++ * qio_channel_detach_aio_context:
-+    _make_test_img -b "${TEST_IMG}.base"
++ * @ioc: the channel object
++ *
-     echo
++ * Disable any I/O handlers set by qio_channel_yield().  With the
-     echo "== Launching QEMU, opts: '$opts1' =="
++ * help of aio_co_schedule(), this allows moving a coroutine that was
-diff --git a/tests/qemu-iotests/200 b/tests/qemu-iotests/200
++ * paused by qio_channel_yield() to another context.
-index XXXXXXX..XXXXXXX 100755
++ */
---- a/tests/qemu-iotests/200
++void qio_channel_detach_aio_context(QIOChannel *ioc);
-+++ b/tests/qemu-iotests/200
++
-@@ -XXX,XX +XXX,XX @@ _supported_proto file
++/**
- BACKING_IMG="${TEST_DIR}/backing.img"
+  * qio_channel_yield:
- TEST_IMG="${TEST_DIR}/test.img"
+  * @ioc: the channel object
+  * @condition: the I/O condition to wait for
--${QEMU_IMG} create -f $IMGFMT "${BACKING_IMG}" 512M | _filter_img_create
+  *
--${QEMU_IMG} create -f $IMGFMT -F $IMGFMT "${TEST_IMG}" -b "${BACKING_IMG}" 512M | _filter_img_create
+- * Yields execution from the current coroutine until
-+TEST_IMG="$BACKING_IMG" _make_test_img 512M
+- * the condition indicated by @condition becomes
-+_make_test_img -F $IMGFMT -b "$BACKING_IMG" 512M
+- * available.
++ * Yields execution from the current coroutine until the condition
- ${QEMU_IO} -c "write -P 0xa5 512 300M" "${BACKING_IMG}" | _filter_qemu_io
++ * indicated by @condition becomes available.  @condition must
 + * be either %G_IO_IN or %G_IO_OUT; it cannot contain both.  In
 + * addition, no two coroutine can be waiting on the same condition
 + * and channel at the same time.
   *
   * This must only be called from coroutine context
   */
 diff --git a/io/channel.c b/io/channel.c
 index XXXXXXX..XXXXXXX 100644
 --- a/io/channel.c
 +++ b/io/channel.c
@@ -XXX,XX +XXX,XX @@
  #include "qemu/osdep.h"
  #include "io/channel.h"
  #include "qapi/error.h"
 -#include "qemu/coroutine.h"
 +#include "qemu/main-loop.h"
  bool qio_channel_has_feature(QIOChannel *ioc,
                               QIOChannelFeature feature)
@@ -XXX,XX +XXX,XX @@ off_t qio_channel_io_seek(QIOChannel *ioc,
  }
 -typedef struct QIOChannelYieldData QIOChannelYieldData;
 -struct QIOChannelYieldData {
 -    QIOChannel *ioc;
 -    Coroutine *co;
 -};
 +static void qio_channel_set_aio_fd_handlers(QIOChannel *ioc);
 +static void qio_channel_restart_read(void *opaque)
 +{
 +    QIOChannel *ioc = opaque;
 +    Coroutine *co = ioc->read_coroutine;
 +
 +    ioc->read_coroutine = NULL;
 +    qio_channel_set_aio_fd_handlers(ioc);
 +    aio_co_wake(co);
 +}
 -static gboolean qio_channel_yield_enter(QIOChannel *ioc,
 -                                        GIOCondition condition,
 -                                        gpointer opaque)
 +static void qio_channel_restart_write(void *opaque)
  {
 -    QIOChannelYieldData *data = opaque;
 -    qemu_coroutine_enter(data->co);
 -    return FALSE;
 +    QIOChannel *ioc = opaque;
 +    Coroutine *co = ioc->write_coroutine;
 +
 +    ioc->write_coroutine = NULL;
 +    qio_channel_set_aio_fd_handlers(ioc);
 +    aio_co_wake(co);
  }
 +static void qio_channel_set_aio_fd_handlers(QIOChannel *ioc)
 +{
 +    IOHandler *rd_handler = NULL, *wr_handler = NULL;
 +    AioContext *ctx;
 +
 +    if (ioc->read_coroutine) {
 +        rd_handler = qio_channel_restart_read;
 +    }
 +    if (ioc->write_coroutine) {
 +        wr_handler = qio_channel_restart_write;
 +    }
 +
 +    ctx = ioc->ctx ? ioc->ctx : iohandler_get_aio_context();
 +    qio_channel_set_aio_fd_handler(ioc, ctx, rd_handler, wr_handler, ioc);
 +}
 +
 +void qio_channel_attach_aio_context(QIOChannel *ioc,
 +                                    AioContext *ctx)
 +{
 +    AioContext *old_ctx;
 +    if (ioc->ctx == ctx) {
 +        return;
 +    }
 +
 +    old_ctx = ioc->ctx ? ioc->ctx : iohandler_get_aio_context();
 +    qio_channel_set_aio_fd_handler(ioc, old_ctx, NULL, NULL, NULL);
 +    ioc->ctx = ctx;
 +    qio_channel_set_aio_fd_handlers(ioc);
 +}
 +
 +void qio_channel_detach_aio_context(QIOChannel *ioc)
 +{
 +    ioc->read_coroutine = NULL;
 +    ioc->write_coroutine = NULL;
 +    qio_channel_set_aio_fd_handlers(ioc);
 +    ioc->ctx = NULL;
 +}
  void coroutine_fn qio_channel_yield(QIOChannel *ioc,
                                      GIOCondition condition)
  {
 -    QIOChannelYieldData data;
 -
      assert(qemu_in_coroutine());
 -    data.ioc = ioc;
 -    data.co = qemu_coroutine_self();
 -    qio_channel_add_watch(ioc,
 -                          condition,
 -                          qio_channel_yield_enter,
 -                          &data,
 -                          NULL);
 +    if (condition == G_IO_IN) {
 +        assert(!ioc->read_coroutine);
 +        ioc->read_coroutine = qemu_coroutine_self();
 +    } else if (condition == G_IO_OUT) {
 +        assert(!ioc->write_coroutine);
 +        ioc->write_coroutine = qemu_coroutine_self();
 +    } else {
 +        abort();
 +    }
 +    qio_channel_set_aio_fd_handlers(ioc);
      qemu_coroutine_yield();
  }
 --
-.24.1
+.9.3

-[PULL 32/34] tests/qemu-iotests: add case to write compressed data of multiple clusters
+[Qemu-devel] [PULL v2 07/24] nbd: convert to use qio_channel_yield
-From: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-Add the case to the iotest #214 that checks possibility of writing
+In the client, read the reply headers from a coroutine, switching the
-compressed data of more than one cluster size. The test case involves
+read side between the "read header" coroutine and the I/O coroutine that
-the compress filter driver showing a sample usage of that.
+reads the body of the reply.
-Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
+In the server, if the server can read more requests it will create a new
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
+"read request" coroutine as soon as a request has been read.  Otherwise,
-Reviewed-by: Max Reitz <mreitz@redhat.com>
+the new coroutine is created in nbd_request_put.
-Message-id: 1575288906-551879-4-git-send-email-andrey.shinkevich@virtuozzo.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-8-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/214     | 43 ++++++++++++++++++++++++++++++++++++++
+ block/nbd-client.h |   2 +-
- tests/qemu-iotests/214.out | 14 +++++++++++++
+ block/nbd-client.c | 117 ++++++++++++++++++++++++-----------------------------
-files changed, 57 insertions(+)
+ nbd/client.c       |   2 +-
  nbd/common.c       |   9 +----
  nbd/server.c       |  94 +++++++++++++-----------------------------
 files changed, 83 insertions(+), 141 deletions(-)
-diff --git a/tests/qemu-iotests/214 b/tests/qemu-iotests/214
+diff --git a/block/nbd-client.h b/block/nbd-client.h
-index XXXXXXX..XXXXXXX 100755
+index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/214
+--- a/block/nbd-client.h
-+++ b/tests/qemu-iotests/214
++++ b/block/nbd-client.h
-@@ -XXX,XX +XXX,XX @@ _check_test_img -r all
+@@ -XXX,XX +XXX,XX @@ typedef struct NBDClientSession {
- $QEMU_IO -c "read  -P 0x11  0 4M" "$TEST_IMG" 2>&1 | _filter_qemu_io | _filter_testdir
- $QEMU_IO -c "read  -P 0x22 4M 4M" "$TEST_IMG" 2>&1 | _filter_qemu_io | _filter_testdir
+     CoMutex send_mutex;
+     CoQueue free_sema;
-+echo
+-    Coroutine *send_coroutine;
-+echo "=== Write compressed data of multiple clusters ==="
++    Coroutine *read_reply_co;
-+echo
+     int in_flight;
-+cluster_size=0x10000
-+_make_test_img 2M -o cluster_size=$cluster_size
+     Coroutine *recv_coroutine[MAX_NBD_REQUESTS];
 diff --git a/block/nbd-client.c b/block/nbd-client.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/nbd-client.c
 +++ b/block/nbd-client.c
@@ -XXX,XX +XXX,XX @@
  #define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs))
  #define INDEX_TO_HANDLE(bs, index)  ((index)  ^ ((uint64_t)(intptr_t)bs))
 -static void nbd_recv_coroutines_enter_all(NBDClientSession *s)
 +static void nbd_recv_coroutines_enter_all(BlockDriverState *bs)
  {
 +    NBDClientSession *s = nbd_get_client_session(bs);
      int i;
      for (i = 0; i < MAX_NBD_REQUESTS; i++) {
@@ -XXX,XX +XXX,XX @@ static void nbd_recv_coroutines_enter_all(NBDClientSession *s)
              qemu_coroutine_enter(s->recv_coroutine[i]);
          }
      }
 +    BDRV_POLL_WHILE(bs, s->read_reply_co);
  }
  static void nbd_teardown_connection(BlockDriverState *bs)
@@ -XXX,XX +XXX,XX @@ static void nbd_teardown_connection(BlockDriverState *bs)
      qio_channel_shutdown(client->ioc,
                           QIO_CHANNEL_SHUTDOWN_BOTH,
                           NULL);
 -    nbd_recv_coroutines_enter_all(client);
 +    nbd_recv_coroutines_enter_all(bs);
      nbd_client_detach_aio_context(bs);
      object_unref(OBJECT(client->sioc));
@@ -XXX,XX +XXX,XX @@ static void nbd_teardown_connection(BlockDriverState *bs)
      client->ioc = NULL;
  }
 -static void nbd_reply_ready(void *opaque)
 +static coroutine_fn void nbd_read_reply_entry(void *opaque)
  {
 -    BlockDriverState *bs = opaque;
 -    NBDClientSession *s = nbd_get_client_session(bs);
 +    NBDClientSession *s = opaque;
      uint64_t i;
      int ret;
 -    if (!s->ioc) { /* Already closed */
 -        return;
 -    }
 -
 -    if (s->reply.handle == 0) {
 -        /* No reply already in flight.  Fetch a header.  It is possible
 -         * that another thread has done the same thing in parallel, so
 -         * the socket is not readable anymore.
 -         */
 +    for (;;) {
 +        assert(s->reply.handle == 0);
          ret = nbd_receive_reply(s->ioc, &s->reply);
 -        if (ret == -EAGAIN) {
 -            return;
 -        }
          if (ret < 0) {
 -            s->reply.handle = 0;
 -            goto fail;
 +            break;
          }
 -    }
 -    /* There's no need for a mutex on the receive side, because the
 -     * handler acts as a synchronization point and ensures that only
 -     * one coroutine is called until the reply finishes.  */
 -    i = HANDLE_TO_INDEX(s, s->reply.handle);
 -    if (i >= MAX_NBD_REQUESTS) {
 -        goto fail;
 -    }
 +        /* There's no need for a mutex on the receive side, because the
 +         * handler acts as a synchronization point and ensures that only
 +         * one coroutine is called until the reply finishes.
 +         */
 +        i = HANDLE_TO_INDEX(s, s->reply.handle);
 +        if (i >= MAX_NBD_REQUESTS || !s->recv_coroutine[i]) {
 +            break;
 +        }
 -    if (s->recv_coroutine[i]) {
 -        qemu_coroutine_enter(s->recv_coroutine[i]);
 -        return;
 +        /* We're woken up by the recv_coroutine itself.  Note that there
 +         * is no race between yielding and reentering read_reply_co.  This
 +         * is because:
 +         *
 +         * - if recv_coroutine[i] runs on the same AioContext, it is only
 +         *   entered after we yield
 +         *
 +         * - if recv_coroutine[i] runs on a different AioContext, reentering
 +         *   read_reply_co happens through a bottom half, which can only
 +         *   run after we yield.
 +         */
 +        aio_co_wake(s->recv_coroutine[i]);
 +        qemu_coroutine_yield();
      }
 -
 -fail:
 -    nbd_teardown_connection(bs);
 -}
 -
 -static void nbd_restart_write(void *opaque)
 -{
 -    BlockDriverState *bs = opaque;
 -
 -    qemu_coroutine_enter(nbd_get_client_session(bs)->send_coroutine);
 +    s->read_reply_co = NULL;
  }
  static int nbd_co_send_request(BlockDriverState *bs,
@@ -XXX,XX +XXX,XX @@ static int nbd_co_send_request(BlockDriverState *bs,
                                 QEMUIOVector *qiov)
  {
      NBDClientSession *s = nbd_get_client_session(bs);
 -    AioContext *aio_context;
      int rc, ret, i;
      qemu_co_mutex_lock(&s->send_mutex);
@@ -XXX,XX +XXX,XX @@ static int nbd_co_send_request(BlockDriverState *bs,
          return -EPIPE;
      }
 -    s->send_coroutine = qemu_coroutine_self();
 -    aio_context = bdrv_get_aio_context(bs);
 -
 -    aio_set_fd_handler(aio_context, s->sioc->fd, false,
 -                       nbd_reply_ready, nbd_restart_write, NULL, bs);
      if (qiov) {
          qio_channel_set_cork(s->ioc, true);
          rc = nbd_send_request(s->ioc, request);
@@ -XXX,XX +XXX,XX @@ static int nbd_co_send_request(BlockDriverState *bs,
      } else {
          rc = nbd_send_request(s->ioc, request);
      }
 -    aio_set_fd_handler(aio_context, s->sioc->fd, false,
 -                       nbd_reply_ready, NULL, NULL, bs);
 -    s->send_coroutine = NULL;
      qemu_co_mutex_unlock(&s->send_mutex);
      return rc;
  }
@@ -XXX,XX +XXX,XX @@ static void nbd_co_receive_reply(NBDClientSession *s,
  {
      int ret;
 -    /* Wait until we're woken up by the read handler.  TODO: perhaps
 -     * peek at the next reply and avoid yielding if it's ours?  */
 +    /* Wait until we're woken up by nbd_read_reply_entry.  */
      qemu_coroutine_yield();
      *reply = s->reply;
      if (reply->handle != request->handle ||
@@ -XXX,XX +XXX,XX @@ static void nbd_coroutine_start(NBDClientSession *s,
      /* s->recv_coroutine[i] is set as soon as we get the send_lock.  */
  }
 -static void nbd_coroutine_end(NBDClientSession *s,
 +static void nbd_coroutine_end(BlockDriverState *bs,
                                NBDRequest *request)
  {
 +    NBDClientSession *s = nbd_get_client_session(bs);
      int i = HANDLE_TO_INDEX(s, request->handle);
 +
-+echo "Write uncompressed data:"
+     s->recv_coroutine[i] = NULL;
-+let data_size="8 * $cluster_size"
+-    if (s->in_flight-- == MAX_NBD_REQUESTS) {
-+$QEMU_IO -c "write -P 0xaa 0 $data_size" "$TEST_IMG" \
+-        qemu_co_queue_next(&s->free_sema);
-+         2>&1 | _filter_qemu_io | _filter_testdir
++    s->in_flight--;
-+sizeA=$($QEMU_IMG info --output=json "$TEST_IMG" |
++    qemu_co_queue_next(&s->free_sema);
 +        sed -n '/"actual-size":/ s/[^0-9]//gp')
 +
-+_make_test_img 2M -o cluster_size=$cluster_size
++    /* Kick the read_reply_co to get the next reply.  */
-+echo "Write compressed data:"
++    if (s->read_reply_co) {
-+let data_size="3 * $cluster_size + $cluster_size / 2"
++        aio_co_wake(s->read_reply_co);
-+# Set compress on. That will align the written data
+     }
-+# by the cluster size and will write them compressed.
+ }
-+QEMU_IO_OPTIONS=$QEMU_IO_OPTIONS_NO_FMT \
-+$QEMU_IO -c "write -P 0xbb 0 $data_size" --image-opts \
+@@ -XXX,XX +XXX,XX @@ int nbd_client_co_preadv(BlockDriverState *bs, uint64_t offset,
-+         "driver=compress,file.driver=$IMGFMT,file.file.driver=file,file.file.filename=$TEST_IMG" \
+     } else {
-+         2>&1 | _filter_qemu_io | _filter_testdir
+         nbd_co_receive_reply(client, &request, &reply, qiov);
      }
 -    nbd_coroutine_end(client, &request);
 +    nbd_coroutine_end(bs, &request);
      return -reply.error;
  }
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_pwritev(BlockDriverState *bs, uint64_t offset,
      } else {
          nbd_co_receive_reply(client, &request, &reply, NULL);
      }
 -    nbd_coroutine_end(client, &request);
 +    nbd_coroutine_end(bs, &request);
      return -reply.error;
  }
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,
      } else {
          nbd_co_receive_reply(client, &request, &reply, NULL);
      }
 -    nbd_coroutine_end(client, &request);
 +    nbd_coroutine_end(bs, &request);
      return -reply.error;
  }
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_flush(BlockDriverState *bs)
      } else {
          nbd_co_receive_reply(client, &request, &reply, NULL);
      }
 -    nbd_coroutine_end(client, &request);
 +    nbd_coroutine_end(bs, &request);
      return -reply.error;
  }
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_pdiscard(BlockDriverState *bs, int64_t offset, int count)
      } else {
          nbd_co_receive_reply(client, &request, &reply, NULL);
      }
 -    nbd_coroutine_end(client, &request);
 +    nbd_coroutine_end(bs, &request);
      return -reply.error;
  }
  void nbd_client_detach_aio_context(BlockDriverState *bs)
  {
 -    aio_set_fd_handler(bdrv_get_aio_context(bs),
 -                       nbd_get_client_session(bs)->sioc->fd,
 -                       false, NULL, NULL, NULL, NULL);
 +    NBDClientSession *client = nbd_get_client_session(bs);
 +    qio_channel_detach_aio_context(QIO_CHANNEL(client->sioc));
  }
  void nbd_client_attach_aio_context(BlockDriverState *bs,
                                     AioContext *new_context)
  {
 -    aio_set_fd_handler(new_context, nbd_get_client_session(bs)->sioc->fd,
 -                       false, nbd_reply_ready, NULL, NULL, bs);
 +    NBDClientSession *client = nbd_get_client_session(bs);
 +    qio_channel_attach_aio_context(QIO_CHANNEL(client->sioc), new_context);
 +    aio_co_schedule(new_context, client->read_reply_co);
  }
  void nbd_client_close(BlockDriverState *bs)
@@ -XXX,XX +XXX,XX @@ int nbd_client_init(BlockDriverState *bs,
      /* Now that we're connected, set the socket to be non-blocking and
       * kick the reply mechanism.  */
      qio_channel_set_blocking(QIO_CHANNEL(sioc), false, NULL);
 -
 +    client->read_reply_co = qemu_coroutine_create(nbd_read_reply_entry, client);
      nbd_client_attach_aio_context(bs, bdrv_get_aio_context(bs));
      logout("Established connection with NBD server\n");
 diff --git a/nbd/client.c b/nbd/client.c
 index XXXXXXX..XXXXXXX 100644
 --- a/nbd/client.c
 +++ b/nbd/client.c
@@ -XXX,XX +XXX,XX @@ ssize_t nbd_receive_reply(QIOChannel *ioc, NBDReply *reply)
      ssize_t ret;
      ret = read_sync(ioc, buf, sizeof(buf));
 -    if (ret < 0) {
 +    if (ret <= 0) {
          return ret;
      }
 diff --git a/nbd/common.c b/nbd/common.c
 index XXXXXXX..XXXXXXX 100644
 --- a/nbd/common.c
 +++ b/nbd/common.c
@@ -XXX,XX +XXX,XX @@ ssize_t nbd_wr_syncv(QIOChannel *ioc,
          }
          if (len == QIO_CHANNEL_ERR_BLOCK) {
              if (qemu_in_coroutine()) {
 -                /* XXX figure out if we can create a variant on
 -                 * qio_channel_yield() that works with AIO contexts
 -                 * and consider using that in this branch */
 -                qemu_coroutine_yield();
 -            } else if (done) {
 -                /* XXX this is needed by nbd_reply_ready.  */
 -                qio_channel_wait(ioc,
 -                                 do_read ? G_IO_IN : G_IO_OUT);
 +                qio_channel_yield(ioc, do_read ? G_IO_IN : G_IO_OUT);
              } else {
                  return -EAGAIN;
              }
 diff --git a/nbd/server.c b/nbd/server.c
 index XXXXXXX..XXXXXXX 100644
 --- a/nbd/server.c
 +++ b/nbd/server.c
@@ -XXX,XX +XXX,XX @@ struct NBDClient {
      CoMutex send_lock;
      Coroutine *send_coroutine;
 -    bool can_read;
 -
      QTAILQ_ENTRY(NBDClient) next;
      int nb_requests;
      bool closing;
@@ -XXX,XX +XXX,XX @@ struct NBDClient {
  /* That's all folks */
 -static void nbd_set_handlers(NBDClient *client);
 -static void nbd_unset_handlers(NBDClient *client);
 -static void nbd_update_can_read(NBDClient *client);
 +static void nbd_client_receive_next_request(NBDClient *client);
  static gboolean nbd_negotiate_continue(QIOChannel *ioc,
                                         GIOCondition condition,
@@ -XXX,XX +XXX,XX @@ void nbd_client_put(NBDClient *client)
           */
          assert(client->closing);
 -        nbd_unset_handlers(client);
 +        qio_channel_detach_aio_context(client->ioc);
          object_unref(OBJECT(client->sioc));
          object_unref(OBJECT(client->ioc));
          if (client->tlscreds) {
@@ -XXX,XX +XXX,XX @@ static NBDRequestData *nbd_request_get(NBDClient *client)
      assert(client->nb_requests <= MAX_NBD_REQUESTS - 1);
      client->nb_requests++;
 -    nbd_update_can_read(client);
      req = g_new0(NBDRequestData, 1);
      nbd_client_get(client);
@@ -XXX,XX +XXX,XX @@ static void nbd_request_put(NBDRequestData *req)
      g_free(req);
      client->nb_requests--;
 -    nbd_update_can_read(client);
 +    nbd_client_receive_next_request(client);
 +
-+let offset="4 * $cluster_size + $cluster_size / 4"
+     nbd_client_put(client);
-+QEMU_IO_OPTIONS=$QEMU_IO_OPTIONS_NO_FMT \
+ }
-+$QEMU_IO -c "write -P 0xcc $offset $data_size" "json:{\
-+    'driver': 'compress',
+@@ -XXX,XX +XXX,XX @@ static void blk_aio_attached(AioContext *ctx, void *opaque)
-+    'file': {'driver': '$IMGFMT',
+     exp->ctx = ctx;
-+             'file': {'driver': 'file',
-+                      'filename': '$TEST_IMG'}}}" | \
+     QTAILQ_FOREACH(client, &exp->clients, next) {
-+                          _filter_qemu_io | _filter_testdir
+-        nbd_set_handlers(client);
 +        qio_channel_attach_aio_context(client->ioc, ctx);
 +        if (client->recv_coroutine) {
 +            aio_co_schedule(ctx, client->recv_coroutine);
 +        }
 +        if (client->send_coroutine) {
 +            aio_co_schedule(ctx, client->send_coroutine);
 +        }
      }
  }
@@ -XXX,XX +XXX,XX @@ static void blk_aio_detach(void *opaque)
      TRACE("Export %s: Detaching clients from AIO context %p\n", exp->name, exp->ctx);
      QTAILQ_FOREACH(client, &exp->clients, next) {
 -        nbd_unset_handlers(client);
 +        qio_channel_detach_aio_context(client->ioc);
      }
      exp->ctx = NULL;
@@ -XXX,XX +XXX,XX @@ static ssize_t nbd_co_send_reply(NBDRequestData *req, NBDReply *reply,
      g_assert(qemu_in_coroutine());
      qemu_co_mutex_lock(&client->send_lock);
      client->send_coroutine = qemu_coroutine_self();
 -    nbd_set_handlers(client);
      if (!len) {
          rc = nbd_send_reply(client->ioc, reply);
@@ -XXX,XX +XXX,XX @@ static ssize_t nbd_co_send_reply(NBDRequestData *req, NBDReply *reply,
      }
      client->send_coroutine = NULL;
 -    nbd_set_handlers(client);
      qemu_co_mutex_unlock(&client->send_lock);
      return rc;
  }
@@ -XXX,XX +XXX,XX @@ static ssize_t nbd_co_receive_request(NBDRequestData *req,
      ssize_t rc;
      g_assert(qemu_in_coroutine());
 -    client->recv_coroutine = qemu_coroutine_self();
 -    nbd_update_can_read(client);
 -
 +    assert(client->recv_coroutine == qemu_coroutine_self());
      rc = nbd_receive_request(client->ioc, request);
      if (rc < 0) {
          if (rc != -EAGAIN) {
@@ -XXX,XX +XXX,XX @@ static ssize_t nbd_co_receive_request(NBDRequestData *req,
  out:
      client->recv_coroutine = NULL;
 -    nbd_update_can_read(client);
 +    nbd_client_receive_next_request(client);
      return rc;
  }
 -static void nbd_trip(void *opaque)
 +/* Owns a reference to the NBDClient passed as opaque.  */
 +static coroutine_fn void nbd_trip(void *opaque)
  {
      NBDClient *client = opaque;
      NBDExport *exp = client->exp;
      NBDRequestData *req;
 -    NBDRequest request;
 +    NBDRequest request = { 0 };    /* GCC thinks it can be used uninitialized */
      NBDReply reply;
      ssize_t ret;
      int flags;
      TRACE("Reading request.");
      if (client->closing) {
 +        nbd_client_put(client);
          return;
      }
@@ -XXX,XX +XXX,XX @@ static void nbd_trip(void *opaque)
  done:
      nbd_request_put(req);
 +    nbd_client_put(client);
      return;
  out:
      nbd_request_put(req);
      client_close(client);
 +    nbd_client_put(client);
  }
 -static void nbd_read(void *opaque)
 +static void nbd_client_receive_next_request(NBDClient *client)
  {
 -    NBDClient *client = opaque;
 -
 -    if (client->recv_coroutine) {
 -        qemu_coroutine_enter(client->recv_coroutine);
 -    } else {
 -        qemu_coroutine_enter(qemu_coroutine_create(nbd_trip, client));
 -    }
 -}
 -
 -static void nbd_restart_write(void *opaque)
 -{
 -    NBDClient *client = opaque;
 -
 -    qemu_coroutine_enter(client->send_coroutine);
 -}
 -
 -static void nbd_set_handlers(NBDClient *client)
 -{
 -    if (client->exp && client->exp->ctx) {
 -        aio_set_fd_handler(client->exp->ctx, client->sioc->fd, true,
 -                           client->can_read ? nbd_read : NULL,
 -                           client->send_coroutine ? nbd_restart_write : NULL,
 -                           NULL, client);
 -    }
 -}
 -
 -static void nbd_unset_handlers(NBDClient *client)
 -{
 -    if (client->exp && client->exp->ctx) {
 -        aio_set_fd_handler(client->exp->ctx, client->sioc->fd, true, NULL,
 -                           NULL, NULL, NULL);
 -    }
 -}
 -
 -static void nbd_update_can_read(NBDClient *client)
 -{
 -    bool can_read = client->recv_coroutine ||
 -                    client->nb_requests < MAX_NBD_REQUESTS;
 -
 -    if (can_read != client->can_read) {
 -        client->can_read = can_read;
 -        nbd_set_handlers(client);
 -
 -        /* There is no need to invoke aio_notify(), since aio_set_fd_handler()
 -         * in nbd_set_handlers() will have taken care of that */
 +    if (!client->recv_coroutine && client->nb_requests < MAX_NBD_REQUESTS) {
 +        nbd_client_get(client);
 +        client->recv_coroutine = qemu_coroutine_create(nbd_trip, client);
 +        aio_co_schedule(client->exp->ctx, client->recv_coroutine);
      }
  }
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void nbd_co_client_start(void *opaque)
          goto out;
      }
      qemu_co_mutex_init(&client->send_lock);
 -    nbd_set_handlers(client);
      if (exp) {
          QTAILQ_INSERT_TAIL(&exp->clients, client, next);
      }
 +
-+sizeB=$($QEMU_IMG info --output=json "$TEST_IMG" |
++    nbd_client_receive_next_request(client);
 +        sed -n '/"actual-size":/ s/[^0-9]//gp')
 +
-+if [ $sizeA -le $sizeB ]
+ out:
-+then
+     g_free(data);
-+    echo "Compression ERROR"
+ }
-+fi
+@@ -XXX,XX +XXX,XX @@ void nbd_client_new(NBDExport *exp,
-+
+     object_ref(OBJECT(client->sioc));
-+$QEMU_IMG check --output=json "$TEST_IMG" |
+     client->ioc = QIO_CHANNEL(sioc);
-+          sed -n 's/,$//; /"compressed-clusters":/ s/^ *//p'
+     object_ref(OBJECT(client->ioc));
-+
+-    client->can_read = true;
- # success, all done
+     client->close = close_fn;
- echo '*** done'
- rm -f $seq.full
+     data->client = client;
 diff --git a/tests/qemu-iotests/214.out b/tests/qemu-iotests/214.out
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/qemu-iotests/214.out
 +++ b/tests/qemu-iotests/214.out
@@ -XXX,XX +XXX,XX @@ read 4194304/4194304 bytes at offset 0
 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
  read 4194304/4194304 bytes at offset 4194304
 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 +
 +=== Write compressed data of multiple clusters ===
 +
 +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2097152
 +Write uncompressed data:
 +wrote 524288/524288 bytes at offset 0
 +512 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2097152
 +Write compressed data:
 +wrote 229376/229376 bytes at offset 0
 +224 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 +wrote 229376/229376 bytes at offset 278528
 +224 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 +"compressed-clusters": 8
  *** done
 --
-.24.1
+.9.3

-[PULL 28/34] iotests: Disable data_file where it cannot be used
+[Qemu-devel] [PULL v2 08/24] coroutine-lock: reschedule coroutine on the AioContext it was running on
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-Message-id: 20191107163708.833192-22-mreitz@redhat.com
-[mreitz: Also disable 273]
+As a small step towards the introduction of multiqueue, we want
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+coroutines to remain on the same AioContext that started them,
 unless they are moved explicitly with e.g. aio_co_schedule.  This patch
 avoids that coroutines switch AioContext when they use a CoMutex.
 For now it does not make much of a difference, because the CoMutex
 is not thread-safe and the AioContext itself is used to protect the
 CoMutex from concurrent access.  However, this is going to change.
 Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-9-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/007 | 5 +++--
+ util/qemu-coroutine-lock.c | 5 ++---
- tests/qemu-iotests/014 | 2 ++
+ util/trace-events          | 1 -
- tests/qemu-iotests/015 | 5 +++--
+files changed, 2 insertions(+), 4 deletions(-)
  tests/qemu-iotests/026 | 5 ++++-
  tests/qemu-iotests/029 | 5 +++--
  tests/qemu-iotests/031 | 6 +++---
  tests/qemu-iotests/036 | 5 +++--
  tests/qemu-iotests/039 | 3 +++
  tests/qemu-iotests/046 | 2 ++
  tests/qemu-iotests/048 | 2 ++
  tests/qemu-iotests/051 | 5 +++--
  tests/qemu-iotests/058 | 5 +++--
  tests/qemu-iotests/060 | 6 ++++--
  tests/qemu-iotests/061 | 6 ++++--
  tests/qemu-iotests/062 | 2 +-
  tests/qemu-iotests/066 | 4 +++-
  tests/qemu-iotests/067 | 6 ++++--
  tests/qemu-iotests/068 | 5 +++--
  tests/qemu-iotests/071 | 3 +++
  tests/qemu-iotests/073 | 4 ++++
  tests/qemu-iotests/074 | 2 ++
  tests/qemu-iotests/080 | 5 +++--
  tests/qemu-iotests/090 | 2 ++
  tests/qemu-iotests/098 | 6 ++++--
  tests/qemu-iotests/099 | 3 ++-
  tests/qemu-iotests/103 | 5 +++--
  tests/qemu-iotests/108 | 6 ++++--
  tests/qemu-iotests/112 | 5 +++--
  tests/qemu-iotests/114 | 2 ++
  tests/qemu-iotests/121 | 3 +++
  tests/qemu-iotests/138 | 3 +++
  tests/qemu-iotests/156 | 2 ++
  tests/qemu-iotests/176 | 7 +++++--
  tests/qemu-iotests/191 | 2 ++
  tests/qemu-iotests/201 | 6 +++---
  tests/qemu-iotests/214 | 3 ++-
  tests/qemu-iotests/217 | 3 ++-
  tests/qemu-iotests/220 | 5 +++--
  tests/qemu-iotests/243 | 6 ++++--
  tests/qemu-iotests/244 | 5 +++--
  tests/qemu-iotests/250 | 2 ++
  tests/qemu-iotests/261 | 3 ++-
  tests/qemu-iotests/267 | 5 +++--
  tests/qemu-iotests/273 | 3 +++
 files changed, 127 insertions(+), 53 deletions(-)
-diff --git a/tests/qemu-iotests/007 b/tests/qemu-iotests/007
+diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
-index XXXXXXX..XXXXXXX 100755
+index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/007
+--- a/util/qemu-coroutine-lock.c
-+++ b/tests/qemu-iotests/007
++++ b/util/qemu-coroutine-lock.c
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
+@@ -XXX,XX +XXX,XX @@
- _supported_fmt qcow2
+ #include "qemu/coroutine.h"
- _supported_proto generic
+ #include "qemu/coroutine_int.h"
- # refcount_bits must be at least 4 so we can create ten internal snapshots
+ #include "qemu/queue.h"
--# (1 bit supports none, 2 bits support two, 4 bits support 14)
++#include "block/aio.h"
--_unsupported_imgopts 'refcount_bits=\(1\|2\)[^0-9]'
+ #include "trace.h"
-+# (1 bit supports none, 2 bits support two, 4 bits support 14);
-+# snapshot are generally impossible with external data files
+ void qemu_co_queue_init(CoQueue *queue)
-+_unsupported_imgopts 'refcount_bits=\(1\|2\)[^0-9]' data_file
+@@ -XXX,XX +XXX,XX @@ void qemu_co_queue_run_restart(Coroutine *co)
- echo
+ static bool qemu_co_queue_do_restart(CoQueue *queue, bool single)
  echo "creating image"
 diff --git a/tests/qemu-iotests/014 b/tests/qemu-iotests/014
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/014
 +++ b/tests/qemu-iotests/014
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# Compression and snapshots do not work with external data files
 +_unsupported_imgopts data_file
  TEST_OFFSETS="0 4294967296"
  TEST_OPS="writev read write readv"
 diff --git a/tests/qemu-iotests/015 b/tests/qemu-iotests/015
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/015
 +++ b/tests/qemu-iotests/015
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  # actually any format that supports snapshots
  _supported_fmt qcow2
  _supported_proto generic
 -# Internal snapshots are (currently) impossible with refcount_bits=1
 -_unsupported_imgopts 'refcount_bits=1[^0-9]'
 +# Internal snapshots are (currently) impossible with refcount_bits=1,
 +# and generally impossible with external data files
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
  echo
  echo "creating image"
 diff --git a/tests/qemu-iotests/026 b/tests/qemu-iotests/026
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/026
 +++ b/tests/qemu-iotests/026
@@ -XXX,XX +XXX,XX @@ _supported_cache_modes writethrough none
  # 32 and 64 bits do not work either, however, due to different leaked cluster
  # count on error.
  # Thus, the only remaining option is refcount_bits=16.
 -_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
 +#
 +# As for data_file, none of the refcount tests can work for it.
 +_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)' \
 +    data_file
  echo "Errors while writing 128 kB"
  echo
 diff --git a/tests/qemu-iotests/029 b/tests/qemu-iotests/029
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/029
 +++ b/tests/qemu-iotests/029
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto generic
  _unsupported_proto vxhs
 -# Internal snapshots are (currently) impossible with refcount_bits=1
 -_unsupported_imgopts 'refcount_bits=1[^0-9]'
 +# Internal snapshots are (currently) impossible with refcount_bits=1,
 +# and generally impossible with external data files
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
  offset_size=24
  offset_l1_size=36
 diff --git a/tests/qemu-iotests/031 b/tests/qemu-iotests/031
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/031
 +++ b/tests/qemu-iotests/031
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  # This tests qcow2-specific low-level functionality
  _supported_fmt qcow2
  _supported_proto file
 -# We want to test compat=0.10, which does not support refcount widths
 -# other than 16
 -_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
 +# We want to test compat=0.10, which does not support external data
 +# files or refcount widths other than 16
 +_unsupported_imgopts data_file 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
  CLUSTER_SIZE=65536
 diff --git a/tests/qemu-iotests/036 b/tests/qemu-iotests/036
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/036
 +++ b/tests/qemu-iotests/036
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  # This tests qcow2-specific low-level functionality
  _supported_fmt qcow2
  _supported_proto file
 -# Only qcow2v3 and later supports feature bits
 -_unsupported_imgopts 'compat=0.10'
 +# Only qcow2v3 and later supports feature bits;
 +# qcow2.py does not support external data files
 +_unsupported_imgopts 'compat=0.10' data_file
  echo
  echo === Image with unknown incompatible feature bit ===
 diff --git a/tests/qemu-iotests/039 b/tests/qemu-iotests/039
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/039
 +++ b/tests/qemu-iotests/039
@@ -XXX,XX +XXX,XX @@ _supported_proto file
  _supported_os Linux
  _default_cache_mode writethrough
  _supported_cache_modes writethrough
 +# Some of these test cases expect no external data file so that all
 +# clusters are part of the qcow2 image and refcounted
 +_unsupported_imgopts data_file
  size=128M
 diff --git a/tests/qemu-iotests/046 b/tests/qemu-iotests/046
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/046
 +++ b/tests/qemu-iotests/046
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
 +# data_file does not support compressed clusters
 +_unsupported_imgopts data_file
  CLUSTER_SIZE=64k
  size=128M
 diff --git a/tests/qemu-iotests/048 b/tests/qemu-iotests/048
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/048
 +++ b/tests/qemu-iotests/048
@@ -XXX,XX +XXX,XX @@ _compare()
  _supported_fmt raw qcow2 qed luks
  _supported_proto file
  _supported_os Linux
 +# Using 'cp' is incompatible with external data files
 +_unsupported_imgopts data_file
  # Remove once all tests are fixed to use TEST_IMG_FILE
  # correctly and common.rc sets it unconditionally
 diff --git a/tests/qemu-iotests/051 b/tests/qemu-iotests/051
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/051
 +++ b/tests/qemu-iotests/051
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  # A compat=0.10 image is created in this test which does not support anything
 -# other than refcount_bits=16
 -_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
 +# other than refcount_bits=16;
 +# it also will not support an external data file
 +_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)' data_file
  _require_drivers nbd
  do_run_qemu()
 diff --git a/tests/qemu-iotests/058 b/tests/qemu-iotests/058
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/058
 +++ b/tests/qemu-iotests/058
@@ -XXX,XX +XXX,XX @@ _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
  _require_command QEMU_NBD
 -# Internal snapshots are (currently) impossible with refcount_bits=1
 -_unsupported_imgopts 'refcount_bits=1[^0-9]'
 +# Internal snapshots are (currently) impossible with refcount_bits=1,
 +# and generally impossible with external data files
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
  nbd_snapshot_img="nbd:unix:$nbd_unix_socket"
 diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/060
 +++ b/tests/qemu-iotests/060
@@ -XXX,XX +XXX,XX @@ _filter_io_error()
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 -# These tests only work for compat=1.1 images with refcount_bits=16
 -_unsupported_imgopts 'compat=0.10' 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
 +# These tests only work for compat=1.1 images without an external
 +# data file with refcount_bits=16
 +_unsupported_imgopts 'compat=0.10' data_file \
 +    'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
  # The repair process will create a large file - so check for availability first
  _require_large_file 64G
 diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/061
 +++ b/tests/qemu-iotests/061
@@ -XXX,XX +XXX,XX @@ _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
  # Conversion between different compat versions can only really work
 -# with refcount_bits=16
 -_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
 +# with refcount_bits=16;
 +# we have explicit tests for data_file here, but the whole test does
 +# not work with it
 +_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)' data_file
  echo
  echo "=== Testing version downgrade with zero expansion ==="
 diff --git a/tests/qemu-iotests/062 b/tests/qemu-iotests/062
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/062
 +++ b/tests/qemu-iotests/062
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto generic
  # We need zero clusters and snapshots
 -_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
 +_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]' data_file
  IMG_SIZE=64M
 diff --git a/tests/qemu-iotests/066 b/tests/qemu-iotests/066
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/066
 +++ b/tests/qemu-iotests/066
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto generic
  # We need zero clusters and snapshots
 -_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
 +# (TODO: Consider splitting the snapshot part into a separate test
 +#        file, so this one runs with refcount_bits=1 and data_file)
 +_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]' data_file
  # Intentionally create an unaligned image
  IMG_SIZE=$((64 * 1024 * 1024 + 512))
 diff --git a/tests/qemu-iotests/067 b/tests/qemu-iotests/067
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/067
 +++ b/tests/qemu-iotests/067
@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
  _supported_fmt qcow2
  _supported_proto file
 -# Because anything other than 16 would change the output of query-block
 -_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
 +# Because anything other than 16 would change the output of query-block,
 +# and external data files would change the output of
 +# query-named-block-nodes
 +_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)' data_file
  do_run_qemu()
  {
-diff --git a/tests/qemu-iotests/068 b/tests/qemu-iotests/068
+-    Coroutine *self = qemu_coroutine_self();
-index XXXXXXX..XXXXXXX 100755
+     Coroutine *next;
---- a/tests/qemu-iotests/068
-+++ b/tests/qemu-iotests/068
+     if (QSIMPLEQ_EMPTY(&queue->entries)) {
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
+@@ -XXX,XX +XXX,XX @@ static bool qemu_co_queue_do_restart(CoQueue *queue, bool single)
- # This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
+     while ((next = QSIMPLEQ_FIRST(&queue->entries)) != NULL) {
- _supported_proto generic
+         QSIMPLEQ_REMOVE_HEAD(&queue->entries, co_queue_next);
--# Internal snapshots are (currently) impossible with refcount_bits=1
+-        QSIMPLEQ_INSERT_TAIL(&self->co_queue_wakeup, next, co_queue_next);
--_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
+-        trace_qemu_co_queue_next(next);
-+# Internal snapshots are (currently) impossible with refcount_bits=1,
++        aio_co_wake(next);
-+# and generally impossible with external data files
+         if (single) {
-+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]' data_file
+             break;
+         }
- IMG_SIZE=128K
+diff --git a/util/trace-events b/util/trace-events
+index XXXXXXX..XXXXXXX 100644
-diff --git a/tests/qemu-iotests/071 b/tests/qemu-iotests/071
+--- a/util/trace-events
-index XXXXXXX..XXXXXXX 100755
++++ b/util/trace-events
---- a/tests/qemu-iotests/071
+@@ -XXX,XX +XXX,XX @@ qemu_coroutine_terminate(void *co) "self %p"
-+++ b/tests/qemu-iotests/071
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
+ # util/qemu-coroutine-lock.c
- _supported_fmt qcow2
+ qemu_co_queue_run_restart(void *co) "co %p"
- _supported_proto file
+-qemu_co_queue_next(void *nxt) "next %p"
- _require_drivers blkdebug blkverify
+ qemu_co_mutex_lock_entry(void *mutex, void *self) "mutex %p self %p"
-+# blkdebug can only inject errors on bs->file, not on the data_file,
+ qemu_co_mutex_lock_return(void *mutex, void *self) "mutex %p self %p"
-+# so thie test does not work with external data files
+ qemu_co_mutex_unlock_entry(void *mutex, void *self) "mutex %p self %p"
 +_unsupported_imgopts data_file
  do_run_qemu()
  {
 diff --git a/tests/qemu-iotests/073 b/tests/qemu-iotests/073
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/073
 +++ b/tests/qemu-iotests/073
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto generic
  _unsupported_proto vxhs
 +# External data files do not support compressed clusters
 +# (TODO: Consider writing a version for external data files that does
 +#        not test compressed clusters)
 +_unsupported_imgopts data_file
  CLUSTER_SIZE=64k
  size=128M
 diff --git a/tests/qemu-iotests/074 b/tests/qemu-iotests/074
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/074
 +++ b/tests/qemu-iotests/074
@@ -XXX,XX +XXX,XX @@ _compare()
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# blkdebug can only inject errors on bs->file
 +_unsupported_imgopts data_file
  # Setup test basic parameters
  TEST_IMG2=$TEST_IMG.2
 diff --git a/tests/qemu-iotests/080 b/tests/qemu-iotests/080
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/080
 +++ b/tests/qemu-iotests/080
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 -# - Internal snapshots are (currently) impossible with refcount_bits=1
 +# - Internal snapshots are (currently) impossible with refcount_bits=1,
 +#   and generally impossible with external data files
  # - This is generally a test for compat=1.1 images
 -_unsupported_imgopts 'refcount_bits=1[^0-9]' 'compat=0.10'
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file 'compat=0.10'
  header_size=104
 diff --git a/tests/qemu-iotests/090 b/tests/qemu-iotests/090
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/090
 +++ b/tests/qemu-iotests/090
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file nfs
 +# External data files do not support compressed clusters
 +_unsupported_imgopts data_file
  IMG_SIZE=128K
 diff --git a/tests/qemu-iotests/098 b/tests/qemu-iotests/098
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/098
 +++ b/tests/qemu-iotests/098
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
 -# The code path we want to test here only works for compat=1.1 images
 -_unsupported_imgopts 'compat=0.10'
 +# The code path we want to test here only works for compat=1.1 images;
 +# blkdebug can only inject errors on bs->file, so external data files
 +# do not work with this test
 +_unsupported_imgopts 'compat=0.10' data_file
  for event in l1_update empty_image_prepare reftable_update refblock_alloc; do
 diff --git a/tests/qemu-iotests/099 b/tests/qemu-iotests/099
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/099
 +++ b/tests/qemu-iotests/099
@@ -XXX,XX +XXX,XX @@ _supported_fmt qcow qcow2 qed vdi vhdx vmdk vpc
  _supported_proto file
  _supported_os Linux
  _require_drivers blkdebug blkverify
 +# data_file would change the json:{} filenames
  _unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" \
 -    "subformat=twoGbMaxExtentSparse"
 +    "subformat=twoGbMaxExtentSparse" data_file
  do_run_qemu()
  {
 diff --git a/tests/qemu-iotests/103 b/tests/qemu-iotests/103
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/103
 +++ b/tests/qemu-iotests/103
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file nfs
 -# Internal snapshots are (currently) impossible with refcount_bits=1
 -_unsupported_imgopts 'refcount_bits=1[^0-9]'
 +# Internal snapshots are (currently) impossible with refcount_bits=1,
 +# and generally impossible with external data files
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
  IMG_SIZE=64K
 diff --git a/tests/qemu-iotests/108 b/tests/qemu-iotests/108
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/108
 +++ b/tests/qemu-iotests/108
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 -# This test directly modifies a refblock so it relies on refcount_bits being 16
 -_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
 +# This test directly modifies a refblock so it relies on refcount_bits being 16;
 +# and the low-level modification it performs are not tuned for external data
 +# files
 +_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)' data_file
  echo
  echo '=== Repairing an image without any refcount table ==='
 diff --git a/tests/qemu-iotests/112 b/tests/qemu-iotests/112
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/112
 +++ b/tests/qemu-iotests/112
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  # This test will set refcount_bits on its own which would conflict with the
 -# manual setting; compat will be overridden as well
 -_unsupported_imgopts refcount_bits 'compat=0.10'
 +# manual setting; compat will be overridden as well;
 +# and external data files do not work well with our refcount testing
 +_unsupported_imgopts refcount_bits 'compat=0.10' data_file
  print_refcount_bits()
  {
 diff --git a/tests/qemu-iotests/114 b/tests/qemu-iotests/114
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/114
 +++ b/tests/qemu-iotests/114
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto generic
  _unsupported_proto vxhs
 +# qcow2.py does not work too well with external data files
 +_unsupported_imgopts data_file
  TEST_IMG="$TEST_IMG.base" _make_test_img 64M
 diff --git a/tests/qemu-iotests/121 b/tests/qemu-iotests/121
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/121
 +++ b/tests/qemu-iotests/121
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# Refcount structures are used much differently with external data
 +# files
 +_unsupported_imgopts data_file
  echo
  echo '=== New refcount structures may not conflict with existing structures ==='
 diff --git a/tests/qemu-iotests/138 b/tests/qemu-iotests/138
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/138
 +++ b/tests/qemu-iotests/138
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# With an external data file, data clusters are not refcounted
 +# (and so qemu-img check does not check their refcount)
 +_unsupported_imgopts data_file
  echo
  echo '=== Check on an image with a multiple of 2^32 clusters ==='
 diff --git a/tests/qemu-iotests/156 b/tests/qemu-iotests/156
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/156
 +++ b/tests/qemu-iotests/156
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2 qed
  _supported_proto generic
  _unsupported_proto vxhs
 +# Copying files around with cp does not work with external data files
 +_unsupported_imgopts data_file
  # Create source disk
  TEST_IMG="$TEST_IMG.backing" _make_test_img 1M
 diff --git a/tests/qemu-iotests/176 b/tests/qemu-iotests/176
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/176
 +++ b/tests/qemu-iotests/176
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 -# Persistent dirty bitmaps require compat=1.1
 -_unsupported_imgopts 'compat=0.10'
 +# Persistent dirty bitmaps require compat=1.1;
 +# Internal snapshots forbid using an external data file
 +# (they work with refcount_bits=1 here, though, because there actually
 +# is no data when creating the snapshot)
 +_unsupported_imgopts 'compat=0.10' data_file
  run_qemu()
  {
 diff --git a/tests/qemu-iotests/191 b/tests/qemu-iotests/191
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/191
 +++ b/tests/qemu-iotests/191
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
 +# An external data file would change the query-named-block-nodes output
 +_unsupported_imgopts data_file
  size=64M
 diff --git a/tests/qemu-iotests/201 b/tests/qemu-iotests/201
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/201
 +++ b/tests/qemu-iotests/201
@@ -XXX,XX +XXX,XX @@ _supported_fmt qcow2
  _supported_proto generic
  _supported_os Linux
 -# Internal snapshots are (currently) impossible with refcount_bits=1
 -# This was taken from test 080
 -_unsupported_imgopts 'refcount_bits=1[^0-9]'
 +# Internal snapshots are (currently) impossible with refcount_bits=1,
 +# and generally impossible with external data files
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
  size=64M
  _make_test_img $size
 diff --git a/tests/qemu-iotests/214 b/tests/qemu-iotests/214
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/214
 +++ b/tests/qemu-iotests/214
@@ -XXX,XX +XXX,XX @@ _supported_proto file
  # Repairing the corrupted image requires qemu-img check to store a
  # refcount up to 3, which requires at least two refcount bits.
 -_unsupported_imgopts 'refcount_bits=1[^0-9]'
 +# External data files do not support compressed clusters.
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
  echo
 diff --git a/tests/qemu-iotests/217 b/tests/qemu-iotests/217
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/217
 +++ b/tests/qemu-iotests/217
@@ -XXX,XX +XXX,XX @@ _supported_proto file
  # This test needs clusters with at least a refcount of 2 so that
  # OFLAG_COPIED is not set.  refcount_bits=1 is therefore unsupported.
 -_unsupported_imgopts 'refcount_bits=1[^0-9]'
 +# (As are external data files.)
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
  echo
  echo '=== Simulating an I/O error during snapshot deletion ==='
 diff --git a/tests/qemu-iotests/220 b/tests/qemu-iotests/220
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/220
 +++ b/tests/qemu-iotests/220
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 -# To use a different refcount width but 16 bits we need compat=1.1
 -_unsupported_imgopts 'compat=0.10'
 +# To use a different refcount width but 16 bits we need compat=1.1,
 +# and external data files do not support compressed clusters.
 +_unsupported_imgopts 'compat=0.10' data_file
  echo "== Creating huge file =="
 diff --git a/tests/qemu-iotests/243 b/tests/qemu-iotests/243
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/243
 +++ b/tests/qemu-iotests/243
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 -# External data files do not work with compat=0.10
 -_unsupported_imgopts 'compat=0.10'
 +# External data files do not work with compat=0.10, and because there
 +# is an explicit case for external data files here, we cannot allow
 +# the user to specify whether to use one
 +_unsupported_imgopts 'compat=0.10' data_file
  for mode in off metadata falloc full; do
 diff --git a/tests/qemu-iotests/244 b/tests/qemu-iotests/244
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/244
 +++ b/tests/qemu-iotests/244
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 -# External data files do not work with compat=0.10
 -_unsupported_imgopts 'compat=0.10'
 +# External data files do not work with compat=0.10, and because we use
 +# our own external data file, we cannot let the user specify one
 +_unsupported_imgopts 'compat=0.10' data_file
  echo
  echo "=== Create and open image with external data file ==="
 diff --git a/tests/qemu-iotests/250 b/tests/qemu-iotests/250
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/250
 +++ b/tests/qemu-iotests/250
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# This test does not make much sense with external data files
 +_unsupported_imgopts data_file
  # This test checks that qcow2_process_discards does not truncate a discard
  # request > 2G.
 diff --git a/tests/qemu-iotests/261 b/tests/qemu-iotests/261
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/261
 +++ b/tests/qemu-iotests/261
@@ -XXX,XX +XXX,XX @@ _supported_os Linux
  # (1) We create a v2 image that supports nothing but refcount_bits=16
  # (2) We do some refcount management on our own which expects
  #     refcount_bits=16
 -_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
 +# As for data files, they do not support snapshots at all.
 +_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)' data_file
  # Parameters:
  #   $1: image filename
 diff --git a/tests/qemu-iotests/267 b/tests/qemu-iotests/267
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/267
 +++ b/tests/qemu-iotests/267
@@ -XXX,XX +XXX,XX @@ _supported_proto file
  _supported_os Linux
  _require_drivers copy-on-read
 -# Internal snapshots are (currently) impossible with refcount_bits=1
 -_unsupported_imgopts 'refcount_bits=1[^0-9]'
 +# Internal snapshots are (currently) impossible with refcount_bits=1,
 +# and generally impossible with external data files
 +_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
  do_run_qemu()
  {
 diff --git a/tests/qemu-iotests/273 b/tests/qemu-iotests/273
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/273
 +++ b/tests/qemu-iotests/273
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# External data files would add nodes to the block graph, so it would
 +# not match the reference output
 +_unsupported_imgopts data_file
  do_run_qemu()
  {
 --
-.24.1
+.9.3

-[PULL 03/34] blkdebug: Allow taking/unsharing permissions
+[Qemu-devel] [PULL v2 09/24] blkdebug: reschedule coroutine on the AioContext it is running on
-Sometimes it is useful to be able to add a node to the block graph that
+From: Paolo Bonzini <pbonzini@redhat.com>
 takes or unshare a certain set of permissions for debugging purposes.
 This patch adds this capability to blkdebug.
-(Note that you cannot make blkdebug release or share permissions that it
+Keep the coroutine on the same AioContext.  Without this change,
-needs to take or cannot share, because this might result in assertion
+there would be a race between yielding the coroutine and reentering it.
-failures in the block layer.  But if the blkdebug node has no parents,
+While the race cannot happen now, because the code only runs from a single
-it will not take any permissions and share everything by default, so you
+AioContext, this will change with multiqueue support in the block layer.
 can then freely choose what permissions to take and share.)
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+While doing the change, replace custom bottom half with aio_co_schedule.
-Message-id: 20191108123455.39445-4-mreitz@redhat.com
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Fam Zheng <famz@redhat.com>
 Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
 Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-10-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- block/blkdebug.c     | 93 +++++++++++++++++++++++++++++++++++++++++++-
+ block/blkdebug.c | 9 +--------
- qapi/block-core.json | 14 ++++++-
+file changed, 1 insertion(+), 8 deletions(-)
 files changed, 105 insertions(+), 2 deletions(-)
 diff --git a/block/blkdebug.c b/block/blkdebug.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/blkdebug.c
 +++ b/block/blkdebug.c
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ out:
- #include "qemu/cutils.h"
+     return ret;
  #include "qemu/config-file.h"
  #include "block/block_int.h"
 +#include "block/qdict.h"
  #include "qemu/module.h"
  #include "qemu/option.h"
 +#include "qapi/qapi-visit-block-core.h"
  #include "qapi/qmp/qdict.h"
 +#include "qapi/qmp/qlist.h"
  #include "qapi/qmp/qstring.h"
 +#include "qapi/qobject-input-visitor.h"
  #include "sysemu/qtest.h"
  typedef struct BDRVBlkdebugState {
@@ -XXX,XX +XXX,XX @@ typedef struct BDRVBlkdebugState {
      uint64_t opt_discard;
      uint64_t max_discard;
 +    uint64_t take_child_perms;
 +    uint64_t unshare_child_perms;
 +
      /* For blkdebug_refresh_filename() */
      char *config_file;
@@ -XXX,XX +XXX,XX @@ static void blkdebug_parse_filename(const char *filename, QDict *options,
      qdict_put_str(options, "x-image", filename);
  }
-+static int blkdebug_parse_perm_list(uint64_t *dest, QDict *options,
+-static void error_callback_bh(void *opaque)
-+                                    const char *prefix, Error **errp)
+-{
-+{
+-    Coroutine *co = opaque;
-+    int ret = 0;
+-    qemu_coroutine_enter(co);
-+    QDict *subqdict = NULL;
+-}
-+    QObject *crumpled_subqdict = NULL;
+-
-+    Visitor *v = NULL;
+ static int inject_error(BlockDriverState *bs, BlkdebugRule *rule)
-+    BlockPermissionList *perm_list = NULL, *element;
+ {
-+    Error *local_err = NULL;
+     BDRVBlkdebugState *s = bs->opaque;
-+
+@@ -XXX,XX +XXX,XX @@ static int inject_error(BlockDriverState *bs, BlkdebugRule *rule)
-+    *dest = 0;
+     }
-+
-+    qdict_extract_subqdict(options, &subqdict, prefix);
+     if (!immediately) {
-+    if (!qdict_size(subqdict)) {
+-        aio_bh_schedule_oneshot(bdrv_get_aio_context(bs), error_callback_bh,
-+        goto out;
+-                                qemu_coroutine_self());
-+    }
++        aio_co_schedule(qemu_get_current_aio_context(), qemu_coroutine_self());
-+
+         qemu_coroutine_yield();
-+    crumpled_subqdict = qdict_crumple(subqdict, errp);
+     }
-+    if (!crumpled_subqdict) {
 +        ret = -EINVAL;
 +        goto out;
 +    }
 +
 +    v = qobject_input_visitor_new(crumpled_subqdict);
 +    visit_type_BlockPermissionList(v, NULL, &perm_list, &local_err);
 +    if (local_err) {
 +        error_propagate(errp, local_err);
 +        ret = -EINVAL;
 +        goto out;
 +    }
 +
 +    for (element = perm_list; element; element = element->next) {
 +        *dest |= bdrv_qapi_perm_to_blk_perm(element->value);
 +    }
 +
 +out:
 +    qapi_free_BlockPermissionList(perm_list);
 +    visit_free(v);
 +    qobject_unref(subqdict);
 +    qobject_unref(crumpled_subqdict);
 +    return ret;
 +}
 +
 +static int blkdebug_parse_perms(BDRVBlkdebugState *s, QDict *options,
 +                                Error **errp)
 +{
 +    int ret;
 +
 +    ret = blkdebug_parse_perm_list(&s->take_child_perms, options,
 +                                   "take-child-perms.", errp);
 +    if (ret < 0) {
 +        return ret;
 +    }
 +
 +    ret = blkdebug_parse_perm_list(&s->unshare_child_perms, options,
 +                                   "unshare-child-perms.", errp);
 +    if (ret < 0) {
 +        return ret;
 +    }
 +
 +    return 0;
 +}
 +
  static QemuOptsList runtime_opts = {
      .name = "blkdebug",
      .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
@@ -XXX,XX +XXX,XX @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
      /* Set initial state */
      s->state = 1;
 +    /* Parse permissions modifiers before opening the image file */
 +    ret = blkdebug_parse_perms(s, options, errp);
 +    if (ret < 0) {
 +        goto out;
 +    }
 +
      /* Open the image file */
      bs->file = bdrv_open_child(qemu_opt_get(opts, "x-image"), options, "image",
                                 bs, &child_file, false, &local_err);
@@ -XXX,XX +XXX,XX @@ static int blkdebug_reopen_prepare(BDRVReopenState *reopen_state,
      return 0;
  }
 +static void blkdebug_child_perm(BlockDriverState *bs, BdrvChild *c,
 +                                const BdrvChildRole *role,
 +                                BlockReopenQueue *reopen_queue,
 +                                uint64_t perm, uint64_t shared,
 +                                uint64_t *nperm, uint64_t *nshared)
 +{
 +    BDRVBlkdebugState *s = bs->opaque;
 +
 +    bdrv_filter_default_perms(bs, c, role, reopen_queue, perm, shared,
 +                              nperm, nshared);
 +
 +    *nperm |= s->take_child_perms;
 +    *nshared &= ~s->unshare_child_perms;
 +}
 +
  static const char *const blkdebug_strong_runtime_opts[] = {
      "config",
      "inject-error.",
@@ -XXX,XX +XXX,XX @@ static BlockDriver bdrv_blkdebug = {
      .bdrv_file_open         = blkdebug_open,
      .bdrv_close             = blkdebug_close,
      .bdrv_reopen_prepare    = blkdebug_reopen_prepare,
 -    .bdrv_child_perm        = bdrv_filter_default_perms,
 +    .bdrv_child_perm        = blkdebug_child_perm,
      .bdrv_getlength         = blkdebug_getlength,
      .bdrv_refresh_filename  = blkdebug_refresh_filename,
 diff --git a/qapi/block-core.json b/qapi/block-core.json
 index XXXXXXX..XXXXXXX 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
@@ -XXX,XX +XXX,XX @@
  #
  # @set-state:       array of state-change descriptions
  #
 +# @take-child-perms: Permissions to take on @image in addition to what
 +#                    is necessary anyway (which depends on how the
 +#                    blkdebug node is used).  Defaults to none.
 +#                    (since 5.0)
 +#
 +# @unshare-child-perms: Permissions not to share on @image in addition
 +#                       to what cannot be shared anyway (which depends
 +#                       on how the blkdebug node is used).  Defaults
 +#                       to none.  (since 5.0)
 +#
  # Since: 2.9
  ##
  { 'struct': 'BlockdevOptionsBlkdebug',
@@ -XXX,XX +XXX,XX @@
              '*opt-write-zero': 'int32', '*max-write-zero': 'int32',
              '*opt-discard': 'int32', '*max-discard': 'int32',
              '*inject-error': ['BlkdebugInjectErrorOptions'],
 -            '*set-state': ['BlkdebugSetStateOptions'] } }
 +            '*set-state': ['BlkdebugSetStateOptions'],
 +            '*take-child-perms': ['BlockPermission'],
 +            '*unshare-child-perms': ['BlockPermission'] } }
  ##
  # @BlockdevOptionsBlklogwrites:
 --
-.24.1
+.9.3

-[PULL 22/34] iotests: Use _rm_test_img for deleting test images
+[Qemu-devel] [PULL v2 10/24] qed: introduce qed_aio_start_io and qed_aio_next_io_cb
-Just rm will not delete external data files.  Use _rm_test_img every
+From: Paolo Bonzini <pbonzini@redhat.com>
 time we delete a test image.
-(In the process, clean up the indentation of every _cleanup() this patch
+qed_aio_start_io and qed_aio_next_io will not have to acquire/release
-touches.)
+the AioContext, while qed_aio_next_io_cb will.  Split the functionality
 and gain a little type-safety in the process.
-((Also, use quotes consistently.  I am happy to see unquoted instances
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-like "rm -rf $TEST_DIR/..." go.))
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-11-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
  block/qed.c | 39 +++++++++++++++++++++++++--------------
 file changed, 25 insertions(+), 14 deletions(-)
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+diff --git a/block/qed.c b/block/qed.c
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+index XXXXXXX..XXXXXXX 100644
-Message-id: 20191107163708.833192-16-mreitz@redhat.com
+--- a/block/qed.c
-Signed-off-by: Max Reitz <mreitz@redhat.com>
++++ b/block/qed.c
----
+@@ -XXX,XX +XXX,XX @@ static CachedL2Table *qed_new_l2_table(BDRVQEDState *s)
- tests/qemu-iotests/019 |  6 +++---
+     return l2_table;
- tests/qemu-iotests/020 |  6 +++---
+ }
- tests/qemu-iotests/024 | 10 +++++-----
- tests/qemu-iotests/028 |  2 +-
+-static void qed_aio_next_io(void *opaque, int ret);
- tests/qemu-iotests/029 |  2 +-
++static void qed_aio_next_io(QEDAIOCB *acb, int ret);
- tests/qemu-iotests/043 |  4 +++-
++
- tests/qemu-iotests/048 |  2 +-
++static void qed_aio_start_io(QEDAIOCB *acb)
- tests/qemu-iotests/050 |  4 ++--
++{
- tests/qemu-iotests/053 |  4 ++--
++    qed_aio_next_io(acb, 0);
- tests/qemu-iotests/058 |  2 +-
++}
- tests/qemu-iotests/059 |  2 +-
++
- tests/qemu-iotests/061 |  2 +-
++static void qed_aio_next_io_cb(void *opaque, int ret)
- tests/qemu-iotests/063 |  6 ++++--
++{
- tests/qemu-iotests/069 |  2 +-
++    QEDAIOCB *acb = opaque;
- tests/qemu-iotests/074 |  2 +-
++
- tests/qemu-iotests/080 |  2 +-
++    qed_aio_next_io(acb, ret);
- tests/qemu-iotests/081 |  6 +++---
++}
- tests/qemu-iotests/085 |  9 ++++++---
- tests/qemu-iotests/088 |  2 +-
+ static void qed_plug_allocating_write_reqs(BDRVQEDState *s)
  tests/qemu-iotests/092 |  2 +-
  tests/qemu-iotests/094 |  2 +-
  tests/qemu-iotests/095 |  5 +++--
  tests/qemu-iotests/099 |  7 ++++---
  tests/qemu-iotests/109 |  4 ++--
  tests/qemu-iotests/110 |  4 ++--
  tests/qemu-iotests/122 |  6 ++++--
  tests/qemu-iotests/123 |  2 +-
  tests/qemu-iotests/141 |  4 +++-
  tests/qemu-iotests/142 |  2 +-
  tests/qemu-iotests/144 |  4 +++-
  tests/qemu-iotests/153 | 10 +++-------
  tests/qemu-iotests/156 |  8 ++++++--
  tests/qemu-iotests/159 |  2 +-
  tests/qemu-iotests/160 |  3 ++-
  tests/qemu-iotests/161 |  4 ++--
  tests/qemu-iotests/170 |  2 +-
  tests/qemu-iotests/172 |  6 +++---
  tests/qemu-iotests/173 |  3 ++-
  tests/qemu-iotests/178 |  2 +-
  tests/qemu-iotests/182 |  2 +-
  tests/qemu-iotests/183 |  2 +-
  tests/qemu-iotests/185 |  4 ++--
  tests/qemu-iotests/187 |  6 +++---
  tests/qemu-iotests/190 |  2 +-
  tests/qemu-iotests/191 |  6 +++---
  tests/qemu-iotests/195 |  2 +-
  tests/qemu-iotests/197 |  2 +-
  tests/qemu-iotests/200 |  3 ++-
  tests/qemu-iotests/215 |  2 +-
  tests/qemu-iotests/225 |  2 +-
  tests/qemu-iotests/229 |  3 ++-
  tests/qemu-iotests/232 |  4 +++-
  tests/qemu-iotests/243 |  2 +-
  tests/qemu-iotests/244 |  4 ++--
  tests/qemu-iotests/247 |  4 +++-
  tests/qemu-iotests/249 |  4 ++--
  tests/qemu-iotests/252 |  2 +-
 files changed, 118 insertions(+), 95 deletions(-)
 diff --git a/tests/qemu-iotests/019 b/tests/qemu-iotests/019
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/019
 +++ b/tests/qemu-iotests/019
@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
  _cleanup()
  {
--    _cleanup_test_img
+@@ -XXX,XX +XXX,XX @@ static void qed_unplug_allocating_write_reqs(BDRVQEDState *s)
--    rm -f "$TEST_IMG.base"
--    rm -f "$TEST_IMG.orig"
+     acb = QSIMPLEQ_FIRST(&s->allocating_write_reqs);
-+    _cleanup_test_img
+     if (acb) {
-+    _rm_test_img "$TEST_IMG.base"
+-        qed_aio_next_io(acb, 0);
-+    _rm_test_img "$TEST_IMG.orig"
++        qed_aio_start_io(acb);
      }
  }
- trap "_cleanup; exit \$status" 0 1 2 3 15
+@@ -XXX,XX +XXX,XX @@ static void qed_aio_complete(QEDAIOCB *acb, int ret)
-diff --git a/tests/qemu-iotests/020 b/tests/qemu-iotests/020
+         QSIMPLEQ_REMOVE_HEAD(&s->allocating_write_reqs, next);
-index XXXXXXX..XXXXXXX 100755
+         acb = QSIMPLEQ_FIRST(&s->allocating_write_reqs);
---- a/tests/qemu-iotests/020
+         if (acb) {
-+++ b/tests/qemu-iotests/020
+-            qed_aio_next_io(acb, 0);
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
++            qed_aio_start_io(acb);
+         } else if (s->header.features & QED_F_NEED_CHECK) {
- _cleanup()
+             qed_start_need_check_timer(s);
          }
@@ -XXX,XX +XXX,XX @@ static void qed_commit_l2_update(void *opaque, int ret)
      acb->request.l2_table = qed_find_l2_cache_entry(&s->l2_cache, l2_offset);
      assert(acb->request.l2_table != NULL);
 -    qed_aio_next_io(opaque, ret);
 +    qed_aio_next_io(acb, ret);
  }
  /**
@@ -XXX,XX +XXX,XX @@ static void qed_aio_write_l2_update(QEDAIOCB *acb, int ret, uint64_t offset)
      if (need_alloc) {
          /* Write out the whole new L2 table */
          qed_write_l2_table(s, &acb->request, 0, s->table_nelems, true,
 -                            qed_aio_write_l1_update, acb);
 +                           qed_aio_write_l1_update, acb);
      } else {
          /* Write out only the updated part of the L2 table */
          qed_write_l2_table(s, &acb->request, index, acb->cur_nclusters, false,
 -                            qed_aio_next_io, acb);
 +                           qed_aio_next_io_cb, acb);
      }
      return;
@@ -XXX,XX +XXX,XX @@ static void qed_aio_write_main(void *opaque, int ret)
      }
      if (acb->find_cluster_ret == QED_CLUSTER_FOUND) {
 -        next_fn = qed_aio_next_io;
 +        next_fn = qed_aio_next_io_cb;
      } else {
          if (s->bs->backing) {
              next_fn = qed_aio_write_flush_before_l2_update;
@@ -XXX,XX +XXX,XX @@ static void qed_aio_write_alloc(QEDAIOCB *acb, size_t len)
      if (acb->flags & QED_AIOCB_ZERO) {
          /* Skip ahead if the clusters are already zero */
          if (acb->find_cluster_ret == QED_CLUSTER_ZERO) {
 -            qed_aio_next_io(acb, 0);
 +            qed_aio_start_io(acb);
              return;
          }
@@ -XXX,XX +XXX,XX @@ static void qed_aio_read_data(void *opaque, int ret,
      /* Handle zero cluster and backing file reads */
      if (ret == QED_CLUSTER_ZERO) {
          qemu_iovec_memset(&acb->cur_qiov, 0, 0, acb->cur_qiov.size);
 -        qed_aio_next_io(acb, 0);
 +        qed_aio_start_io(acb);
          return;
      } else if (ret != QED_CLUSTER_FOUND) {
          qed_read_backing_file(s, acb->cur_pos, &acb->cur_qiov,
 -                              &acb->backing_qiov, qed_aio_next_io, acb);
 +                              &acb->backing_qiov, qed_aio_next_io_cb, acb);
          return;
      }
      BLKDBG_EVENT(bs->file, BLKDBG_READ_AIO);
      bdrv_aio_readv(bs->file, offset / BDRV_SECTOR_SIZE,
                     &acb->cur_qiov, acb->cur_qiov.size / BDRV_SECTOR_SIZE,
 -                   qed_aio_next_io, acb);
 +                   qed_aio_next_io_cb, acb);
      return;
  err:
@@ -XXX,XX +XXX,XX @@ err:
  /**
   * Begin next I/O or complete the request
   */
 -static void qed_aio_next_io(void *opaque, int ret)
 +static void qed_aio_next_io(QEDAIOCB *acb, int ret)
  {
--    _cleanup_test_img
+-    QEDAIOCB *acb = opaque;
--    rm -f "$TEST_IMG.base"
+     BDRVQEDState *s = acb_to_s(acb);
--    rm -f "$TEST_IMG.orig"
+     QEDFindClusterFunc *io_fn = (acb->flags & QED_AIOCB_WRITE) ?
-+    _cleanup_test_img
+                                 qed_aio_write_data : qed_aio_read_data;
-+    _rm_test_img "$TEST_IMG.base"
+@@ -XXX,XX +XXX,XX @@ static BlockAIOCB *qed_aio_setup(BlockDriverState *bs,
-+    _rm_test_img "$TEST_IMG.orig"
+     qemu_iovec_init(&acb->cur_qiov, qiov->niov);
      /* Start request */
 -    qed_aio_next_io(acb, 0);
 +    qed_aio_start_io(acb);
      return &acb->common;
  }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/024 b/tests/qemu-iotests/024
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/024
-+++ b/tests/qemu-iotests/024
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_DIR/t.$IMGFMT.base_old"
--    rm -f "$TEST_DIR/t.$IMGFMT.base_new"
-+    _rm_test_img "$TEST_DIR/t.$IMGFMT.base_old"
-+    _rm_test_img "$TEST_DIR/t.$IMGFMT.base_new"
--    rm -f "$TEST_DIR/subdir/t.$IMGFMT"
--    rm -f "$TEST_DIR/subdir/t.$IMGFMT.base_old"
--    rm -f "$TEST_DIR/subdir/t.$IMGFMT.base_new"
-+    _rm_test_img "$TEST_DIR/subdir/t.$IMGFMT"
-+    _rm_test_img "$TEST_DIR/subdir/t.$IMGFMT.base_old"
-+    _rm_test_img "$TEST_DIR/subdir/t.$IMGFMT.base_new"
-     rmdir "$TEST_DIR/subdir" 2> /dev/null
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/028 b/tests/qemu-iotests/028
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/028
-+++ b/tests/qemu-iotests/028
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_qemu
--    rm -f "${TEST_IMG}.copy"
-+    _rm_test_img "${TEST_IMG}.copy"
-     _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/029 b/tests/qemu-iotests/029
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/029
-+++ b/tests/qemu-iotests/029
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    rm -f $TEST_IMG.snap
-+    _rm_test_img "$TEST_IMG.snap"
-     _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/043 b/tests/qemu-iotests/043
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/043
-+++ b/tests/qemu-iotests/043
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG".[123].base
-+    for img in "$TEST_IMG".[123].base; do
-+        _rm_test_img "$img"
-+    done
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/048 b/tests/qemu-iotests/048
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/048
-+++ b/tests/qemu-iotests/048
-@@ -XXX,XX +XXX,XX @@ _cleanup()
- {
-     echo "Cleanup"
-     _cleanup_test_img
--    rm "${TEST_IMG_FILE2}"
-+    _rm_test_img "${TEST_IMG_FILE2}"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/050 b/tests/qemu-iotests/050
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/050
-+++ b/tests/qemu-iotests/050
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.old"
--    rm -f "$TEST_IMG.new"
-+    _rm_test_img "$TEST_IMG.old"
-+    _rm_test_img "$TEST_IMG.new"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/053 b/tests/qemu-iotests/053
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/053
-+++ b/tests/qemu-iotests/053
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    rm -f "$TEST_IMG.orig"
--    _cleanup_test_img
-+    _rm_test_img "$TEST_IMG.orig"
-+    _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/058 b/tests/qemu-iotests/058
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/058
-+++ b/tests/qemu-iotests/058
-@@ -XXX,XX +XXX,XX @@ _cleanup()
- {
-     nbd_server_stop
-     _cleanup_test_img
--    rm -f "$converted_image"
-+    _rm_test_img "$converted_image"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/059 b/tests/qemu-iotests/059
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/059
-+++ b/tests/qemu-iotests/059
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.qcow2"
-+    IMGFMT=qcow2 _rm_test_img "$TEST_IMG.qcow2"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/061
-+++ b/tests/qemu-iotests/061
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f $TEST_IMG.data
-+    _rm_test_img "$TEST_IMG.data"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/063 b/tests/qemu-iotests/063
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/063
-+++ b/tests/qemu-iotests/063
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    _cleanup_test_img
--    rm -f "$TEST_IMG.orig" "$TEST_IMG.raw1" "$TEST_IMG.raw2"
-+    _cleanup_test_img
-+    for img in "$TEST_IMG".{orig,raw1,raw2,target}; do
-+        _rm_test_img "$img"
-+    done
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/069 b/tests/qemu-iotests/069
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/069
-+++ b/tests/qemu-iotests/069
-@@ -XXX,XX +XXX,XX @@ echo "=== Creating an image with a backing file and deleting that file ==="
- echo
- TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE
- _make_test_img -b "$TEST_IMG.base" $IMG_SIZE
--rm -f "$TEST_IMG.base"
-+_rm_test_img "$TEST_IMG.base"
- # Just open the image and close it right again (this should print an error message)
- $QEMU_IO -c quit "$TEST_IMG" 2>&1 | _filter_testdir | _filter_imgfmt
-diff --git a/tests/qemu-iotests/074 b/tests/qemu-iotests/074
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/074
-+++ b/tests/qemu-iotests/074
-@@ -XXX,XX +XXX,XX @@ _cleanup()
- {
-     echo "Cleanup"
-     _cleanup_test_img
--    rm "${TEST_IMG2}"
-+    _rm_test_img "${TEST_IMG2}"
-     rm -f "$TEST_DIR/blkdebug.conf"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/080 b/tests/qemu-iotests/080
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/080
-+++ b/tests/qemu-iotests/080
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    rm -f $TEST_IMG.snap
-+    _rm_test_img "$TEST_IMG.snap"
-     _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/081 b/tests/qemu-iotests/081
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/081
-+++ b/tests/qemu-iotests/081
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    rm -rf $TEST_DIR/1.raw
--    rm -rf $TEST_DIR/2.raw
--    rm -rf $TEST_DIR/3.raw
-+    _rm_test_img "$TEST_DIR/1.raw"
-+    _rm_test_img "$TEST_DIR/2.raw"
-+    _rm_test_img "$TEST_DIR/3.raw"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/085 b/tests/qemu-iotests/085
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/085
-+++ b/tests/qemu-iotests/085
-@@ -XXX,XX +XXX,XX @@ _cleanup()
-     _cleanup_qemu
-     for i in $(seq 1 ${SNAPSHOTS})
-     do
--        rm -f "${TEST_DIR}/${i}-${snapshot_virt0}"
--        rm -f "${TEST_DIR}/${i}-${snapshot_virt1}"
-+        _rm_test_img "${TEST_DIR}/${i}-${snapshot_virt0}"
-+        _rm_test_img "${TEST_DIR}/${i}-${snapshot_virt1}"
-+    done
-+    for img in "${TEST_IMG}".{1,2,base}
-+    do
-+        _rm_test_img "$img"
-     done
--    rm -f "${TEST_IMG}" "${TEST_IMG}.1" "${TEST_IMG}.2" "${TEST_IMG}.base"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/088 b/tests/qemu-iotests/088
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/088
-+++ b/tests/qemu-iotests/088
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    rm -f $TEST_IMG.snap
-+    _rm_test_img "$TEST_IMG.snap"
-     _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/092 b/tests/qemu-iotests/092
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/092
-+++ b/tests/qemu-iotests/092
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    rm -f $TEST_IMG.snap
-+    _rm_test_img "$TEST_IMG.snap"
-     _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/094 b/tests/qemu-iotests/094
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/094
-+++ b/tests/qemu-iotests/094
-@@ -XXX,XX +XXX,XX @@ _cleanup()
- {
-     _cleanup_qemu
-     _cleanup_test_img
--    rm -f "$TEST_DIR/source.$IMGFMT"
-+    _rm_test_img "$TEST_DIR/source.$IMGFMT"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/095 b/tests/qemu-iotests/095
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/095
-+++ b/tests/qemu-iotests/095
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_qemu
--    rm  -f "${TEST_IMG}.base" "${TEST_IMG}.snp1"
--    _cleanup_test_img
-+    _rm_test_img "${TEST_IMG}.base"
-+    _rm_test_img "${TEST_IMG}.snp1"
-+    _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/099 b/tests/qemu-iotests/099
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/099
-+++ b/tests/qemu-iotests/099
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    _cleanup_test_img
-+    _cleanup_test_img
-+    _rm_test_img "$TEST_IMG.compare"
-+    rm -f "$TEST_DIR/blkdebug.conf"
-+
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-@@ -XXX,XX +XXX,XX @@ echo
- test_qemu "file.driver=blkdebug,file.image.filename=$TEST_IMG"
--rm -f "$TEST_IMG.compare" "$TEST_DIR/blkdebug.conf"
--
- # success, all done
- echo "*** done"
- rm -f $seq.full
-diff --git a/tests/qemu-iotests/109 b/tests/qemu-iotests/109
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/109
-+++ b/tests/qemu-iotests/109
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_qemu
--    rm -f $TEST_IMG.src
--    _cleanup_test_img
-+    _rm_test_img "$TEST_IMG.src"
-+    _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/110 b/tests/qemu-iotests/110
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/110
-+++ b/tests/qemu-iotests/110
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    _cleanup_test_img
--        rm -f "$TEST_IMG.copy"
-+    _cleanup_test_img
-+    _rm_test_img "$TEST_IMG.copy"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/122 b/tests/qemu-iotests/122
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/122
-+++ b/tests/qemu-iotests/122
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    rm -f "$TEST_IMG".[123]
--    _cleanup_test_img
-+    for img in "$TEST_IMG".[123]; do
-+        _rm_test_img "$img"
-+    done
-+    _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/123 b/tests/qemu-iotests/123
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/123
-+++ b/tests/qemu-iotests/123
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$SRC_IMG"
-+    _rm_test_img "$SRC_IMG"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/141 b/tests/qemu-iotests/141
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/141
-+++ b/tests/qemu-iotests/141
-@@ -XXX,XX +XXX,XX @@ _cleanup()
- {
-     _cleanup_qemu
-     _cleanup_test_img
--    rm -f "$TEST_DIR"/{b,m,o}.$IMGFMT
-+    for img in "$TEST_DIR"/{b,m,o}.$IMGFMT; do
-+        _rm_test_img "$img"
-+    done
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/142 b/tests/qemu-iotests/142
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/142
-+++ b/tests/qemu-iotests/142
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f $TEST_IMG.snap
-+    _rm_test_img "$TEST_IMG.snap"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/144 b/tests/qemu-iotests/144
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/144
-+++ b/tests/qemu-iotests/144
-@@ -XXX,XX +XXX,XX @@ TMP_SNAP2=${TEST_DIR}/tmp2.qcow2
- _cleanup()
- {
-     _cleanup_qemu
--    rm -f "${TEST_IMG}" "${TMP_SNAP1}" "${TMP_SNAP2}"
-+    for img in "${TEST_IMG}" "${TMP_SNAP1}" "${TMP_SNAP2}"; do
-+        _rm_test_img "$img"
-+    done
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/153 b/tests/qemu-iotests/153
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/153
-+++ b/tests/qemu-iotests/153
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "${TEST_IMG}.base"
--    rm -f "${TEST_IMG}.overlay"
--    rm -f "${TEST_IMG}.convert"
--    rm -f "${TEST_IMG}.a"
--    rm -f "${TEST_IMG}.b"
--    rm -f "${TEST_IMG}.c"
--    rm -f "${TEST_IMG}.lnk"
-+    for img in "${TEST_IMG}".{base,overlay,convert,a,b,c,lnk}; do
-+        _rm_test_img "$img"
-+    done
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/156 b/tests/qemu-iotests/156
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/156
-+++ b/tests/qemu-iotests/156
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_qemu
--    rm -f "$TEST_IMG"{,.target}{,.backing,.overlay}
-+    for img in "$TEST_IMG"{,.target}{,.backing,.overlay}; do
-+        _rm_test_img "$img"
-+    done
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-@@ -XXX,XX +XXX,XX @@ _send_qemu_cmd $QEMU_HANDLE \
-     '"status": "null"'
- # Remove the source images
--rm -f "$TEST_IMG{,.backing,.overlay}"
-+for img in "$TEST_IMG{,.backing,.overlay}"; do
-+    _rm_test_img "$img"
-+done
- echo
-diff --git a/tests/qemu-iotests/159 b/tests/qemu-iotests/159
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/159
-+++ b/tests/qemu-iotests/159
-@@ -XXX,XX +XXX,XX @@ status=1
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.out"
-+    _rm_test_img "$TEST_IMG.out"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/160 b/tests/qemu-iotests/160
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/160
-+++ b/tests/qemu-iotests/160
-@@ -XXX,XX +XXX,XX @@ status=1
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.out" "$TEST_IMG.out.dd"
-+    _rm_test_img "$TEST_IMG.out"
-+    _rm_test_img "$TEST_IMG.out.dd"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/161 b/tests/qemu-iotests/161
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/161
-+++ b/tests/qemu-iotests/161
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.base"
--    rm -f "$TEST_IMG.int"
-+    _rm_test_img "$TEST_IMG.base"
-+    _rm_test_img "$TEST_IMG.int"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/170 b/tests/qemu-iotests/170
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/170
-+++ b/tests/qemu-iotests/170
-@@ -XXX,XX +XXX,XX @@ status=1
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.out"
-+    _rm_test_img  "$TEST_IMG.out"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/172 b/tests/qemu-iotests/172
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/172
-+++ b/tests/qemu-iotests/172
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    _cleanup_test_img
--    rm -f "$TEST_IMG.2"
--    rm -f "$TEST_IMG.3"
-+    _cleanup_test_img
-+    _rm_test_img "$TEST_IMG.2"
-+    _rm_test_img "$TEST_IMG.3"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/173 b/tests/qemu-iotests/173
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/173
-+++ b/tests/qemu-iotests/173
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_qemu
--    rm  -f "${QEMU_TEST_DIR}/image.base" "${QEMU_TEST_DIR}/image.snp1"
-+    _rm_test_img "${TEST_DIR}/image.base"
-+    _rm_test_img "${TEST_DIR}/image.snp1"
-     _cleanup_test_img
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/178 b/tests/qemu-iotests/178
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/178
-+++ b/tests/qemu-iotests/178
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.converted"
-+    _rm_test_img "$TEST_IMG.converted"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/182 b/tests/qemu-iotests/182
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/182
-+++ b/tests/qemu-iotests/182
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.overlay"
-+    _rm_test_img "$TEST_IMG.overlay"
-     rm -f "$SOCK_DIR/nbd.socket"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/183 b/tests/qemu-iotests/183
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/183
-+++ b/tests/qemu-iotests/183
-@@ -XXX,XX +XXX,XX @@ MIG_SOCKET="${SOCK_DIR}/migrate"
- _cleanup()
- {
-     rm -f "${MIG_SOCKET}"
--    rm -f "${TEST_IMG}.dest"
-+    _rm_test_img "${TEST_IMG}.dest"
-     _cleanup_test_img
-     _cleanup_qemu
- }
-diff --git a/tests/qemu-iotests/185 b/tests/qemu-iotests/185
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/185
-+++ b/tests/qemu-iotests/185
-@@ -XXX,XX +XXX,XX @@ status=1 # failure is the default!
- _cleanup()
- {
--    rm -f "${TEST_IMG}.mid"
--    rm -f "${TEST_IMG}.copy"
-+    _rm_test_img "${TEST_IMG}.mid"
-+    _rm_test_img "${TEST_IMG}.copy"
-     _cleanup_test_img
-     _cleanup_qemu
- }
-diff --git a/tests/qemu-iotests/187 b/tests/qemu-iotests/187
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/187
-+++ b/tests/qemu-iotests/187
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
--    _cleanup_test_img
--    rm -f "$TEST_IMG.2"
--    rm -f "$TEST_IMG.3"
-+    _cleanup_test_img
-+    _rm_test_img "$TEST_IMG.2"
-+    _rm_test_img "$TEST_IMG.3"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/190 b/tests/qemu-iotests/190
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/190
-+++ b/tests/qemu-iotests/190
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.converted"
-+    _rm_test_img "$TEST_IMG.converted"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/191 b/tests/qemu-iotests/191
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/191
-+++ b/tests/qemu-iotests/191
-@@ -XXX,XX +XXX,XX @@ status=1 # failure is the default!
- _cleanup()
- {
--    rm -f "${TEST_IMG}.mid"
--    rm -f "${TEST_IMG}.ovl2"
--    rm -f "${TEST_IMG}.ovl3"
-+    _rm_test_img "${TEST_IMG}.mid"
-+    _rm_test_img "${TEST_IMG}.ovl2"
-+    _rm_test_img "${TEST_IMG}.ovl3"
-     _cleanup_test_img
-     _cleanup_qemu
- }
-diff --git a/tests/qemu-iotests/195 b/tests/qemu-iotests/195
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/195
-+++ b/tests/qemu-iotests/195
-@@ -XXX,XX +XXX,XX @@ status=1 # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.mid"
-+    _rm_test_img "$TEST_IMG.mid"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/197 b/tests/qemu-iotests/197
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/197
-+++ b/tests/qemu-iotests/197
-@@ -XXX,XX +XXX,XX @@ esac
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_WRAP"
-+    _rm_test_img "$TEST_WRAP"
-     rm -f "$BLKDBG_CONF"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/200 b/tests/qemu-iotests/200
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/200
-+++ b/tests/qemu-iotests/200
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_qemu
--    rm -f "${TEST_IMG}" "${BACKING_IMG}"
-+    _rm_test_img "${TEST_IMG}"
-+    _rm_test_img "${BACKING_IMG}"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/215 b/tests/qemu-iotests/215
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/215
-+++ b/tests/qemu-iotests/215
-@@ -XXX,XX +XXX,XX @@ esac
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_WRAP"
-+    _rm_test_img "$TEST_WRAP"
-     rm -f "$BLKDBG_CONF"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/225 b/tests/qemu-iotests/225
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/225
-+++ b/tests/qemu-iotests/225
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.not_base"
-+    _rm_test_img "$TEST_IMG.not_base"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/229 b/tests/qemu-iotests/229
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/229
-+++ b/tests/qemu-iotests/229
-@@ -XXX,XX +XXX,XX @@ _cleanup()
- {
-     _cleanup_qemu
-     _cleanup_test_img
--    rm -f "$TEST_IMG" "$DEST_IMG"
-+    _rm_test_img "$TEST_IMG"
-+    _rm_test_img "$DEST_IMG"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/232 b/tests/qemu-iotests/232
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/232
-+++ b/tests/qemu-iotests/232
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f $TEST_IMG.[01234]
-+    for img in "$TEST_IMG".[01234]; do
-+        _rm_test_img "$img"
-+    done
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/243 b/tests/qemu-iotests/243
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/243
-+++ b/tests/qemu-iotests/243
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f $TEST_IMG.data
-+    _rm_test_img "$TEST_IMG.data"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/244 b/tests/qemu-iotests/244
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/244
-+++ b/tests/qemu-iotests/244
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f $TEST_IMG.data
--    rm -f $TEST_IMG.src
-+    _rm_test_img "$TEST_IMG.data"
-+    _rm_test_img "$TEST_IMG.src"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/247 b/tests/qemu-iotests/247
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/247
-+++ b/tests/qemu-iotests/247
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f $TEST_IMG.[01234]
-+    for img in "$TEST_IMG".[01234]; do
-+        _rm_test_img "$img"
-+    done
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/249 b/tests/qemu-iotests/249
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/249
-+++ b/tests/qemu-iotests/249
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.base"
--    rm -f "$TEST_IMG.int"
-+    _rm_test_img "$TEST_IMG.base"
-+    _rm_test_img "$TEST_IMG.int"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
-diff --git a/tests/qemu-iotests/252 b/tests/qemu-iotests/252
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/252
-+++ b/tests/qemu-iotests/252
-@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
- _cleanup()
- {
-     _cleanup_test_img
--    rm -f "$TEST_IMG.base_new"
-+    _rm_test_img "$TEST_IMG.base_new"
- }
- trap "_cleanup; exit \$status" 0 1 2 3 15
 --
-.24.1
+.9.3

-[PULL 27/34] iotests: Make 198 work with data_file
+[Qemu-devel] [PULL v2 11/24] aio: push aio_context_acquire/release down to dispatching
-We do not care about the json:{} filenames here, so we can just filter
+From: Paolo Bonzini <pbonzini@redhat.com>
 them out and thus make the test work both with and without external data
 files.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+The AioContext data structures are now protected by list_lock and/or
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+they are walked with FOREACH_RCU primitives.  There is no need anymore
-Message-id: 20191107163708.833192-21-mreitz@redhat.com
+to acquire the AioContext for the entire duration of aio_dispatch.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Instead, just acquire it before and after invoking the callbacks.
 The next step is then to push it further down.
 Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-12-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/198     | 6 ++++--
+ util/aio-posix.c | 25 +++++++++++--------------
- tests/qemu-iotests/198.out | 4 ++--
+ util/aio-win32.c | 15 +++++++--------
-files changed, 6 insertions(+), 4 deletions(-)
+ util/async.c     |  2 ++
 files changed, 20 insertions(+), 22 deletions(-)
-diff --git a/tests/qemu-iotests/198 b/tests/qemu-iotests/198
+diff --git a/util/aio-posix.c b/util/aio-posix.c
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/198
 +++ b/tests/qemu-iotests/198
@@ -XXX,XX +XXX,XX @@ echo
  echo "== checking image base =="
  $QEMU_IMG info --image-opts $IMGSPECBASE | _filter_img_info --format-specific \
      | sed -e "/^disk size:/ D" -e '/refcount bits:/ D' -e '/compat:/ D' \
 -          -e '/lazy refcounts:/ D' -e '/corrupt:/ D'
 +          -e '/lazy refcounts:/ D' -e '/corrupt:/ D' -e '/^\s*data file/ D' \
 +    | _filter_json_filename
  echo
  echo "== checking image layer =="
  $QEMU_IMG info --image-opts $IMGSPECLAYER | _filter_img_info --format-specific \
      | sed -e "/^disk size:/ D" -e '/refcount bits:/ D' -e '/compat:/ D' \
 -          -e '/lazy refcounts:/ D' -e '/corrupt:/ D'
 +          -e '/lazy refcounts:/ D' -e '/corrupt:/ D' -e '/^\s*data file/ D' \
 +    | _filter_json_filename
  # success, all done
 diff --git a/tests/qemu-iotests/198.out b/tests/qemu-iotests/198.out
 index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/198.out
+--- a/util/aio-posix.c
-+++ b/tests/qemu-iotests/198.out
++++ b/util/aio-posix.c
-@@ -XXX,XX +XXX,XX @@ read 16777216/16777216 bytes at offset 0
+@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
-MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+             (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR)) &&
+             aio_node_check(ctx, node->is_external) &&
- == checking image base ==
+             node->io_read) {
--image: json:{"encrypt.key-secret": "sec0", "driver": "IMGFMT", "file": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT.base"}}
++            aio_context_acquire(ctx);
-+image: json:{ /* filtered */ }
+             node->io_read(node->opaque);
- file format: IMGFMT
++            aio_context_release(ctx);
- virtual size: 16 MiB (16777216 bytes)
- Format specific information:
+             /* aio_notify() does not count as progress */
-@@ -XXX,XX +XXX,XX @@ Format specific information:
+             if (node->opaque != &ctx->notifier) {
-         master key iters: 1024
+@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
+             (revents & (G_IO_OUT | G_IO_ERR)) &&
- == checking image layer ==
+             aio_node_check(ctx, node->is_external) &&
--image: json:{"encrypt.key-secret": "sec1", "driver": "IMGFMT", "file": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}}
+             node->io_write) {
-+image: json:{ /* filtered */ }
++            aio_context_acquire(ctx);
- file format: IMGFMT
+             node->io_write(node->opaque);
- virtual size: 16 MiB (16777216 bytes)
++            aio_context_release(ctx);
- backing file: TEST_DIR/t.IMGFMT.base
+             progress = true;
          }
@@ -XXX,XX +XXX,XX @@ bool aio_dispatch(AioContext *ctx, bool dispatch_fds)
      }
      /* Run our timers */
 +    aio_context_acquire(ctx);
      progress |= timerlistgroup_run_timers(&ctx->tlg);
 +    aio_context_release(ctx);
      return progress;
  }
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
      int64_t timeout;
      int64_t start = 0;
 -    aio_context_acquire(ctx);
 -    progress = false;
 -
      /* aio_notify can avoid the expensive event_notifier_set if
       * everything (file descriptors, bottom halves, timers) will
       * be re-evaluated before the next blocking poll().  This is
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
          start = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
      }
 -    if (try_poll_mode(ctx, blocking)) {
 -        progress = true;
 -    } else {
 +    aio_context_acquire(ctx);
 +    progress = try_poll_mode(ctx, blocking);
 +    aio_context_release(ctx);
 +
 +    if (!progress) {
          assert(npfd == 0);
          /* fill pollfds */
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
          timeout = blocking ? aio_compute_timeout(ctx) : 0;
          /* wait until next event */
 -        if (timeout) {
 -            aio_context_release(ctx);
 -        }
          if (aio_epoll_check_poll(ctx, pollfds, npfd, timeout)) {
              AioHandler epoll_handler;
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
          } else  {
              ret = qemu_poll_ns(pollfds, npfd, timeout);
          }
 -        if (timeout) {
 -            aio_context_acquire(ctx);
 -        }
      }
      if (blocking) {
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
          progress = true;
      }
 -    aio_context_release(ctx);
 -
      return progress;
  }
 diff --git a/util/aio-win32.c b/util/aio-win32.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/aio-win32.c
 +++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
              (revents || event_notifier_get_handle(node->e) == event) &&
              node->io_notify) {
              node->pfd.revents = 0;
 +            aio_context_acquire(ctx);
              node->io_notify(node->e);
 +            aio_context_release(ctx);
              /* aio_notify() does not count as progress */
              if (node->e != &ctx->notifier) {
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
              (node->io_read || node->io_write)) {
              node->pfd.revents = 0;
              if ((revents & G_IO_IN) && node->io_read) {
 +                aio_context_acquire(ctx);
                  node->io_read(node->opaque);
 +                aio_context_release(ctx);
                  progress = true;
              }
              if ((revents & G_IO_OUT) && node->io_write) {
 +                aio_context_acquire(ctx);
                  node->io_write(node->opaque);
 +                aio_context_release(ctx);
                  progress = true;
              }
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
      int count;
      int timeout;
 -    aio_context_acquire(ctx);
      progress = false;
      /* aio_notify can avoid the expensive event_notifier_set if
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
          timeout = blocking && !have_select_revents
              ? qemu_timeout_ns_to_ms(aio_compute_timeout(ctx)) : 0;
 -        if (timeout) {
 -            aio_context_release(ctx);
 -        }
          ret = WaitForMultipleObjects(count, events, FALSE, timeout);
          if (blocking) {
              assert(first);
              atomic_sub(&ctx->notify_me, 2);
          }
 -        if (timeout) {
 -            aio_context_acquire(ctx);
 -        }
          if (first) {
              aio_notify_accept(ctx);
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
          progress |= aio_dispatch_handlers(ctx, event);
      } while (count > 0);
 +    aio_context_acquire(ctx);
      progress |= timerlistgroup_run_timers(&ctx->tlg);
 -
      aio_context_release(ctx);
      return progress;
  }
 diff --git a/util/async.c b/util/async.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/async.c
 +++ b/util/async.c
@@ -XXX,XX +XXX,XX @@ int aio_bh_poll(AioContext *ctx)
                  ret = 1;
              }
              bh->idle = 0;
 +            aio_context_acquire(ctx);
              aio_bh_call(bh);
 +            aio_context_release(ctx);
          }
          if (bh->deleted) {
              deleted = true;
 --
-.24.1
+.9.3

-[PULL 06/34] throttle-groups: fix memory leak in throttle_group_set_limit:
+[Qemu-devel] [PULL v2 12/24] block: explicitly acquire aiocontext in timers that need it
-From: PanNengyuan <pannengyuan@huawei.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-This avoid a memory leak when qom-set is called to set throttle_group
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-limits, here is an easy way to reproduce:
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+Reviewed-by: Fam Zheng <famz@redhat.com>
-. run qemu-iotests as follow and check the result with asan:
+Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
-       ./check -qcow2 184
+Message-id: 20170213135235.12274-13-pbonzini@redhat.com
+Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 Following is the asan output backtrack:
 Direct leak of 912 byte(s) in 3 object(s) allocated from:
     #0 0xffff8d7ab3c3 in __interceptor_calloc   (/lib64/libasan.so.4+0xd33c3)
     #1 0xffff8d4c31cb in g_malloc0   (/lib64/libglib-2.0.so.0+0x571cb)
     #2 0x190c857 in qobject_input_start_struct  /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qapi/qobject-input-visitor.c:295
     #3 0x19070df in visit_start_struct   /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qapi/qapi-visit-core.c:49
     #4 0x1948b87 in visit_type_ThrottleLimits   qapi/qapi-visit-block-core.c:3759
     #5 0x17e4aa3 in throttle_group_set_limits   /mnt/sdc/qemu-master/qemu-4.2.0-rc0/block/throttle-groups.c:900
     #6 0x1650eff in object_property_set     /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qom/object.c:1272
     #7 0x1658517 in object_property_set_qobject   /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qom/qom-qobject.c:26
     #8 0x15880bb in qmp_qom_set  /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qom/qom-qmp-cmds.c:74
     #9 0x157e3e3 in qmp_marshal_qom_set  qapi/qapi-commands-qom.c:154
 Reported-by: Euler Robot <euler.robot@huawei.com>
 Signed-off-by: PanNengyuan <pannengyuan@huawei.com>
 Message-id: 1574835614-42028-1-git-send-email-pannengyuan@huawei.com
 Signed-off-by: Max Reitz <mreitz@redhat.com>
 ---
- block/throttle-groups.c | 4 ++--
+ block/qed.h                 |  3 +++
-file changed, 2 insertions(+), 2 deletions(-)
+ block/curl.c                |  2 ++
+ block/io.c                  |  5 +++++
  block/iscsi.c               |  8 ++++++--
  block/null.c                |  4 ++++
  block/qed.c                 | 12 ++++++++++++
  block/throttle-groups.c     |  2 ++
  util/aio-posix.c            |  2 --
  util/aio-win32.c            |  2 --
  util/qemu-coroutine-sleep.c |  2 +-
 files changed, 35 insertions(+), 7 deletions(-)
 diff --git a/block/qed.h b/block/qed.h
 index XXXXXXX..XXXXXXX 100644
 --- a/block/qed.h
 +++ b/block/qed.h
@@ -XXX,XX +XXX,XX @@ enum {
   */
  typedef void QEDFindClusterFunc(void *opaque, int ret, uint64_t offset, size_t len);
 +void qed_acquire(BDRVQEDState *s);
 +void qed_release(BDRVQEDState *s);
 +
  /**
   * Generic callback for chaining async callbacks
   */
 diff --git a/block/curl.c b/block/curl.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/curl.c
 +++ b/block/curl.c
@@ -XXX,XX +XXX,XX @@ static void curl_multi_timeout_do(void *arg)
          return;
      }
 +    aio_context_acquire(s->aio_context);
      curl_multi_socket_action(s->multi, CURL_SOCKET_TIMEOUT, 0, &running);
      curl_multi_check_completion(s);
 +    aio_context_release(s->aio_context);
  #else
      abort();
  #endif
 diff --git a/block/io.c b/block/io.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/io.c
 +++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ void bdrv_aio_cancel(BlockAIOCB *acb)
          if (acb->aiocb_info->get_aio_context) {
              aio_poll(acb->aiocb_info->get_aio_context(acb), true);
          } else if (acb->bs) {
 +            /* qemu_aio_ref and qemu_aio_unref are not thread-safe, so
 +             * assert that we're not using an I/O thread.  Thread-safe
 +             * code should use bdrv_aio_cancel_async exclusively.
 +             */
 +            assert(bdrv_get_aio_context(acb->bs) == qemu_get_aio_context());
              aio_poll(bdrv_get_aio_context(acb->bs), true);
          } else {
              abort();
 diff --git a/block/iscsi.c b/block/iscsi.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/iscsi.c
 +++ b/block/iscsi.c
@@ -XXX,XX +XXX,XX @@ static void iscsi_retry_timer_expired(void *opaque)
      struct IscsiTask *iTask = opaque;
      iTask->complete = 1;
      if (iTask->co) {
 -        qemu_coroutine_enter(iTask->co);
 +        aio_co_wake(iTask->co);
      }
  }
@@ -XXX,XX +XXX,XX @@ static void iscsi_nop_timed_event(void *opaque)
  {
      IscsiLun *iscsilun = opaque;
 +    aio_context_acquire(iscsilun->aio_context);
      if (iscsi_get_nops_in_flight(iscsilun->iscsi) >= MAX_NOP_FAILURES) {
          error_report("iSCSI: NOP timeout. Reconnecting...");
          iscsilun->request_timed_out = true;
      } else if (iscsi_nop_out_async(iscsilun->iscsi, NULL, NULL, 0, NULL) != 0) {
          error_report("iSCSI: failed to sent NOP-Out. Disabling NOP messages.");
 -        return;
 +        goto out;
      }
      timer_mod(iscsilun->nop_timer, qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + NOP_INTERVAL);
      iscsi_set_events(iscsilun);
 +
 +out:
 +    aio_context_release(iscsilun->aio_context);
  }
  static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp)
 diff --git a/block/null.c b/block/null.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/null.c
 +++ b/block/null.c
@@ -XXX,XX +XXX,XX @@ static void null_bh_cb(void *opaque)
  static void null_timer_cb(void *opaque)
  {
      NullAIOCB *acb = opaque;
 +    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 +
 +    aio_context_acquire(ctx);
      acb->common.cb(acb->common.opaque, 0);
 +    aio_context_release(ctx);
      timer_deinit(&acb->timer);
      qemu_aio_unref(acb);
  }
 diff --git a/block/qed.c b/block/qed.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/qed.c
 +++ b/block/qed.c
@@ -XXX,XX +XXX,XX @@ static void qed_need_check_timer_cb(void *opaque)
      trace_qed_need_check_timer_cb(s);
 +    qed_acquire(s);
      qed_plug_allocating_write_reqs(s);
      /* Ensure writes are on disk before clearing flag */
      bdrv_aio_flush(s->bs->file->bs, qed_clear_need_check, s);
 +    qed_release(s);
 +}
 +
 +void qed_acquire(BDRVQEDState *s)
 +{
 +    aio_context_acquire(bdrv_get_aio_context(s->bs));
 +}
 +
 +void qed_release(BDRVQEDState *s)
 +{
 +    aio_context_release(bdrv_get_aio_context(s->bs));
  }
  static void qed_start_need_check_timer(BDRVQEDState *s)
 diff --git a/block/throttle-groups.c b/block/throttle-groups.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/throttle-groups.c
 +++ b/block/throttle-groups.c
-@@ -XXX,XX +XXX,XX @@ static void throttle_group_set_limits(Object *obj, Visitor *v,
+@@ -XXX,XX +XXX,XX @@ static void timer_cb(BlockBackend *blk, bool is_write)
      qemu_mutex_unlock(&tg->lock);
      /* Run the request that was waiting for this timer */
 +    aio_context_acquire(blk_get_aio_context(blk));
      empty_queue = !qemu_co_enter_next(&blkp->throttled_reqs[is_write]);
 +    aio_context_release(blk_get_aio_context(blk));
      /* If the request queue was empty then we have to take care of
       * scheduling the next one */
 diff --git a/util/aio-posix.c b/util/aio-posix.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/aio-posix.c
 +++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@ bool aio_dispatch(AioContext *ctx, bool dispatch_fds)
      }
      /* Run our timers */
 -    aio_context_acquire(ctx);
      progress |= timerlistgroup_run_timers(&ctx->tlg);
 -    aio_context_release(ctx);
      return progress;
  }
 diff --git a/util/aio-win32.c b/util/aio-win32.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/aio-win32.c
 +++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
          progress |= aio_dispatch_handlers(ctx, event);
      } while (count > 0);
 -    aio_context_acquire(ctx);
      progress |= timerlistgroup_run_timers(&ctx->tlg);
 -    aio_context_release(ctx);
      return progress;
  }
 diff --git a/util/qemu-coroutine-sleep.c b/util/qemu-coroutine-sleep.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/qemu-coroutine-sleep.c
 +++ b/util/qemu-coroutine-sleep.c
@@ -XXX,XX +XXX,XX @@ static void co_sleep_cb(void *opaque)
  {
-     ThrottleGroup *tg = THROTTLE_GROUP(obj);
+     CoSleepCB *sleep_cb = opaque;
-     ThrottleConfig cfg;
--    ThrottleLimits arg = { 0 };
+-    qemu_coroutine_enter(sleep_cb->co);
--    ThrottleLimits *argp = &arg;
++    aio_co_wake(sleep_cb->co);
-+    ThrottleLimits *argp;
+ }
-     Error *local_err = NULL;
+ void coroutine_fn co_aio_sleep_ns(AioContext *ctx, QEMUClockType type,
      visit_type_ThrottleLimits(v, name, &argp, &local_err);
@@ -XXX,XX +XXX,XX @@ static void throttle_group_set_limits(Object *obj, Visitor *v,
  unlock:
      qemu_mutex_unlock(&tg->lock);
  ret:
 +    qapi_free_ThrottleLimits(argp);
      error_propagate(errp, local_err);
      return;
  }
 --
-.24.1
+.9.3

-[PULL 01/34] block: Add bdrv_qapi_perm_to_blk_perm()
+[Qemu-devel] [PULL v2 13/24] block: explicitly acquire aiocontext in callbacks that need it
-We need some way to correlate QAPI BlockPermission values with
+From: Paolo Bonzini <pbonzini@redhat.com>
 BLK_PERM_* flags.  We could:
-(1) have the same order in the QAPI definition as the the BLK_PERM_*
+This covers both file descriptor callbacks and polling callbacks,
-    flags are in LSb-first order.  However, then there is no guarantee
+since they execute related code.
     that they actually match (e.g. when someone modifies the QAPI schema
     without thinking of the BLK_PERM_* definitions).
     We could add static assertions, but these would break what’s good
     about this solution, namely its simplicity.
-(2) define the BLK_PERM_* flags based on the BlockPermission values.
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-    But this way whenever someone were to modify the QAPI order
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-    (perfectly sensible in theory), the BLK_PERM_* values would change.
+Reviewed-by: Fam Zheng <famz@redhat.com>
-    Because these values are used for file locking, this might break
+Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
-    file locking between different qemu versions.
+Message-id: 20170213135235.12274-14-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
  block/curl.c          | 16 +++++++++++++---
  block/iscsi.c         |  4 ++++
  block/linux-aio.c     |  4 ++++
  block/nfs.c           |  6 ++++++
  block/sheepdog.c      | 29 +++++++++++++++--------------
  block/ssh.c           | 29 +++++++++--------------------
  block/win32-aio.c     | 10 ++++++----
  hw/block/virtio-blk.c |  5 ++++-
  hw/scsi/virtio-scsi.c |  7 +++++++
  util/aio-posix.c      |  7 -------
  util/aio-win32.c      |  6 ------
 files changed, 68 insertions(+), 55 deletions(-)
-Therefore, go the slightly more cumbersome way: Add a function to
+diff --git a/block/curl.c b/block/curl.c
-translate from the QAPI constants to the BLK_PERM_* flags.
+index XXXXXXX..XXXXXXX 100644
+--- a/block/curl.c
-Signed-off-by: Max Reitz <mreitz@redhat.com>
++++ b/block/curl.c
-Message-id: 20191108123455.39445-2-mreitz@redhat.com
+@@ -XXX,XX +XXX,XX @@ static void curl_multi_check_completion(BDRVCURLState *s)
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
+     }
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+ }
----
- block.c               | 18 ++++++++++++++++++
+-static void curl_multi_do(void *arg)
- include/block/block.h |  1 +
++static void curl_multi_do_locked(CURLState *s)
-files changed, 19 insertions(+)
+ {
+-    CURLState *s = (CURLState *)arg;
-diff --git a/block.c b/block.c
+     CURLSocket *socket, *next_socket;
-index XXXXXXX..XXXXXXX 100644
+     int running;
---- a/block.c
+     int r;
-+++ b/block.c
+@@ -XXX,XX +XXX,XX @@ static void curl_multi_do(void *arg)
-@@ -XXX,XX +XXX,XX @@ void bdrv_format_default_perms(BlockDriverState *bs, BdrvChild *c,
+     }
-     *nshared = shared;
+ }
- }
++static void curl_multi_do(void *arg)
 +uint64_t bdrv_qapi_perm_to_blk_perm(BlockPermission qapi_perm)
 +{
-+    static const uint64_t permissions[] = {
++    CURLState *s = (CURLState *)arg;
-+        [BLOCK_PERMISSION_CONSISTENT_READ]  = BLK_PERM_CONSISTENT_READ,
++
-+        [BLOCK_PERMISSION_WRITE]            = BLK_PERM_WRITE,
++    aio_context_acquire(s->s->aio_context);
-+        [BLOCK_PERMISSION_WRITE_UNCHANGED]  = BLK_PERM_WRITE_UNCHANGED,
++    curl_multi_do_locked(s);
-+        [BLOCK_PERMISSION_RESIZE]           = BLK_PERM_RESIZE,
++    aio_context_release(s->s->aio_context);
 +        [BLOCK_PERMISSION_GRAPH_MOD]        = BLK_PERM_GRAPH_MOD,
 +    };
 +
 +    QEMU_BUILD_BUG_ON(ARRAY_SIZE(permissions) != BLOCK_PERMISSION__MAX);
 +    QEMU_BUILD_BUG_ON(1UL << ARRAY_SIZE(permissions) != BLK_PERM_ALL + 1);
 +
 +    assert(qapi_perm < BLOCK_PERMISSION__MAX);
 +
 +    return permissions[qapi_perm];
 +}
 +
- static void bdrv_replace_child_noperm(BdrvChild *child,
+ static void curl_multi_read(void *arg)
-                                       BlockDriverState *new_bs)
+ {
- {
+     CURLState *s = (CURLState *)arg;
-diff --git a/include/block/block.h b/include/block/block.h
-index XXXXXXX..XXXXXXX 100644
+-    curl_multi_do(arg);
---- a/include/block/block.h
++    aio_context_acquire(s->s->aio_context);
-+++ b/include/block/block.h
++    curl_multi_do_locked(s);
-@@ -XXX,XX +XXX,XX @@ enum {
+     curl_multi_check_completion(s->s);
 +    aio_context_release(s->s->aio_context);
  }
  static void curl_multi_timeout_do(void *arg)
 diff --git a/block/iscsi.c b/block/iscsi.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/iscsi.c
 +++ b/block/iscsi.c
@@ -XXX,XX +XXX,XX @@ iscsi_process_read(void *arg)
      IscsiLun *iscsilun = arg;
      struct iscsi_context *iscsi = iscsilun->iscsi;
 +    aio_context_acquire(iscsilun->aio_context);
      iscsi_service(iscsi, POLLIN);
      iscsi_set_events(iscsilun);
 +    aio_context_release(iscsilun->aio_context);
  }
  static void
@@ -XXX,XX +XXX,XX @@ iscsi_process_write(void *arg)
      IscsiLun *iscsilun = arg;
      struct iscsi_context *iscsi = iscsilun->iscsi;
 +    aio_context_acquire(iscsilun->aio_context);
      iscsi_service(iscsi, POLLOUT);
      iscsi_set_events(iscsilun);
 +    aio_context_release(iscsilun->aio_context);
  }
  static int64_t sector_lun2qemu(int64_t sector, IscsiLun *iscsilun)
 diff --git a/block/linux-aio.c b/block/linux-aio.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/linux-aio.c
 +++ b/block/linux-aio.c
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_completion_cb(EventNotifier *e)
      LinuxAioState *s = container_of(e, LinuxAioState, e);
      if (event_notifier_test_and_clear(&s->e)) {
 +        aio_context_acquire(s->aio_context);
          qemu_laio_process_completions_and_submit(s);
 +        aio_context_release(s->aio_context);
      }
  }
@@ -XXX,XX +XXX,XX @@ static bool qemu_laio_poll_cb(void *opaque)
          return false;
      }
 +    aio_context_acquire(s->aio_context);
      qemu_laio_process_completions_and_submit(s);
 +    aio_context_release(s->aio_context);
      return true;
  }
 diff --git a/block/nfs.c b/block/nfs.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/nfs.c
 +++ b/block/nfs.c
@@ -XXX,XX +XXX,XX @@ static void nfs_set_events(NFSClient *client)
  static void nfs_process_read(void *arg)
  {
      NFSClient *client = arg;
 +
 +    aio_context_acquire(client->aio_context);
      nfs_service(client->context, POLLIN);
      nfs_set_events(client);
 +    aio_context_release(client->aio_context);
  }
  static void nfs_process_write(void *arg)
  {
      NFSClient *client = arg;
 +
 +    aio_context_acquire(client->aio_context);
      nfs_service(client->context, POLLOUT);
      nfs_set_events(client);
 +    aio_context_release(client->aio_context);
  }
  static void nfs_co_init_task(BlockDriverState *bs, NFSRPC *task)
 diff --git a/block/sheepdog.c b/block/sheepdog.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/sheepdog.c
 +++ b/block/sheepdog.c
@@ -XXX,XX +XXX,XX @@ static coroutine_fn int send_co_req(int sockfd, SheepdogReq *hdr, void *data,
      return ret;
  }
 -static void restart_co_req(void *opaque)
 -{
 -    Coroutine *co = opaque;
 -
 -    qemu_coroutine_enter(co);
 -}
 -
  typedef struct SheepdogReqCo {
      int sockfd;
      BlockDriverState *bs;
@@ -XXX,XX +XXX,XX @@ typedef struct SheepdogReqCo {
      unsigned int *rlen;
      int ret;
      bool finished;
 +    Coroutine *co;
  } SheepdogReqCo;
 +static void restart_co_req(void *opaque)
 +{
 +    SheepdogReqCo *srco = opaque;
 +
 +    aio_co_wake(srco->co);
 +}
 +
  static coroutine_fn void do_co_req(void *opaque)
  {
      int ret;
 -    Coroutine *co;
      SheepdogReqCo *srco = opaque;
      int sockfd = srco->sockfd;
      SheepdogReq *hdr = srco->hdr;
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void do_co_req(void *opaque)
      unsigned int *wlen = srco->wlen;
      unsigned int *rlen = srco->rlen;
 -    co = qemu_coroutine_self();
 +    srco->co = qemu_coroutine_self();
      aio_set_fd_handler(srco->aio_context, sockfd, false,
 -                       NULL, restart_co_req, NULL, co);
 +                       NULL, restart_co_req, NULL, srco);
      ret = send_co_req(sockfd, hdr, data, wlen);
      if (ret < 0) {
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void do_co_req(void *opaque)
      }
      aio_set_fd_handler(srco->aio_context, sockfd, false,
 -                       restart_co_req, NULL, NULL, co);
 +                       restart_co_req, NULL, NULL, srco);
      ret = qemu_co_recv(sockfd, hdr, sizeof(*hdr));
      if (ret != sizeof(*hdr)) {
@@ -XXX,XX +XXX,XX @@ out:
      aio_set_fd_handler(srco->aio_context, sockfd, false,
                         NULL, NULL, NULL, NULL);
 +    srco->co = NULL;
      srco->ret = ret;
      srco->finished = true;
      if (srco->bs) {
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn aio_read_response(void *opaque)
           * We've finished all requests which belong to the AIOCB, so
           * we can switch back to sd_co_readv/writev now.
           */
 -        qemu_coroutine_enter(acb->coroutine);
 +        aio_co_wake(acb->coroutine);
      }
      return;
@@ -XXX,XX +XXX,XX @@ static void co_read_response(void *opaque)
          s->co_recv = qemu_coroutine_create(aio_read_response, opaque);
      }
 -    qemu_coroutine_enter(s->co_recv);
 +    aio_co_wake(s->co_recv);
  }
  static void co_write_request(void *opaque)
  {
      BDRVSheepdogState *s = opaque;
 -    qemu_coroutine_enter(s->co_send);
 +    aio_co_wake(s->co_send);
  }
  /*
 diff --git a/block/ssh.c b/block/ssh.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/ssh.c
 +++ b/block/ssh.c
@@ -XXX,XX +XXX,XX @@ static void restart_coroutine(void *opaque)
      DPRINTF("co=%p", co);
 -    qemu_coroutine_enter(co);
 +    aio_co_wake(co);
  }
 -static coroutine_fn void set_fd_handler(BDRVSSHState *s, BlockDriverState *bs)
 +/* A non-blocking call returned EAGAIN, so yield, ensuring the
 + * handlers are set up so that we'll be rescheduled when there is an
 + * interesting event on the socket.
 + */
 +static coroutine_fn void co_yield(BDRVSSHState *s, BlockDriverState *bs)
  {
      int r;
      IOHandler *rd_handler = NULL, *wr_handler = NULL;
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void set_fd_handler(BDRVSSHState *s, BlockDriverState *bs)
      aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock,
                         false, rd_handler, wr_handler, NULL, co);
 -}
 -
 -static coroutine_fn void clear_fd_handler(BDRVSSHState *s,
 -                                          BlockDriverState *bs)
 -{
 -    DPRINTF("s->sock=%d", s->sock);
 -    aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock,
 -                       false, NULL, NULL, NULL, NULL);
 -}
 -
 -/* A non-blocking call returned EAGAIN, so yield, ensuring the
 - * handlers are set up so that we'll be rescheduled when there is an
 - * interesting event on the socket.
 - */
 -static coroutine_fn void co_yield(BDRVSSHState *s, BlockDriverState *bs)
 -{
 -    set_fd_handler(s, bs);
      qemu_coroutine_yield();
 -    clear_fd_handler(s, bs);
 +    DPRINTF("s->sock=%d - back", s->sock);
 +    aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock, false,
 +                       NULL, NULL, NULL, NULL);
  }
  /* SFTP has a function `libssh2_sftp_seek64' which seeks to a position
 diff --git a/block/win32-aio.c b/block/win32-aio.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/win32-aio.c
 +++ b/block/win32-aio.c
@@ -XXX,XX +XXX,XX @@ struct QEMUWin32AIOState {
      HANDLE hIOCP;
      EventNotifier e;
      int count;
 -    bool is_aio_context_attached;
 +    AioContext *aio_ctx;
  };
- char *bdrv_perm_names(uint64_t perm);
+ typedef struct QEMUWin32AIOCB {
-+uint64_t bdrv_qapi_perm_to_blk_perm(BlockPermission qapi_perm);
+@@ -XXX,XX +XXX,XX @@ static void win32_aio_process_completion(QEMUWin32AIOState *s,
+     }
- /* disk I/O throttling */
- void bdrv_init(void);
 +    aio_context_acquire(s->aio_ctx);
      waiocb->common.cb(waiocb->common.opaque, ret);
 +    aio_context_release(s->aio_ctx);
      qemu_aio_unref(waiocb);
  }
@@ -XXX,XX +XXX,XX @@ void win32_aio_detach_aio_context(QEMUWin32AIOState *aio,
                                    AioContext *old_context)
  {
      aio_set_event_notifier(old_context, &aio->e, false, NULL, NULL);
 -    aio->is_aio_context_attached = false;
 +    aio->aio_ctx = NULL;
  }
  void win32_aio_attach_aio_context(QEMUWin32AIOState *aio,
                                    AioContext *new_context)
  {
 -    aio->is_aio_context_attached = true;
 +    aio->aio_ctx = new_context;
      aio_set_event_notifier(new_context, &aio->e, false,
                             win32_aio_completion_cb, NULL);
  }
@@ -XXX,XX +XXX,XX @@ out_free_state:
  void win32_aio_cleanup(QEMUWin32AIOState *aio)
  {
 -    assert(!aio->is_aio_context_attached);
 +    assert(!aio->aio_ctx);
      CloseHandle(aio->hIOCP);
      event_notifier_cleanup(&aio->e);
      g_free(aio);
 diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/block/virtio-blk.c
 +++ b/hw/block/virtio-blk.c
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_ioctl_complete(void *opaque, int status)
  {
      VirtIOBlockIoctlReq *ioctl_req = opaque;
      VirtIOBlockReq *req = ioctl_req->req;
 -    VirtIODevice *vdev = VIRTIO_DEVICE(req->dev);
 +    VirtIOBlock *s = req->dev;
 +    VirtIODevice *vdev = VIRTIO_DEVICE(s);
      struct virtio_scsi_inhdr *scsi;
      struct sg_io_hdr *hdr;
@@ -XXX,XX +XXX,XX @@ bool virtio_blk_handle_vq(VirtIOBlock *s, VirtQueue *vq)
      MultiReqBuffer mrb = {};
      bool progress = false;
 +    aio_context_acquire(blk_get_aio_context(s->blk));
      blk_io_plug(s->blk);
      do {
@@ -XXX,XX +XXX,XX @@ bool virtio_blk_handle_vq(VirtIOBlock *s, VirtQueue *vq)
      }
      blk_io_unplug(s->blk);
 +    aio_context_release(blk_get_aio_context(s->blk));
      return progress;
  }
 diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/scsi/virtio-scsi.c
 +++ b/hw/scsi/virtio-scsi.c
@@ -XXX,XX +XXX,XX @@ bool virtio_scsi_handle_ctrl_vq(VirtIOSCSI *s, VirtQueue *vq)
      VirtIOSCSIReq *req;
      bool progress = false;
 +    virtio_scsi_acquire(s);
      while ((req = virtio_scsi_pop_req(s, vq))) {
          progress = true;
          virtio_scsi_handle_ctrl_req(s, req);
      }
 +    virtio_scsi_release(s);
      return progress;
  }
@@ -XXX,XX +XXX,XX @@ bool virtio_scsi_handle_cmd_vq(VirtIOSCSI *s, VirtQueue *vq)
      QTAILQ_HEAD(, VirtIOSCSIReq) reqs = QTAILQ_HEAD_INITIALIZER(reqs);
 +    virtio_scsi_acquire(s);
      do {
          virtio_queue_set_notification(vq, 0);
@@ -XXX,XX +XXX,XX @@ bool virtio_scsi_handle_cmd_vq(VirtIOSCSI *s, VirtQueue *vq)
      QTAILQ_FOREACH_SAFE(req, &reqs, next, next) {
          virtio_scsi_handle_cmd_req_submit(s, req);
      }
 +    virtio_scsi_release(s);
      return progress;
  }
@@ -XXX,XX +XXX,XX @@ out:
  bool virtio_scsi_handle_event_vq(VirtIOSCSI *s, VirtQueue *vq)
  {
 +    virtio_scsi_acquire(s);
      if (s->events_dropped) {
          virtio_scsi_push_event(s, NULL, VIRTIO_SCSI_T_NO_EVENT, 0);
 +        virtio_scsi_release(s);
          return true;
      }
 +    virtio_scsi_release(s);
      return false;
  }
 diff --git a/util/aio-posix.c b/util/aio-posix.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/aio-posix.c
 +++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
              (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR)) &&
              aio_node_check(ctx, node->is_external) &&
              node->io_read) {
 -            aio_context_acquire(ctx);
              node->io_read(node->opaque);
 -            aio_context_release(ctx);
              /* aio_notify() does not count as progress */
              if (node->opaque != &ctx->notifier) {
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
              (revents & (G_IO_OUT | G_IO_ERR)) &&
              aio_node_check(ctx, node->is_external) &&
              node->io_write) {
 -            aio_context_acquire(ctx);
              node->io_write(node->opaque);
 -            aio_context_release(ctx);
              progress = true;
          }
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
          start = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
      }
 -    aio_context_acquire(ctx);
      progress = try_poll_mode(ctx, blocking);
 -    aio_context_release(ctx);
 -
      if (!progress) {
          assert(npfd == 0);
 diff --git a/util/aio-win32.c b/util/aio-win32.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/aio-win32.c
 +++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
              (revents || event_notifier_get_handle(node->e) == event) &&
              node->io_notify) {
              node->pfd.revents = 0;
 -            aio_context_acquire(ctx);
              node->io_notify(node->e);
 -            aio_context_release(ctx);
              /* aio_notify() does not count as progress */
              if (node->e != &ctx->notifier) {
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
              (node->io_read || node->io_write)) {
              node->pfd.revents = 0;
              if ((revents & G_IO_IN) && node->io_read) {
 -                aio_context_acquire(ctx);
                  node->io_read(node->opaque);
 -                aio_context_release(ctx);
                  progress = true;
              }
              if ((revents & G_IO_OUT) && node->io_write) {
 -                aio_context_acquire(ctx);
                  node->io_write(node->opaque);
 -                aio_context_release(ctx);
                  progress = true;
              }
 --
-.24.1
+.9.3

-[PULL 26/34] iotests: Make 137 work with data_file
+[Qemu-devel] [PULL v2 14/24] block: explicitly acquire aiocontext in bottom halves that need it
-When using an external data file, there are no refcounts for data
+From: Paolo Bonzini <pbonzini@redhat.com>
 clusters.  We thus have to adjust the corruption test in this patch to
 not be based around a data cluster allocation, but the L2 table
 allocation (L2 tables are still refcounted with external data files).
-Furthermore, we should not print qcow2.py's list of incompatible
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-features because it differs depending on whether there is an external
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-data file or not.
+Reviewed-by: Fam Zheng <famz@redhat.com>
 Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-15-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
  block/archipelago.c   |  3 +++
  block/blkreplay.c     |  2 +-
  block/block-backend.c |  6 ++++++
  block/curl.c          | 26 ++++++++++++++++++--------
  block/gluster.c       |  9 +--------
  block/io.c            |  6 +++++-
  block/iscsi.c         |  6 +++++-
  block/linux-aio.c     | 15 +++++++++------
  block/nfs.c           |  3 ++-
  block/null.c          |  4 ++++
  block/qed.c           |  3 +++
  block/rbd.c           |  4 ++++
  dma-helpers.c         |  2 ++
  hw/block/virtio-blk.c |  2 ++
  hw/scsi/scsi-bus.c    |  2 ++
  util/async.c          |  4 ++--
  util/thread-pool.c    |  2 ++
 files changed, 71 insertions(+), 28 deletions(-)
-With those two changes, the test will work both with and without
+diff --git a/block/archipelago.c b/block/archipelago.c
-external data files (once that options works with the iotests at all).
+index XXXXXXX..XXXXXXX 100644
+--- a/block/archipelago.c
-Signed-off-by: Max Reitz <mreitz@redhat.com>
++++ b/block/archipelago.c
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+@@ -XXX,XX +XXX,XX @@ static void qemu_archipelago_complete_aio(void *opaque)
-Message-id: 20191107163708.833192-20-mreitz@redhat.com
+ {
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+     AIORequestData *reqdata = (AIORequestData *) opaque;
----
+     ArchipelagoAIOCB *aio_cb = (ArchipelagoAIOCB *) reqdata->aio_cb;
- tests/qemu-iotests/137     | 15 +++++++++++----
++    AioContext *ctx = bdrv_get_aio_context(aio_cb->common.bs);
- tests/qemu-iotests/137.out |  6 ++----
-files changed, 13 insertions(+), 8 deletions(-)
++    aio_context_acquire(ctx);
+     aio_cb->common.cb(aio_cb->common.opaque, aio_cb->ret);
-diff --git a/tests/qemu-iotests/137 b/tests/qemu-iotests/137
++    aio_context_release(ctx);
      aio_cb->status = 0;
      qemu_aio_unref(aio_cb);
 diff --git a/block/blkreplay.c b/block/blkreplay.c
 index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/137
+--- a/block/blkreplay.c
-+++ b/tests/qemu-iotests/137
++++ b/block/blkreplay.c
-@@ -XXX,XX +XXX,XX @@ $QEMU_IO \
+@@ -XXX,XX +XXX,XX @@ static int64_t blkreplay_getlength(BlockDriverState *bs)
-     "$TEST_IMG" 2>&1 | _filter_qemu_io
+ static void blkreplay_bh_cb(void *opaque)
+ {
- # The dirty bit must not be set
+     Request *req = opaque;
--$PYTHON qcow2.py "$TEST_IMG" dump-header | grep incompatible_features
+-    qemu_coroutine_enter(req->co);
-+# (Filter the external data file bit)
++    aio_co_wake(req->co);
-+if $PYTHON qcow2.py "$TEST_IMG" dump-header | grep incompatible_features \
+     qemu_bh_delete(req->bh);
-+    | grep -q '\<0\>'
+     g_free(req);
-+then
+ }
-+    echo 'ERROR: Dirty bit set'
+diff --git a/block/block-backend.c b/block/block-backend.c
-+else
+index XXXXXXX..XXXXXXX 100644
-+    echo 'OK: Dirty bit not set'
+--- a/block/block-backend.c
-+fi
++++ b/block/block-backend.c
+@@ -XXX,XX +XXX,XX @@ int blk_make_zero(BlockBackend *blk, BdrvRequestFlags flags)
- # Similarly we can test whether corruption detection has been enabled:
+ static void error_callback_bh(void *opaque)
--# Create L1/L2, overwrite first entry in refcount block, allocate something.
+ {
-+# Create L1, overwrite refcounts, force allocation of L2 by writing
+     struct BlockBackendAIOCB *acb = opaque;
-+# data.
++    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
- # Disabling the checks should fail, so the corruption must be detected.
- _make_test_img 64M
+     bdrv_dec_in_flight(acb->common.bs);
--$QEMU_IO -c "write 0 64k" "$TEST_IMG" | _filter_qemu_io
++    aio_context_acquire(ctx);
--poke_file "$TEST_IMG" "$((0x20000))" "\x00\x00"
+     acb->common.cb(acb->common.opaque, acb->ret);
-+poke_file "$TEST_IMG" "$((0x20000))" "\x00\x00\x00\x00\x00\x00\x00\x00"
++    aio_context_release(ctx);
- $QEMU_IO \
+     qemu_aio_unref(acb);
-     -c "reopen -o overlap-check=none,lazy-refcounts=42" \
+ }
-     -c "write 64k 64k" \
-diff --git a/tests/qemu-iotests/137.out b/tests/qemu-iotests/137.out
+@@ -XXX,XX +XXX,XX @@ static void blk_aio_complete(BlkAioEmAIOCB *acb)
-index XXXXXXX..XXXXXXX 100644
+ static void blk_aio_complete_bh(void *opaque)
---- a/tests/qemu-iotests/137.out
+ {
-+++ b/tests/qemu-iotests/137.out
+     BlkAioEmAIOCB *acb = opaque;
-@@ -XXX,XX +XXX,XX @@ qemu-io: Unsupported value 'blubb' for qcow2 option 'overlap-check'. Allowed are
++    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+     assert(acb->has_returned);
- ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
++    aio_context_acquire(ctx);
--incompatible_features     []
+     blk_aio_complete(acb);
-+OK: Dirty bit not set
++    aio_context_release(ctx);
- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
+ }
--wrote 65536/65536 bytes at offset 0
--64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+ static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes,
- qemu-io: Parameter 'lazy-refcounts' expects 'on' or 'off'
+diff --git a/block/curl.c b/block/curl.c
--qcow2: Marking image as corrupt: Preventing invalid write on metadata (overlaps with qcow2_header); further corruption events will be suppressed
+index XXXXXXX..XXXXXXX 100644
-+qcow2: Marking image as corrupt: Preventing invalid allocation of L2 table at offset 0; further corruption events will be suppressed
+--- a/block/curl.c
- write failed: Input/output error
++++ b/block/curl.c
- *** done
+@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
  {
      CURLState *state;
      int running;
 +    int ret = -EINPROGRESS;
      CURLAIOCB *acb = p;
 -    BDRVCURLState *s = acb->common.bs->opaque;
 +    BlockDriverState *bs = acb->common.bs;
 +    BDRVCURLState *s = bs->opaque;
 +    AioContext *ctx = bdrv_get_aio_context(bs);
      size_t start = acb->sector_num * BDRV_SECTOR_SIZE;
      size_t end;
 +    aio_context_acquire(ctx);
 +
      // In case we have the requested data already (e.g. read-ahead),
      // we can just call the callback and be done.
      switch (curl_find_buf(s, start, acb->nb_sectors * BDRV_SECTOR_SIZE, acb)) {
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
              qemu_aio_unref(acb);
              // fall through
          case FIND_RET_WAIT:
 -            return;
 +            goto out;
          default:
              break;
      }
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
      // No cache found, so let's start a new request
      state = curl_init_state(acb->common.bs, s);
      if (!state) {
 -        acb->common.cb(acb->common.opaque, -EIO);
 -        qemu_aio_unref(acb);
 -        return;
 +        ret = -EIO;
 +        goto out;
      }
      acb->start = 0;
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
      state->orig_buf = g_try_malloc(state->buf_len);
      if (state->buf_len && state->orig_buf == NULL) {
          curl_clean_state(state);
 -        acb->common.cb(acb->common.opaque, -ENOMEM);
 -        qemu_aio_unref(acb);
 -        return;
 +        ret = -ENOMEM;
 +        goto out;
      }
      state->acb[0] = acb;
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
      /* Tell curl it needs to kick things off */
      curl_multi_socket_action(s->multi, CURL_SOCKET_TIMEOUT, 0, &running);
 +
 +out:
 +    if (ret != -EINPROGRESS) {
 +        acb->common.cb(acb->common.opaque, ret);
 +        qemu_aio_unref(acb);
 +    }
 +    aio_context_release(ctx);
  }
  static BlockAIOCB *curl_aio_readv(BlockDriverState *bs,
 diff --git a/block/gluster.c b/block/gluster.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/gluster.c
 +++ b/block/gluster.c
@@ -XXX,XX +XXX,XX @@ static struct glfs *qemu_gluster_init(BlockdevOptionsGluster *gconf,
      return qemu_gluster_glfs_init(gconf, errp);
  }
 -static void qemu_gluster_complete_aio(void *opaque)
 -{
 -    GlusterAIOCB *acb = (GlusterAIOCB *)opaque;
 -
 -    qemu_coroutine_enter(acb->coroutine);
 -}
 -
  /*
   * AIO callback routine called from GlusterFS thread.
   */
@@ -XXX,XX +XXX,XX @@ static void gluster_finish_aiocb(struct glfs_fd *fd, ssize_t ret, void *arg)
          acb->ret = -EIO; /* Partial read/write - fail it */
      }
 -    aio_bh_schedule_oneshot(acb->aio_context, qemu_gluster_complete_aio, acb);
 +    aio_co_schedule(acb->aio_context, acb->coroutine);
  }
  static void qemu_gluster_parse_flags(int bdrv_flags, int *open_flags)
 diff --git a/block/io.c b/block/io.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/io.c
 +++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ static void bdrv_co_drain_bh_cb(void *opaque)
      bdrv_dec_in_flight(bs);
      bdrv_drained_begin(bs);
      data->done = true;
 -    qemu_coroutine_enter(co);
 +    aio_co_wake(co);
  }
  static void coroutine_fn bdrv_co_yield_to_drain(BlockDriverState *bs)
@@ -XXX,XX +XXX,XX @@ static void bdrv_co_complete(BlockAIOCBCoroutine *acb)
  static void bdrv_co_em_bh(void *opaque)
  {
      BlockAIOCBCoroutine *acb = opaque;
 +    BlockDriverState *bs = acb->common.bs;
 +    AioContext *ctx = bdrv_get_aio_context(bs);
      assert(!acb->need_bh);
 +    aio_context_acquire(ctx);
      bdrv_co_complete(acb);
 +    aio_context_release(ctx);
  }
  static void bdrv_co_maybe_schedule_bh(BlockAIOCBCoroutine *acb)
 diff --git a/block/iscsi.c b/block/iscsi.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/iscsi.c
 +++ b/block/iscsi.c
@@ -XXX,XX +XXX,XX @@ static void
  iscsi_bh_cb(void *p)
  {
      IscsiAIOCB *acb = p;
 +    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
      qemu_bh_delete(acb->bh);
      g_free(acb->buf);
      acb->buf = NULL;
 +    aio_context_acquire(ctx);
      acb->common.cb(acb->common.opaque, acb->status);
 +    aio_context_release(ctx);
      if (acb->task != NULL) {
          scsi_free_scsi_task(acb->task);
@@ -XXX,XX +XXX,XX @@ iscsi_schedule_bh(IscsiAIOCB *acb)
  static void iscsi_co_generic_bh_cb(void *opaque)
  {
      struct IscsiTask *iTask = opaque;
 +
      iTask->complete = 1;
 -    qemu_coroutine_enter(iTask->co);
 +    aio_co_wake(iTask->co);
  }
  static void iscsi_retry_timer_expired(void *opaque)
 diff --git a/block/linux-aio.c b/block/linux-aio.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/linux-aio.c
 +++ b/block/linux-aio.c
@@ -XXX,XX +XXX,XX @@ struct LinuxAioState {
      io_context_t ctx;
      EventNotifier e;
 -    /* io queue for submit at batch */
 +    /* io queue for submit at batch.  Protected by AioContext lock. */
      LaioQueue io_q;
 -    /* I/O completion processing */
 +    /* I/O completion processing.  Only runs in I/O thread.  */
      QEMUBH *completion_bh;
      int event_idx;
      int event_max;
@@ -XXX,XX +XXX,XX @@ static inline ssize_t io_event_ret(struct io_event *ev)
   */
  static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
  {
 +    LinuxAioState *s = laiocb->ctx;
      int ret;
      ret = laiocb->ret;
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
      }
      laiocb->ret = ret;
 +    aio_context_acquire(s->aio_context);
      if (laiocb->co) {
          /* If the coroutine is already entered it must be in ioq_submit() and
           * will notice laio->ret has been filled in when it eventually runs
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
          laiocb->common.cb(laiocb->common.opaque, ret);
          qemu_aio_unref(laiocb);
      }
 +    aio_context_release(s->aio_context);
  }
  /**
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completions(LinuxAioState *s)
  static void qemu_laio_process_completions_and_submit(LinuxAioState *s)
  {
      qemu_laio_process_completions(s);
 +
 +    aio_context_acquire(s->aio_context);
      if (!s->io_q.plugged && !QSIMPLEQ_EMPTY(&s->io_q.pending)) {
          ioq_submit(s);
      }
 +    aio_context_release(s->aio_context);
  }
  static void qemu_laio_completion_bh(void *opaque)
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_completion_cb(EventNotifier *e)
      LinuxAioState *s = container_of(e, LinuxAioState, e);
      if (event_notifier_test_and_clear(&s->e)) {
 -        aio_context_acquire(s->aio_context);
          qemu_laio_process_completions_and_submit(s);
 -        aio_context_release(s->aio_context);
      }
  }
@@ -XXX,XX +XXX,XX @@ static bool qemu_laio_poll_cb(void *opaque)
          return false;
      }
 -    aio_context_acquire(s->aio_context);
      qemu_laio_process_completions_and_submit(s);
 -    aio_context_release(s->aio_context);
      return true;
  }
@@ -XXX,XX +XXX,XX @@ void laio_detach_aio_context(LinuxAioState *s, AioContext *old_context)
  {
      aio_set_event_notifier(old_context, &s->e, false, NULL, NULL);
      qemu_bh_delete(s->completion_bh);
 +    s->aio_context = NULL;
  }
  void laio_attach_aio_context(LinuxAioState *s, AioContext *new_context)
 diff --git a/block/nfs.c b/block/nfs.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/nfs.c
 +++ b/block/nfs.c
@@ -XXX,XX +XXX,XX @@ static void nfs_co_init_task(BlockDriverState *bs, NFSRPC *task)
  static void nfs_co_generic_bh_cb(void *opaque)
  {
      NFSRPC *task = opaque;
 +
      task->complete = 1;
 -    qemu_coroutine_enter(task->co);
 +    aio_co_wake(task->co);
  }
  static void
 diff --git a/block/null.c b/block/null.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/null.c
 +++ b/block/null.c
@@ -XXX,XX +XXX,XX @@ static const AIOCBInfo null_aiocb_info = {
  static void null_bh_cb(void *opaque)
  {
      NullAIOCB *acb = opaque;
 +    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 +
 +    aio_context_acquire(ctx);
      acb->common.cb(acb->common.opaque, 0);
 +    aio_context_release(ctx);
      qemu_aio_unref(acb);
  }
 diff --git a/block/qed.c b/block/qed.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/qed.c
 +++ b/block/qed.c
@@ -XXX,XX +XXX,XX @@ static void qed_update_l2_table(BDRVQEDState *s, QEDTable *table, int index,
  static void qed_aio_complete_bh(void *opaque)
  {
      QEDAIOCB *acb = opaque;
 +    BDRVQEDState *s = acb_to_s(acb);
      BlockCompletionFunc *cb = acb->common.cb;
      void *user_opaque = acb->common.opaque;
      int ret = acb->bh_ret;
@@ -XXX,XX +XXX,XX @@ static void qed_aio_complete_bh(void *opaque)
      qemu_aio_unref(acb);
      /* Invoke callback */
 +    qed_acquire(s);
      cb(user_opaque, ret);
 +    qed_release(s);
  }
  static void qed_aio_complete(QEDAIOCB *acb, int ret)
 diff --git a/block/rbd.c b/block/rbd.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/rbd.c
 +++ b/block/rbd.c
@@ -XXX,XX +XXX,XX @@ shutdown:
  static void qemu_rbd_complete_aio(RADOSCB *rcb)
  {
      RBDAIOCB *acb = rcb->acb;
 +    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
      int64_t r;
      r = rcb->ret;
@@ -XXX,XX +XXX,XX @@ static void qemu_rbd_complete_aio(RADOSCB *rcb)
          qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
      }
      qemu_vfree(acb->bounce);
 +
 +    aio_context_acquire(ctx);
      acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
 +    aio_context_release(ctx);
      qemu_aio_unref(acb);
  }
 diff --git a/dma-helpers.c b/dma-helpers.c
 index XXXXXXX..XXXXXXX 100644
 --- a/dma-helpers.c
 +++ b/dma-helpers.c
@@ -XXX,XX +XXX,XX @@ static void dma_blk_cb(void *opaque, int ret)
                                  QEMU_ALIGN_DOWN(dbs->iov.size, dbs->align));
      }
 +    aio_context_acquire(dbs->ctx);
      dbs->acb = dbs->io_func(dbs->offset, &dbs->iov,
                              dma_blk_cb, dbs, dbs->io_func_opaque);
 +    aio_context_release(dbs->ctx);
      assert(dbs->acb);
  }
 diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/block/virtio-blk.c
 +++ b/hw/block/virtio-blk.c
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_dma_restart_bh(void *opaque)
      s->rq = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->conf.conf.blk));
      while (req) {
          VirtIOBlockReq *next = req->next;
          if (virtio_blk_handle_request(req, &mrb)) {
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_dma_restart_bh(void *opaque)
      if (mrb.num_reqs) {
          virtio_blk_submit_multireq(s->blk, &mrb);
      }
 +    aio_context_release(blk_get_aio_context(s->conf.conf.blk));
  }
  static void virtio_blk_dma_restart_cb(void *opaque, int running,
 diff --git a/hw/scsi/scsi-bus.c b/hw/scsi/scsi-bus.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/scsi/scsi-bus.c
 +++ b/hw/scsi/scsi-bus.c
@@ -XXX,XX +XXX,XX @@ static void scsi_dma_restart_bh(void *opaque)
      qemu_bh_delete(s->bh);
      s->bh = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->conf.blk));
      QTAILQ_FOREACH_SAFE(req, &s->requests, next, next) {
          scsi_req_ref(req);
          if (req->retry) {
@@ -XXX,XX +XXX,XX @@ static void scsi_dma_restart_bh(void *opaque)
          }
          scsi_req_unref(req);
      }
 +    aio_context_release(blk_get_aio_context(s->conf.blk));
  }
  void scsi_req_retry(SCSIRequest *req)
 diff --git a/util/async.c b/util/async.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/async.c
 +++ b/util/async.c
@@ -XXX,XX +XXX,XX @@ int aio_bh_poll(AioContext *ctx)
                  ret = 1;
              }
              bh->idle = 0;
 -            aio_context_acquire(ctx);
              aio_bh_call(bh);
 -            aio_context_release(ctx);
          }
          if (bh->deleted) {
              deleted = true;
@@ -XXX,XX +XXX,XX @@ static void co_schedule_bh_cb(void *opaque)
          Coroutine *co = QSLIST_FIRST(&straight);
          QSLIST_REMOVE_HEAD(&straight, co_scheduled_next);
          trace_aio_co_schedule_bh_cb(ctx, co);
 +        aio_context_acquire(ctx);
          qemu_coroutine_enter(co);
 +        aio_context_release(ctx);
      }
  }
 diff --git a/util/thread-pool.c b/util/thread-pool.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/thread-pool.c
 +++ b/util/thread-pool.c
@@ -XXX,XX +XXX,XX @@ static void thread_pool_completion_bh(void *opaque)
      ThreadPool *pool = opaque;
      ThreadPoolElement *elem, *next;
 +    aio_context_acquire(pool->ctx);
  restart:
      QLIST_FOREACH_SAFE(elem, &pool->head, all, next) {
          if (elem->state != THREAD_DONE) {
@@ -XXX,XX +XXX,XX @@ restart:
              qemu_aio_unref(elem);
          }
      }
 +    aio_context_release(pool->ctx);
  }
  static void thread_pool_cancel(BlockAIOCB *acb)
 --
-.24.1
+.9.3

-[PULL 25/34] iotests: Make 110 work with data_file
+[Qemu-devel] [PULL v2 15/24] block: explicitly acquire aiocontext in aio callbacks that need it
-The only difference is that the json:{} filename of the image looks
+From: Paolo Bonzini <pbonzini@redhat.com>
 different.  We actually do not care about that filename in this test, we
 are only interested in (1) that there is a json:{} filename, and (2)
 whether the backing filename can be constructed.
-So just filter out the json:{} data, thus making this test pass both
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-with and without data_file.
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-16-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
  block/archipelago.c    |  3 ---
  block/block-backend.c  |  7 -------
  block/curl.c           |  2 +-
  block/io.c             |  6 +-----
  block/iscsi.c          |  3 ---
  block/linux-aio.c      |  5 +----
  block/mirror.c         | 12 +++++++++---
  block/null.c           |  8 --------
  block/qed-cluster.c    |  2 ++
  block/qed-table.c      | 12 ++++++++++--
  block/qed.c            |  4 ++--
  block/rbd.c            |  4 ----
  block/win32-aio.c      |  3 ---
  hw/block/virtio-blk.c  | 12 +++++++++++-
  hw/scsi/scsi-disk.c    | 15 +++++++++++++++
  hw/scsi/scsi-generic.c | 20 +++++++++++++++++---
  util/thread-pool.c     |  4 +++-
 files changed, 72 insertions(+), 50 deletions(-)
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+diff --git a/block/archipelago.c b/block/archipelago.c
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+index XXXXXXX..XXXXXXX 100644
-Message-id: 20191107163708.833192-19-mreitz@redhat.com
+--- a/block/archipelago.c
-Signed-off-by: Max Reitz <mreitz@redhat.com>
++++ b/block/archipelago.c
----
+@@ -XXX,XX +XXX,XX @@ static void qemu_archipelago_complete_aio(void *opaque)
- tests/qemu-iotests/110     | 7 +++++--
+ {
- tests/qemu-iotests/110.out | 4 ++--
+     AIORequestData *reqdata = (AIORequestData *) opaque;
-files changed, 7 insertions(+), 4 deletions(-)
+     ArchipelagoAIOCB *aio_cb = (ArchipelagoAIOCB *) reqdata->aio_cb;
+-    AioContext *ctx = bdrv_get_aio_context(aio_cb->common.bs);
-diff --git a/tests/qemu-iotests/110 b/tests/qemu-iotests/110
-index XXXXXXX..XXXXXXX 100755
+-    aio_context_acquire(ctx);
---- a/tests/qemu-iotests/110
+     aio_cb->common.cb(aio_cb->common.opaque, aio_cb->ret);
-+++ b/tests/qemu-iotests/110
+-    aio_context_release(ctx);
-@@ -XXX,XX +XXX,XX @@ echo
+     aio_cb->status = 0;
- # Across blkdebug without a config file, you cannot reconstruct filenames, so
- # qemu is incapable of knowing the directory of the top image from the filename
+     qemu_aio_unref(aio_cb);
- # alone. However, using bdrv_dirname(), it should still work.
+diff --git a/block/block-backend.c b/block/block-backend.c
-+# (Filter out the json:{} filename so this test works with external data files)
+index XXXXXXX..XXXXXXX 100644
- TEST_IMG="json:{
+--- a/block/block-backend.c
-     'driver': '$IMGFMT',
++++ b/block/block-backend.c
-     'file': {
+@@ -XXX,XX +XXX,XX @@ int blk_make_zero(BlockBackend *blk, BdrvRequestFlags flags)
-@@ -XXX,XX +XXX,XX @@ TEST_IMG="json:{
+ static void error_callback_bh(void *opaque)
-             }
+ {
-         ]
+     struct BlockBackendAIOCB *acb = opaque;
-     }
+-    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
--}" _img_info | _filter_img_info | grep -v 'backing file format'
-+}" _img_info | _filter_img_info | grep -v 'backing file format' \
+     bdrv_dec_in_flight(acb->common.bs);
-+    | _filter_json_filename
+-    aio_context_acquire(ctx);
+     acb->common.cb(acb->common.opaque, acb->ret);
- echo
+-    aio_context_release(ctx);
- echo '=== Backing name is always relative to the backed image ==='
+     qemu_aio_unref(acb);
-@@ -XXX,XX +XXX,XX @@ TEST_IMG="json:{
+ }
-             }
-         ]
+@@ -XXX,XX +XXX,XX @@ static void blk_aio_complete(BlkAioEmAIOCB *acb)
-     }
+ static void blk_aio_complete_bh(void *opaque)
--}" _img_info | _filter_img_info | grep -v 'backing file format'
+ {
-+}" _img_info | _filter_img_info | grep -v 'backing file format' \
+     BlkAioEmAIOCB *acb = opaque;
-+    | _filter_json_filename
+-    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
+-
+     assert(acb->has_returned);
- # success, all done
+-    aio_context_acquire(ctx);
-diff --git a/tests/qemu-iotests/110.out b/tests/qemu-iotests/110.out
+     blk_aio_complete(acb);
-index XXXXXXX..XXXXXXX 100644
+-    aio_context_release(ctx);
---- a/tests/qemu-iotests/110.out
+ }
-+++ b/tests/qemu-iotests/110.out
-@@ -XXX,XX +XXX,XX @@ backing file: t.IMGFMT.base (actual path: TEST_DIR/t.IMGFMT.base)
+ static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes,
+diff --git a/block/curl.c b/block/curl.c
- === Non-reconstructable filename ===
+index XXXXXXX..XXXXXXX 100644
+--- a/block/curl.c
--image: json:{"driver": "IMGFMT", "file": {"set-state.0.event": "read_aio", "image": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}, "driver": "blkdebug", "set-state.0.new_state": 42}}
++++ b/block/curl.c
-+image: json:{ /* filtered */ }
+@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
- file format: IMGFMT
+     curl_multi_socket_action(s->multi, CURL_SOCKET_TIMEOUT, 0, &running);
- virtual size: 64 MiB (67108864 bytes)
- backing file: t.IMGFMT.base (actual path: TEST_DIR/t.IMGFMT.base)
+ out:
-@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 backing_file=t.IMGFMT.b
++    aio_context_release(ctx);
+     if (ret != -EINPROGRESS) {
- === Nodes without a common directory ===
+         acb->common.cb(acb->common.opaque, ret);
+         qemu_aio_unref(acb);
--image: json:{"driver": "IMGFMT", "file": {"children": [{"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}, {"driver": "file", "filename": "TEST_DIR/t.IMGFMT.copy"}], "driver": "quorum", "vote-threshold": 1}}
+     }
-+image: json:{ /* filtered */ }
+-    aio_context_release(ctx);
- file format: IMGFMT
+ }
- virtual size: 64 MiB (67108864 bytes)
- backing file: t.IMGFMT.base (cannot determine actual path)
+ static BlockAIOCB *curl_aio_readv(BlockDriverState *bs,
 diff --git a/block/io.c b/block/io.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/io.c
 +++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ static void bdrv_co_io_em_complete(void *opaque, int ret)
      CoroutineIOCompletion *co = opaque;
      co->ret = ret;
 -    qemu_coroutine_enter(co->coroutine);
 +    aio_co_wake(co->coroutine);
  }
  static int coroutine_fn bdrv_driver_preadv(BlockDriverState *bs,
@@ -XXX,XX +XXX,XX @@ static void bdrv_co_complete(BlockAIOCBCoroutine *acb)
  static void bdrv_co_em_bh(void *opaque)
  {
      BlockAIOCBCoroutine *acb = opaque;
 -    BlockDriverState *bs = acb->common.bs;
 -    AioContext *ctx = bdrv_get_aio_context(bs);
      assert(!acb->need_bh);
 -    aio_context_acquire(ctx);
      bdrv_co_complete(acb);
 -    aio_context_release(ctx);
  }
  static void bdrv_co_maybe_schedule_bh(BlockAIOCBCoroutine *acb)
 diff --git a/block/iscsi.c b/block/iscsi.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/iscsi.c
 +++ b/block/iscsi.c
@@ -XXX,XX +XXX,XX @@ static void
  iscsi_bh_cb(void *p)
  {
      IscsiAIOCB *acb = p;
 -    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
      qemu_bh_delete(acb->bh);
      g_free(acb->buf);
      acb->buf = NULL;
 -    aio_context_acquire(ctx);
      acb->common.cb(acb->common.opaque, acb->status);
 -    aio_context_release(ctx);
      if (acb->task != NULL) {
          scsi_free_scsi_task(acb->task);
 diff --git a/block/linux-aio.c b/block/linux-aio.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/linux-aio.c
 +++ b/block/linux-aio.c
@@ -XXX,XX +XXX,XX @@ static inline ssize_t io_event_ret(struct io_event *ev)
   */
  static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
  {
 -    LinuxAioState *s = laiocb->ctx;
      int ret;
      ret = laiocb->ret;
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
      }
      laiocb->ret = ret;
 -    aio_context_acquire(s->aio_context);
      if (laiocb->co) {
          /* If the coroutine is already entered it must be in ioq_submit() and
           * will notice laio->ret has been filled in when it eventually runs
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
           * that!
           */
          if (!qemu_coroutine_entered(laiocb->co)) {
 -            qemu_coroutine_enter(laiocb->co);
 +            aio_co_wake(laiocb->co);
          }
      } else {
          laiocb->common.cb(laiocb->common.opaque, ret);
          qemu_aio_unref(laiocb);
      }
 -    aio_context_release(s->aio_context);
  }
  /**
 diff --git a/block/mirror.c b/block/mirror.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
@@ -XXX,XX +XXX,XX @@ static void mirror_write_complete(void *opaque, int ret)
  {
      MirrorOp *op = opaque;
      MirrorBlockJob *s = op->s;
 +
 +    aio_context_acquire(blk_get_aio_context(s->common.blk));
      if (ret < 0) {
          BlockErrorAction action;
@@ -XXX,XX +XXX,XX @@ static void mirror_write_complete(void *opaque, int ret)
          }
      }
      mirror_iteration_done(op, ret);
 +    aio_context_release(blk_get_aio_context(s->common.blk));
  }
  static void mirror_read_complete(void *opaque, int ret)
  {
      MirrorOp *op = opaque;
      MirrorBlockJob *s = op->s;
 +
 +    aio_context_acquire(blk_get_aio_context(s->common.blk));
      if (ret < 0) {
          BlockErrorAction action;
@@ -XXX,XX +XXX,XX @@ static void mirror_read_complete(void *opaque, int ret)
          }
          mirror_iteration_done(op, ret);
 -        return;
 +    } else {
 +        blk_aio_pwritev(s->target, op->sector_num * BDRV_SECTOR_SIZE, &op->qiov,
 +                        0, mirror_write_complete, op);
      }
 -    blk_aio_pwritev(s->target, op->sector_num * BDRV_SECTOR_SIZE, &op->qiov,
 -                    0, mirror_write_complete, op);
 +    aio_context_release(blk_get_aio_context(s->common.blk));
  }
  static inline void mirror_clip_sectors(MirrorBlockJob *s,
 diff --git a/block/null.c b/block/null.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/null.c
 +++ b/block/null.c
@@ -XXX,XX +XXX,XX @@ static const AIOCBInfo null_aiocb_info = {
  static void null_bh_cb(void *opaque)
  {
      NullAIOCB *acb = opaque;
 -    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 -
 -    aio_context_acquire(ctx);
      acb->common.cb(acb->common.opaque, 0);
 -    aio_context_release(ctx);
      qemu_aio_unref(acb);
  }
  static void null_timer_cb(void *opaque)
  {
      NullAIOCB *acb = opaque;
 -    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 -
 -    aio_context_acquire(ctx);
      acb->common.cb(acb->common.opaque, 0);
 -    aio_context_release(ctx);
      timer_deinit(&acb->timer);
      qemu_aio_unref(acb);
  }
 diff --git a/block/qed-cluster.c b/block/qed-cluster.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/qed-cluster.c
 +++ b/block/qed-cluster.c
@@ -XXX,XX +XXX,XX @@ static void qed_find_cluster_cb(void *opaque, int ret)
      unsigned int index;
      unsigned int n;
 +    qed_acquire(s);
      if (ret) {
          goto out;
      }
@@ -XXX,XX +XXX,XX @@ static void qed_find_cluster_cb(void *opaque, int ret)
  out:
      find_cluster_cb->cb(find_cluster_cb->opaque, ret, offset, len);
 +    qed_release(s);
      g_free(find_cluster_cb);
  }
 diff --git a/block/qed-table.c b/block/qed-table.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/qed-table.c
 +++ b/block/qed-table.c
@@ -XXX,XX +XXX,XX @@ static void qed_read_table_cb(void *opaque, int ret)
  {
      QEDReadTableCB *read_table_cb = opaque;
      QEDTable *table = read_table_cb->table;
 +    BDRVQEDState *s = read_table_cb->s;
      int noffsets = read_table_cb->qiov.size / sizeof(uint64_t);
      int i;
@@ -XXX,XX +XXX,XX @@ static void qed_read_table_cb(void *opaque, int ret)
      }
      /* Byteswap offsets */
 +    qed_acquire(s);
      for (i = 0; i < noffsets; i++) {
          table->offsets[i] = le64_to_cpu(table->offsets[i]);
      }
 +    qed_release(s);
  out:
      /* Completion */
 -    trace_qed_read_table_cb(read_table_cb->s, read_table_cb->table, ret);
 +    trace_qed_read_table_cb(s, read_table_cb->table, ret);
      gencb_complete(&read_table_cb->gencb, ret);
  }
@@ -XXX,XX +XXX,XX @@ typedef struct {
  static void qed_write_table_cb(void *opaque, int ret)
  {
      QEDWriteTableCB *write_table_cb = opaque;
 +    BDRVQEDState *s = write_table_cb->s;
 -    trace_qed_write_table_cb(write_table_cb->s,
 +    trace_qed_write_table_cb(s,
                               write_table_cb->orig_table,
                               write_table_cb->flush,
                               ret);
@@ -XXX,XX +XXX,XX @@ static void qed_write_table_cb(void *opaque, int ret)
      if (write_table_cb->flush) {
          /* We still need to flush first */
          write_table_cb->flush = false;
 +        qed_acquire(s);
          bdrv_aio_flush(write_table_cb->s->bs, qed_write_table_cb,
                         write_table_cb);
 +        qed_release(s);
          return;
      }
@@ -XXX,XX +XXX,XX @@ static void qed_read_l2_table_cb(void *opaque, int ret)
      CachedL2Table *l2_table = request->l2_table;
      uint64_t l2_offset = read_l2_table_cb->l2_offset;
 +    qed_acquire(s);
      if (ret) {
          /* can't trust loaded L2 table anymore */
          qed_unref_l2_cache_entry(l2_table);
@@ -XXX,XX +XXX,XX @@ static void qed_read_l2_table_cb(void *opaque, int ret)
          request->l2_table = qed_find_l2_cache_entry(&s->l2_cache, l2_offset);
          assert(request->l2_table != NULL);
      }
 +    qed_release(s);
      gencb_complete(&read_l2_table_cb->gencb, ret);
  }
 diff --git a/block/qed.c b/block/qed.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/qed.c
 +++ b/block/qed.c
@@ -XXX,XX +XXX,XX @@ static void qed_is_allocated_cb(void *opaque, int ret, uint64_t offset, size_t l
      }
      if (cb->co) {
 -        qemu_coroutine_enter(cb->co);
 +        aio_co_wake(cb->co);
      }
  }
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn qed_co_pwrite_zeroes_cb(void *opaque, int ret)
      cb->done = true;
      cb->ret = ret;
      if (cb->co) {
 -        qemu_coroutine_enter(cb->co);
 +        aio_co_wake(cb->co);
      }
  }
 diff --git a/block/rbd.c b/block/rbd.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/rbd.c
 +++ b/block/rbd.c
@@ -XXX,XX +XXX,XX @@ shutdown:
  static void qemu_rbd_complete_aio(RADOSCB *rcb)
  {
      RBDAIOCB *acb = rcb->acb;
 -    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
      int64_t r;
      r = rcb->ret;
@@ -XXX,XX +XXX,XX @@ static void qemu_rbd_complete_aio(RADOSCB *rcb)
          qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
      }
      qemu_vfree(acb->bounce);
 -
 -    aio_context_acquire(ctx);
      acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
 -    aio_context_release(ctx);
      qemu_aio_unref(acb);
  }
 diff --git a/block/win32-aio.c b/block/win32-aio.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/win32-aio.c
 +++ b/block/win32-aio.c
@@ -XXX,XX +XXX,XX @@ static void win32_aio_process_completion(QEMUWin32AIOState *s,
          qemu_vfree(waiocb->buf);
      }
 -
 -    aio_context_acquire(s->aio_ctx);
      waiocb->common.cb(waiocb->common.opaque, ret);
 -    aio_context_release(s->aio_ctx);
      qemu_aio_unref(waiocb);
  }
 diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/block/virtio-blk.c
 +++ b/hw/block/virtio-blk.c
@@ -XXX,XX +XXX,XX @@ static int virtio_blk_handle_rw_error(VirtIOBlockReq *req, int error,
  static void virtio_blk_rw_complete(void *opaque, int ret)
  {
      VirtIOBlockReq *next = opaque;
 +    VirtIOBlock *s = next->dev;
 +    aio_context_acquire(blk_get_aio_context(s->conf.conf.blk));
      while (next) {
          VirtIOBlockReq *req = next;
          next = req->mr_next;
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_rw_complete(void *opaque, int ret)
          block_acct_done(blk_get_stats(req->dev->blk), &req->acct);
          virtio_blk_free_request(req);
      }
 +    aio_context_release(blk_get_aio_context(s->conf.conf.blk));
  }
  static void virtio_blk_flush_complete(void *opaque, int ret)
  {
      VirtIOBlockReq *req = opaque;
 +    VirtIOBlock *s = req->dev;
 +    aio_context_acquire(blk_get_aio_context(s->conf.conf.blk));
      if (ret) {
          if (virtio_blk_handle_rw_error(req, -ret, 0)) {
 -            return;
 +            goto out;
          }
      }
      virtio_blk_req_complete(req, VIRTIO_BLK_S_OK);
      block_acct_done(blk_get_stats(req->dev->blk), &req->acct);
      virtio_blk_free_request(req);
 +
 +out:
 +    aio_context_release(blk_get_aio_context(s->conf.conf.blk));
  }
  #ifdef __linux__
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_ioctl_complete(void *opaque, int status)
      virtio_stl_p(vdev, &scsi->data_len, hdr->dxfer_len);
  out:
 +    aio_context_acquire(blk_get_aio_context(s->conf.conf.blk));
      virtio_blk_req_complete(req, status);
      virtio_blk_free_request(req);
 +    aio_context_release(blk_get_aio_context(s->conf.conf.blk));
      g_free(ioctl_req);
  }
 diff --git a/hw/scsi/scsi-disk.c b/hw/scsi/scsi-disk.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/scsi/scsi-disk.c
 +++ b/hw/scsi/scsi-disk.c
@@ -XXX,XX +XXX,XX @@ static void scsi_aio_complete(void *opaque, int ret)
      assert(r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
      if (scsi_disk_req_check_error(r, ret, true)) {
          goto done;
      }
@@ -XXX,XX +XXX,XX @@ static void scsi_aio_complete(void *opaque, int ret)
      scsi_req_complete(&r->req, GOOD);
  done:
 +    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
      scsi_req_unref(&r->req);
  }
@@ -XXX,XX +XXX,XX @@ static void scsi_dma_complete(void *opaque, int ret)
      assert(r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
      if (ret < 0) {
          block_acct_failed(blk_get_stats(s->qdev.conf.blk), &r->acct);
      } else {
          block_acct_done(blk_get_stats(s->qdev.conf.blk), &r->acct);
      }
      scsi_dma_complete_noio(r, ret);
 +    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
  }
  static void scsi_read_complete(void * opaque, int ret)
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
      assert(r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
      if (scsi_disk_req_check_error(r, ret, true)) {
          goto done;
      }
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
  done:
      scsi_req_unref(&r->req);
 +    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
  }
  /* Actually issue a read to the block device.  */
@@ -XXX,XX +XXX,XX @@ static void scsi_do_read_cb(void *opaque, int ret)
      assert (r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
      if (ret < 0) {
          block_acct_failed(blk_get_stats(s->qdev.conf.blk), &r->acct);
      } else {
          block_acct_done(blk_get_stats(s->qdev.conf.blk), &r->acct);
      }
      scsi_do_read(opaque, ret);
 +    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
  }
  /* Read more data from scsi device into buffer.  */
@@ -XXX,XX +XXX,XX @@ static void scsi_write_complete(void * opaque, int ret)
      assert (r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
      if (ret < 0) {
          block_acct_failed(blk_get_stats(s->qdev.conf.blk), &r->acct);
      } else {
          block_acct_done(blk_get_stats(s->qdev.conf.blk), &r->acct);
      }
      scsi_write_complete_noio(r, ret);
 +    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
  }
  static void scsi_write_data(SCSIRequest *req)
@@ -XXX,XX +XXX,XX @@ static void scsi_unmap_complete(void *opaque, int ret)
  {
      UnmapCBData *data = opaque;
      SCSIDiskReq *r = data->r;
 +    SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev);
      assert(r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
      scsi_unmap_complete_noio(data, ret);
 +    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
  }
  static void scsi_disk_emulate_unmap(SCSIDiskReq *r, uint8_t *inbuf)
@@ -XXX,XX +XXX,XX @@ static void scsi_write_same_complete(void *opaque, int ret)
      assert(r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
      if (scsi_disk_req_check_error(r, ret, true)) {
          goto done;
      }
@@ -XXX,XX +XXX,XX @@ done:
      scsi_req_unref(&r->req);
      qemu_vfree(data->iov.iov_base);
      g_free(data);
 +    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
  }
  static void scsi_disk_emulate_write_same(SCSIDiskReq *r, uint8_t *inbuf)
 diff --git a/hw/scsi/scsi-generic.c b/hw/scsi/scsi-generic.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/scsi/scsi-generic.c
 +++ b/hw/scsi/scsi-generic.c
@@ -XXX,XX +XXX,XX @@ done:
  static void scsi_command_complete(void *opaque, int ret)
  {
      SCSIGenericReq *r = (SCSIGenericReq *)opaque;
 +    SCSIDevice *s = r->req.dev;
      assert(r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +
 +    aio_context_acquire(blk_get_aio_context(s->conf.blk));
      scsi_command_complete_noio(r, ret);
 +    aio_context_release(blk_get_aio_context(s->conf.blk));
  }
  static int execute_command(BlockBackend *blk,
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
      assert(r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->conf.blk));
 +
      if (ret || r->req.io_canceled) {
          scsi_command_complete_noio(r, ret);
 -        return;
 +        goto done;
      }
      len = r->io_header.dxfer_len - r->io_header.resid;
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
      r->len = -1;
      if (len == 0) {
          scsi_command_complete_noio(r, 0);
 -        return;
 +        goto done;
      }
      /* Snoop READ CAPACITY output to set the blocksize.  */
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
      }
      scsi_req_data(&r->req, len);
      scsi_req_unref(&r->req);
 +
 +done:
 +    aio_context_release(blk_get_aio_context(s->conf.blk));
  }
  /* Read more data from scsi device into buffer.  */
@@ -XXX,XX +XXX,XX @@ static void scsi_write_complete(void * opaque, int ret)
      assert(r->req.aiocb != NULL);
      r->req.aiocb = NULL;
 +    aio_context_acquire(blk_get_aio_context(s->conf.blk));
 +
      if (ret || r->req.io_canceled) {
          scsi_command_complete_noio(r, ret);
 -        return;
 +        goto done;
      }
      if (r->req.cmd.buf[0] == MODE_SELECT && r->req.cmd.buf[4] == 12 &&
@@ -XXX,XX +XXX,XX @@ static void scsi_write_complete(void * opaque, int ret)
      }
      scsi_command_complete_noio(r, ret);
 +
 +done:
 +    aio_context_release(blk_get_aio_context(s->conf.blk));
  }
  /* Write data to a scsi device.  Returns nonzero on failure.
 diff --git a/util/thread-pool.c b/util/thread-pool.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/thread-pool.c
 +++ b/util/thread-pool.c
@@ -XXX,XX +XXX,XX @@ restart:
               */
              qemu_bh_schedule(pool->completion_bh);
 +            aio_context_release(pool->ctx);
              elem->common.cb(elem->common.opaque, elem->ret);
 +            aio_context_acquire(pool->ctx);
              qemu_aio_unref(elem);
              goto restart;
          } else {
@@ -XXX,XX +XXX,XX @@ static void thread_pool_co_cb(void *opaque, int ret)
      ThreadPoolCo *co = opaque;
      co->ret = ret;
 -    qemu_coroutine_enter(co->co);
 +    aio_co_wake(co->co);
  }
  int coroutine_fn thread_pool_submit_co(ThreadPool *pool, ThreadPoolFunc *func,
 --
-.24.1
+.9.3

-[PULL 31/34] qcow2: Allow writing compressed data of multiple clusters
+[Qemu-devel] [PULL v2 16/24] aio-posix: partially inline aio_dispatch into aio_poll
-From: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-QEMU currently supports writing compressed data of the size equal to
+This patch prepares for the removal of unnecessary lockcnt inc/dec pairs.
-one cluster. This patch allows writing QCOW2 compressed data that
+Extract the dispatching loop for file descriptor handlers into a new
-exceed one cluster. Now, we split buffered data into separate clusters
+function aio_dispatch_handlers, and then inline aio_dispatch into
-and write them compressed using the block/aio_task API.
+aio_poll.
-Suggested-by: Pavel Butsykin <pbutsykin@virtuozzo.com>
+aio_dispatch can now become void.
-Suggested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
-Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-Reviewed-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Fam Zheng <famz@redhat.com>
-Message-id: 1575288906-551879-3-git-send-email-andrey.shinkevich@virtuozzo.com
+Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Message-id: 20170213135235.12274-17-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- block/qcow2.c | 102 +++++++++++++++++++++++++++++++++++++-------------
+ include/block/aio.h |  6 +-----
-file changed, 75 insertions(+), 27 deletions(-)
+ util/aio-posix.c    | 44 ++++++++++++++------------------------------
  util/aio-win32.c    | 13 ++++---------
  util/async.c        |  2 +-
 files changed, 20 insertions(+), 45 deletions(-)
-diff --git a/block/qcow2.c b/block/qcow2.c
+diff --git a/include/block/aio.h b/include/block/aio.h
 index XXXXXXX..XXXXXXX 100644
---- a/block/qcow2.c
+--- a/include/block/aio.h
-+++ b/block/qcow2.c
++++ b/include/block/aio.h
-@@ -XXX,XX +XXX,XX @@ fail:
+@@ -XXX,XX +XXX,XX @@ bool aio_pending(AioContext *ctx);
-     return ret;
+ /* Dispatch any pending callbacks from the GSource attached to the AioContext.
   *
   * This is used internally in the implementation of the GSource.
 - *
 - * @dispatch_fds: true to process fds, false to skip them
 - *                (can be used as an optimization by callers that know there
 - *                are no fds ready)
   */
 -bool aio_dispatch(AioContext *ctx, bool dispatch_fds);
 +void aio_dispatch(AioContext *ctx);
  /* Progress in completing AIO work to occur.  This can issue new pending
   * aio as a result of executing I/O completion or bh callbacks.
 diff --git a/util/aio-posix.c b/util/aio-posix.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/aio-posix.c
 +++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
      AioHandler *node, *tmp;
      bool progress = false;
 -    /*
 -     * We have to walk very carefully in case aio_set_fd_handler is
 -     * called while we're walking.
 -     */
 -    qemu_lockcnt_inc(&ctx->list_lock);
 -
      QLIST_FOREACH_SAFE_RCU(node, &ctx->aio_handlers, node, tmp) {
          int revents;
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
          }
      }
 -    qemu_lockcnt_dec(&ctx->list_lock);
      return progress;
  }
--/* XXX: put compressed sectors first, then all the cluster aligned
+-/*
--   tables to avoid losing bytes in alignment */
+- * Note that dispatch_fds == false has the side-effect of post-poning the
- static coroutine_fn int
+- * freeing of deleted handlers.
--qcow2_co_pwritev_compressed_part(BlockDriverState *bs,
+- */
-+qcow2_co_pwritev_compressed_task(BlockDriverState *bs,
+-bool aio_dispatch(AioContext *ctx, bool dispatch_fds)
-                                  uint64_t offset, uint64_t bytes,
++void aio_dispatch(AioContext *ctx)
                                   QEMUIOVector *qiov, size_t qiov_offset)
  {
-@@ -XXX,XX +XXX,XX @@ qcow2_co_pwritev_compressed_part(BlockDriverState *bs,
+-    bool progress;
-     uint8_t *buf, *out_buf;
++    aio_bh_poll(ctx);
-     uint64_t cluster_offset;
+-    /*
--    if (has_data_file(bs)) {
+-     * If there are callbacks left that have been queued, we need to call them.
--        return -ENOTSUP;
+-     * Do not call select in this case, because it is possible that the caller
 -     * does not need a complete flush (as is the case for aio_poll loops).
 -     */
 -    progress = aio_bh_poll(ctx);
 +    qemu_lockcnt_inc(&ctx->list_lock);
 +    aio_dispatch_handlers(ctx);
 +    qemu_lockcnt_dec(&ctx->list_lock);
 -    if (dispatch_fds) {
 -        progress |= aio_dispatch_handlers(ctx);
 -    }
 -
--    if (bytes == 0) {
+-    /* Run our timers */
--        /* align end of file to a sector boundary to ease reading with
+-    progress |= timerlistgroup_run_timers(&ctx->tlg);
--           sector based I/Os */
+-
--        int64_t len = bdrv_getlength(bs->file->bs);
+-    return progress;
--        if (len < 0) {
++    timerlistgroup_run_timers(&ctx->tlg);
--            return len;
+ }
--        }
--        return bdrv_co_truncate(bs->file, len, false, PREALLOC_MODE_OFF, NULL);
+ /* These thread-local variables are used only in a small part of aio_poll
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
      npfd = 0;
      qemu_lockcnt_dec(&ctx->list_lock);
 -    /* Run dispatch even if there were no readable fds to run timers */
 -    if (aio_dispatch(ctx, ret > 0)) {
 -        progress = true;
 +    progress |= aio_bh_poll(ctx);
 +
 +    if (ret > 0) {
 +        qemu_lockcnt_inc(&ctx->list_lock);
 +        progress |= aio_dispatch_handlers(ctx);
 +        qemu_lockcnt_dec(&ctx->list_lock);
      }
 +    progress |= timerlistgroup_run_timers(&ctx->tlg);
 +
      return progress;
  }
 diff --git a/util/aio-win32.c b/util/aio-win32.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/aio-win32.c
 +++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
      return progress;
  }
 -bool aio_dispatch(AioContext *ctx, bool dispatch_fds)
 +void aio_dispatch(AioContext *ctx)
  {
 -    bool progress;
 -
 -    progress = aio_bh_poll(ctx);
 -    if (dispatch_fds) {
 -        progress |= aio_dispatch_handlers(ctx, INVALID_HANDLE_VALUE);
 -    }
--
+-    progress |= timerlistgroup_run_timers(&ctx->tlg);
--    if (offset_into_cluster(s, offset)) {
+-    return progress;
--        return -EINVAL;
++    aio_bh_poll(ctx);
--    }
++    aio_dispatch_handlers(ctx, INVALID_HANDLE_VALUE);
-+    assert(bytes == s->cluster_size || (bytes < s->cluster_size &&
++    timerlistgroup_run_timers(&ctx->tlg);
 +           (offset + bytes == bs->total_sectors << BDRV_SECTOR_BITS)));
      buf = qemu_blockalign(bs, s->cluster_size);
 -    if (bytes != s->cluster_size) {
 -        if (bytes > s->cluster_size ||
 -            offset + bytes != bs->total_sectors << BDRV_SECTOR_BITS)
 -        {
 -            qemu_vfree(buf);
 -            return -EINVAL;
 -        }
 +    if (bytes < s->cluster_size) {
          /* Zero-pad last write if image size is not cluster aligned */
          memset(buf + bytes, 0, s->cluster_size - bytes);
      }
@@ -XXX,XX +XXX,XX @@ fail:
      return ret;
  }
-+static coroutine_fn int qcow2_co_pwritev_compressed_task_entry(AioTask *task)
+ bool aio_poll(AioContext *ctx, bool blocking)
-+{
+diff --git a/util/async.c b/util/async.c
-+    Qcow2AioTask *t = container_of(task, Qcow2AioTask, task);
+index XXXXXXX..XXXXXXX 100644
-+
+--- a/util/async.c
-+    assert(!t->cluster_type && !t->l2meta);
++++ b/util/async.c
-+
+@@ -XXX,XX +XXX,XX @@ aio_ctx_dispatch(GSource     *source,
-+    return qcow2_co_pwritev_compressed_task(t->bs, t->offset, t->bytes, t->qiov,
+     AioContext *ctx = (AioContext *) source;
-+                                            t->qiov_offset);
-+}
+     assert(callback == NULL);
-+
+-    aio_dispatch(ctx, true);
-+/*
++    aio_dispatch(ctx);
-+ * XXX: put compressed sectors first, then all the cluster aligned
+     return true;
-+ * tables to avoid losing bytes in alignment
+ }
-+ */
 +static coroutine_fn int
 +qcow2_co_pwritev_compressed_part(BlockDriverState *bs,
 +                                 uint64_t offset, uint64_t bytes,
 +                                 QEMUIOVector *qiov, size_t qiov_offset)
 +{
 +    BDRVQcow2State *s = bs->opaque;
 +    AioTaskPool *aio = NULL;
 +    int ret = 0;
 +
 +    if (has_data_file(bs)) {
 +        return -ENOTSUP;
 +    }
 +
 +    if (bytes == 0) {
 +        /*
 +         * align end of file to a sector boundary to ease reading with
 +         * sector based I/Os
 +         */
 +        int64_t len = bdrv_getlength(bs->file->bs);
 +        if (len < 0) {
 +            return len;
 +        }
 +        return bdrv_co_truncate(bs->file, len, false, PREALLOC_MODE_OFF, NULL);
 +    }
 +
 +    if (offset_into_cluster(s, offset)) {
 +        return -EINVAL;
 +    }
 +
 +    while (bytes && aio_task_pool_status(aio) == 0) {
 +        uint64_t chunk_size = MIN(bytes, s->cluster_size);
 +
 +        if (!aio && chunk_size != bytes) {
 +            aio = aio_task_pool_new(QCOW2_MAX_WORKERS);
 +        }
 +
 +        ret = qcow2_add_task(bs, aio, qcow2_co_pwritev_compressed_task_entry,
 +                             0, 0, offset, chunk_size, qiov, qiov_offset, NULL);
 +        if (ret < 0) {
 +            break;
 +        }
 +        qiov_offset += chunk_size;
 +        offset += chunk_size;
 +        bytes -= chunk_size;
 +    }
 +
 +    if (aio) {
 +        aio_task_pool_wait_all(aio);
 +        if (ret == 0) {
 +            ret = aio_task_pool_status(aio);
 +        }
 +        g_free(aio);
 +    }
 +
 +    return ret;
 +}
 +
  static int coroutine_fn
  qcow2_co_preadv_compressed(BlockDriverState *bs,
                             uint64_t file_cluster_offset,
 --
-.24.1
+.9.3

-[PULL 02/34] block: Use bdrv_qapi_perm_to_blk_perm()
+[Qemu-devel] [PULL v2 17/24] async: remove unnecessary inc/dec pairs
-We can save some LoC in xdbg_graph_add_edge() by using
+From: Paolo Bonzini <pbonzini@redhat.com>
 bdrv_qapi_perm_to_blk_perm().
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Pull the increment/decrement pair out of aio_bh_poll and into the
-Message-id: 20191108123455.39445-3-mreitz@redhat.com
+callers.
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-18-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- block.c | 29 ++++++++---------------------
+ util/aio-posix.c |  8 +++-----
-file changed, 8 insertions(+), 21 deletions(-)
+ util/aio-win32.c |  8 ++++----
  util/async.c     | 12 ++++++------
 files changed, 13 insertions(+), 15 deletions(-)
-diff --git a/block.c b/block.c
+diff --git a/util/aio-posix.c b/util/aio-posix.c
 index XXXXXXX..XXXXXXX 100644
---- a/block.c
+--- a/util/aio-posix.c
-+++ b/block.c
++++ b/util/aio-posix.c
-@@ -XXX,XX +XXX,XX @@ static void xdbg_graph_add_node(XDbgBlockGraphConstructor *gr, void *node,
+@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
- static void xdbg_graph_add_edge(XDbgBlockGraphConstructor *gr, void *parent,
-                                 const BdrvChild *child)
+ void aio_dispatch(AioContext *ctx)
  {
--    typedef struct {
++    qemu_lockcnt_inc(&ctx->list_lock);
--        unsigned int flag;
+     aio_bh_poll(ctx);
 -        BlockPermission num;
 -    } PermissionMap;
 -
--    static const PermissionMap permissions[] = {
+-    qemu_lockcnt_inc(&ctx->list_lock);
--        { BLK_PERM_CONSISTENT_READ, BLOCK_PERMISSION_CONSISTENT_READ },
+     aio_dispatch_handlers(ctx);
--        { BLK_PERM_WRITE,           BLOCK_PERMISSION_WRITE },
+     qemu_lockcnt_dec(&ctx->list_lock);
--        { BLK_PERM_WRITE_UNCHANGED, BLOCK_PERMISSION_WRITE_UNCHANGED },
--        { BLK_PERM_RESIZE,          BLOCK_PERMISSION_RESIZE },
+@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
--        { BLK_PERM_GRAPH_MOD,       BLOCK_PERMISSION_GRAPH_MOD },
+     }
--        { 0, 0 }
--    };
+     npfd = 0;
--    const PermissionMap *p;
+-    qemu_lockcnt_dec(&ctx->list_lock);
-+    BlockPermission qapi_perm;
-     XDbgBlockGraphEdge *edge;
+     progress |= aio_bh_poll(ctx);
--    QEMU_BUILD_BUG_ON(1UL << (ARRAY_SIZE(permissions) - 1) != BLK_PERM_ALL + 1);
+     if (ret > 0) {
 -        qemu_lockcnt_inc(&ctx->list_lock);
          progress |= aio_dispatch_handlers(ctx);
 -        qemu_lockcnt_dec(&ctx->list_lock);
      }
 +    qemu_lockcnt_dec(&ctx->list_lock);
 +
      progress |= timerlistgroup_run_timers(&ctx->tlg);
      return progress;
 diff --git a/util/aio-win32.c b/util/aio-win32.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/aio-win32.c
 +++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
      bool progress = false;
      AioHandler *tmp;
 -    qemu_lockcnt_inc(&ctx->list_lock);
 -
-     edge = g_new0(XDbgBlockGraphEdge, 1);
+     /*
+      * We have to walk very carefully in case aio_set_fd_handler is
-     edge->parent = xdbg_graph_node_num(gr, parent);
+      * called while we're walking.
-     edge->child = xdbg_graph_node_num(gr, child->bs);
+@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
      edge->name = g_strdup(child->name);
 -    for (p = permissions; p->flag; p++) {
 -        if (p->flag & child->perm) {
 -            QAPI_LIST_ADD(edge->perm, p->num);
 +    for (qapi_perm = 0; qapi_perm < BLOCK_PERMISSION__MAX; qapi_perm++) {
 +        uint64_t flag = bdrv_qapi_perm_to_blk_perm(qapi_perm);
 +
 +        if (flag & child->perm) {
 +            QAPI_LIST_ADD(edge->perm, qapi_perm);
          }
 -        if (p->flag & child->shared_perm) {
 -            QAPI_LIST_ADD(edge->shared_perm, p->num);
 +        if (flag & child->shared_perm) {
 +            QAPI_LIST_ADD(edge->shared_perm, qapi_perm);
          }
      }
+-    qemu_lockcnt_dec(&ctx->list_lock);
+     return progress;
+ }
+ void aio_dispatch(AioContext *ctx)
+ {
++    qemu_lockcnt_inc(&ctx->list_lock);
+     aio_bh_poll(ctx);
+     aio_dispatch_handlers(ctx, INVALID_HANDLE_VALUE);
++    qemu_lockcnt_dec(&ctx->list_lock);
+     timerlistgroup_run_timers(&ctx->tlg);
+ }
+@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
+         }
+     }
+-    qemu_lockcnt_dec(&ctx->list_lock);
+     first = true;
+     /* ctx->notifier is always registered.  */
+@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
+         progress |= aio_dispatch_handlers(ctx, event);
+     } while (count > 0);
++    qemu_lockcnt_dec(&ctx->list_lock);
++
+     progress |= timerlistgroup_run_timers(&ctx->tlg);
+     return progress;
+ }
+diff --git a/util/async.c b/util/async.c
+index XXXXXXX..XXXXXXX 100644
+--- a/util/async.c
++++ b/util/async.c
+@@ -XXX,XX +XXX,XX @@ void aio_bh_call(QEMUBH *bh)
+     bh->cb(bh->opaque);
+ }
+-/* Multiple occurrences of aio_bh_poll cannot be called concurrently */
++/* Multiple occurrences of aio_bh_poll cannot be called concurrently.
++ * The count in ctx->list_lock is incremented before the call, and is
++ * not affected by the call.
++ */
+ int aio_bh_poll(AioContext *ctx)
+ {
+     QEMUBH *bh, **bhp, *next;
+     int ret;
+     bool deleted = false;
+-    qemu_lockcnt_inc(&ctx->list_lock);
+-
+     ret = 0;
+     for (bh = atomic_rcu_read(&ctx->first_bh); bh; bh = next) {
+         next = atomic_rcu_read(&bh->next);
+@@ -XXX,XX +XXX,XX @@ int aio_bh_poll(AioContext *ctx)
+     /* remove deleted bhs */
+     if (!deleted) {
+-        qemu_lockcnt_dec(&ctx->list_lock);
+         return ret;
+     }
+-    if (qemu_lockcnt_dec_and_lock(&ctx->list_lock)) {
++    if (qemu_lockcnt_dec_if_lock(&ctx->list_lock)) {
+         bhp = &ctx->first_bh;
+         while (*bhp) {
+             bh = *bhp;
+@@ -XXX,XX +XXX,XX @@ int aio_bh_poll(AioContext *ctx)
+                 bhp = &bh->next;
+             }
+         }
+-        qemu_lockcnt_unlock(&ctx->list_lock);
++        qemu_lockcnt_inc_and_unlock(&ctx->list_lock);
+     }
+     return ret;
+ }
 --
-.24.1
+.9.3

-[PULL 04/34] iotests: Add @error to wait_until_completed
+Deleted patch
-Callers can use this new parameter to expect failure during the
-completion process.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: John Snow <jsnow@redhat.com>
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
-Message-id: 20191108123455.39445-5-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/iotests.py | 18 ++++++++++++------
-file changed, 12 insertions(+), 6 deletions(-)
-diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/iotests.py
-+++ b/tests/qemu-iotests/iotests.py
-@@ -XXX,XX +XXX,XX @@ class QMPTestCase(unittest.TestCase):
-         self.assert_no_active_block_jobs()
-         return result
--    def wait_until_completed(self, drive='drive0', check_offset=True, wait=60.0):
-+    def wait_until_completed(self, drive='drive0', check_offset=True, wait=60.0,
-+                             error=None):
-         '''Wait for a block job to finish, returning the event'''
-         while True:
-             for event in self.vm.get_qmp_events(wait=wait):
-                 if event['event'] == 'BLOCK_JOB_COMPLETED':
-                     self.assert_qmp(event, 'data/device', drive)
--                    self.assert_qmp_absent(event, 'data/error')
--                    if check_offset:
--                        self.assert_qmp(event, 'data/offset', event['data']['len'])
-+                    if error is None:
-+                        self.assert_qmp_absent(event, 'data/error')
-+                        if check_offset:
-+                            self.assert_qmp(event, 'data/offset',
-+                                            event['data']['len'])
-+                    else:
-+                        self.assert_qmp(event, 'data/error', error)
-                     self.assert_no_active_block_jobs()
-                     return event
-                 elif event['event'] == 'JOB_STATUS_CHANGE':
-@@ -XXX,XX +XXX,XX @@ class QMPTestCase(unittest.TestCase):
-         self.assert_qmp(event, 'data/type', 'mirror')
-         self.assert_qmp(event, 'data/offset', event['data']['len'])
--    def complete_and_wait(self, drive='drive0', wait_ready=True):
-+    def complete_and_wait(self, drive='drive0', wait_ready=True,
-+                          completion_error=None):
-         '''Complete a block job and wait for it to finish'''
-         if wait_ready:
-             self.wait_ready(drive=drive)
-@@ -XXX,XX +XXX,XX @@ class QMPTestCase(unittest.TestCase):
-         result = self.vm.qmp('block-job-complete', device=drive)
-         self.assert_qmp(result, 'return', {})
--        event = self.wait_until_completed(drive=drive)
-+        event = self.wait_until_completed(drive=drive, error=completion_error)
-         self.assert_qmp(event, 'data/type', 'mirror')
-     def pause_wait(self, job_id='job0'):
---
-.24.1

-[PULL 05/34] iotests: Add test for failing mirror complete
+Deleted patch
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
-Tested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
-Reviewed-by: John Snow <jsnow@redhat.com>
-Message-id: 20191108123455.39445-6-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/041     | 44 ++++++++++++++++++++++++++++++++++++++
- tests/qemu-iotests/041.out |  4 ++--
-files changed, 46 insertions(+), 2 deletions(-)
-diff --git a/tests/qemu-iotests/041 b/tests/qemu-iotests/041
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/041
-+++ b/tests/qemu-iotests/041
-@@ -XXX,XX +XXX,XX @@ class TestOrphanedSource(iotests.QMPTestCase):
-                              target='dest-ro')
-         self.assert_qmp(result, 'error/class', 'GenericError')
-+    def test_failing_permission_in_complete(self):
-+        self.assert_no_active_block_jobs()
-+
-+        # Unshare consistent-read on the target
-+        # (The mirror job does not care)
-+        result = self.vm.qmp('blockdev-add',
-+                             driver='blkdebug',
-+                             node_name='dest-perm',
-+                             image='dest',
-+                             unshare_child_perms=['consistent-read'])
-+        self.assert_qmp(result, 'return', {})
-+
-+        result = self.vm.qmp('blockdev-mirror', job_id='job', device='src',
-+                             sync='full', target='dest',
-+                             filter_node_name='mirror-filter')
-+        self.assert_qmp(result, 'return', {})
-+
-+        # Require consistent-read on the source
-+        # (We can only add this node once the job has started, or it
-+        # will complain that it does not want to run on non-root nodes)
-+        result = self.vm.qmp('blockdev-add',
-+                             driver='blkdebug',
-+                             node_name='src-perm',
-+                             image='src',
-+                             take_child_perms=['consistent-read'])
-+        self.assert_qmp(result, 'return', {})
-+
-+        # While completing, mirror will attempt to replace src by
-+        # dest, which must fail because src-perm requires
-+        # consistent-read but dest-perm does not share it; thus
-+        # aborting the job when it is supposed to complete
-+        self.complete_and_wait('job',
-+                               completion_error='Operation not permitted')
-+
-+        # Assert that all of our nodes are still there (except for the
-+        # mirror filter, which should be gone despite the failure)
-+        nodes = self.vm.qmp('query-named-block-nodes')['return']
-+        nodes = [node['node-name'] for node in nodes]
-+
-+        for expect in ('src', 'src-perm', 'dest', 'dest-perm'):
-+            self.assertTrue(expect in nodes, '%s disappeared' % expect)
-+        self.assertFalse('mirror-filter' in nodes,
-+                         'Mirror filter node did not disappear')
-+
- if __name__ == '__main__':
-     iotests.main(supported_fmts=['qcow2', 'qed'],
-                  supported_protocols=['file'])
-diff --git a/tests/qemu-iotests/041.out b/tests/qemu-iotests/041.out
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/041.out
-+++ b/tests/qemu-iotests/041.out
-@@ -XXX,XX +XXX,XX @@
--..........................................................................................
-+...........................................................................................
- ----------------------------------------------------------------------
--Ran 90 tests
-+Ran 91 tests
- OK
---
-.24.1

-[PULL 24/34] iotests: Make 091 work with data_file
+[Qemu-devel] [PULL v2 18/24] block: document fields protected by AioContext lock
-The image end offset as reported by qemu-img check is different when
+From: Paolo Bonzini <pbonzini@redhat.com>
 using an external data file; we do not care about its value here, so we
 can just filter it.  Incidentally, common.rc already has _check_test_img
 for us which does exactly that.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-Message-id: 20191107163708.833192-18-mreitz@redhat.com
+Reviewed-by: Fam Zheng <famz@redhat.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
 Message-id: 20170213135235.12274-19-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/091     | 2 +-
+ include/block/block_int.h      | 64 +++++++++++++++++++++++++-----------------
- tests/qemu-iotests/091.out | 2 --
+ include/sysemu/block-backend.h | 14 ++++++---
-files changed, 1 insertion(+), 3 deletions(-)
+files changed, 49 insertions(+), 29 deletions(-)
-diff --git a/tests/qemu-iotests/091 b/tests/qemu-iotests/091
+diff --git a/include/block/block_int.h b/include/block/block_int.h
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/091
 +++ b/tests/qemu-iotests/091
@@ -XXX,XX +XXX,XX @@ echo "Check image pattern"
  ${QEMU_IO} -c "read -P 0x22 0 4M" "${TEST_IMG}" | _filter_testdir | _filter_qemu_io
  echo "Running 'qemu-img check -r all \$TEST_IMG'"
 -"${QEMU_IMG}" check -r all "${TEST_IMG}" 2>&1 | _filter_testdir | _filter_qemu
 +_check_test_img -r all
  echo "*** done"
  rm -f $seq.full
 diff --git a/tests/qemu-iotests/091.out b/tests/qemu-iotests/091.out
 index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/091.out
+--- a/include/block/block_int.h
-+++ b/tests/qemu-iotests/091.out
++++ b/include/block/block_int.h
-@@ -XXX,XX +XXX,XX @@ read 4194304/4194304 bytes at offset 0
+@@ -XXX,XX +XXX,XX @@ struct BdrvChild {
-MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+  * copied as well.
- Running 'qemu-img check -r all $TEST_IMG'
+  */
- No errors were found on the image.
+ struct BlockDriverState {
--80/16384 = 0.49% allocated, 0.00% fragmented, 0.00% compressed clusters
+-    int64_t total_sectors; /* if we are reading a disk image, give its
--Image end offset: 5570560
+-                              size in sectors */
- *** done
++    /* Protected by big QEMU lock or read-only after opening.  No special
 +     * locking needed during I/O...
 +     */
      int open_flags; /* flags used to open the file, re-used for re-open */
      bool read_only; /* if true, the media is read only */
      bool encrypted; /* if true, the media is encrypted */
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
      bool sg;        /* if true, the device is a /dev/sg* */
      bool probed;    /* if true, format was probed rather than specified */
 -    int copy_on_read; /* if nonzero, copy read backing sectors into image.
 -                         note this is a reference count */
 -
 -    CoQueue flush_queue;            /* Serializing flush queue */
 -    bool active_flush_req;          /* Flush request in flight? */
 -    unsigned int write_gen;         /* Current data generation */
 -    unsigned int flushed_gen;       /* Flushed write generation */
 -
      BlockDriver *drv; /* NULL means no media */
      void *opaque;
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
      BdrvChild *backing;
      BdrvChild *file;
 -    /* Callback before write request is processed */
 -    NotifierWithReturnList before_write_notifiers;
 -
 -    /* number of in-flight requests; overall and serialising */
 -    unsigned int in_flight;
 -    unsigned int serialising_in_flight;
 -
 -    bool wakeup;
 -
 -    /* Offset after the highest byte written to */
 -    uint64_t wr_highest_offset;
 -
      /* I/O Limits */
      BlockLimits bl;
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
      QTAILQ_ENTRY(BlockDriverState) bs_list;
      /* element of the list of monitor-owned BDS */
      QTAILQ_ENTRY(BlockDriverState) monitor_list;
 -    QLIST_HEAD(, BdrvDirtyBitmap) dirty_bitmaps;
      int refcnt;
 -    QLIST_HEAD(, BdrvTrackedRequest) tracked_requests;
 -
      /* operation blockers */
      QLIST_HEAD(, BdrvOpBlocker) op_blockers[BLOCK_OP_TYPE_MAX];
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
      /* The error object in use for blocking operations on backing_hd */
      Error *backing_blocker;
 +    /* Protected by AioContext lock */
 +
 +    /* If true, copy read backing sectors into image.  Can be >1 if more
 +     * than one client has requested copy-on-read.
 +     */
 +    int copy_on_read;
 +
 +    /* If we are reading a disk image, give its size in sectors.
 +     * Generally read-only; it is written to by load_vmstate and save_vmstate,
 +     * but the block layer is quiescent during those.
 +     */
 +    int64_t total_sectors;
 +
 +    /* Callback before write request is processed */
 +    NotifierWithReturnList before_write_notifiers;
 +
 +    /* number of in-flight requests; overall and serialising */
 +    unsigned int in_flight;
 +    unsigned int serialising_in_flight;
 +
 +    bool wakeup;
 +
 +    /* Offset after the highest byte written to */
 +    uint64_t wr_highest_offset;
 +
      /* threshold limit for writes, in bytes. "High water mark". */
      uint64_t write_threshold_offset;
      NotifierWithReturn write_threshold_notifier;
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
      /* counter for nested bdrv_io_plug */
      unsigned io_plugged;
 +    QLIST_HEAD(, BdrvTrackedRequest) tracked_requests;
 +    CoQueue flush_queue;                  /* Serializing flush queue */
 +    bool active_flush_req;                /* Flush request in flight? */
 +    unsigned int write_gen;               /* Current data generation */
 +    unsigned int flushed_gen;             /* Flushed write generation */
 +
 +    QLIST_HEAD(, BdrvDirtyBitmap) dirty_bitmaps;
 +
 +    /* do we need to tell the quest if we have a volatile write cache? */
 +    int enable_write_cache;
 +
      int quiesce_counter;
  };
 diff --git a/include/sysemu/block-backend.h b/include/sysemu/block-backend.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/sysemu/block-backend.h
 +++ b/include/sysemu/block-backend.h
@@ -XXX,XX +XXX,XX @@ typedef struct BlockDevOps {
   * fields that must be public. This is in particular for QLIST_ENTRY() and
   * friends so that BlockBackends can be kept in lists outside block-backend.c */
  typedef struct BlockBackendPublic {
 -    /* I/O throttling.
 -     * throttle_state tells us if this BlockBackend has I/O limits configured.
 -     * io_limits_disabled tells us if they are currently being enforced */
 +    /* I/O throttling has its own locking, but also some fields are
 +     * protected by the AioContext lock.
 +     */
 +
 +    /* Protected by AioContext lock.  */
      CoQueue      throttled_reqs[2];
 +
 +    /* Nonzero if the I/O limits are currently being ignored; generally
 +     * it is zero.  */
      unsigned int io_limits_disabled;
      /* The following fields are protected by the ThrottleGroup lock.
 -     * See the ThrottleGroup documentation for details. */
 +     * See the ThrottleGroup documentation for details.
 +     * throttle_state tells us if I/O limits are configured. */
      ThrottleState *throttle_state;
      ThrottleTimers throttle_timers;
      unsigned       pending_reqs[2];
 --
-.24.1
+.9.3

-[PULL 07/34] qcow2-bitmaps: fix qcow2_can_store_new_dirty_bitmap
+[Qemu-devel] [PULL v2 19/24] coroutine-lock: make CoMutex thread-safe
-From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-qcow2_can_store_new_dirty_bitmap works wrong, as it considers only
+This uses the lock-free mutex described in the paper '"Blocking without
-bitmaps already stored in the qcow2 image and ignores persistent
+Locking", or LFTHREADS: A lock-free thread library' by Gidenstam and
-BdrvDirtyBitmap objects.
+Papatriantafilou.  The same technique is used in OSv, and in fact
+the code is essentially a conversion to C of OSv's code.
-So, let's instead count persistent BdrvDirtyBitmaps. We load all qcow2
-bitmaps on open, so there should not be any bitmap in the image for
+[Added missing coroutine_fn in tests/test-aio-multithread.c.
-which we don't have BdrvDirtyBitmaps version. If it is - it's a kind of
+--Stefan]
-corruption, and no reason to check for corruptions here (open() and
-close() are better places for it).
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+Reviewed-by: Fam Zheng <famz@redhat.com>
-Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
+Message-id: 20170213181244.16297-2-pbonzini@redhat.com
-Message-id: 20191014115126.15360-2-vsementsov@virtuozzo.com
+Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 Reviewed-by: Max Reitz <mreitz@redhat.com>
 Cc: qemu-stable@nongnu.org
 Signed-off-by: Max Reitz <mreitz@redhat.com>
 ---
- block/qcow2-bitmap.c | 41 ++++++++++++++++++-----------------------
+ include/qemu/coroutine.h     |  17 ++++-
-file changed, 18 insertions(+), 23 deletions(-)
+ tests/test-aio-multithread.c |  86 ++++++++++++++++++++++++
+ util/qemu-coroutine-lock.c   | 155 ++++++++++++++++++++++++++++++++++++++++---
-diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c
+ util/trace-events            |   1 +
 files changed, 246 insertions(+), 13 deletions(-)
 diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
 index XXXXXXX..XXXXXXX 100644
---- a/block/qcow2-bitmap.c
+--- a/include/qemu/coroutine.h
-+++ b/block/qcow2-bitmap.c
++++ b/include/qemu/coroutine.h
-@@ -XXX,XX +XXX,XX @@ bool coroutine_fn qcow2_co_can_store_new_dirty_bitmap(BlockDriverState *bs,
+@@ -XXX,XX +XXX,XX @@ bool qemu_co_queue_empty(CoQueue *queue);
-                                                       Error **errp)
+ /**
   * Provides a mutex that can be used to synchronise coroutines
   */
 +struct CoWaitRecord;
  typedef struct CoMutex {
 -    bool locked;
 +    /* Count of pending lockers; 0 for a free mutex, 1 for an
 +     * uncontended mutex.
 +     */
 +    unsigned locked;
 +
 +    /* A queue of waiters.  Elements are added atomically in front of
 +     * from_push.  to_pop is only populated, and popped from, by whoever
 +     * is in charge of the next wakeup.  This can be an unlocker or,
 +     * through the handoff protocol, a locker that is about to go to sleep.
 +     */
 +    QSLIST_HEAD(, CoWaitRecord) from_push, to_pop;
 +
 +    unsigned handoff, sequence;
 +
      Coroutine *holder;
 -    CoQueue queue;
  } CoMutex;
  /**
 diff --git a/tests/test-aio-multithread.c b/tests/test-aio-multithread.c
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/test-aio-multithread.c
 +++ b/tests/test-aio-multithread.c
@@ -XXX,XX +XXX,XX @@ static void test_multi_co_schedule_10(void)
      test_multi_co_schedule(10);
  }
 +/* CoMutex thread-safety.  */
 +
 +static uint32_t atomic_counter;
 +static uint32_t running;
 +static uint32_t counter;
 +static CoMutex comutex;
 +
 +static void coroutine_fn test_multi_co_mutex_entry(void *opaque)
 +{
 +    while (!atomic_mb_read(&now_stopping)) {
 +        qemu_co_mutex_lock(&comutex);
 +        counter++;
 +        qemu_co_mutex_unlock(&comutex);
 +
 +        /* Increase atomic_counter *after* releasing the mutex.  Otherwise
 +         * there is a chance (it happens about 1 in 3 runs) that the iothread
 +         * exits before the coroutine is woken up, causing a spurious
 +         * assertion failure.
 +         */
 +        atomic_inc(&atomic_counter);
 +    }
 +    atomic_dec(&running);
 +}
 +
 +static void test_multi_co_mutex(int threads, int seconds)
 +{
 +    int i;
 +
 +    qemu_co_mutex_init(&comutex);
 +    counter = 0;
 +    atomic_counter = 0;
 +    now_stopping = false;
 +
 +    create_aio_contexts();
 +    assert(threads <= NUM_CONTEXTS);
 +    running = threads;
 +    for (i = 0; i < threads; i++) {
 +        Coroutine *co1 = qemu_coroutine_create(test_multi_co_mutex_entry, NULL);
 +        aio_co_schedule(ctx[i], co1);
 +    }
 +
 +    g_usleep(seconds * 1000000);
 +
 +    atomic_mb_set(&now_stopping, true);
 +    while (running > 0) {
 +        g_usleep(100000);
 +    }
 +
 +    join_aio_contexts();
 +    g_test_message("%d iterations/second\n", counter / seconds);
 +    g_assert_cmpint(counter, ==, atomic_counter);
 +}
 +
 +/* Testing with NUM_CONTEXTS threads focuses on the queue.  The mutex however
 + * is too contended (and the threads spend too much time in aio_poll)
 + * to actually stress the handoff protocol.
 + */
 +static void test_multi_co_mutex_1(void)
 +{
 +    test_multi_co_mutex(NUM_CONTEXTS, 1);
 +}
 +
 +static void test_multi_co_mutex_10(void)
 +{
 +    test_multi_co_mutex(NUM_CONTEXTS, 10);
 +}
 +
 +/* Testing with fewer threads stresses the handoff protocol too.  Still, the
 + * case where the locker _can_ pick up a handoff is very rare, happening
 + * about 10 times in 1 million, so increase the runtime a bit compared to
 + * other "quick" testcases that only run for 1 second.
 + */
 +static void test_multi_co_mutex_2_3(void)
 +{
 +    test_multi_co_mutex(2, 3);
 +}
 +
 +static void test_multi_co_mutex_2_30(void)
 +{
 +    test_multi_co_mutex(2, 30);
 +}
 +
  /* End of tests.  */
  int main(int argc, char **argv)
@@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv)
      g_test_add_func("/aio/multi/lifecycle", test_lifecycle);
      if (g_test_quick()) {
          g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_1);
 +        g_test_add_func("/aio/multi/mutex/contended", test_multi_co_mutex_1);
 +        g_test_add_func("/aio/multi/mutex/handoff", test_multi_co_mutex_2_3);
      } else {
          g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_10);
 +        g_test_add_func("/aio/multi/mutex/contended", test_multi_co_mutex_10);
 +        g_test_add_func("/aio/multi/mutex/handoff", test_multi_co_mutex_2_30);
      }
      return g_test_run();
  }
 diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/qemu-coroutine-lock.c
 +++ b/util/qemu-coroutine-lock.c
@@ -XXX,XX +XXX,XX @@
   * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
   * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
   * THE SOFTWARE.
 + *
 + * The lock-free mutex implementation is based on OSv
 + * (core/lfmutex.cc, include/lockfree/mutex.hh).
 + * Copyright (C) 2013 Cloudius Systems, Ltd.
   */
  #include "qemu/osdep.h"
@@ -XXX,XX +XXX,XX @@ bool qemu_co_queue_empty(CoQueue *queue)
      return QSIMPLEQ_FIRST(&queue->entries) == NULL;
  }
 +/* The wait records are handled with a multiple-producer, single-consumer
 + * lock-free queue.  There cannot be two concurrent pop_waiter() calls
 + * because pop_waiter() can only be called while mutex->handoff is zero.
 + * This can happen in three cases:
 + * - in qemu_co_mutex_unlock, before the hand-off protocol has started.
 + *   In this case, qemu_co_mutex_lock will see mutex->handoff == 0 and
 + *   not take part in the handoff.
 + * - in qemu_co_mutex_lock, if it steals the hand-off responsibility from
 + *   qemu_co_mutex_unlock.  In this case, qemu_co_mutex_unlock will fail
 + *   the cmpxchg (it will see either 0 or the next sequence value) and
 + *   exit.  The next hand-off cannot begin until qemu_co_mutex_lock has
 + *   woken up someone.
 + * - in qemu_co_mutex_unlock, if it takes the hand-off token itself.
 + *   In this case another iteration starts with mutex->handoff == 0;
 + *   a concurrent qemu_co_mutex_lock will fail the cmpxchg, and
 + *   qemu_co_mutex_unlock will go back to case (1).
 + *
 + * The following functions manage this queue.
 + */
 +typedef struct CoWaitRecord {
 +    Coroutine *co;
 +    QSLIST_ENTRY(CoWaitRecord) next;
 +} CoWaitRecord;
 +
 +static void push_waiter(CoMutex *mutex, CoWaitRecord *w)
 +{
 +    w->co = qemu_coroutine_self();
 +    QSLIST_INSERT_HEAD_ATOMIC(&mutex->from_push, w, next);
 +}
 +
 +static void move_waiters(CoMutex *mutex)
 +{
 +    QSLIST_HEAD(, CoWaitRecord) reversed;
 +    QSLIST_MOVE_ATOMIC(&reversed, &mutex->from_push);
 +    while (!QSLIST_EMPTY(&reversed)) {
 +        CoWaitRecord *w = QSLIST_FIRST(&reversed);
 +        QSLIST_REMOVE_HEAD(&reversed, next);
 +        QSLIST_INSERT_HEAD(&mutex->to_pop, w, next);
 +    }
 +}
 +
 +static CoWaitRecord *pop_waiter(CoMutex *mutex)
 +{
 +    CoWaitRecord *w;
 +
 +    if (QSLIST_EMPTY(&mutex->to_pop)) {
 +        move_waiters(mutex);
 +        if (QSLIST_EMPTY(&mutex->to_pop)) {
 +            return NULL;
 +        }
 +    }
 +    w = QSLIST_FIRST(&mutex->to_pop);
 +    QSLIST_REMOVE_HEAD(&mutex->to_pop, next);
 +    return w;
 +}
 +
 +static bool has_waiters(CoMutex *mutex)
 +{
 +    return QSLIST_EMPTY(&mutex->to_pop) || QSLIST_EMPTY(&mutex->from_push);
 +}
 +
  void qemu_co_mutex_init(CoMutex *mutex)
  {
-     BDRVQcow2State *s = bs->opaque;
+     memset(mutex, 0, sizeof(*mutex));
--    bool found;
+-    qemu_co_queue_init(&mutex->queue);
--    Qcow2BitmapList *bm_list;
+ }
-+    BdrvDirtyBitmap *bitmap;
-+    uint64_t bitmap_directory_size = 0;
+-void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex)
-+    uint32_t nb_bitmaps = 0;
++static void coroutine_fn qemu_co_mutex_lock_slowpath(CoMutex *mutex)
-+
+ {
-+    if (bdrv_find_dirty_bitmap(bs, name)) {
+     Coroutine *self = qemu_coroutine_self();
-+        error_setg(errp, "Bitmap already exists: %s", name);
++    CoWaitRecord w;
-+        return false;
++    unsigned old_handoff;
-+    }
+     trace_qemu_co_mutex_lock_entry(mutex, self);
-     if (s->qcow_version < 3) {
++    w.co = self;
-         /* Without autoclear_features, we would always have to assume
++    push_waiter(mutex, &w);
-@@ -XXX,XX +XXX,XX @@ bool coroutine_fn qcow2_co_can_store_new_dirty_bitmap(BlockDriverState *bs,
-         goto fail;
+-    while (mutex->locked) {
 -        qemu_co_queue_wait(&mutex->queue);
 +    /* This is the "Responsibility Hand-Off" protocol; a lock() picks from
 +     * a concurrent unlock() the responsibility of waking somebody up.
 +     */
 +    old_handoff = atomic_mb_read(&mutex->handoff);
 +    if (old_handoff &&
 +        has_waiters(mutex) &&
 +        atomic_cmpxchg(&mutex->handoff, old_handoff, 0) == old_handoff) {
 +        /* There can be no concurrent pops, because there can be only
 +         * one active handoff at a time.
 +         */
 +        CoWaitRecord *to_wake = pop_waiter(mutex);
 +        Coroutine *co = to_wake->co;
 +        if (co == self) {
 +            /* We got the lock ourselves!  */
 +            assert(to_wake == &w);
 +            return;
 +        }
 +
 +        aio_co_wake(co);
      }
--    if (s->nb_bitmaps == 0) {
+-    mutex->locked = true;
--        return true;
+-    mutex->holder = self;
-+    FOR_EACH_DIRTY_BITMAP(bs, bitmap) {
+-    self->locks_held++;
 +        if (bdrv_dirty_bitmap_get_persistence(bitmap)) {
 +            nb_bitmaps++;
 +            bitmap_directory_size +=
 +                calc_dir_entry_size(strlen(bdrv_dirty_bitmap_name(bitmap)), 0);
 +        }
      }
 +    nb_bitmaps++;
 +    bitmap_directory_size += calc_dir_entry_size(strlen(name), 0);
 -    if (s->nb_bitmaps >= QCOW2_MAX_BITMAPS) {
 +    if (nb_bitmaps > QCOW2_MAX_BITMAPS) {
          error_setg(errp,
                     "Maximum number of persistent bitmaps is already reached");
          goto fail;
      }
 -    if (s->bitmap_directory_size + calc_dir_entry_size(strlen(name), 0) >
 -        QCOW2_MAX_BITMAP_DIRECTORY_SIZE)
 -    {
 +    if (bitmap_directory_size > QCOW2_MAX_BITMAP_DIRECTORY_SIZE) {
          error_setg(errp, "Not enough space in the bitmap directory");
          goto fail;
      }
 -    qemu_co_mutex_lock(&s->lock);
 -    bm_list = bitmap_list_load(bs, s->bitmap_directory_offset,
 -                               s->bitmap_directory_size, errp);
 -    qemu_co_mutex_unlock(&s->lock);
 -    if (bm_list == NULL) {
 -        goto fail;
 -    }
 -
--    found = find_bitmap_by_name(bm_list, name);
++    qemu_coroutine_yield();
--    bitmap_list_free(bm_list);
+     trace_qemu_co_mutex_lock_return(mutex, self);
--    if (found) {
+ }
--        error_setg(errp, "Bitmap with the same name is already stored");
--        goto fail;
++void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex)
--    }
++{
--
++    Coroutine *self = qemu_coroutine_self();
-     return true;
++
++    if (atomic_fetch_inc(&mutex->locked) == 0) {
- fail:
++        /* Uncontended.  */
 +        trace_qemu_co_mutex_lock_uncontended(mutex, self);
 +    } else {
 +        qemu_co_mutex_lock_slowpath(mutex);
 +    }
 +    mutex->holder = self;
 +    self->locks_held++;
 +}
 +
  void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex)
  {
      Coroutine *self = qemu_coroutine_self();
      trace_qemu_co_mutex_unlock_entry(mutex, self);
 -    assert(mutex->locked == true);
 +    assert(mutex->locked);
      assert(mutex->holder == self);
      assert(qemu_in_coroutine());
 -    mutex->locked = false;
      mutex->holder = NULL;
      self->locks_held--;
 -    qemu_co_queue_next(&mutex->queue);
 +    if (atomic_fetch_dec(&mutex->locked) == 1) {
 +        /* No waiting qemu_co_mutex_lock().  Pfew, that was easy!  */
 +        return;
 +    }
 +
 +    for (;;) {
 +        CoWaitRecord *to_wake = pop_waiter(mutex);
 +        unsigned our_handoff;
 +
 +        if (to_wake) {
 +            Coroutine *co = to_wake->co;
 +            aio_co_wake(co);
 +            break;
 +        }
 +
 +        /* Some concurrent lock() is in progress (we know this because
 +         * mutex->locked was >1) but it hasn't yet put itself on the wait
 +         * queue.  Pick a sequence number for the handoff protocol (not 0).
 +         */
 +        if (++mutex->sequence == 0) {
 +            mutex->sequence = 1;
 +        }
 +
 +        our_handoff = mutex->sequence;
 +        atomic_mb_set(&mutex->handoff, our_handoff);
 +        if (!has_waiters(mutex)) {
 +            /* The concurrent lock has not added itself yet, so it
 +             * will be able to pick our handoff.
 +             */
 +            break;
 +        }
 +
 +        /* Try to do the handoff protocol ourselves; if somebody else has
 +         * already taken it, however, we're done and they're responsible.
 +         */
 +        if (atomic_cmpxchg(&mutex->handoff, our_handoff, 0) != our_handoff) {
 +            break;
 +        }
 +    }
      trace_qemu_co_mutex_unlock_return(mutex, self);
  }
 diff --git a/util/trace-events b/util/trace-events
 index XXXXXXX..XXXXXXX 100644
 --- a/util/trace-events
 +++ b/util/trace-events
@@ -XXX,XX +XXX,XX @@ qemu_coroutine_terminate(void *co) "self %p"
  # util/qemu-coroutine-lock.c
  qemu_co_queue_run_restart(void *co) "co %p"
 +qemu_co_mutex_lock_uncontended(void *mutex, void *self) "mutex %p self %p"
  qemu_co_mutex_lock_entry(void *mutex, void *self) "mutex %p self %p"
  qemu_co_mutex_lock_return(void *mutex, void *self) "mutex %p self %p"
  qemu_co_mutex_unlock_entry(void *mutex, void *self) "mutex %p self %p"
 --
-.24.1
+.9.3

-[PULL 08/34] iotests: s/qocw2/qcow2/
+Deleted patch
-Probably due to blind copy-pasting, we have several instances of "qocw2"
-in our iotests.  Fix them.
-Reported-by: Maxim Levitsky <mlevitsk@redhat.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Message-id: 20191107163708.833192-2-mreitz@redhat.com
-Reviewed-by: Eric Blake <eblake@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/060 | 2 +-
- tests/qemu-iotests/061 | 2 +-
- tests/qemu-iotests/062 | 2 +-
- tests/qemu-iotests/066 | 2 +-
- tests/qemu-iotests/068 | 2 +-
- tests/qemu-iotests/108 | 2 +-
- tests/qemu-iotests/138 | 2 +-
- tests/qemu-iotests/261 | 2 +-
-files changed, 8 insertions(+), 8 deletions(-)
-diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/060
-+++ b/tests/qemu-iotests/060
-@@ -XXX,XX +XXX,XX @@ _filter_io_error()
- . ./common.rc
- . ./common.filter
--# This tests qocw2-specific low-level functionality
-+# This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto file
- _supported_os Linux
-diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/061
-+++ b/tests/qemu-iotests/061
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- . ./common.rc
- . ./common.filter
--# This tests qocw2-specific low-level functionality
-+# This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto file
- _supported_os Linux
-diff --git a/tests/qemu-iotests/062 b/tests/qemu-iotests/062
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/062
-+++ b/tests/qemu-iotests/062
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- . ./common.rc
- . ./common.filter
--# This tests qocw2-specific low-level functionality
-+# This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto generic
-diff --git a/tests/qemu-iotests/066 b/tests/qemu-iotests/066
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/066
-+++ b/tests/qemu-iotests/066
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- . ./common.rc
- . ./common.filter
--# This tests qocw2-specific low-level functionality
-+# This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto generic
-diff --git a/tests/qemu-iotests/068 b/tests/qemu-iotests/068
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/068
-+++ b/tests/qemu-iotests/068
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- . ./common.rc
- . ./common.filter
--# This tests qocw2-specific low-level functionality
-+# This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto generic
-diff --git a/tests/qemu-iotests/108 b/tests/qemu-iotests/108
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/108
-+++ b/tests/qemu-iotests/108
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- . ./common.rc
- . ./common.filter
--# This tests qocw2-specific low-level functionality
-+# This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto file
- _supported_os Linux
-diff --git a/tests/qemu-iotests/138 b/tests/qemu-iotests/138
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/138
-+++ b/tests/qemu-iotests/138
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- . ./common.rc
- . ./common.filter
--# This tests qocw2-specific low-level functionality
-+# This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto file
- _supported_os Linux
-diff --git a/tests/qemu-iotests/261 b/tests/qemu-iotests/261
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/261
-+++ b/tests/qemu-iotests/261
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- . ./common.rc
- . ./common.filter
--# This tests qocw2-specific low-level functionality
-+# This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto file
- _supported_os Linux
---
-.24.1

-[PULL 09/34] iotests/qcow2.py: Add dump-header-exts
+Deleted patch
-This is useful for tests that want to whitelist fields from dump-header
-(with grep) but still print all header extensions.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-3-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/qcow2.py | 5 +++++
-file changed, 5 insertions(+)
-diff --git a/tests/qemu-iotests/qcow2.py b/tests/qemu-iotests/qcow2.py
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/qcow2.py
-+++ b/tests/qemu-iotests/qcow2.py
-@@ -XXX,XX +XXX,XX @@ def cmd_dump_header(fd):
-     h.dump()
-     h.dump_extensions()
-+def cmd_dump_header_exts(fd):
-+    h = QcowHeader(fd)
-+    h.dump_extensions()
-+
- def cmd_set_header(fd, name, value):
-     try:
-         value = int(value, 0)
-@@ -XXX,XX +XXX,XX @@ def cmd_set_feature_bit(fd, group, bit):
- cmds = [
-     [ 'dump-header',          cmd_dump_header,          0, 'Dump image header and header extensions' ],
-+    [ 'dump-header-exts',     cmd_dump_header_exts,     0, 'Dump image header extensions' ],
-     [ 'set-header',           cmd_set_header,           2, 'Set a field in the header'],
-     [ 'add-header-ext',       cmd_add_header_ext,       2, 'Add a header extension' ],
-     [ 'add-header-ext-stdio', cmd_add_header_ext_stdio, 1, 'Add a header extension, data from stdin' ],
---
-.24.1

-[PULL 10/34] iotests/qcow2.py: Split feature fields into bits
+Deleted patch
-Print the feature fields as a set of bits so that filtering is easier.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-4-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/031.out  | 36 +++++++++----------
- tests/qemu-iotests/036.out  | 18 +++++-----
- tests/qemu-iotests/039.out  | 22 ++++++------
- tests/qemu-iotests/060.out  | 20 +++++------
- tests/qemu-iotests/061.out  | 72 ++++++++++++++++++-------------------
- tests/qemu-iotests/137.out  |  2 +-
- tests/qemu-iotests/qcow2.py | 18 +++++++---
-files changed, 99 insertions(+), 89 deletions(-)
-diff --git a/tests/qemu-iotests/031.out b/tests/qemu-iotests/031.out
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/031.out
-+++ b/tests/qemu-iotests/031.out
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             72
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             72
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             72
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             104
-diff --git a/tests/qemu-iotests/036.out b/tests/qemu-iotests/036.out
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/036.out
-+++ b/tests/qemu-iotests/036.out
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x8000000000000000
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     [63]
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x8000000000000000
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        [63]
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             104
-diff --git a/tests/qemu-iotests/039.out b/tests/qemu-iotests/039.out
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/039.out
-+++ b/tests/qemu-iotests/039.out
-@@ -XXX,XX +XXX,XX @@ QA output created by 039
- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
--incompatible_features     0x0
-+incompatible_features     []
- No errors were found on the image.
- == Creating a dirty image file ==
-@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
- ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
--incompatible_features     0x1
-+incompatible_features     [0]
- ERROR cluster 5 refcount=0 reference=1
- ERROR OFLAG_COPIED data cluster: l2_entry=8000000000050000 refcount=0
-@@ -XXX,XX +XXX,XX @@ Data may be corrupted, or further writes to the image may corrupt it.
- == Read-only access must still work ==
- read 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
--incompatible_features     0x1
-+incompatible_features     [0]
- == Repairing the image file must succeed ==
- ERROR cluster 5 refcount=0 reference=1
-@@ -XXX,XX +XXX,XX @@ The following inconsistencies were found and repaired:
- Double checking the fixed image now...
- No errors were found on the image.
--incompatible_features     0x0
-+incompatible_features     []
- == Data should still be accessible after repair ==
- read 512/512 bytes at offset 0
-@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
- ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
--incompatible_features     0x1
-+incompatible_features     [0]
- ERROR cluster 5 refcount=0 reference=1
- Rebuilding refcount structure
- Repairing cluster 1 refcount=1 reference=0
- Repairing cluster 2 refcount=1 reference=0
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
--incompatible_features     0x0
-+incompatible_features     []
- == Creating an image file with lazy_refcounts=off ==
- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
- ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
--incompatible_features     0x0
-+incompatible_features     []
- No errors were found on the image.
- == Committing to a backing file with lazy_refcounts=on ==
-@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
- Image committed.
--incompatible_features     0x0
--incompatible_features     0x0
-+incompatible_features     []
-+incompatible_features     []
- No errors were found on the image.
- No errors were found on the image.
-@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
- ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
--incompatible_features     0x1
-+incompatible_features     [0]
- ERROR cluster 5 refcount=0 reference=1
- ERROR OFLAG_COPIED data cluster: l2_entry=8000000000050000 refcount=0
-@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
- ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
--incompatible_features     0x0
-+incompatible_features     []
- No errors were found on the image.
- *** done
-diff --git a/tests/qemu-iotests/060.out b/tests/qemu-iotests/060.out
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/060.out
-+++ b/tests/qemu-iotests/060.out
-@@ -XXX,XX +XXX,XX @@ ERROR cluster 3 refcount=1 reference=3
-errors were found on the image.
- Data may be corrupted, or further writes to the image may corrupt it.
--incompatible_features     0x0
-+incompatible_features     []
- qcow2: Marking image as corrupt: Preventing invalid write on metadata (overlaps with active L1 table); further corruption events will be suppressed
- write failed: Input/output error
--incompatible_features     0x2
-+incompatible_features     [1]
- image: TEST_DIR/t.IMGFMT
- file format: IMGFMT
- virtual size: 64 MiB (67108864 bytes)
-@@ -XXX,XX +XXX,XX @@ ERROR cluster 2 refcount=1 reference=2
-errors were found on the image.
- Data may be corrupted, or further writes to the image may corrupt it.
--incompatible_features     0x0
-+incompatible_features     []
- qcow2: Marking image as corrupt: Preventing invalid write on metadata (overlaps with refcount block); further corruption events will be suppressed
- write failed: Input/output error
--incompatible_features     0x2
-+incompatible_features     [1]
- ERROR refcount block 0 refcount=2
- ERROR cluster 2 refcount=1 reference=2
- Rebuilding refcount structure
-@@ -XXX,XX +XXX,XX @@ The following inconsistencies were found and repaired:
- Double checking the fixed image now...
- No errors were found on the image.
--incompatible_features     0x0
-+incompatible_features     []
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
--incompatible_features     0x0
-+incompatible_features     []
- === Testing cluster data reference into inactive L2 table ===
-@@ -XXX,XX +XXX,XX @@ Data may be corrupted, or further writes to the image may corrupt it.
-leaked clusters were found on the image.
- This means waste of disk space, but no harm to data.
--incompatible_features     0x0
-+incompatible_features     []
- qcow2: Marking image as corrupt: Preventing invalid write on metadata (overlaps with inactive L2 table); further corruption events will be suppressed
- write failed: Input/output error
--incompatible_features     0x2
-+incompatible_features     [1]
- ERROR cluster 4 refcount=1 reference=2
- Leaked cluster 9 refcount=1 reference=0
- Repairing cluster 4 refcount=1 reference=2
-@@ -XXX,XX +XXX,XX @@ The following inconsistencies were found and repaired:
- Double checking the fixed image now...
- No errors were found on the image.
--incompatible_features     0x0
-+incompatible_features     []
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
--incompatible_features     0x0
-+incompatible_features     []
- read 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
- No errors were found on the image.
-diff --git a/tests/qemu-iotests/061.out b/tests/qemu-iotests/061.out
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/061.out
-+++ b/tests/qemu-iotests/061.out
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x1
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       [0]
-+autoclear_features        []
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             72
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x1
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       [0]
-+autoclear_features        []
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             72
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x1
--compatible_features       0x1
--autoclear_features        0x0
-+incompatible_features     [0]
-+compatible_features       [0]
-+autoclear_features        []
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x80000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             72
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x40000000000
--autoclear_features        0x40000000000
-+incompatible_features     []
-+compatible_features       [42]
-+autoclear_features        [42]
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             72
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             72
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x1
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       [0]
-+autoclear_features        []
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x1
--compatible_features       0x1
--autoclear_features        0x0
-+incompatible_features     [0]
-+compatible_features       [0]
-+autoclear_features        []
- refcount_order            4
- header_length             104
-@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x80000
- refcount_table_clusters   1
- nb_snapshots              0
- snapshot_offset           0x0
--incompatible_features     0x0
--compatible_features       0x0
--autoclear_features        0x0
-+incompatible_features     []
-+compatible_features       []
-+autoclear_features        []
- refcount_order            4
- header_length             104
-diff --git a/tests/qemu-iotests/137.out b/tests/qemu-iotests/137.out
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/137.out
-+++ b/tests/qemu-iotests/137.out
-@@ -XXX,XX +XXX,XX @@ qemu-io: Unsupported value 'blubb' for qcow2 option 'overlap-check'. Allowed are
- wrote 512/512 bytes at offset 0
-bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
- ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
--incompatible_features     0x0
-+incompatible_features     []
- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
- wrote 65536/65536 bytes at offset 0
-KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-diff --git a/tests/qemu-iotests/qcow2.py b/tests/qemu-iotests/qcow2.py
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/qcow2.py
-+++ b/tests/qemu-iotests/qcow2.py
-@@ -XXX,XX +XXX,XX @@ class QcowHeader:
-         [ uint64_t, '%#x',  'snapshot_offset' ],
-         # Version 3 header fields
--        [ uint64_t, '%#x',  'incompatible_features' ],
--        [ uint64_t, '%#x',  'compatible_features' ],
--        [ uint64_t, '%#x',  'autoclear_features' ],
-+        [ uint64_t, 'mask', 'incompatible_features' ],
-+        [ uint64_t, 'mask', 'compatible_features' ],
-+        [ uint64_t, 'mask', 'autoclear_features' ],
-         [ uint32_t, '%d',   'refcount_order' ],
-         [ uint32_t, '%d',   'header_length' ],
-     ];
-@@ -XXX,XX +XXX,XX @@ class QcowHeader:
-     def dump(self):
-         for f in QcowHeader.fields:
--            print("%-25s" % f[2], f[1] % self.__dict__[f[2]])
-+            value = self.__dict__[f[2]]
-+            if f[1] == 'mask':
-+                bits = []
-+                for bit in range(64):
-+                    if value & (1 << bit):
-+                        bits.append(bit)
-+                value_str = str(bits)
-+            else:
-+                value_str = f[1] % value
-+
-+            print("%-25s" % f[2], value_str)
-         print("")
-     def dump_extensions(self):
---
-.24.1

-[PULL 11/34] iotests: Add _filter_json_filename
+[Qemu-devel] [PULL v2 20/24] coroutine-lock: add limited spinning to CoMutex
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+From: Paolo Bonzini <pbonzini@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-5-mreitz@redhat.com
+Running a very small critical section on pthread_mutex_t and CoMutex
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+shows that pthread_mutex_t is much faster because it doesn't actually
 go to sleep.  What happens is that the critical section is shorter
 than the latency of entering the kernel and thus FUTEX_WAIT always
 fails.  With CoMutex there is no such latency but you still want to
 avoid wait and wakeup.  So introduce it artificially.
 This only works with one waiters; because CoMutex is fair, it will
 always have more waits and wakeups than a pthread_mutex_t.
 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Message-id: 20170213181244.16297-3-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/common.filter | 24 ++++++++++++++++++++++++
+ include/qemu/coroutine.h   |  5 +++++
-file changed, 24 insertions(+)
+ util/qemu-coroutine-lock.c | 51 ++++++++++++++++++++++++++++++++++++++++------
  util/qemu-coroutine.c      |  2 +-
 files changed, 51 insertions(+), 7 deletions(-)
-diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
+diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
 index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/common.filter
+--- a/include/qemu/coroutine.h
-+++ b/tests/qemu-iotests/common.filter
++++ b/include/qemu/coroutine.h
-@@ -XXX,XX +XXX,XX @@ _filter_qmp_empty_return()
+@@ -XXX,XX +XXX,XX @@ typedef struct CoMutex {
-     grep -v '{"return": {}}'
+      */
      unsigned locked;
 +    /* Context that is holding the lock.  Useful to avoid spinning
 +     * when two coroutines on the same AioContext try to get the lock. :)
 +     */
 +    AioContext *ctx;
 +
      /* A queue of waiters.  Elements are added atomically in front of
       * from_push.  to_pop is only populated, and popped from, by whoever
       * is in charge of the next wakeup.  This can be an unlocker or,
 diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/qemu-coroutine-lock.c
 +++ b/util/qemu-coroutine-lock.c
@@ -XXX,XX +XXX,XX @@
  #include "qemu-common.h"
  #include "qemu/coroutine.h"
  #include "qemu/coroutine_int.h"
 +#include "qemu/processor.h"
  #include "qemu/queue.h"
  #include "block/aio.h"
  #include "trace.h"
@@ -XXX,XX +XXX,XX @@ void qemu_co_mutex_init(CoMutex *mutex)
      memset(mutex, 0, sizeof(*mutex));
  }
-+_filter_json_filename()
+-static void coroutine_fn qemu_co_mutex_lock_slowpath(CoMutex *mutex)
 +static void coroutine_fn qemu_co_mutex_wake(CoMutex *mutex, Coroutine *co)
 +{
-+    $PYTHON -c 'import sys
++    /* Read co before co->ctx; pairs with smp_wmb() in
-+result, *fnames = sys.stdin.read().split("json:{")
++     * qemu_coroutine_enter().
-+depth = 0
++     */
-+for fname in fnames:
++    smp_read_barrier_depends();
-+    depth += 1 # For the opening brace in the split separator
++    mutex->ctx = co->ctx;
-+    for chr_i, chr in enumerate(fname):
++    aio_co_wake(co);
 +        if chr == "{":
 +            depth += 1
 +        elif chr == "}":
 +            depth -= 1
 +            if depth == 0:
 +                break
 +
 +    # json:{} filenames may be nested; filter out everything from
 +    # inside the outermost one
 +    if depth == 0:
 +        chr_i += 1 # First character past the filename
 +        result += "json:{ /* filtered */ }" + fname[chr_i:]
 +
 +sys.stdout.write(result)'
 +}
 +
- # make sure this script returns success
++static void coroutine_fn qemu_co_mutex_lock_slowpath(AioContext *ctx,
- true
++                                                     CoMutex *mutex)
  {
      Coroutine *self = qemu_coroutine_self();
      CoWaitRecord w;
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn qemu_co_mutex_lock_slowpath(CoMutex *mutex)
          if (co == self) {
              /* We got the lock ourselves!  */
              assert(to_wake == &w);
 +            mutex->ctx = ctx;
              return;
          }
 -        aio_co_wake(co);
 +        qemu_co_mutex_wake(mutex, co);
      }
      qemu_coroutine_yield();
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn qemu_co_mutex_lock_slowpath(CoMutex *mutex)
  void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex)
  {
 +    AioContext *ctx = qemu_get_current_aio_context();
      Coroutine *self = qemu_coroutine_self();
 +    int waiters, i;
 -    if (atomic_fetch_inc(&mutex->locked) == 0) {
 +    /* Running a very small critical section on pthread_mutex_t and CoMutex
 +     * shows that pthread_mutex_t is much faster because it doesn't actually
 +     * go to sleep.  What happens is that the critical section is shorter
 +     * than the latency of entering the kernel and thus FUTEX_WAIT always
 +     * fails.  With CoMutex there is no such latency but you still want to
 +     * avoid wait and wakeup.  So introduce it artificially.
 +     */
 +    i = 0;
 +retry_fast_path:
 +    waiters = atomic_cmpxchg(&mutex->locked, 0, 1);
 +    if (waiters != 0) {
 +        while (waiters == 1 && ++i < 1000) {
 +            if (atomic_read(&mutex->ctx) == ctx) {
 +                break;
 +            }
 +            if (atomic_read(&mutex->locked) == 0) {
 +                goto retry_fast_path;
 +            }
 +            cpu_relax();
 +        }
 +        waiters = atomic_fetch_inc(&mutex->locked);
 +    }
 +
 +    if (waiters == 0) {
          /* Uncontended.  */
          trace_qemu_co_mutex_lock_uncontended(mutex, self);
 +        mutex->ctx = ctx;
      } else {
 -        qemu_co_mutex_lock_slowpath(mutex);
 +        qemu_co_mutex_lock_slowpath(ctx, mutex);
      }
      mutex->holder = self;
      self->locks_held++;
@@ -XXX,XX +XXX,XX @@ void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex)
      assert(mutex->holder == self);
      assert(qemu_in_coroutine());
 +    mutex->ctx = NULL;
      mutex->holder = NULL;
      self->locks_held--;
      if (atomic_fetch_dec(&mutex->locked) == 1) {
@@ -XXX,XX +XXX,XX @@ void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex)
          unsigned our_handoff;
          if (to_wake) {
 -            Coroutine *co = to_wake->co;
 -            aio_co_wake(co);
 +            qemu_co_mutex_wake(mutex, to_wake->co);
              break;
          }
 diff --git a/util/qemu-coroutine.c b/util/qemu-coroutine.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/qemu-coroutine.c
 +++ b/util/qemu-coroutine.c
@@ -XXX,XX +XXX,XX @@ void qemu_coroutine_enter(Coroutine *co)
      co->ctx = qemu_get_current_aio_context();
      /* Store co->ctx before anything that stores co.  Matches
 -     * barrier in aio_co_wake.
 +     * barrier in aio_co_wake and qemu_co_mutex_wake.
       */
      smp_wmb();
 --
-.24.1
+.9.3

-[PULL 12/34] iotests: Filter refcount_order in 036
+Deleted patch
-This test can run just fine with other values for refcount_bits, so we
-should filter the value from qcow2.py's dump-header.  In fact, we can
-filter everything but the feature bits and header extensions, because
-that is what the test is about.
-(036 currently ignores user-specified image options, but that will be
-fixed in the next patch.)
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-6-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/036     |  9 ++++---
- tests/qemu-iotests/036.out | 48 --------------------------------------
-files changed, 6 insertions(+), 51 deletions(-)
-diff --git a/tests/qemu-iotests/036 b/tests/qemu-iotests/036
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/036
-+++ b/tests/qemu-iotests/036
-@@ -XXX,XX +XXX,XX @@ $PYTHON qcow2.py "$TEST_IMG" set-feature-bit incompatible 63
- # Without feature table
- $PYTHON qcow2.py "$TEST_IMG" del-header-ext 0x6803f857
--$PYTHON qcow2.py "$TEST_IMG" dump-header
-+$PYTHON qcow2.py "$TEST_IMG" dump-header | grep features
-+$PYTHON qcow2.py "$TEST_IMG" dump-header-exts
- _img_info
- # With feature table containing bit 63
-@@ -XXX,XX +XXX,XX @@ echo === Create image with unknown autoclear feature bit ===
- echo
- _make_test_img 64M
- $PYTHON qcow2.py "$TEST_IMG" set-feature-bit autoclear 63
--$PYTHON qcow2.py "$TEST_IMG" dump-header
-+$PYTHON qcow2.py "$TEST_IMG" dump-header | grep features
-+$PYTHON qcow2.py "$TEST_IMG" dump-header-exts
- echo
- echo === Repair image ===
- echo
- _check_test_img -r all
--$PYTHON qcow2.py "$TEST_IMG" dump-header
-+$PYTHON qcow2.py "$TEST_IMG" dump-header | grep features
-+$PYTHON qcow2.py "$TEST_IMG" dump-header-exts
- # success, all done
- echo "*** done"
-diff --git a/tests/qemu-iotests/036.out b/tests/qemu-iotests/036.out
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/036.out
-+++ b/tests/qemu-iotests/036.out
-@@ -XXX,XX +XXX,XX @@ QA output created by 036
- === Image with unknown incompatible feature bit ===
- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
--magic                     0x514649fb
--version                   3
--backing_file_offset       0x0
--backing_file_size         0x0
--cluster_bits              16
--size                      67108864
--crypt_method              0
--l1_size                   1
--l1_table_offset           0x30000
--refcount_table_offset     0x10000
--refcount_table_clusters   1
--nb_snapshots              0
--snapshot_offset           0x0
- incompatible_features     [63]
- compatible_features       []
- autoclear_features        []
--refcount_order            4
--header_length             104
--
- qemu-img: Could not open 'TEST_DIR/t.IMGFMT': Unsupported IMGFMT feature(s): Unknown incompatible feature: 8000000000000000
- qemu-img: Could not open 'TEST_DIR/t.IMGFMT': Unsupported IMGFMT feature(s): Test feature
-@@ -XXX,XX +XXX,XX @@ qemu-img: Could not open 'TEST_DIR/t.IMGFMT': Unsupported IMGFMT feature(s): tes
- === Create image with unknown autoclear feature bit ===
- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
--magic                     0x514649fb
--version                   3
--backing_file_offset       0x0
--backing_file_size         0x0
--cluster_bits              16
--size                      67108864
--crypt_method              0
--l1_size                   1
--l1_table_offset           0x30000
--refcount_table_offset     0x10000
--refcount_table_clusters   1
--nb_snapshots              0
--snapshot_offset           0x0
- incompatible_features     []
- compatible_features       []
- autoclear_features        [63]
--refcount_order            4
--header_length             104
--
- Header extension:
- magic                     0x6803f857
- length                    192
-@@ -XXX,XX +XXX,XX @@ data                      <binary>
- === Repair image ===
- No errors were found on the image.
--magic                     0x514649fb
--version                   3
--backing_file_offset       0x0
--backing_file_size         0x0
--cluster_bits              16
--size                      67108864
--crypt_method              0
--l1_size                   1
--l1_table_offset           0x30000
--refcount_table_offset     0x10000
--refcount_table_clusters   1
--nb_snapshots              0
--snapshot_offset           0x0
- incompatible_features     []
- compatible_features       []
- autoclear_features        []
--refcount_order            4
--header_length             104
--
- Header extension:
- magic                     0x6803f857
- length                    192
---
-.24.1

-[PULL 13/34] iotests: Replace IMGOPTS by _unsupported_imgopts
+Deleted patch
-Some tests require compat=1.1 and thus set IMGOPTS='compat=1.1'
-globally.  That is not how it should be done; instead, they should
-simply set _unsupported_imgopts to compat=0.10 (compat=1.1 is the
-default anyway).
-This makes the tests heed user-specified $IMGOPTS.  Some do not work
-with all image options, though, so we need to disable them accordingly.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsky@redhat.com>
-Message-id: 20191107163708.833192-7-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/036 | 3 +--
- tests/qemu-iotests/060 | 4 ++--
- tests/qemu-iotests/062 | 3 ++-
- tests/qemu-iotests/066 | 3 ++-
- tests/qemu-iotests/068 | 3 ++-
- tests/qemu-iotests/098 | 4 ++--
-files changed, 11 insertions(+), 9 deletions(-)
-diff --git a/tests/qemu-iotests/036 b/tests/qemu-iotests/036
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/036
-+++ b/tests/qemu-iotests/036
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- # This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto file
--
- # Only qcow2v3 and later supports feature bits
--IMGOPTS="compat=1.1"
-+_unsupported_imgopts 'compat=0.10'
- echo
- echo === Image with unknown incompatible feature bit ===
-diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/060
-+++ b/tests/qemu-iotests/060
-@@ -XXX,XX +XXX,XX @@ _filter_io_error()
- _supported_fmt qcow2
- _supported_proto file
- _supported_os Linux
-+# These tests only work for compat=1.1 images with refcount_bits=16
-+_unsupported_imgopts 'compat=0.10' 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
- # The repair process will create a large file - so check for availability first
- _require_large_file 64G
-@@ -XXX,XX +XXX,XX @@ l1_offset=196608 # 0x30000 (XXX: just an assumption)
- l2_offset=262144 # 0x40000 (XXX: just an assumption)
- l2_offset_after_snapshot=524288 # 0x80000 (XXX: just an assumption)
--IMGOPTS="compat=1.1"
--
- OPEN_RW="open -o overlap-check=all $TEST_IMG"
- # Overlap checks are done before write operations only, therefore opening an
- # image read-only makes the overlap-check option irrelevant
-diff --git a/tests/qemu-iotests/062 b/tests/qemu-iotests/062
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/062
-+++ b/tests/qemu-iotests/062
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- # This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto generic
-+# We need zero clusters and snapshots
-+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
--IMGOPTS="compat=1.1"
- IMG_SIZE=64M
- echo
-diff --git a/tests/qemu-iotests/066 b/tests/qemu-iotests/066
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/066
-+++ b/tests/qemu-iotests/066
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- # This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto generic
-+# We need zero clusters and snapshots
-+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
- # Intentionally create an unaligned image
--IMGOPTS="compat=1.1"
- IMG_SIZE=$((64 * 1024 * 1024 + 512))
- echo
-diff --git a/tests/qemu-iotests/068 b/tests/qemu-iotests/068
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/068
-+++ b/tests/qemu-iotests/068
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- # This tests qcow2-specific low-level functionality
- _supported_fmt qcow2
- _supported_proto generic
-+# Internal snapshots are (currently) impossible with refcount_bits=1
-+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
--IMGOPTS="compat=1.1"
- IMG_SIZE=128K
- case "$QEMU_DEFAULT_MACHINE" in
-diff --git a/tests/qemu-iotests/098 b/tests/qemu-iotests/098
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/098
-+++ b/tests/qemu-iotests/098
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- _supported_fmt qcow2
- _supported_proto file
--
--IMGOPTS="compat=1.1"
-+# The code path we want to test here only works for compat=1.1 images
-+_unsupported_imgopts 'compat=0.10'
- for event in l1_update empty_image_prepare reftable_update refblock_alloc; do
---
-.24.1

-[PULL 14/34] iotests: Drop compat=1.1 in 050
+Deleted patch
-IMGOPTS can never be empty for qcow2, because the check scripts adds
-compat=1.1 unless the user specified any compat option themselves.
-Thus, this block does not do anything and can be dropped.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-8-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/050 | 4 ----
-file changed, 4 deletions(-)
-diff --git a/tests/qemu-iotests/050 b/tests/qemu-iotests/050
-index XXXXXXX..XXXXXXX 100755
---- a/tests/qemu-iotests/050
-+++ b/tests/qemu-iotests/050
-@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
- _supported_fmt qcow2 qed
- _supported_proto file
--if test "$IMGFMT" = qcow2 && test $IMGOPTS = ""; then
--  IMGOPTS=compat=1.1
--fi
--
- echo
- echo "== Creating images =="
---
-.24.1

-[PULL 15/34] iotests: Let _make_test_img parse its parameters
+Deleted patch
-This will allow us to add more options than just -b.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-9-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/common.rc | 28 ++++++++++++++++++++--------
-file changed, 20 insertions(+), 8 deletions(-)
-diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/common.rc
-+++ b/tests/qemu-iotests/common.rc
-@@ -XXX,XX +XXX,XX @@ _make_test_img()
-     # extra qemu-img options can be added by tests
-     # at least one argument (the image size) needs to be added
-     local extra_img_options=""
--    local image_size=$*
-     local optstr=""
-     local img_name=""
-     local use_backing=0
-     local backing_file=""
-     local object_options=""
-+    local misc_params=()
-     if [ -n "$TEST_IMG_FILE" ]; then
-         img_name=$TEST_IMG_FILE
-@@ -XXX,XX +XXX,XX @@ _make_test_img()
-         optstr=$(_optstr_add "$optstr" "key-secret=keysec0")
-     fi
--    if [ "$1" = "-b" ]; then
--        use_backing=1
--        backing_file=$2
--        image_size=$3
--    fi
-+    for param; do
-+        if [ "$use_backing" = "1" -a -z "$backing_file" ]; then
-+            backing_file=$param
-+            continue
-+        fi
-+
-+        case "$param" in
-+            -b)
-+                use_backing=1
-+                ;;
-+
-+            *)
-+                misc_params=("${misc_params[@]}" "$param")
-+                ;;
-+        esac
-+    done
-+
-     if [ \( "$IMGFMT" = "qcow2" -o "$IMGFMT" = "qed" \) -a -n "$CLUSTER_SIZE" ]; then
-         optstr=$(_optstr_add "$optstr" "cluster_size=$CLUSTER_SIZE")
-     fi
-@@ -XXX,XX +XXX,XX @@ _make_test_img()
-     # XXX(hch): have global image options?
-     (
-      if [ $use_backing = 1 ]; then
--        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options -b "$backing_file" "$img_name" $image_size 2>&1
-+        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options -b "$backing_file" "$img_name" "${misc_params[@]}" 2>&1
-      else
--        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options "$img_name" $image_size 2>&1
-+        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options "$img_name" "${misc_params[@]}" 2>&1
-      fi
-     ) | _filter_img_create
---
-.24.1

-[PULL 16/34] iotests: Add -o and --no-opts to _make_test_img
+Deleted patch
-Blindly overriding IMGOPTS is suboptimal as this discards user-specified
-options.  Whatever options the test needs should simply be appended.
-Some tests do this (with IMGOPTS=$(_optstr_add "$IMGOPTS" "...")), but
-that is cumbersome.  It’s simpler to just give _make_test_img an -o
-parameter with which tests can add options.
-Some tests actually must override the user-specified options, though,
-for example when creating an image in a different format than the test
-$IMGFMT.  For such cases, --no-opts allows clearing the current option
-list.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-10-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
----
- tests/qemu-iotests/common.rc | 13 +++++++++++++
-file changed, 13 insertions(+)
-diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
-index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/common.rc
-+++ b/tests/qemu-iotests/common.rc
-@@ -XXX,XX +XXX,XX @@ _make_test_img()
-     local use_backing=0
-     local backing_file=""
-     local object_options=""
-+    local opts_param=false
-     local misc_params=()
-     if [ -n "$TEST_IMG_FILE" ]; then
-@@ -XXX,XX +XXX,XX @@ _make_test_img()
-         if [ "$use_backing" = "1" -a -z "$backing_file" ]; then
-             backing_file=$param
-             continue
-+        elif $opts_param; then
-+            optstr=$(_optstr_add "$optstr" "$param")
-+            opts_param=false
-+            continue
-         fi
-         case "$param" in
-@@ -XXX,XX +XXX,XX @@ _make_test_img()
-                 use_backing=1
-                 ;;
-+            -o)
-+                opts_param=true
-+                ;;
-+
-+            --no-opts)
-+                optstr=""
-+                ;;
-+
-             *)
-                 misc_params=("${misc_params[@]}" "$param")
-                 ;;
---
-.24.1

-[PULL 29/34] iotests: Allow check -o data_file
+[Qemu-devel] [PULL v2 21/24] test-aio-multithread: add performance comparison with thread-based mutexes
-The problem with allowing the data_file option is that you want to use a
+From: Paolo Bonzini <pbonzini@redhat.com>
-different data file per image used in the test.  Therefore, we need to
-allow patterns like -o data_file='$TEST_IMG.data_file'.
+Add two implementations of the same benchmark as the previous patch,
+but using pthreads.  One uses a normal QemuMutex, the other is Linux
-Then, we need to filter it out from qemu-img map, qemu-img create, and
+only and implements a fair mutex based on MCS locks and futexes.
-remove the data file in _rm_test_img.
+This shows that the slower performance of the 5-thread case is due to
+the fairness of CoMutex, rather than to coroutines.  If fairness does
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+not matter, as is the case with two threads, CoMutex can actually be
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+faster than pthreads.
-Message-id: 20191107163708.833192-23-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Message-id: 20170213181244.16297-4-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/common.filter | 23 +++++++++++++++++++++--
+ tests/test-aio-multithread.c | 164 +++++++++++++++++++++++++++++++++++++++++++
- tests/qemu-iotests/common.rc     | 22 +++++++++++++++++++++-
+file changed, 164 insertions(+)
-files changed, 42 insertions(+), 3 deletions(-)
+diff --git a/tests/test-aio-multithread.c b/tests/test-aio-multithread.c
 diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
 index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/common.filter
+--- a/tests/test-aio-multithread.c
-+++ b/tests/qemu-iotests/common.filter
++++ b/tests/test-aio-multithread.c
-@@ -XXX,XX +XXX,XX @@ _filter_actual_image_size()
+@@ -XXX,XX +XXX,XX @@ static void test_multi_co_mutex_2_30(void)
- # replace driver-specific options in the "Formatting..." line
+     test_multi_co_mutex(2, 30);
  _filter_img_create()
  {
 -    $SED -e "s#$REMOTE_TEST_DIR#TEST_DIR#g" \
 +    data_file_filter=()
 +    if data_file=$(_get_data_file "$TEST_IMG"); then
 +        data_file_filter=(-e "s# data_file=$data_file##")
 +    fi
 +
 +    $SED "${data_file_filter[@]}" \
 +        -e "s#$REMOTE_TEST_DIR#TEST_DIR#g" \
          -e "s#$IMGPROTO:$TEST_DIR#TEST_DIR#g" \
          -e "s#$TEST_DIR#TEST_DIR#g" \
          -e "s#$SOCK_DIR#SOCK_DIR#g" \
@@ -XXX,XX +XXX,XX @@ _filter_img_info()
  # human and json output
  _filter_qemu_img_map()
  {
 +    # Assuming the data_file value in $IMGOPTS contains a '$TEST_IMG',
 +    # create a filter that replaces the data file name by $TEST_IMG.
 +    # Example:
 +    #   In $IMGOPTS: 'data_file=$TEST_IMG.data_file'
 +    #   Then data_file_pattern == '\(.*\).data_file'
 +    #   And  data_file_filter  == -e 's#\(.*\).data_file#\1#
 +    data_file_filter=()
 +    if data_file_pattern=$(_get_data_file '\\(.*\\)'); then
 +        data_file_filter=(-e "s#$data_file_pattern#\\1#")
 +    fi
 +
      $SED -e 's/\([0-9a-fx]* *[0-9a-fx]* *\)[0-9a-fx]* */\1/g' \
          -e 's/"offset": [0-9]\+/"offset": OFFSET/g' \
 -        -e 's/Mapped to *//' | _filter_testdir | _filter_imgfmt
 +        -e 's/Mapped to *//' \
 +        "${data_file_filter[@]}" \
 +        | _filter_testdir | _filter_imgfmt
  }
- _filter_nbd()
++/* Same test with fair mutexes, for performance comparison.  */
-diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
++
-index XXXXXXX..XXXXXXX 100644
++#ifdef CONFIG_LINUX
---- a/tests/qemu-iotests/common.rc
++#include "qemu/futex.h"
-+++ b/tests/qemu-iotests/common.rc
++
-@@ -XXX,XX +XXX,XX @@ _stop_nbd_server()
++/* The nodes for the mutex reside in this structure (on which we try to avoid
-     fi
++ * false sharing).  The head of the mutex is in the "mutex_head" variable.
- }
++ */
++static struct {
-+# Gets the data_file value from IMGOPTS and replaces the '$TEST_IMG'
++    int next, locked;
-+# pattern by '$1'
++    int padding[14];
-+# Caution: The replacement is done with sed, so $1 must be escaped
++} nodes[NUM_CONTEXTS] __attribute__((__aligned__(64)));
-+#          properly.  (The delimiter is '#'.)
++
-+_get_data_file()
++static int mutex_head = -1;
-+{
++
-+    if ! echo "$IMGOPTS" | grep -q 'data_file='; then
++static void mcs_mutex_lock(void)
-+        return 1
++{
-+    fi
++    int prev;
 +
-+    echo "$IMGOPTS" | sed -e 's/.*data_file=\([^,]*\).*/\1/' \
++    nodes[id].next = -1;
-+                    | sed -e "s#\\\$TEST_IMG#$1#"
++    nodes[id].locked = 1;
-+}
++    prev = atomic_xchg(&mutex_head, id);
-+
++    if (prev != -1) {
- _make_test_img()
++        atomic_set(&nodes[prev].next, id);
- {
++        qemu_futex_wait(&nodes[id].locked, 1);
-     # extra qemu-img options can be added by tests
++    }
-@@ -XXX,XX +XXX,XX @@ _make_test_img()
++}
-     fi
++
++static void mcs_mutex_unlock(void)
-     if [ -n "$IMGOPTS" ]; then
++{
--        optstr=$(_optstr_add "$optstr" "$IMGOPTS")
++    int next;
-+        imgopts_expanded=$(echo "$IMGOPTS" | sed -e "s#\\\$TEST_IMG#$img_name#")
++    if (nodes[id].next == -1) {
-+        optstr=$(_optstr_add "$optstr" "$imgopts_expanded")
++        if (atomic_read(&mutex_head) == id &&
-     fi
++            atomic_cmpxchg(&mutex_head, id, -1) == id) {
-     if [ -n "$IMGKEYSECRET" ]; then
++            /* Last item in the list, exit.  */
-         object_options="--object secret,id=keysec0,data=$IMGKEYSECRET"
++            return;
-@@ -XXX,XX +XXX,XX @@ _rm_test_img()
++        }
-         # Remove all the extents for vmdk
++        while (atomic_read(&nodes[id].next) == -1) {
-         "$QEMU_IMG" info "$img" 2>/dev/null | grep 'filename:' | cut -f 2 -d: \
++            /* mcs_mutex_lock did the xchg, but has not updated
-             | xargs -I {} rm -f "{}"
++             * nodes[prev].next yet.
-+    elif [ "$IMGFMT" = "qcow2" ]; then
++             */
-+        # Remove external data file
++        }
-+        if data_file=$(_get_data_file "$img"); then
++    }
-+            rm -f "$data_file"
++
-+        fi
++    /* Wake up the next in line.  */
-     fi
++    next = nodes[id].next;
-     rm -f "$img"
++    nodes[next].locked = 0;
 +    qemu_futex_wake(&nodes[next].locked, 1);
 +}
 +
 +static void test_multi_fair_mutex_entry(void *opaque)
 +{
 +    while (!atomic_mb_read(&now_stopping)) {
 +        mcs_mutex_lock();
 +        counter++;
 +        mcs_mutex_unlock();
 +        atomic_inc(&atomic_counter);
 +    }
 +    atomic_dec(&running);
 +}
 +
 +static void test_multi_fair_mutex(int threads, int seconds)
 +{
 +    int i;
 +
 +    assert(mutex_head == -1);
 +    counter = 0;
 +    atomic_counter = 0;
 +    now_stopping = false;
 +
 +    create_aio_contexts();
 +    assert(threads <= NUM_CONTEXTS);
 +    running = threads;
 +    for (i = 0; i < threads; i++) {
 +        Coroutine *co1 = qemu_coroutine_create(test_multi_fair_mutex_entry, NULL);
 +        aio_co_schedule(ctx[i], co1);
 +    }
 +
 +    g_usleep(seconds * 1000000);
 +
 +    atomic_mb_set(&now_stopping, true);
 +    while (running > 0) {
 +        g_usleep(100000);
 +    }
 +
 +    join_aio_contexts();
 +    g_test_message("%d iterations/second\n", counter / seconds);
 +    g_assert_cmpint(counter, ==, atomic_counter);
 +}
 +
 +static void test_multi_fair_mutex_1(void)
 +{
 +    test_multi_fair_mutex(NUM_CONTEXTS, 1);
 +}
 +
 +static void test_multi_fair_mutex_10(void)
 +{
 +    test_multi_fair_mutex(NUM_CONTEXTS, 10);
 +}
 +#endif
 +
 +/* Same test with pthread mutexes, for performance comparison and
 + * portability.  */
 +
 +static QemuMutex mutex;
 +
 +static void test_multi_mutex_entry(void *opaque)
 +{
 +    while (!atomic_mb_read(&now_stopping)) {
 +        qemu_mutex_lock(&mutex);
 +        counter++;
 +        qemu_mutex_unlock(&mutex);
 +        atomic_inc(&atomic_counter);
 +    }
 +    atomic_dec(&running);
 +}
 +
 +static void test_multi_mutex(int threads, int seconds)
 +{
 +    int i;
 +
 +    qemu_mutex_init(&mutex);
 +    counter = 0;
 +    atomic_counter = 0;
 +    now_stopping = false;
 +
 +    create_aio_contexts();
 +    assert(threads <= NUM_CONTEXTS);
 +    running = threads;
 +    for (i = 0; i < threads; i++) {
 +        Coroutine *co1 = qemu_coroutine_create(test_multi_mutex_entry, NULL);
 +        aio_co_schedule(ctx[i], co1);
 +    }
 +
 +    g_usleep(seconds * 1000000);
 +
 +    atomic_mb_set(&now_stopping, true);
 +    while (running > 0) {
 +        g_usleep(100000);
 +    }
 +
 +    join_aio_contexts();
 +    g_test_message("%d iterations/second\n", counter / seconds);
 +    g_assert_cmpint(counter, ==, atomic_counter);
 +}
 +
 +static void test_multi_mutex_1(void)
 +{
 +    test_multi_mutex(NUM_CONTEXTS, 1);
 +}
 +
 +static void test_multi_mutex_10(void)
 +{
 +    test_multi_mutex(NUM_CONTEXTS, 10);
 +}
 +
  /* End of tests.  */
  int main(int argc, char **argv)
@@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv)
          g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_1);
          g_test_add_func("/aio/multi/mutex/contended", test_multi_co_mutex_1);
          g_test_add_func("/aio/multi/mutex/handoff", test_multi_co_mutex_2_3);
 +#ifdef CONFIG_LINUX
 +        g_test_add_func("/aio/multi/mutex/mcs", test_multi_fair_mutex_1);
 +#endif
 +        g_test_add_func("/aio/multi/mutex/pthread", test_multi_mutex_1);
      } else {
          g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_10);
          g_test_add_func("/aio/multi/mutex/contended", test_multi_co_mutex_10);
          g_test_add_func("/aio/multi/mutex/handoff", test_multi_co_mutex_2_30);
 +#ifdef CONFIG_LINUX
 +        g_test_add_func("/aio/multi/mutex/mcs", test_multi_fair_mutex_10);
 +#endif
 +        g_test_add_func("/aio/multi/mutex/pthread", test_multi_mutex_10);
      }
      return g_test_run();
  }
 --
-.24.1
+.9.3

-[PULL 17/34] iotests: Inject space into -ocompat=0.10 in 051
+[Qemu-devel] [PULL v2 22/24] coroutine-lock: place CoMutex before CoQueue in header
-It did not matter before, but now that _make_test_img understands -o, we
+From: Paolo Bonzini <pbonzini@redhat.com>
 should use it properly here.
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+This will avoid forward references in the next patch.  It is also
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+more logical because CoQueue is not anymore the basic primitive.
-Message-id: 20191107163708.833192-11-mreitz@redhat.com
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
 Reviewed-by: Fam Zheng <famz@redhat.com>
 Message-id: 20170213181244.16297-5-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/051 | 2 +-
+ include/qemu/coroutine.h | 89 ++++++++++++++++++++++++------------------------
-file changed, 1 insertion(+), 1 deletion(-)
+file changed, 44 insertions(+), 45 deletions(-)
-diff --git a/tests/qemu-iotests/051 b/tests/qemu-iotests/051
+diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
-index XXXXXXX..XXXXXXX 100755
+index XXXXXXX..XXXXXXX 100644
---- a/tests/qemu-iotests/051
+--- a/include/qemu/coroutine.h
-+++ b/tests/qemu-iotests/051
++++ b/include/qemu/coroutine.h
-@@ -XXX,XX +XXX,XX @@ echo
+@@ -XXX,XX +XXX,XX @@ bool qemu_in_coroutine(void);
- echo === With version 2 images enabling lazy refcounts must fail ===
+  */
- echo
+ bool qemu_coroutine_entered(Coroutine *co);
--_make_test_img -ocompat=0.10 $size
+-
-+_make_test_img -o compat=0.10 $size
+-/**
+- * CoQueues are a mechanism to queue coroutines in order to continue executing
- run_qemu -drive file="$TEST_IMG",format=qcow2,lazy-refcounts=on
+- * them later. They provide the fundamental primitives on which coroutine locks
- run_qemu -drive file="$TEST_IMG",format=qcow2,lazy-refcounts=off
+- * are built.
 - */
 -typedef struct CoQueue {
 -    QSIMPLEQ_HEAD(, Coroutine) entries;
 -} CoQueue;
 -
 -/**
 - * Initialise a CoQueue. This must be called before any other operation is used
 - * on the CoQueue.
 - */
 -void qemu_co_queue_init(CoQueue *queue);
 -
 -/**
 - * Adds the current coroutine to the CoQueue and transfers control to the
 - * caller of the coroutine.
 - */
 -void coroutine_fn qemu_co_queue_wait(CoQueue *queue);
 -
 -/**
 - * Restarts the next coroutine in the CoQueue and removes it from the queue.
 - *
 - * Returns true if a coroutine was restarted, false if the queue is empty.
 - */
 -bool coroutine_fn qemu_co_queue_next(CoQueue *queue);
 -
 -/**
 - * Restarts all coroutines in the CoQueue and leaves the queue empty.
 - */
 -void coroutine_fn qemu_co_queue_restart_all(CoQueue *queue);
 -
 -/**
 - * Enter the next coroutine in the queue
 - */
 -bool qemu_co_enter_next(CoQueue *queue);
 -
 -/**
 - * Checks if the CoQueue is empty.
 - */
 -bool qemu_co_queue_empty(CoQueue *queue);
 -
 -
  /**
   * Provides a mutex that can be used to synchronise coroutines
   */
@@ -XXX,XX +XXX,XX @@ void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex);
   */
  void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex);
 +
 +/**
 + * CoQueues are a mechanism to queue coroutines in order to continue executing
 + * them later.
 + */
 +typedef struct CoQueue {
 +    QSIMPLEQ_HEAD(, Coroutine) entries;
 +} CoQueue;
 +
 +/**
 + * Initialise a CoQueue. This must be called before any other operation is used
 + * on the CoQueue.
 + */
 +void qemu_co_queue_init(CoQueue *queue);
 +
 +/**
 + * Adds the current coroutine to the CoQueue and transfers control to the
 + * caller of the coroutine.
 + */
 +void coroutine_fn qemu_co_queue_wait(CoQueue *queue);
 +
 +/**
 + * Restarts the next coroutine in the CoQueue and removes it from the queue.
 + *
 + * Returns true if a coroutine was restarted, false if the queue is empty.
 + */
 +bool coroutine_fn qemu_co_queue_next(CoQueue *queue);
 +
 +/**
 + * Restarts all coroutines in the CoQueue and leaves the queue empty.
 + */
 +void coroutine_fn qemu_co_queue_restart_all(CoQueue *queue);
 +
 +/**
 + * Enter the next coroutine in the queue
 + */
 +bool qemu_co_enter_next(CoQueue *queue);
 +
 +/**
 + * Checks if the CoQueue is empty.
 + */
 +bool qemu_co_queue_empty(CoQueue *queue);
 +
 +
  typedef struct CoRwlock {
      bool writer;
      int reader;
 --
-.24.1
+.9.3

-[PULL 23/34] iotests: Avoid cp/mv of test images
+[Qemu-devel] [PULL v2 23/24] coroutine-lock: add mutex argument to CoQueue APIs
-This will not work with external data files, so try to get tests working
+From: Paolo Bonzini <pbonzini@redhat.com>
-without it as far as possible.
+All that CoQueue needs in order to become thread-safe is help
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+from an external mutex.  Add this to the API.
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Message-id: 20191107163708.833192-17-mreitz@redhat.com
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Fam Zheng <famz@redhat.com>
 Message-id: 20170213181244.16297-6-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
- tests/qemu-iotests/063     | 12 ++++--------
+ include/qemu/coroutine.h   |  8 +++++---
- tests/qemu-iotests/063.out |  3 ++-
+ block/backup.c             |  2 +-
- tests/qemu-iotests/085     |  9 +++------
+ block/io.c                 |  4 ++--
- tests/qemu-iotests/085.out |  8 ++++----
+ block/nbd-client.c         |  2 +-
-files changed, 13 insertions(+), 19 deletions(-)
+ block/qcow2-cluster.c      |  4 +---
+ block/sheepdog.c           |  2 +-
-diff --git a/tests/qemu-iotests/063 b/tests/qemu-iotests/063
+ block/throttle-groups.c    |  2 +-
-index XXXXXXX..XXXXXXX 100755
+ hw/9pfs/9p.c               |  2 +-
---- a/tests/qemu-iotests/063
+ util/qemu-coroutine-lock.c | 24 +++++++++++++++++++++---
-+++ b/tests/qemu-iotests/063
+files changed, 34 insertions(+), 16 deletions(-)
-@@ -XXX,XX +XXX,XX @@ _unsupported_imgopts "subformat=monolithicFlat" \
- _make_test_img 4M
+diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
+index XXXXXXX..XXXXXXX 100644
- echo "== Testing conversion with -n fails with no target file =="
+--- a/include/qemu/coroutine.h
--# check .orig file does not exist
++++ b/include/qemu/coroutine.h
--rm -f "$TEST_IMG.orig"
+@@ -XXX,XX +XXX,XX @@ void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex);
- if $QEMU_IMG convert -f $IMGFMT -O $IMGFMT -n "$TEST_IMG" "$TEST_IMG.orig" >/dev/null 2>&1; then
-     exit 1
+ /**
- fi
+  * CoQueues are a mechanism to queue coroutines in order to continue executing
+- * them later.
- echo "== Testing conversion with -n succeeds with a target file =="
++ * them later.  They are similar to condition variables, but they need help
--rm -f "$TEST_IMG.orig"
++ * from an external mutex in order to maintain thread-safety.
--cp "$TEST_IMG" "$TEST_IMG.orig"
+  */
-+_rm_test_img "$TEST_IMG.orig"
+ typedef struct CoQueue {
-+TEST_IMG="$TEST_IMG.orig" _make_test_img 4M
+     QSIMPLEQ_HEAD(, Coroutine) entries;
- if ! $QEMU_IMG convert -f $IMGFMT -O $IMGFMT -n "$TEST_IMG" "$TEST_IMG.orig" ; then
+@@ -XXX,XX +XXX,XX @@ void qemu_co_queue_init(CoQueue *queue);
-     exit 1
- fi
+ /**
-@@ -XXX,XX +XXX,XX @@ fi
+  * Adds the current coroutine to the CoQueue and transfers control to the
- _check_test_img
+- * caller of the coroutine.
++ * caller of the coroutine.  The mutex is unlocked during the wait and
- echo "== Testing conversion to a smaller file fails =="
++ * locked again afterwards.
--rm -f "$TEST_IMG.orig"
+  */
--mv "$TEST_IMG" "$TEST_IMG.orig"
+-void coroutine_fn qemu_co_queue_wait(CoQueue *queue);
--_make_test_img 2M
++void coroutine_fn qemu_co_queue_wait(CoQueue *queue, CoMutex *mutex);
--if $QEMU_IMG convert -f $IMGFMT -O $IMGFMT -n "$TEST_IMG.orig" "$TEST_IMG" >/dev/null 2>&1; then
-+TEST_IMG="$TEST_IMG.target" _make_test_img 2M
+ /**
-+if $QEMU_IMG convert -f $IMGFMT -O $IMGFMT -n "$TEST_IMG" "$TEST_IMG.target" >/dev/null 2>&1; then
+  * Restarts the next coroutine in the CoQueue and removes it from the queue.
-     exit 1
+diff --git a/block/backup.c b/block/backup.c
- fi
+index XXXXXXX..XXXXXXX 100644
+--- a/block/backup.c
-diff --git a/tests/qemu-iotests/063.out b/tests/qemu-iotests/063.out
++++ b/block/backup.c
-index XXXXXXX..XXXXXXX 100644
+@@ -XXX,XX +XXX,XX @@ static void coroutine_fn wait_for_overlapping_requests(BackupBlockJob *job,
---- a/tests/qemu-iotests/063.out
+         retry = false;
-+++ b/tests/qemu-iotests/063.out
+         QLIST_FOREACH(req, &job->inflight_reqs, list) {
-@@ -XXX,XX +XXX,XX @@ QA output created by 063
+             if (end > req->start && start < req->end) {
- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=4194304
+-                qemu_co_queue_wait(&req->wait_queue);
- == Testing conversion with -n fails with no target file ==
++                qemu_co_queue_wait(&req->wait_queue, NULL);
- == Testing conversion with -n succeeds with a target file ==
+                 retry = true;
-+Formatting 'TEST_DIR/t.IMGFMT.orig', fmt=IMGFMT size=4194304
+                 break;
- == Testing conversion to raw is the same after conversion with -n ==
+             }
- == Testing conversion back to original format ==
+diff --git a/block/io.c b/block/io.c
- No errors were found on the image.
+index XXXXXXX..XXXXXXX 100644
- == Testing conversion to a smaller file fails ==
+--- a/block/io.c
--Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2097152
++++ b/block/io.c
-+Formatting 'TEST_DIR/t.IMGFMT.target', fmt=IMGFMT size=2097152
+@@ -XXX,XX +XXX,XX @@ static bool coroutine_fn wait_serialising_requests(BdrvTrackedRequest *self)
- == Regression testing for copy offloading bug ==
+                  * (instead of producing a deadlock in the former case). */
- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576
+                 if (!req->waiting_for) {
- Formatting 'TEST_DIR/t.IMGFMT.target', fmt=IMGFMT size=1048576
+                     self->waiting_for = req;
-diff --git a/tests/qemu-iotests/085 b/tests/qemu-iotests/085
+-                    qemu_co_queue_wait(&req->wait_queue);
-index XXXXXXX..XXXXXXX 100755
++                    qemu_co_queue_wait(&req->wait_queue, NULL);
---- a/tests/qemu-iotests/085
+                     self->waiting_for = NULL;
-+++ b/tests/qemu-iotests/085
+                     retry = true;
-@@ -XXX,XX +XXX,XX @@ add_snapshot_image()
+                     waited = true;
@@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_flush(BlockDriverState *bs)
      /* Wait until any previous flushes are completed */
      while (bs->active_flush_req) {
 -        qemu_co_queue_wait(&bs->flush_queue);
 +        qemu_co_queue_wait(&bs->flush_queue, NULL);
      }
      bs->active_flush_req = true;
 diff --git a/block/nbd-client.c b/block/nbd-client.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/nbd-client.c
 +++ b/block/nbd-client.c
@@ -XXX,XX +XXX,XX @@ static void nbd_coroutine_start(NBDClientSession *s,
      /* Poor man semaphore.  The free_sema is locked when no other request
       * can be accepted, and unlocked after receiving one reply.  */
      if (s->in_flight == MAX_NBD_REQUESTS) {
 -        qemu_co_queue_wait(&s->free_sema);
 +        qemu_co_queue_wait(&s->free_sema, NULL);
          assert(s->in_flight < MAX_NBD_REQUESTS);
      }
      s->in_flight++;
 diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/qcow2-cluster.c
 +++ b/block/qcow2-cluster.c
@@ -XXX,XX +XXX,XX @@ static int handle_dependencies(BlockDriverState *bs, uint64_t guest_offset,
              if (bytes == 0) {
                  /* Wait for the dependency to complete. We need to recheck
                   * the free/allocated clusters when we continue. */
 -                qemu_co_mutex_unlock(&s->lock);
 -                qemu_co_queue_wait(&old_alloc->dependent_requests);
 -                qemu_co_mutex_lock(&s->lock);
 +                qemu_co_queue_wait(&old_alloc->dependent_requests, &s->lock);
                  return -EAGAIN;
              }
          }
 diff --git a/block/sheepdog.c b/block/sheepdog.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/sheepdog.c
 +++ b/block/sheepdog.c
@@ -XXX,XX +XXX,XX @@ static void wait_for_overlapping_aiocb(BDRVSheepdogState *s, SheepdogAIOCB *acb)
  retry:
      QLIST_FOREACH(cb, &s->inflight_aiocb_head, aiocb_siblings) {
          if (AIOCBOverlapping(acb, cb)) {
 -            qemu_co_queue_wait(&s->overlapping_queue);
 +            qemu_co_queue_wait(&s->overlapping_queue, NULL);
              goto retry;
          }
      }
 diff --git a/block/throttle-groups.c b/block/throttle-groups.c
 index XXXXXXX..XXXXXXX 100644
 --- a/block/throttle-groups.c
 +++ b/block/throttle-groups.c
@@ -XXX,XX +XXX,XX @@ void coroutine_fn throttle_group_co_io_limits_intercept(BlockBackend *blk,
      if (must_wait || blkp->pending_reqs[is_write]) {
          blkp->pending_reqs[is_write]++;
          qemu_mutex_unlock(&tg->lock);
 -        qemu_co_queue_wait(&blkp->throttled_reqs[is_write]);
 +        qemu_co_queue_wait(&blkp->throttled_reqs[is_write], NULL);
          qemu_mutex_lock(&tg->lock);
          blkp->pending_reqs[is_write]--;
      }
 diff --git a/hw/9pfs/9p.c b/hw/9pfs/9p.c
 index XXXXXXX..XXXXXXX 100644
 --- a/hw/9pfs/9p.c
 +++ b/hw/9pfs/9p.c
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn v9fs_flush(void *opaque)
          /*
           * Wait for pdu to complete.
           */
 -        qemu_co_queue_wait(&cancel_pdu->complete);
 +        qemu_co_queue_wait(&cancel_pdu->complete, NULL);
          cancel_pdu->cancelled = 0;
          pdu_free(cancel_pdu);
      }
 diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/qemu-coroutine-lock.c
 +++ b/util/qemu-coroutine-lock.c
@@ -XXX,XX +XXX,XX @@ void qemu_co_queue_init(CoQueue *queue)
      QSIMPLEQ_INIT(&queue->entries);
  }
 -void coroutine_fn qemu_co_queue_wait(CoQueue *queue)
 +void coroutine_fn qemu_co_queue_wait(CoQueue *queue, CoMutex *mutex)
  {
-     base_image="${TEST_DIR}/$((${1}-1))-${snapshot_virt0}"
+     Coroutine *self = qemu_coroutine_self();
-     snapshot_file="${TEST_DIR}/${1}-${snapshot_virt0}"
+     QSIMPLEQ_INSERT_TAIL(&queue->entries, self, co_queue_next);
--    _make_test_img -u -b "${base_image}" "$size"
++
--    mv "${TEST_IMG}" "${snapshot_file}"
++    if (mutex) {
-+    TEST_IMG=$snapshot_file _make_test_img -u -b "${base_image}" "$size"
++        qemu_co_mutex_unlock(mutex);
-     do_blockdev_add "$1" "'backing': null, " "${snapshot_file}"
++    }
 +
 +    /* There is no race condition here.  Other threads will call
 +     * aio_co_schedule on our AioContext, which can reenter this
 +     * coroutine but only after this yield and after the main loop
 +     * has gone through the next iteration.
 +     */
      qemu_coroutine_yield();
      assert(qemu_in_coroutine());
 +
 +    /* TODO: OSv implements wait morphing here, where the wakeup
 +     * primitive automatically places the woken coroutine on the
 +     * mutex's queue.  This avoids the thundering herd effect.
 +     */
 +    if (mutex) {
 +        qemu_co_mutex_lock(mutex);
 +    }
  }
-@@ -XXX,XX +XXX,XX @@ blockdev_snapshot()
+ /**
+@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_rdlock(CoRwlock *lock)
- size=128M
+     Coroutine *self = qemu_coroutine_self();
--_make_test_img $size
+     while (lock->writer) {
--mv "${TEST_IMG}" "${TEST_IMG}.1"
+-        qemu_co_queue_wait(&lock->queue);
--_make_test_img $size
++        qemu_co_queue_wait(&lock->queue, NULL);
--mv "${TEST_IMG}" "${TEST_IMG}.2"
+     }
-+TEST_IMG="$TEST_IMG.1" _make_test_img $size
+     lock->reader++;
-+TEST_IMG="$TEST_IMG.2" _make_test_img $size
+     self->locks_held++;
+@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_wrlock(CoRwlock *lock)
- echo
+     Coroutine *self = qemu_coroutine_self();
- echo === Running QEMU ===
-diff --git a/tests/qemu-iotests/085.out b/tests/qemu-iotests/085.out
+     while (lock->writer || lock->reader) {
-index XXXXXXX..XXXXXXX 100644
+-        qemu_co_queue_wait(&lock->queue);
---- a/tests/qemu-iotests/085.out
++        qemu_co_queue_wait(&lock->queue, NULL);
-+++ b/tests/qemu-iotests/085.out
+     }
-@@ -XXX,XX +XXX,XX @@
+     lock->writer = true;
- QA output created by 085
+     self->locks_held++;
 -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 +Formatting 'TEST_DIR/t.IMGFMT.1', fmt=IMGFMT size=134217728
 +Formatting 'TEST_DIR/t.IMGFMT.2', fmt=IMGFMT size=134217728
  === Running QEMU ===
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/10-snapshot-v1.qcow2', fmt=qcow2 size=134217728 backing_fil
  === Create a couple of snapshots using blockdev-snapshot ===
 -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/10-snapshot-v0.IMGFMT
 +Formatting 'TEST_DIR/11-snapshot-v0.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/10-snapshot-v0.IMGFMT
  { 'execute': 'blockdev-add', 'arguments': { 'driver': 'IMGFMT', 'node-name': 'snap_11', 'backing': null, 'file': { 'driver': 'file', 'filename': 'TEST_DIR/11-snapshot-v0.IMGFMT', 'node-name': 'file_11' } } }
  {"return": {}}
  { 'execute': 'blockdev-snapshot', 'arguments': { 'node': 'virtio0', 'overlay':'snap_11' } }
  {"return": {}}
 -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/11-snapshot-v0.IMGFMT
 +Formatting 'TEST_DIR/12-snapshot-v0.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/11-snapshot-v0.IMGFMT
  { 'execute': 'blockdev-add', 'arguments': { 'driver': 'IMGFMT', 'node-name': 'snap_12', 'backing': null, 'file': { 'driver': 'file', 'filename': 'TEST_DIR/12-snapshot-v0.IMGFMT', 'node-name': 'file_12' } } }
  {"return": {}}
  { 'execute': 'blockdev-snapshot', 'arguments': { 'node': 'virtio0', 'overlay':'snap_12' } }
 --
-.24.1
+.9.3

-[PULL 18/34] iotests: Replace IMGOPTS= by -o
+[Qemu-devel] [PULL v2 24/24] coroutine-lock: make CoRwlock thread-safe and fair
-Tests should not overwrite all user-supplied image options, but only add
+From: Paolo Bonzini <pbonzini@redhat.com>
 to it (which will effectively overwrite conflicting values).  Accomplish
 this by passing options to _make_test_img via -o instead of $IMGOPTS.
-For some tests, there is no functional change because they already only
+This adds a CoMutex around the existing CoQueue.  Because the write-side
-appended options to IMGOPTS.  For these, this patch is just a
+can just take CoMutex, the old "writer" field is not necessary anymore.
-simplification.
+Instead of removing it altogether, count the number of pending writers
 during a read-side critical section and forbid further readers from
 entering.
-For others, this is a change, so they now heed user-specified $IMGOPTS.
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-Some of those tests do not work with all image options, though, so we
+Reviewed-by: Fam Zheng <famz@redhat.com>
-need to disable them accordingly.
+Message-id: 20170213181244.16297-7-pbonzini@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
  include/qemu/coroutine.h   |  3 ++-
  util/qemu-coroutine-lock.c | 35 ++++++++++++++++++++++++-----------
 files changed, 26 insertions(+), 12 deletions(-)
-Signed-off-by: Max Reitz <mreitz@redhat.com>
+diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
+index XXXXXXX..XXXXXXX 100644
-Message-id: 20191107163708.833192-12-mreitz@redhat.com
+--- a/include/qemu/coroutine.h
-Signed-off-by: Max Reitz <mreitz@redhat.com>
++++ b/include/qemu/coroutine.h
----
+@@ -XXX,XX +XXX,XX @@ bool qemu_co_queue_empty(CoQueue *queue);
- tests/qemu-iotests/031 |  9 ++++---
- tests/qemu-iotests/039 | 24 ++++++------------
- tests/qemu-iotests/059 | 18 ++++++-------
+ typedef struct CoRwlock {
- tests/qemu-iotests/060 |  6 ++---
+-    bool writer;
- tests/qemu-iotests/061 | 57 ++++++++++++++++++++++--------------------
++    int pending_writer;
- tests/qemu-iotests/079 |  3 +--
+     int reader;
- tests/qemu-iotests/106 |  2 +-
++    CoMutex mutex;
- tests/qemu-iotests/108 |  2 +-
+     CoQueue queue;
- tests/qemu-iotests/112 | 32 ++++++++++++------------
+ } CoRwlock;
- tests/qemu-iotests/115 |  3 +--
- tests/qemu-iotests/121 |  6 ++---
+diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
- tests/qemu-iotests/125 |  2 +-
+index XXXXXXX..XXXXXXX 100644
- tests/qemu-iotests/137 |  2 +-
+--- a/util/qemu-coroutine-lock.c
- tests/qemu-iotests/138 |  3 +--
++++ b/util/qemu-coroutine-lock.c
- tests/qemu-iotests/175 |  2 +-
+@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_init(CoRwlock *lock)
- tests/qemu-iotests/190 |  2 +-
+ {
- tests/qemu-iotests/191 |  3 +--
+     memset(lock, 0, sizeof(*lock));
- tests/qemu-iotests/220 |  4 ++-
+     qemu_co_queue_init(&lock->queue);
- tests/qemu-iotests/243 |  6 +++--
++    qemu_co_mutex_init(&lock->mutex);
  tests/qemu-iotests/244 | 10 +++++---
  tests/qemu-iotests/250 |  3 +--
  tests/qemu-iotests/265 |  2 +-
 files changed, 100 insertions(+), 101 deletions(-)
 diff --git a/tests/qemu-iotests/031 b/tests/qemu-iotests/031
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/031
 +++ b/tests/qemu-iotests/031
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  # This tests qcow2-specific low-level functionality
  _supported_fmt qcow2
  _supported_proto file
 +# We want to test compat=0.10, which does not support refcount widths
 +# other than 16
 +_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
  CLUSTER_SIZE=65536
  # qcow2.py output depends on the exact options used, so override the command
  # line here as an exception
 -for IMGOPTS in "compat=0.10" "compat=1.1"; do
 +for compat in "compat=0.10" "compat=1.1"; do
      echo
 -    echo ===== Testing with -o $IMGOPTS =====
 +    echo ===== Testing with -o $compat =====
      echo
      echo === Create image with unknown header extension ===
      echo
 -    _make_test_img 64M
 +    _make_test_img -o $compat 64M
      $PYTHON qcow2.py "$TEST_IMG" add-header-ext 0x12345678 "This is a test header extension"
      $PYTHON qcow2.py "$TEST_IMG" dump-header
      _check_test_img
 diff --git a/tests/qemu-iotests/039 b/tests/qemu-iotests/039
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/039
 +++ b/tests/qemu-iotests/039
@@ -XXX,XX +XXX,XX @@ size=128M
  echo
  echo "== Checking that image is clean on shutdown =="
 -IMGOPTS="compat=1.1,lazy_refcounts=on"
 -_make_test_img $size
 +_make_test_img -o "compat=1.1,lazy_refcounts=on" $size
  $QEMU_IO -c "write -P 0x5a 0 512" "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "== Creating a dirty image file =="
 -IMGOPTS="compat=1.1,lazy_refcounts=on"
 -_make_test_img $size
 +_make_test_img -o "compat=1.1,lazy_refcounts=on" $size
  _NO_VALGRIND \
  $QEMU_IO -c "write -P 0x5a 0 512" \
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0x5a 0 512" "$TEST_IMG" | _filter_qemu_io
  echo
  echo "== Opening a dirty image read/write should repair it =="
 -IMGOPTS="compat=1.1,lazy_refcounts=on"
 -_make_test_img $size
 +_make_test_img -o "compat=1.1,lazy_refcounts=on" $size
  _NO_VALGRIND \
  $QEMU_IO -c "write -P 0x5a 0 512" \
@@ -XXX,XX +XXX,XX @@ $PYTHON qcow2.py "$TEST_IMG" dump-header | grep incompatible_features
  echo
  echo "== Creating an image file with lazy_refcounts=off =="
 -IMGOPTS="compat=1.1,lazy_refcounts=off"
 -_make_test_img $size
 +_make_test_img -o "compat=1.1,lazy_refcounts=off" $size
  _NO_VALGRIND \
  $QEMU_IO -c "write -P 0x5a 0 512" \
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "== Committing to a backing file with lazy_refcounts=on =="
 -IMGOPTS="compat=1.1,lazy_refcounts=on"
 -TEST_IMG="$TEST_IMG".base _make_test_img $size
 +TEST_IMG="$TEST_IMG".base _make_test_img -o "compat=1.1,lazy_refcounts=on" $size
 -IMGOPTS="compat=1.1,lazy_refcounts=on,backing_file=$TEST_IMG.base"
 -_make_test_img $size
 +_make_test_img -o "compat=1.1,lazy_refcounts=on,backing_file=$TEST_IMG.base" $size
  $QEMU_IO -c "write 0 512" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IMG commit "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ TEST_IMG="$TEST_IMG".base _check_test_img
  echo
  echo "== Changing lazy_refcounts setting at runtime =="
 -IMGOPTS="compat=1.1,lazy_refcounts=off"
 -_make_test_img $size
 +_make_test_img -o "compat=1.1,lazy_refcounts=off" $size
  _NO_VALGRIND \
  $QEMU_IO -c "reopen -o lazy-refcounts=on" \
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "reopen -o lazy-refcounts=on" \
  $PYTHON qcow2.py "$TEST_IMG" dump-header | grep incompatible_features
  _check_test_img
 -IMGOPTS="compat=1.1,lazy_refcounts=on"
 -_make_test_img $size
 +_make_test_img -o "compat=1.1,lazy_refcounts=on" $size
  _NO_VALGRIND \
  $QEMU_IO -c "reopen -o lazy-refcounts=off" \
 diff --git a/tests/qemu-iotests/059 b/tests/qemu-iotests/059
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/059
 +++ b/tests/qemu-iotests/059
@@ -XXX,XX +XXX,XX @@ poke_file "$TEST_IMG" "$grain_table_size_offset" "\x01\x00\x00\x00"
  echo
  echo "=== Testing monolithicFlat creation and opening ==="
 -IMGOPTS="subformat=monolithicFlat" _make_test_img 2G
 +_make_test_img -o "subformat=monolithicFlat" 2G
  _img_info
  _cleanup_test_img
  echo
  echo "=== Testing monolithicFlat with zeroed_grain ==="
 -IMGOPTS="subformat=monolithicFlat,zeroed_grain=on" _make_test_img 2G
 +_make_test_img -o "subformat=monolithicFlat,zeroed_grain=on" 2G
  _cleanup_test_img
  echo
  echo "=== Testing big twoGbMaxExtentFlat ==="
 -IMGOPTS="subformat=twoGbMaxExtentFlat" _make_test_img 1000G
 +_make_test_img -o "subformat=twoGbMaxExtentFlat" 1000G
  $QEMU_IMG info $TEST_IMG | _filter_testdir | sed -e 's/cid: [0-9]*/cid: XXXXXXXX/'
  _cleanup_test_img
@@ -XXX,XX +XXX,XX @@ _img_info
  echo
  echo "=== Testing truncated sparse ==="
 -IMGOPTS="subformat=monolithicSparse" _make_test_img 100G
 +_make_test_img -o "subformat=monolithicSparse" 100G
  truncate -s 10M $TEST_IMG
  _img_info
  echo
  echo "=== Converting to streamOptimized from image with small cluster size==="
 -TEST_IMG="$TEST_IMG.qcow2" IMGFMT=qcow2 IMGOPTS="cluster_size=4096" _make_test_img 1G
 +TEST_IMG="$TEST_IMG.qcow2" IMGFMT=qcow2 _make_test_img -o "cluster_size=4096" 1G
  $QEMU_IO -f qcow2 -c "write -P 0xa 0 512" "$TEST_IMG.qcow2" | _filter_qemu_io
  $QEMU_IO -f qcow2 -c "write -P 0xb 10240 512" "$TEST_IMG.qcow2" | _filter_qemu_io
  $QEMU_IMG convert -f qcow2 -O vmdk -o subformat=streamOptimized "$TEST_IMG.qcow2" "$TEST_IMG" 2>&1
@@ -XXX,XX +XXX,XX @@ echo "=== Testing monolithicFlat with internally generated JSON file name ==="
  echo '--- blkdebug ---'
  # Should work, because bdrv_dirname() works fine with blkdebug
 -IMGOPTS="subformat=monolithicFlat" _make_test_img 64M
 +_make_test_img -o "subformat=monolithicFlat" 64M
  $QEMU_IO -c "open -o driver=$IMGFMT,file.driver=blkdebug,file.image.filename=$TEST_IMG,file.inject-error.0.event=read_aio" \
           -c info \
 >&1 \
@@ -XXX,XX +XXX,XX @@ _cleanup_test_img
  echo '--- quorum ---'
  # Should not work, because bdrv_dirname() does not work with quorum
 -IMGOPTS="subformat=monolithicFlat" _make_test_img 64M
 +_make_test_img -o "subformat=monolithicFlat" 64M
  cp "$TEST_IMG" "$TEST_IMG.orig"
  filename="json:{
@@ -XXX,XX +XXX,XX @@ _cleanup_test_img
  echo
  echo "=== Testing 4TB monolithicFlat creation and IO ==="
 -IMGOPTS="subformat=monolithicFlat" _make_test_img 4T
 +_make_test_img -o "subformat=monolithicFlat" 4T
  _img_info
  $QEMU_IO -c "write -P 0xa 900G 512" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IO -c "read -v 900G 1024" "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _cleanup_test_img
  echo
  echo "=== Testing qemu-img map on extents ==="
  for fmt in monolithicSparse twoGbMaxExtentSparse; do
 -    IMGOPTS="subformat=$fmt" _make_test_img 31G
 +    _make_test_img -o "subformat=$fmt" 31G
      $QEMU_IO -c "write 65024 1k" "$TEST_IMG" | _filter_qemu_io
      $QEMU_IO -c "write 2147483136 1k" "$TEST_IMG" | _filter_qemu_io
      $QEMU_IO -c "write 5G 1k" "$TEST_IMG" | _filter_qemu_io
 diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/060
 +++ b/tests/qemu-iotests/060
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c 'write 0k 64k' "$BACKING_IMG" | _filter_qemu_io
  # compat=0.10 is required in order to make the following discard actually
  # unallocate the sector rather than make it a zero sector - we want COW, after
  # all.
 -IMGOPTS='compat=0.10' _make_test_img -b "$BACKING_IMG" 1G
 +_make_test_img -o 'compat=0.10' -b "$BACKING_IMG" 1G
  # Write two clusters, the second one enforces creation of an L2 table after
  # the first data cluster.
  $QEMU_IO -c 'write 0k 64k' -c 'write 512M 64k' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
  echo "=== Discarding a non-covered in-bounds refblock ==="
  echo
 -IMGOPTS='refcount_bits=1' _make_test_img 64M
 +_make_test_img -o 'refcount_bits=1' 64M
  # Pretend there's a refblock somewhere where there is no refblock to
  # cover it (but the covering refblock has a valid index in the
@@ -XXX,XX +XXX,XX @@ echo
  echo "=== Discarding a refblock covered by an unaligned refblock ==="
  echo
 -IMGOPTS='refcount_bits=1' _make_test_img 64M
 +_make_test_img -o 'refcount_bits=1' 64M
  # Same as above
  poke_file "$TEST_IMG" "$(($rt_offset+8))" "\x00\x00\x00\x10\x00\x00\x00\x00"
 diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/061
 +++ b/tests/qemu-iotests/061
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# Conversion between different compat versions can only really work
 +# with refcount_bits=16
 +_unsupported_imgopts 'refcount_bits=\([^1]\|.\([^6]\|$\)\)'
  echo
  echo "=== Testing version downgrade with zero expansion ==="
  echo
 -IMGOPTS="compat=1.1,lazy_refcounts=on" _make_test_img 64M
 +_make_test_img -o "compat=1.1,lazy_refcounts=on" 64M
  $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
  $PYTHON qcow2.py "$TEST_IMG" dump-header
  $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing version downgrade with zero expansion and 4K cache entries ==="
  echo
 -IMGOPTS="compat=1.1,lazy_refcounts=on" _make_test_img 64M
 +_make_test_img -o "compat=1.1,lazy_refcounts=on" 64M
  $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IO -c "write -z 32M 128k" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IO -c map "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing dirty version downgrade ==="
  echo
 -IMGOPTS="compat=1.1,lazy_refcounts=on" _make_test_img 64M
 +_make_test_img -o "compat=1.1,lazy_refcounts=on" 64M
  _NO_VALGRIND \
  $QEMU_IO -c "write -P 0x2a 0 128k" -c flush \
           -c "sigraise $(kill -l KILL)" "$TEST_IMG" 2>&1 | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing version downgrade with unknown compat/autoclear flags ==="
  echo
 -IMGOPTS="compat=1.1" _make_test_img 64M
 +_make_test_img -o "compat=1.1" 64M
  $PYTHON qcow2.py "$TEST_IMG" set-feature-bit compatible 42
  $PYTHON qcow2.py "$TEST_IMG" set-feature-bit autoclear 42
  $PYTHON qcow2.py "$TEST_IMG" dump-header
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing version upgrade and resize ==="
  echo
 -IMGOPTS="compat=0.10" _make_test_img 64M
 +_make_test_img -o "compat=0.10" 64M
  $QEMU_IO -c "write -P 0x2a 42M 64k" "$TEST_IMG" | _filter_qemu_io
  $PYTHON qcow2.py "$TEST_IMG" dump-header
  $QEMU_IMG amend -o "compat=1.1,lazy_refcounts=on,size=128M" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing dirty lazy_refcounts=off ==="
  echo
 -IMGOPTS="compat=1.1,lazy_refcounts=on" _make_test_img 64M
 +_make_test_img -o "compat=1.1,lazy_refcounts=on" 64M
  _NO_VALGRIND \
  $QEMU_IO -c "write -P 0x2a 0 128k" -c flush \
           -c "sigraise $(kill -l KILL)" "$TEST_IMG" 2>&1 | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing backing file ==="
  echo
 -IMGOPTS="compat=1.1" _make_test_img 64M
 -IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 64M
 +_make_test_img -o "compat=1.1" 64M
 +TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M
  $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io
  $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IMG amend -o "backing_file=$TEST_IMG.base,backing_fmt=qcow2" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing invalid configurations ==="
  echo
 -IMGOPTS="compat=0.10" _make_test_img 64M
 +_make_test_img -o "compat=0.10" 64M
  $QEMU_IMG amend -o "lazy_refcounts=on" "$TEST_IMG"
  $QEMU_IMG amend -o "compat=1.1" "$TEST_IMG" # actually valid
  $QEMU_IMG amend -o "compat=0.10,lazy_refcounts=on" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ $QEMU_IMG amend -o "preallocation=on" "$TEST_IMG"
  echo
  echo "=== Testing correct handling of unset value ==="
  echo
 -IMGOPTS="compat=1.1,cluster_size=1k" _make_test_img 64M
 +_make_test_img -o "compat=1.1,cluster_size=1k" 64M
  echo "Should work:"
  $QEMU_IMG amend -o "lazy_refcounts=on" "$TEST_IMG"
  echo "Should not work:" # Just to know which of these tests actually fails
@@ -XXX,XX +XXX,XX @@ $QEMU_IMG amend -o "cluster_size=64k" "$TEST_IMG"
  echo
  echo "=== Testing zero expansion on inactive clusters ==="
  echo
 -IMGOPTS="compat=1.1" _make_test_img 64M
 +_make_test_img -o "compat=1.1" 64M
  $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IMG snapshot -c foo "$TEST_IMG"
  $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io
  echo
  echo "=== Testing zero expansion on shared L2 table ==="
  echo
 -IMGOPTS="compat=1.1" _make_test_img 64M
 +_make_test_img -o "compat=1.1" 64M
  $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IMG snapshot -c foo "$TEST_IMG"
  $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io
  echo
  echo "=== Testing zero expansion on backed image ==="
  echo
 -IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 64M
 +TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M
  $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io
 -IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 64M
 +_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 64M
  $QEMU_IO -c "read -P 0x2a 0 128k" -c "write -z 0 64k" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
  _check_test_img
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 64k" -c "read -P 0x2a 64k 64k" "$TEST_IMG" | _filter_qe
  echo
  echo "=== Testing zero expansion on backed inactive clusters ==="
  echo
 -IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 64M
 +TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M
  $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io
 -IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 64M
 +_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 64M
  $QEMU_IO -c "write -z 0 64k" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IMG snapshot -c foo "$TEST_IMG"
  $QEMU_IO -c "write -P 0x42 0 128k" "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 64k" -c "read -P 0x2a 64k 64k" "$TEST_IMG" | _filter_qe
  echo
  echo "=== Testing zero expansion on backed image with shared L2 table ==="
  echo
 -IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 64M
 +TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M
  $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io
 -IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 64M
 +_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 64M
  $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
  $QEMU_IMG snapshot -c foo "$TEST_IMG"
  $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io
  echo
  echo "=== Testing preallocated zero expansion on full image ==="
  echo
 -IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG" _make_test_img 64M
 +TEST_IMG="$TEST_IMG" _make_test_img -o "compat=1.1" 64M
  $QEMU_IO -c "write -P 0x2a 0 64M" "$TEST_IMG" -c "write -z 0 64M" | _filter_qemu_io
  $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
  _check_test_img
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 64M" "$TEST_IMG" | _filter_qemu_io
  echo
  echo "=== Testing progress report without snapshot ==="
  echo
 -IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 4G
 -IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 4G
 +TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 4G
 +_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 4G
  $QEMU_IO -c "write -z 0  64k" \
           -c "write -z 1G 64k" \
           -c "write -z 2G 64k" \
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing progress report with snapshot ==="
  echo
 -IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 4G
 -IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 4G
 +TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 4G
 +_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 4G
  $QEMU_IO -c "write -z 0  64k" \
           -c "write -z 1G 64k" \
           -c "write -z 2G 64k" \
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Testing version downgrade with external data file ==="
  echo
 -IMGOPTS="compat=1.1,data_file=$TEST_IMG.data" _make_test_img 64M
 +_make_test_img -o "compat=1.1,data_file=$TEST_IMG.data" 64M
  $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
  _img_info --format-specific
  _check_test_img
@@ -XXX,XX +XXX,XX @@ _check_test_img
  echo
  echo "=== Try changing the external data file ==="
  echo
 -IMGOPTS="compat=1.1" _make_test_img 64M
 +_make_test_img -o "compat=1.1" 64M
  $QEMU_IMG amend -o "data_file=foo" "$TEST_IMG"
  echo
 -IMGOPTS="compat=1.1,data_file=$TEST_IMG.data" _make_test_img 64M
 +_make_test_img -o "compat=1.1,data_file=$TEST_IMG.data" 64M
  $QEMU_IMG amend -o "data_file=foo" "$TEST_IMG"
  _img_info --format-specific
  TEST_IMG="data-file.filename=$TEST_IMG.data,file.filename=$TEST_IMG" _img_info --format-specific --image-opts
@@ -XXX,XX +XXX,XX @@ TEST_IMG="data-file.filename=$TEST_IMG.data,file.filename=$TEST_IMG" _img_info -
  echo
  echo "=== Clearing and setting data-file-raw ==="
  echo
 -IMGOPTS="compat=1.1,data_file=$TEST_IMG.data,data_file_raw=on" _make_test_img 64M
 +_make_test_img -o "compat=1.1,data_file=$TEST_IMG.data,data_file_raw=on" 64M
  $QEMU_IMG amend -o "data_file_raw=on" "$TEST_IMG"
  _img_info --format-specific
  _check_test_img
 diff --git a/tests/qemu-iotests/079 b/tests/qemu-iotests/079
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/079
 +++ b/tests/qemu-iotests/079
@@ -XXX,XX +XXX,XX @@ echo
  cluster_sizes="16384 32768 65536 131072 262144 524288 1048576 2097152 4194304"
  for s in $cluster_sizes; do
 -    IMGOPTS=$(_optstr_add "$IMGOPTS" "preallocation=metadata,cluster_size=$s") \
 -        _make_test_img 4G
 +    _make_test_img -o "preallocation=metadata,cluster_size=$s" 4G
  done
  # success, all done
 diff --git a/tests/qemu-iotests/106 b/tests/qemu-iotests/106
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/106
 +++ b/tests/qemu-iotests/106
@@ -XXX,XX +XXX,XX @@ for create_mode in off falloc full; do
          echo
          echo "--- create_mode=$create_mode growth_mode=$growth_mode ---"
 -        IMGOPTS="preallocation=$create_mode" _make_test_img ${CREATION_SIZE}K
 +        _make_test_img -o "preallocation=$create_mode" ${CREATION_SIZE}K
          $QEMU_IMG resize -f "$IMGFMT" --preallocation=$growth_mode "$TEST_IMG" +${GROWTH_SIZE}K
          expected_size=0
 diff --git a/tests/qemu-iotests/108 b/tests/qemu-iotests/108
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/108
 +++ b/tests/qemu-iotests/108
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== Repairing unreferenced data cluster in new refblock area ==='
  echo
 -IMGOPTS='cluster_size=512' _make_test_img 64M
 +_make_test_img -o 'cluster_size=512' 64M
  # Allocate the first 128 kB in the image (first refblock)
  $QEMU_IO -c 'write 0 0x1b200' "$TEST_IMG" | _filter_qemu_io
  # should be 131072 == 0x20000
 diff --git a/tests/qemu-iotests/112 b/tests/qemu-iotests/112
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/112
 +++ b/tests/qemu-iotests/112
@@ -XXX,XX +XXX,XX @@ echo '=== refcount_bits limits ==='
  echo
  # Must be positive (non-zero)
 -IMGOPTS="$IMGOPTS,refcount_bits=0" _make_test_img 64M
 +_make_test_img -o "refcount_bits=0" 64M
  # Must be positive (non-negative)
 -IMGOPTS="$IMGOPTS,refcount_bits=-1" _make_test_img 64M
 +_make_test_img -o "refcount_bits=-1" 64M
  # May not exceed 64
 -IMGOPTS="$IMGOPTS,refcount_bits=128" _make_test_img 64M
 +_make_test_img -o "refcount_bits=128" 64M
  # Must be a power of two
 -IMGOPTS="$IMGOPTS,refcount_bits=42" _make_test_img 64M
 +_make_test_img -o "refcount_bits=42" 64M
  # 1 is the minimum
 -IMGOPTS="$IMGOPTS,refcount_bits=1" _make_test_img 64M
 +_make_test_img -o "refcount_bits=1" 64M
  print_refcount_bits
  # 64 is the maximum
 -IMGOPTS="$IMGOPTS,refcount_bits=64" _make_test_img 64M
 +_make_test_img -o "refcount_bits=64" 64M
  print_refcount_bits
  # 16 is the default
@@ -XXX,XX +XXX,XX @@ echo '=== refcount_bits and compat=0.10 ==='
  echo
  # Should work
 -IMGOPTS="$IMGOPTS,compat=0.10,refcount_bits=16" _make_test_img 64M
 +_make_test_img -o "compat=0.10,refcount_bits=16" 64M
  print_refcount_bits
  # Should not work
 -IMGOPTS="$IMGOPTS,compat=0.10,refcount_bits=1" _make_test_img 64M
 -IMGOPTS="$IMGOPTS,compat=0.10,refcount_bits=64" _make_test_img 64M
 +_make_test_img -o "compat=0.10,refcount_bits=1" 64M
 +_make_test_img -o "compat=0.10,refcount_bits=64" 64M
  echo
  echo '=== Snapshot limit on refcount_bits=1 ==='
  echo
 -IMGOPTS="$IMGOPTS,refcount_bits=1" _make_test_img 64M
 +_make_test_img -o "refcount_bits=1" 64M
  print_refcount_bits
  $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== Snapshot limit on refcount_bits=2 ==='
  echo
 -IMGOPTS="$IMGOPTS,refcount_bits=2" _make_test_img 64M
 +_make_test_img -o "refcount_bits=2" 64M
  print_refcount_bits
  $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== Compressed clusters with refcount_bits=1 ==='
  echo
 -IMGOPTS="$IMGOPTS,refcount_bits=1" _make_test_img 64M
 +_make_test_img -o "refcount_bits=1" 64M
  print_refcount_bits
  # Both should fit into a single host cluster; instead of failing to increase the
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== MSb set in 64 bit refcount ==='
  echo
 -IMGOPTS="$IMGOPTS,refcount_bits=64" _make_test_img 64M
 +_make_test_img -o "refcount_bits=64" 64M
  print_refcount_bits
  $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== Snapshot on maximum 64 bit refcount value ==='
  echo
 -IMGOPTS="$IMGOPTS,refcount_bits=64" _make_test_img 64M
 +_make_test_img -o "refcount_bits=64" 64M
  print_refcount_bits
  $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== Testing too many references for check ==='
  echo
 -IMGOPTS="$IMGOPTS,refcount_bits=1" _make_test_img 64M
 +_make_test_img -o "refcount_bits=1" 64M
  print_refcount_bits
  # This cluster should be created at 0x50000
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== Multiple walks necessary during amend ==='
  echo
 -IMGOPTS="$IMGOPTS,refcount_bits=1,cluster_size=512" _make_test_img 64k
 +_make_test_img -o "refcount_bits=1,cluster_size=512" 64k
  # Cluster 0 is the image header, clusters 1 to 4 are used by the L1 table, a
  # single L2 table, the reftable and a single refblock. This creates 58 data
 diff --git a/tests/qemu-iotests/115 b/tests/qemu-iotests/115
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/115
 +++ b/tests/qemu-iotests/115
@@ -XXX,XX +XXX,XX @@ echo
  # least 256 MB. We can achieve that by using preallocation=metadata for an image
  # which has a guest disk size of 256 MB.
 -IMGOPTS="$IMGOPTS,refcount_bits=64,cluster_size=512,preallocation=metadata" \
 -    _make_test_img 256M
 +_make_test_img -o "refcount_bits=64,cluster_size=512,preallocation=metadata" 256M
  # We know for sure that the L1 and refcount tables do not overlap with any other
  # structure because the metadata overlap checks would have caught that case.
 diff --git a/tests/qemu-iotests/121 b/tests/qemu-iotests/121
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/121
 +++ b/tests/qemu-iotests/121
@@ -XXX,XX +XXX,XX @@ echo
  # Preallocation speeds up the write operation, but preallocating everything will
  # destroy the purpose of the write; so preallocate one KB less than what would
  # cause a reftable growth...
 -IMGOPTS='preallocation=metadata,cluster_size=1k' _make_test_img 64512K
 +_make_test_img -o 'preallocation=metadata,cluster_size=1k' 64512K
  # ...and make the image the desired size afterwards.
  $QEMU_IMG resize "$TEST_IMG" 65M
@@ -XXX,XX +XXX,XX @@ echo
  echo '--- Test 2 ---'
  echo
 -IMGOPTS='preallocation=metadata,cluster_size=1k' _make_test_img 64513K
 +_make_test_img -o 'preallocation=metadata,cluster_size=1k' 64513K
  # This results in an L1 table growth which in turn results in some clusters at
  # the start of the image becoming free
  $QEMU_IMG resize "$TEST_IMG" 65M
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== Allocating a new refcount block must not leave holes in the image ==='
  echo
 -IMGOPTS='cluster_size=512,refcount_bits=16' _make_test_img 1M
 +_make_test_img -o 'cluster_size=512,refcount_bits=16' 1M
  # This results in an image with 256 used clusters: the qcow2 header,
  # the refcount table, one refcount block, the L1 table, four L2 tables
 diff --git a/tests/qemu-iotests/125 b/tests/qemu-iotests/125
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/125
 +++ b/tests/qemu-iotests/125
@@ -XXX,XX +XXX,XX @@ for GROWTH_SIZE in 16 48 80; do
          for growth_mode in off metadata falloc full; do
              echo "--- cluster_size=$cluster_size growth_size=$GROWTH_SIZE create_mode=$create_mode growth_mode=$growth_mode ---"
 -            IMGOPTS="preallocation=$create_mode,cluster_size=$cluster_size" _make_test_img ${CREATION_SIZE}
 +            _make_test_img -o "preallocation=$create_mode,cluster_size=$cluster_size" ${CREATION_SIZE}
              $QEMU_IMG resize -f "$IMGFMT" --preallocation=$growth_mode "$TEST_IMG" +${GROWTH_SIZE}K
              host_size_0=$(get_image_size_on_host)
 diff --git a/tests/qemu-iotests/137 b/tests/qemu-iotests/137
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/137
 +++ b/tests/qemu-iotests/137
@@ -XXX,XX +XXX,XX @@ $QEMU_IO \
      -c "reopen -o cache-clean-interval=-1" \
      "$TEST_IMG" | _filter_qemu_io
 -IMGOPTS="cluster_size=256k" _make_test_img 32P
 +_make_test_img -o "cluster_size=256k" 32P
  $QEMU_IO \
      -c "reopen -o l2-cache-entry-size=512,l2-cache-size=1T" \
      "$TEST_IMG" | _filter_qemu_io
 diff --git a/tests/qemu-iotests/138 b/tests/qemu-iotests/138
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/138
 +++ b/tests/qemu-iotests/138
@@ -XXX,XX +XXX,XX @@ echo
  echo '=== Check on an image with a multiple of 2^32 clusters ==='
  echo
 -IMGOPTS=$(_optstr_add "$IMGOPTS" "cluster_size=512") \
 -    _make_test_img 512
 +_make_test_img -o "cluster_size=512" 512
  # Allocate L2 table
  $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
 diff --git a/tests/qemu-iotests/175 b/tests/qemu-iotests/175
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/175
 +++ b/tests/qemu-iotests/175
@@ -XXX,XX +XXX,XX @@ stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_block
  for mode in off full falloc; do
      echo
      echo "== creating image with preallocation $mode =="
 -    IMGOPTS=preallocation=$mode _make_test_img $size | _filter_imgfmt
 +    _make_test_img -o preallocation=$mode $size | _filter_imgfmt
      stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $size
  done
 diff --git a/tests/qemu-iotests/190 b/tests/qemu-iotests/190
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/190
 +++ b/tests/qemu-iotests/190
@@ -XXX,XX +XXX,XX @@ _supported_proto file
  echo "== Huge file =="
  echo
 -IMGOPTS='cluster_size=2M' _make_test_img 2T
 +_make_test_img -o 'cluster_size=2M' 2T
  $QEMU_IMG measure -O raw -f qcow2 "$TEST_IMG"
  $QEMU_IMG measure -O qcow2 -o cluster_size=64k -f qcow2 "$TEST_IMG"
 diff --git a/tests/qemu-iotests/191 b/tests/qemu-iotests/191
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/191
 +++ b/tests/qemu-iotests/191
@@ -XXX,XX +XXX,XX @@ echo === Preparing and starting VM ===
  echo
  TEST_IMG="${TEST_IMG}.base" _make_test_img $size
 -IMGOPTS=$(_optstr_add "$IMGOPTS" "backing_fmt=$IMGFMT") \
 -    TEST_IMG="${TEST_IMG}.mid" _make_test_img -b "${TEST_IMG}.base"
 +TEST_IMG="${TEST_IMG}.mid" _make_test_img -o "backing_fmt=$IMGFMT" -b "${TEST_IMG}.base"
  _make_test_img -b "${TEST_IMG}.mid"
  TEST_IMG="${TEST_IMG}.ovl2" _make_test_img -b "${TEST_IMG}.mid"
 diff --git a/tests/qemu-iotests/220 b/tests/qemu-iotests/220
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/220
 +++ b/tests/qemu-iotests/220
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# To use a different refcount width but 16 bits we need compat=1.1
 +_unsupported_imgopts 'compat=0.10'
  echo "== Creating huge file =="
@@ -XXX,XX +XXX,XX @@ echo "== Creating huge file =="
  # of a HUGE (but very sparse) file.  tmpfs works, ext4 does not.
  _require_large_file 513T
 -IMGOPTS='cluster_size=2M,refcount_bits=1' _make_test_img 513T
 +_make_test_img -o 'cluster_size=2M,refcount_bits=1' 513T
  echo "== Populating refcounts =="
  # We want an image with 256M refcounts * 2M clusters = 512T referenced.
 diff --git a/tests/qemu-iotests/243 b/tests/qemu-iotests/243
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/243
 +++ b/tests/qemu-iotests/243
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# External data files do not work with compat=0.10
 +_unsupported_imgopts 'compat=0.10'
  for mode in off metadata falloc full; do
@@ -XXX,XX +XXX,XX @@ for mode in off metadata falloc full; do
      echo "=== preallocation=$mode ==="
      echo
 -    IMGOPTS="preallocation=$mode" _make_test_img 64M
 +    _make_test_img -o "preallocation=$mode" 64M
      printf "File size: "
      du -b $TEST_IMG | cut -f1
@@ -XXX,XX +XXX,XX @@ for mode in off metadata falloc full; do
      echo "=== External data file: preallocation=$mode ==="
      echo
 -    IMGOPTS="data_file=$TEST_IMG.data,preallocation=$mode" _make_test_img 64M
 +    _make_test_img -o "data_file=$TEST_IMG.data,preallocation=$mode" 64M
      echo -n "qcow2 file size: "
      du -b $TEST_IMG | cut -f1
 diff --git a/tests/qemu-iotests/244 b/tests/qemu-iotests/244
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/244
 +++ b/tests/qemu-iotests/244
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
  _supported_fmt qcow2
  _supported_proto file
  _supported_os Linux
 +# External data files do not work with compat=0.10
 +_unsupported_imgopts 'compat=0.10'
  echo
  echo "=== Create and open image with external data file ==="
  echo
  echo "With data file name in the image:"
 -IMGOPTS="data_file=$TEST_IMG.data" _make_test_img 64M
 +_make_test_img -o "data_file=$TEST_IMG.data" 64M
  _check_test_img
  $QEMU_IO -c "open $TEST_IMG" -c "read -P 0 0 64k" 2>&1 | _filter_qemu_io | _filter_testdir
@@ -XXX,XX +XXX,XX @@ echo
  echo "=== Standalone image with external data file (efficient) ==="
  echo
 -IMGOPTS="data_file=$TEST_IMG.data" _make_test_img 64M
 +_make_test_img -o "data_file=$TEST_IMG.data" 64M
  echo -n "qcow2 file size before I/O: "
  du -b $TEST_IMG | cut -f1
@@ -XXX,XX +XXX,XX @@ echo
  echo "=== Standalone image with external data file (valid raw) ==="
  echo
 -IMGOPTS="data_file=$TEST_IMG.data,data_file_raw=on" _make_test_img 64M
 +_make_test_img -o "data_file=$TEST_IMG.data,data_file_raw=on" 64M
  echo -n "qcow2 file size before I/O: "
  du -b $TEST_IMG | cut -f1
@@ -XXX,XX +XXX,XX @@ echo
  echo "=== bdrv_co_block_status test for file and offset=0 ==="
  echo
 -IMGOPTS="data_file=$TEST_IMG.data" _make_test_img 64M
 +_make_test_img -o "data_file=$TEST_IMG.data" 64M
  $QEMU_IO -c 'write -P 0x11 0 1M' -f $IMGFMT "$TEST_IMG" | _filter_qemu_io
  $QEMU_IO -c 'read -P 0x11 0 1M' -f $IMGFMT "$TEST_IMG" | _filter_qemu_io
 diff --git a/tests/qemu-iotests/250 b/tests/qemu-iotests/250
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/250
 +++ b/tests/qemu-iotests/250
@@ -XXX,XX +XXX,XX @@ disk_usage()
  }
- size=2100M
+ void qemu_co_rwlock_rdlock(CoRwlock *lock)
--IMGOPTS="cluster_size=1M,preallocation=metadata"
+ {
+     Coroutine *self = qemu_coroutine_self();
--_make_test_img $size
-+_make_test_img -o "cluster_size=1M,preallocation=metadata" $size
+-    while (lock->writer) {
- $QEMU_IO -c 'discard 0 10M' -c 'discard 2090M 10M' \
+-        qemu_co_queue_wait(&lock->queue, NULL);
-          -c 'write 2090M 10M' -c 'write 0 10M' "$TEST_IMG" | _filter_qemu_io
++    qemu_co_mutex_lock(&lock->mutex);
++    /* For fairness, wait if a writer is in line.  */
-diff --git a/tests/qemu-iotests/265 b/tests/qemu-iotests/265
++    while (lock->pending_writer) {
-index XXXXXXX..XXXXXXX 100755
++        qemu_co_queue_wait(&lock->queue, &lock->mutex);
---- a/tests/qemu-iotests/265
+     }
-+++ b/tests/qemu-iotests/265
+     lock->reader++;
-@@ -XXX,XX +XXX,XX @@ _supported_os Linux
++    qemu_co_mutex_unlock(&lock->mutex);
- echo '--- Writing to the image ---'
++
++    /* The rest of the read-side critical section is run without the mutex.  */
- # Reduce cluster size so we get more and quicker I/O
+     self->locks_held++;
--IMGOPTS='cluster_size=4096' _make_test_img 1M
+ }
-+_make_test_img -o 'cluster_size=4096' 1M
- (for ((kb = 1024 - 4; kb >= 0; kb -= 4)); do \
+@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_unlock(CoRwlock *lock)
-      echo "aio_write -P 42 $((kb + 1))k 2k"; \
+     Coroutine *self = qemu_coroutine_self();
-  done) \
      assert(qemu_in_coroutine());
 -    if (lock->writer) {
 -        lock->writer = false;
 +    if (!lock->reader) {
 +        /* The critical section started in qemu_co_rwlock_wrlock.  */
          qemu_co_queue_restart_all(&lock->queue);
      } else {
 +        self->locks_held--;
 +
 +        qemu_co_mutex_lock(&lock->mutex);
          lock->reader--;
          assert(lock->reader >= 0);
          /* Wakeup only one waiting writer */
@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_unlock(CoRwlock *lock)
              qemu_co_queue_next(&lock->queue);
          }
      }
 -    self->locks_held--;
 +    qemu_co_mutex_unlock(&lock->mutex);
  }
  void qemu_co_rwlock_wrlock(CoRwlock *lock)
  {
 -    Coroutine *self = qemu_coroutine_self();
 -
 -    while (lock->writer || lock->reader) {
 -        qemu_co_queue_wait(&lock->queue, NULL);
 +    qemu_co_mutex_lock(&lock->mutex);
 +    lock->pending_writer++;
 +    while (lock->reader) {
 +        qemu_co_queue_wait(&lock->queue, &lock->mutex);
      }
 -    lock->writer = true;
 -    self->locks_held++;
 +    lock->pending_writer--;
 +
 +    /* The rest of the write-side critical section is run with
 +     * the mutex taken, so that lock->reader remains zero.
 +     * There is no need to update self->locks_held.
 +     */
  }
 --
-.24.1
+.9.3

The following changes since commit f17783e706ab9c7b3a2b69cf48e4f0ba40664f54:

Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20191220' into staging (2020-01-03 18:50:33 +0000)

are available in the Git repository at:

https://github.com/XanClic/qemu.git tags/pull-block-2020-01-06

for you to fetch changes up to 503ca1262bab2c11c533a4816d1ff4297d4f58a6:

backup-top: Begin drain earlier (2020-01-06 14:26:23 +0100)

----------------------------------------------------------------
Block patches:
- Minor fixes and tests from the freeze period (too minor to be included
  in 4.2)
- Allow many bash iotests to test qcow2's external data file feature
- Add compress filter driver
- Fix Python iotests after 6f6e1698a6
- Fix for the backup job

----------------------------------------------------------------
Andrey Shinkevich (3):
  block: introduce compress filter driver
  qcow2: Allow writing compressed data of multiple clusters
  tests/qemu-iotests: add case to write compressed data of multiple
    clusters

Max Reitz (28):
  block: Add bdrv_qapi_perm_to_blk_perm()
  block: Use bdrv_qapi_perm_to_blk_perm()
  blkdebug: Allow taking/unsharing permissions
  iotests: Add @error to wait_until_completed
  iotests: Add test for failing mirror complete
  iotests: s/qocw2/qcow2/
  iotests/qcow2.py: Add dump-header-exts
  iotests/qcow2.py: Split feature fields into bits
  iotests: Add _filter_json_filename
  iotests: Filter refcount_order in 036
  iotests: Replace IMGOPTS by _unsupported_imgopts
  iotests: Drop compat=1.1 in 050
  iotests: Let _make_test_img parse its parameters
  iotests: Add -o and --no-opts to _make_test_img
  iotests: Inject space into -ocompat=0.10 in 051
  iotests: Replace IMGOPTS= by -o
  iotests: Replace IMGOPTS='' by --no-opts
  iotests: Drop IMGOPTS use in 267
  iotests: Avoid qemu-img create
  iotests: Use _rm_test_img for deleting test images
  iotests: Avoid cp/mv of test images
  iotests: Make 091 work with data_file
  iotests: Make 110 work with data_file
  iotests: Make 137 work with data_file
  iotests: Make 198 work with data_file
  iotests: Disable data_file where it cannot be used
  iotests: Allow check -o data_file
  backup-top: Begin drain earlier

PanNengyuan (1):
  throttle-groups: fix memory leak in throttle_group_set_limit:

Philippe Mathieu-Daudé (1):
  tests/qemu-iotests: Update tests to recent desugarized -accel option

Vladimir Sementsov-Ogievskiy (1):
  qcow2-bitmaps: fix qcow2_can_store_new_dirty_bitmap

-- 
2.24.1

We need some way to correlate QAPI BlockPermission values with
BLK_PERM_* flags.  We could:

(1) have the same order in the QAPI definition as the the BLK_PERM_*
    flags are in LSb-first order.  However, then there is no guarantee
    that they actually match (e.g. when someone modifies the QAPI schema
    without thinking of the BLK_PERM_* definitions).
    We could add static assertions, but these would break what’s good
    about this solution, namely its simplicity.

(2) define the BLK_PERM_* flags based on the BlockPermission values.
    But this way whenever someone were to modify the QAPI order
    (perfectly sensible in theory), the BLK_PERM_* values would change.
    Because these values are used for file locking, this might break
    file locking between different qemu versions.

Therefore, go the slightly more cumbersome way: Add a function to
translate from the QAPI constants to the BLK_PERM_* flags.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20191108123455.39445-2-mreitz@redhat.com
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block.c               | 18 ++++++++++++++++++
 include/block/block.h |  1 +
 2 files changed, 19 insertions(+)

diff --git a/block.c b/block.c
index XXXXXXX..XXXXXXX 100644
--- a/block.c
+++ b/block.c
@@ -XXX,XX +XXX,XX @@ void bdrv_format_default_perms(BlockDriverState *bs, BdrvChild *c,
     *nshared = shared;
 }
 
+uint64_t bdrv_qapi_perm_to_blk_perm(BlockPermission qapi_perm)
+{
+    static const uint64_t permissions[] = {
+        [BLOCK_PERMISSION_CONSISTENT_READ]  = BLK_PERM_CONSISTENT_READ,
+        [BLOCK_PERMISSION_WRITE]            = BLK_PERM_WRITE,
+        [BLOCK_PERMISSION_WRITE_UNCHANGED]  = BLK_PERM_WRITE_UNCHANGED,
+        [BLOCK_PERMISSION_RESIZE]           = BLK_PERM_RESIZE,
+        [BLOCK_PERMISSION_GRAPH_MOD]        = BLK_PERM_GRAPH_MOD,
+    };
+
+    QEMU_BUILD_BUG_ON(ARRAY_SIZE(permissions) != BLOCK_PERMISSION__MAX);
+    QEMU_BUILD_BUG_ON(1UL << ARRAY_SIZE(permissions) != BLK_PERM_ALL + 1);
+
+    assert(qapi_perm < BLOCK_PERMISSION__MAX);
+
+    return permissions[qapi_perm];
+}
+
 static void bdrv_replace_child_noperm(BdrvChild *child,
                                       BlockDriverState *new_bs)
 {
diff --git a/include/block/block.h b/include/block/block.h
index XXXXXXX..XXXXXXX 100644
--- a/include/block/block.h
+++ b/include/block/block.h
@@ -XXX,XX +XXX,XX @@ enum {
 };
 
 char *bdrv_perm_names(uint64_t perm);
+uint64_t bdrv_qapi_perm_to_blk_perm(BlockPermission qapi_perm);
 
 /* disk I/O throttling */
 void bdrv_init(void);
-- 
2.24.1

We can save some LoC in xdbg_graph_add_edge() by using
bdrv_qapi_perm_to_blk_perm().

Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20191108123455.39445-3-mreitz@redhat.com
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block.c | 29 ++++++++---------------------
 1 file changed, 8 insertions(+), 21 deletions(-)

diff --git a/block.c b/block.c
index XXXXXXX..XXXXXXX 100644
--- a/block.c
+++ b/block.c
@@ -XXX,XX +XXX,XX @@ static void xdbg_graph_add_node(XDbgBlockGraphConstructor *gr, void *node,
 static void xdbg_graph_add_edge(XDbgBlockGraphConstructor *gr, void *parent,
                                 const BdrvChild *child)
 {
-    typedef struct {
-        unsigned int flag;
-        BlockPermission num;
-    } PermissionMap;
-
-    static const PermissionMap permissions[] = {
-        { BLK_PERM_CONSISTENT_READ, BLOCK_PERMISSION_CONSISTENT_READ },
-        { BLK_PERM_WRITE,           BLOCK_PERMISSION_WRITE },
-        { BLK_PERM_WRITE_UNCHANGED, BLOCK_PERMISSION_WRITE_UNCHANGED },
-        { BLK_PERM_RESIZE,          BLOCK_PERMISSION_RESIZE },
-        { BLK_PERM_GRAPH_MOD,       BLOCK_PERMISSION_GRAPH_MOD },
-        { 0, 0 }
-    };
-    const PermissionMap *p;
+    BlockPermission qapi_perm;
     XDbgBlockGraphEdge *edge;
 
-    QEMU_BUILD_BUG_ON(1UL << (ARRAY_SIZE(permissions) - 1) != BLK_PERM_ALL + 1);
-
     edge = g_new0(XDbgBlockGraphEdge, 1);
 
     edge->parent = xdbg_graph_node_num(gr, parent);
     edge->child = xdbg_graph_node_num(gr, child->bs);
     edge->name = g_strdup(child->name);
 
-    for (p = permissions; p->flag; p++) {
-        if (p->flag & child->perm) {
-            QAPI_LIST_ADD(edge->perm, p->num);
+    for (qapi_perm = 0; qapi_perm < BLOCK_PERMISSION__MAX; qapi_perm++) {
+        uint64_t flag = bdrv_qapi_perm_to_blk_perm(qapi_perm);
+
+        if (flag & child->perm) {
+            QAPI_LIST_ADD(edge->perm, qapi_perm);
         }
-        if (p->flag & child->shared_perm) {
-            QAPI_LIST_ADD(edge->shared_perm, p->num);
+        if (flag & child->shared_perm) {
+            QAPI_LIST_ADD(edge->shared_perm, qapi_perm);
         }
     }
 
-- 
2.24.1

Sometimes it is useful to be able to add a node to the block graph that
takes or unshare a certain set of permissions for debugging purposes.
This patch adds this capability to blkdebug.

(Note that you cannot make blkdebug release or share permissions that it
needs to take or cannot share, because this might result in assertion
failures in the block layer.  But if the blkdebug node has no parents,
it will not take any permissions and share everything by default, so you
can then freely choose what permissions to take and share.)

Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20191108123455.39445-4-mreitz@redhat.com
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/blkdebug.c     | 93 +++++++++++++++++++++++++++++++++++++++++++-
 qapi/block-core.json | 14 ++++++-
 2 files changed, 105 insertions(+), 2 deletions(-)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index XXXXXXX..XXXXXXX 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/cutils.h"
 #include "qemu/config-file.h"
 #include "block/block_int.h"
+#include "block/qdict.h"
 #include "qemu/module.h"
 #include "qemu/option.h"
+#include "qapi/qapi-visit-block-core.h"
 #include "qapi/qmp/qdict.h"
+#include "qapi/qmp/qlist.h"
 #include "qapi/qmp/qstring.h"
+#include "qapi/qobject-input-visitor.h"
 #include "sysemu/qtest.h"
 
 typedef struct BDRVBlkdebugState {
@@ -XXX,XX +XXX,XX @@ typedef struct BDRVBlkdebugState {
     uint64_t opt_discard;
     uint64_t max_discard;
 
+    uint64_t take_child_perms;
+    uint64_t unshare_child_perms;
+
     /* For blkdebug_refresh_filename() */
     char *config_file;
 
@@ -XXX,XX +XXX,XX @@ static void blkdebug_parse_filename(const char *filename, QDict *options,
     qdict_put_str(options, "x-image", filename);
 }
 
+static int blkdebug_parse_perm_list(uint64_t *dest, QDict *options,
+                                    const char *prefix, Error **errp)
+{
+    int ret = 0;
+    QDict *subqdict = NULL;
+    QObject *crumpled_subqdict = NULL;
+    Visitor *v = NULL;
+    BlockPermissionList *perm_list = NULL, *element;
+    Error *local_err = NULL;
+
+    *dest = 0;
+
+    qdict_extract_subqdict(options, &subqdict, prefix);
+    if (!qdict_size(subqdict)) {
+        goto out;
+    }
+
+    crumpled_subqdict = qdict_crumple(subqdict, errp);
+    if (!crumpled_subqdict) {
+        ret = -EINVAL;
+        goto out;
+    }
+
+    v = qobject_input_visitor_new(crumpled_subqdict);
+    visit_type_BlockPermissionList(v, NULL, &perm_list, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        ret = -EINVAL;
+        goto out;
+    }
+
+    for (element = perm_list; element; element = element->next) {
+        *dest |= bdrv_qapi_perm_to_blk_perm(element->value);
+    }
+
+out:
+    qapi_free_BlockPermissionList(perm_list);
+    visit_free(v);
+    qobject_unref(subqdict);
+    qobject_unref(crumpled_subqdict);
+    return ret;
+}
+
+static int blkdebug_parse_perms(BDRVBlkdebugState *s, QDict *options,
+                                Error **errp)
+{
+    int ret;
+
+    ret = blkdebug_parse_perm_list(&s->take_child_perms, options,
+                                   "take-child-perms.", errp);
+    if (ret < 0) {
+        return ret;
+    }
+
+    ret = blkdebug_parse_perm_list(&s->unshare_child_perms, options,
+                                   "unshare-child-perms.", errp);
+    if (ret < 0) {
+        return ret;
+    }
+
+    return 0;
+}
+
 static QemuOptsList runtime_opts = {
     .name = "blkdebug",
     .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
@@ -XXX,XX +XXX,XX @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
     /* Set initial state */
     s->state = 1;
 
+    /* Parse permissions modifiers before opening the image file */
+    ret = blkdebug_parse_perms(s, options, errp);
+    if (ret < 0) {
+        goto out;
+    }
+
     /* Open the image file */
     bs->file = bdrv_open_child(qemu_opt_get(opts, "x-image"), options, "image",
                                bs, &child_file, false, &local_err);
@@ -XXX,XX +XXX,XX @@ static int blkdebug_reopen_prepare(BDRVReopenState *reopen_state,
     return 0;
 }
 
+static void blkdebug_child_perm(BlockDriverState *bs, BdrvChild *c,
+                                const BdrvChildRole *role,
+                                BlockReopenQueue *reopen_queue,
+                                uint64_t perm, uint64_t shared,
+                                uint64_t *nperm, uint64_t *nshared)
+{
+    BDRVBlkdebugState *s = bs->opaque;
+
+    bdrv_filter_default_perms(bs, c, role, reopen_queue, perm, shared,
+                              nperm, nshared);
+
+    *nperm |= s->take_child_perms;
+    *nshared &= ~s->unshare_child_perms;
+}
+
 static const char *const blkdebug_strong_runtime_opts[] = {
     "config",
     "inject-error.",
@@ -XXX,XX +XXX,XX @@ static BlockDriver bdrv_blkdebug = {
     .bdrv_file_open         = blkdebug_open,
     .bdrv_close             = blkdebug_close,
     .bdrv_reopen_prepare    = blkdebug_reopen_prepare,
-    .bdrv_child_perm        = bdrv_filter_default_perms,
+    .bdrv_child_perm        = blkdebug_child_perm,
 
     .bdrv_getlength         = blkdebug_getlength,
     .bdrv_refresh_filename  = blkdebug_refresh_filename,
diff --git a/qapi/block-core.json b/qapi/block-core.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/block-core.json
+++ b/qapi/block-core.json
@@ -XXX,XX +XXX,XX @@
 #
 # @set-state:       array of state-change descriptions
 #
+# @take-child-perms: Permissions to take on @image in addition to what
+#                    is necessary anyway (which depends on how the
+#                    blkdebug node is used).  Defaults to none.
+#                    (since 5.0)
+#
+# @unshare-child-perms: Permissions not to share on @image in addition
+#                       to what cannot be shared anyway (which depends
+#                       on how the blkdebug node is used).  Defaults
+#                       to none.  (since 5.0)
+#
 # Since: 2.9
 ##
 { 'struct': 'BlockdevOptionsBlkdebug',
@@ -XXX,XX +XXX,XX @@
             '*opt-write-zero': 'int32', '*max-write-zero': 'int32',
             '*opt-discard': 'int32', '*max-discard': 'int32',
             '*inject-error': ['BlkdebugInjectErrorOptions'],
-            '*set-state': ['BlkdebugSetStateOptions'] } }
+            '*set-state': ['BlkdebugSetStateOptions'],
+            '*take-child-perms': ['BlockPermission'],
+            '*unshare-child-perms': ['BlockPermission'] } }
 
 ##
 # @BlockdevOptionsBlklogwrites:
-- 
2.24.1

Callers can use this new parameter to expect failure during the
completion process.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20191108123455.39445-5-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/iotests.py | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/iotests.py
+++ b/tests/qemu-iotests/iotests.py
@@ -XXX,XX +XXX,XX @@ class QMPTestCase(unittest.TestCase):
         self.assert_no_active_block_jobs()
         return result
 
-    def wait_until_completed(self, drive='drive0', check_offset=True, wait=60.0):
+    def wait_until_completed(self, drive='drive0', check_offset=True, wait=60.0,
+                             error=None):
         '''Wait for a block job to finish, returning the event'''
         while True:
             for event in self.vm.get_qmp_events(wait=wait):
                 if event['event'] == 'BLOCK_JOB_COMPLETED':
                     self.assert_qmp(event, 'data/device', drive)
-                    self.assert_qmp_absent(event, 'data/error')
-                    if check_offset:
-                        self.assert_qmp(event, 'data/offset', event['data']['len'])
+                    if error is None:
+                        self.assert_qmp_absent(event, 'data/error')
+                        if check_offset:
+                            self.assert_qmp(event, 'data/offset',
+                                            event['data']['len'])
+                    else:
+                        self.assert_qmp(event, 'data/error', error)
                     self.assert_no_active_block_jobs()
                     return event
                 elif event['event'] == 'JOB_STATUS_CHANGE':
@@ -XXX,XX +XXX,XX @@ class QMPTestCase(unittest.TestCase):
         self.assert_qmp(event, 'data/type', 'mirror')
         self.assert_qmp(event, 'data/offset', event['data']['len'])
 
-    def complete_and_wait(self, drive='drive0', wait_ready=True):
+    def complete_and_wait(self, drive='drive0', wait_ready=True,
+                          completion_error=None):
         '''Complete a block job and wait for it to finish'''
         if wait_ready:
             self.wait_ready(drive=drive)
@@ -XXX,XX +XXX,XX @@ class QMPTestCase(unittest.TestCase):
         result = self.vm.qmp('block-job-complete', device=drive)
         self.assert_qmp(result, 'return', {})
 
-        event = self.wait_until_completed(drive=drive)
+        event = self.wait_until_completed(drive=drive, error=completion_error)
         self.assert_qmp(event, 'data/type', 'mirror')
 
     def pause_wait(self, job_id='job0'):
-- 
2.24.1

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Tested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Message-id: 20191108123455.39445-6-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/041     | 44 ++++++++++++++++++++++++++++++++++++++
 tests/qemu-iotests/041.out |  4 ++--
 2 files changed, 46 insertions(+), 2 deletions(-)

diff --git a/tests/qemu-iotests/041 b/tests/qemu-iotests/041
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/041
+++ b/tests/qemu-iotests/041
@@ -XXX,XX +XXX,XX @@ class TestOrphanedSource(iotests.QMPTestCase):
                              target='dest-ro')
         self.assert_qmp(result, 'error/class', 'GenericError')
 
+    def test_failing_permission_in_complete(self):
+        self.assert_no_active_block_jobs()
+
+        # Unshare consistent-read on the target
+        # (The mirror job does not care)
+        result = self.vm.qmp('blockdev-add',
+                             driver='blkdebug',
+                             node_name='dest-perm',
+                             image='dest',
+                             unshare_child_perms=['consistent-read'])
+        self.assert_qmp(result, 'return', {})
+
+        result = self.vm.qmp('blockdev-mirror', job_id='job', device='src',
+                             sync='full', target='dest',
+                             filter_node_name='mirror-filter')
+        self.assert_qmp(result, 'return', {})
+
+        # Require consistent-read on the source
+        # (We can only add this node once the job has started, or it
+        # will complain that it does not want to run on non-root nodes)
+        result = self.vm.qmp('blockdev-add',
+                             driver='blkdebug',
+                             node_name='src-perm',
+                             image='src',
+                             take_child_perms=['consistent-read'])
+        self.assert_qmp(result, 'return', {})
+
+        # While completing, mirror will attempt to replace src by
+        # dest, which must fail because src-perm requires
+        # consistent-read but dest-perm does not share it; thus
+        # aborting the job when it is supposed to complete
+        self.complete_and_wait('job',
+                               completion_error='Operation not permitted')
+
+        # Assert that all of our nodes are still there (except for the
+        # mirror filter, which should be gone despite the failure)
+        nodes = self.vm.qmp('query-named-block-nodes')['return']
+        nodes = [node['node-name'] for node in nodes]
+
+        for expect in ('src', 'src-perm', 'dest', 'dest-perm'):
+            self.assertTrue(expect in nodes, '%s disappeared' % expect)
+        self.assertFalse('mirror-filter' in nodes,
+                         'Mirror filter node did not disappear')
+
 if __name__ == '__main__':
     iotests.main(supported_fmts=['qcow2', 'qed'],
                  supported_protocols=['file'])
diff --git a/tests/qemu-iotests/041.out b/tests/qemu-iotests/041.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/041.out
+++ b/tests/qemu-iotests/041.out
@@ -XXX,XX +XXX,XX @@
-..........................................................................................
+...........................................................................................
 ----------------------------------------------------------------------
-Ran 90 tests
+Ran 91 tests
 
 OK
-- 
2.24.1

From: PanNengyuan <pannengyuan@huawei.com>

This avoid a memory leak when qom-set is called to set throttle_group
limits, here is an easy way to reproduce:

1. run qemu-iotests as follow and check the result with asan:
       ./check -qcow2 184

Following is the asan output backtrack:
Direct leak of 912 byte(s) in 3 object(s) allocated from:
    #0 0xffff8d7ab3c3 in __interceptor_calloc   (/lib64/libasan.so.4+0xd33c3)
    #1 0xffff8d4c31cb in g_malloc0   (/lib64/libglib-2.0.so.0+0x571cb)
    #2 0x190c857 in qobject_input_start_struct  /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qapi/qobject-input-visitor.c:295
    #3 0x19070df in visit_start_struct   /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qapi/qapi-visit-core.c:49
    #4 0x1948b87 in visit_type_ThrottleLimits   qapi/qapi-visit-block-core.c:3759
    #5 0x17e4aa3 in throttle_group_set_limits   /mnt/sdc/qemu-master/qemu-4.2.0-rc0/block/throttle-groups.c:900
    #6 0x1650eff in object_property_set     /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qom/object.c:1272
    #7 0x1658517 in object_property_set_qobject   /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qom/qom-qobject.c:26
    #8 0x15880bb in qmp_qom_set  /mnt/sdc/qemu-master/qemu-4.2.0-rc0/qom/qom-qmp-cmds.c:74
    #9 0x157e3e3 in qmp_marshal_qom_set  qapi/qapi-commands-qom.c:154

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: PanNengyuan <pannengyuan@huawei.com>
Message-id: 1574835614-42028-1-git-send-email-pannengyuan@huawei.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/throttle-groups.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/block/throttle-groups.c b/block/throttle-groups.c
index XXXXXXX..XXXXXXX 100644
--- a/block/throttle-groups.c
+++ b/block/throttle-groups.c
@@ -XXX,XX +XXX,XX @@ static void throttle_group_set_limits(Object *obj, Visitor *v,
 {
     ThrottleGroup *tg = THROTTLE_GROUP(obj);
     ThrottleConfig cfg;
-    ThrottleLimits arg = { 0 };
-    ThrottleLimits *argp = &arg;
+    ThrottleLimits *argp;
     Error *local_err = NULL;
 
     visit_type_ThrottleLimits(v, name, &argp, &local_err);
@@ -XXX,XX +XXX,XX @@ static void throttle_group_set_limits(Object *obj, Visitor *v,
 unlock:
     qemu_mutex_unlock(&tg->lock);
 ret:
+    qapi_free_ThrottleLimits(argp);
     error_propagate(errp, local_err);
     return;
 }
-- 
2.24.1

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

qcow2_can_store_new_dirty_bitmap works wrong, as it considers only
bitmaps already stored in the qcow2 image and ignores persistent
BdrvDirtyBitmap objects.

So, let's instead count persistent BdrvDirtyBitmaps. We load all qcow2
bitmaps on open, so there should not be any bitmap in the image for
which we don't have BdrvDirtyBitmaps version. If it is - it's a kind of
corruption, and no reason to check for corruptions here (open() and
close() are better places for it).

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20191014115126.15360-2-vsementsov@virtuozzo.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/qcow2-bitmap.c | 41 ++++++++++++++++++-----------------------
 1 file changed, 18 insertions(+), 23 deletions(-)

diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qcow2-bitmap.c
+++ b/block/qcow2-bitmap.c
@@ -XXX,XX +XXX,XX @@ bool coroutine_fn qcow2_co_can_store_new_dirty_bitmap(BlockDriverState *bs,
                                                       Error **errp)
 {
     BDRVQcow2State *s = bs->opaque;
-    bool found;
-    Qcow2BitmapList *bm_list;
+    BdrvDirtyBitmap *bitmap;
+    uint64_t bitmap_directory_size = 0;
+    uint32_t nb_bitmaps = 0;
+
+    if (bdrv_find_dirty_bitmap(bs, name)) {
+        error_setg(errp, "Bitmap already exists: %s", name);
+        return false;
+    }
 
     if (s->qcow_version < 3) {
         /* Without autoclear_features, we would always have to assume
@@ -XXX,XX +XXX,XX @@ bool coroutine_fn qcow2_co_can_store_new_dirty_bitmap(BlockDriverState *bs,
         goto fail;
     }
 
-    if (s->nb_bitmaps == 0) {
-        return true;
+    FOR_EACH_DIRTY_BITMAP(bs, bitmap) {
+        if (bdrv_dirty_bitmap_get_persistence(bitmap)) {
+            nb_bitmaps++;
+            bitmap_directory_size +=
+                calc_dir_entry_size(strlen(bdrv_dirty_bitmap_name(bitmap)), 0);
+        }
     }
+    nb_bitmaps++;
+    bitmap_directory_size += calc_dir_entry_size(strlen(name), 0);
 
-    if (s->nb_bitmaps >= QCOW2_MAX_BITMAPS) {
+    if (nb_bitmaps > QCOW2_MAX_BITMAPS) {
         error_setg(errp,
                    "Maximum number of persistent bitmaps is already reached");
         goto fail;
     }
 
-    if (s->bitmap_directory_size + calc_dir_entry_size(strlen(name), 0) >
-        QCOW2_MAX_BITMAP_DIRECTORY_SIZE)
-    {
+    if (bitmap_directory_size > QCOW2_MAX_BITMAP_DIRECTORY_SIZE) {
         error_setg(errp, "Not enough space in the bitmap directory");
         goto fail;
     }
 
-    qemu_co_mutex_lock(&s->lock);
-    bm_list = bitmap_list_load(bs, s->bitmap_directory_offset,
-                               s->bitmap_directory_size, errp);
-    qemu_co_mutex_unlock(&s->lock);
-    if (bm_list == NULL) {
-        goto fail;
-    }
-
-    found = find_bitmap_by_name(bm_list, name);
-    bitmap_list_free(bm_list);
-    if (found) {
-        error_setg(errp, "Bitmap with the same name is already stored");
-        goto fail;
-    }
-
     return true;
 
 fail:
-- 
2.24.1

Probably due to blind copy-pasting, we have several instances of "qocw2"
in our iotests.  Fix them.

Reported-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20191107163708.833192-2-mreitz@redhat.com
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/060 | 2 +-
 tests/qemu-iotests/061 | 2 +-
 tests/qemu-iotests/062 | 2 +-
 tests/qemu-iotests/066 | 2 +-
 tests/qemu-iotests/068 | 2 +-
 tests/qemu-iotests/108 | 2 +-
 tests/qemu-iotests/138 | 2 +-
 tests/qemu-iotests/261 | 2 +-
 8 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/060
+++ b/tests/qemu-iotests/060
@@ -XXX,XX +XXX,XX @@ _filter_io_error()
 . ./common.rc
 . ./common.filter
 
-# This tests qocw2-specific low-level functionality
+# This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/061
+++ b/tests/qemu-iotests/061
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.rc
 . ./common.filter
 
-# This tests qocw2-specific low-level functionality
+# This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
diff --git a/tests/qemu-iotests/062 b/tests/qemu-iotests/062
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/062
+++ b/tests/qemu-iotests/062
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.rc
 . ./common.filter
 
-# This tests qocw2-specific low-level functionality
+# This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto generic
 
diff --git a/tests/qemu-iotests/066 b/tests/qemu-iotests/066
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/066
+++ b/tests/qemu-iotests/066
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.rc
 . ./common.filter
 
-# This tests qocw2-specific low-level functionality
+# This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto generic
 
diff --git a/tests/qemu-iotests/068 b/tests/qemu-iotests/068
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/068
+++ b/tests/qemu-iotests/068
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.rc
 . ./common.filter
 
-# This tests qocw2-specific low-level functionality
+# This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto generic
 
diff --git a/tests/qemu-iotests/108 b/tests/qemu-iotests/108
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/108
+++ b/tests/qemu-iotests/108
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.rc
 . ./common.filter
 
-# This tests qocw2-specific low-level functionality
+# This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
diff --git a/tests/qemu-iotests/138 b/tests/qemu-iotests/138
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/138
+++ b/tests/qemu-iotests/138
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.rc
 . ./common.filter
 
-# This tests qocw2-specific low-level functionality
+# This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
diff --git a/tests/qemu-iotests/261 b/tests/qemu-iotests/261
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/261
+++ b/tests/qemu-iotests/261
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.rc
 . ./common.filter
 
-# This tests qocw2-specific low-level functionality
+# This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
-- 
2.24.1

This is useful for tests that want to whitelist fields from dump-header
(with grep) but still print all header extensions.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-3-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/qcow2.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/tests/qemu-iotests/qcow2.py b/tests/qemu-iotests/qcow2.py
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/qcow2.py
+++ b/tests/qemu-iotests/qcow2.py
@@ -XXX,XX +XXX,XX @@ def cmd_dump_header(fd):
     h.dump()
     h.dump_extensions()
 
+def cmd_dump_header_exts(fd):
+    h = QcowHeader(fd)
+    h.dump_extensions()
+
 def cmd_set_header(fd, name, value):
     try:
         value = int(value, 0)
@@ -XXX,XX +XXX,XX @@ def cmd_set_feature_bit(fd, group, bit):
 
 cmds = [
     [ 'dump-header',          cmd_dump_header,          0, 'Dump image header and header extensions' ],
+    [ 'dump-header-exts',     cmd_dump_header_exts,     0, 'Dump image header extensions' ],
     [ 'set-header',           cmd_set_header,           2, 'Set a field in the header'],
     [ 'add-header-ext',       cmd_add_header_ext,       2, 'Add a header extension' ],
     [ 'add-header-ext-stdio', cmd_add_header_ext_stdio, 1, 'Add a header extension, data from stdin' ],
-- 
2.24.1

Print the feature fields as a set of bits so that filtering is easier.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-4-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/031.out  | 36 +++++++++----------
 tests/qemu-iotests/036.out  | 18 +++++-----
 tests/qemu-iotests/039.out  | 22 ++++++------
 tests/qemu-iotests/060.out  | 20 +++++------
 tests/qemu-iotests/061.out  | 72 ++++++++++++++++++-------------------
 tests/qemu-iotests/137.out  |  2 +-
 tests/qemu-iotests/qcow2.py | 18 +++++++---
 7 files changed, 99 insertions(+), 89 deletions(-)

diff --git a/tests/qemu-iotests/031.out b/tests/qemu-iotests/031.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/031.out
+++ b/tests/qemu-iotests/031.out
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             72
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             72
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             72
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             104
 
diff --git a/tests/qemu-iotests/036.out b/tests/qemu-iotests/036.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/036.out
+++ b/tests/qemu-iotests/036.out
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x8000000000000000
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     [63]
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x8000000000000000
+incompatible_features     []
+compatible_features       []
+autoclear_features        [63]
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             104
 
diff --git a/tests/qemu-iotests/039.out b/tests/qemu-iotests/039.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/039.out
+++ b/tests/qemu-iotests/039.out
@@ -XXX,XX +XXX,XX @@ QA output created by 039
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-incompatible_features     0x0
+incompatible_features     []
 No errors were found on the image.
 
 == Creating a dirty image file ==
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
-incompatible_features     0x1
+incompatible_features     [0]
 ERROR cluster 5 refcount=0 reference=1
 ERROR OFLAG_COPIED data cluster: l2_entry=8000000000050000 refcount=0
 
@@ -XXX,XX +XXX,XX @@ Data may be corrupted, or further writes to the image may corrupt it.
 == Read-only access must still work ==
 read 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-incompatible_features     0x1
+incompatible_features     [0]
 
 == Repairing the image file must succeed ==
 ERROR cluster 5 refcount=0 reference=1
@@ -XXX,XX +XXX,XX @@ The following inconsistencies were found and repaired:
 
 Double checking the fixed image now...
 No errors were found on the image.
-incompatible_features     0x0
+incompatible_features     []
 
 == Data should still be accessible after repair ==
 read 512/512 bytes at offset 0
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
-incompatible_features     0x1
+incompatible_features     [0]
 ERROR cluster 5 refcount=0 reference=1
 Rebuilding refcount structure
 Repairing cluster 1 refcount=1 reference=0
 Repairing cluster 2 refcount=1 reference=0
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-incompatible_features     0x0
+incompatible_features     []
 
 == Creating an image file with lazy_refcounts=off ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
-incompatible_features     0x0
+incompatible_features     []
 No errors were found on the image.
 
 == Committing to a backing file with lazy_refcounts=on ==
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 Image committed.
-incompatible_features     0x0
-incompatible_features     0x0
+incompatible_features     []
+incompatible_features     []
 No errors were found on the image.
 No errors were found on the image.
 
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
-incompatible_features     0x1
+incompatible_features     [0]
 ERROR cluster 5 refcount=0 reference=1
 ERROR OFLAG_COPIED data cluster: l2_entry=8000000000050000 refcount=0
 
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
-incompatible_features     0x0
+incompatible_features     []
 No errors were found on the image.
 *** done
diff --git a/tests/qemu-iotests/060.out b/tests/qemu-iotests/060.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/060.out
+++ b/tests/qemu-iotests/060.out
@@ -XXX,XX +XXX,XX @@ ERROR cluster 3 refcount=1 reference=3
 
 1 errors were found on the image.
 Data may be corrupted, or further writes to the image may corrupt it.
-incompatible_features     0x0
+incompatible_features     []
 qcow2: Marking image as corrupt: Preventing invalid write on metadata (overlaps with active L1 table); further corruption events will be suppressed
 write failed: Input/output error
-incompatible_features     0x2
+incompatible_features     [1]
 image: TEST_DIR/t.IMGFMT
 file format: IMGFMT
 virtual size: 64 MiB (67108864 bytes)
@@ -XXX,XX +XXX,XX @@ ERROR cluster 2 refcount=1 reference=2
 
 2 errors were found on the image.
 Data may be corrupted, or further writes to the image may corrupt it.
-incompatible_features     0x0
+incompatible_features     []
 qcow2: Marking image as corrupt: Preventing invalid write on metadata (overlaps with refcount block); further corruption events will be suppressed
 write failed: Input/output error
-incompatible_features     0x2
+incompatible_features     [1]
 ERROR refcount block 0 refcount=2
 ERROR cluster 2 refcount=1 reference=2
 Rebuilding refcount structure
@@ -XXX,XX +XXX,XX @@ The following inconsistencies were found and repaired:
 
 Double checking the fixed image now...
 No errors were found on the image.
-incompatible_features     0x0
+incompatible_features     []
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-incompatible_features     0x0
+incompatible_features     []
 
 === Testing cluster data reference into inactive L2 table ===
 
@@ -XXX,XX +XXX,XX @@ Data may be corrupted, or further writes to the image may corrupt it.
 
 1 leaked clusters were found on the image.
 This means waste of disk space, but no harm to data.
-incompatible_features     0x0
+incompatible_features     []
 qcow2: Marking image as corrupt: Preventing invalid write on metadata (overlaps with inactive L2 table); further corruption events will be suppressed
 write failed: Input/output error
-incompatible_features     0x2
+incompatible_features     [1]
 ERROR cluster 4 refcount=1 reference=2
 Leaked cluster 9 refcount=1 reference=0
 Repairing cluster 4 refcount=1 reference=2
@@ -XXX,XX +XXX,XX @@ The following inconsistencies were found and repaired:
 
 Double checking the fixed image now...
 No errors were found on the image.
-incompatible_features     0x0
+incompatible_features     []
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-incompatible_features     0x0
+incompatible_features     []
 read 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 No errors were found on the image.
diff --git a/tests/qemu-iotests/061.out b/tests/qemu-iotests/061.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/061.out
+++ b/tests/qemu-iotests/061.out
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x1
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       [0]
+autoclear_features        []
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             72
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x1
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       [0]
+autoclear_features        []
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             72
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x1
-compatible_features       0x1
-autoclear_features        0x0
+incompatible_features     [0]
+compatible_features       [0]
+autoclear_features        []
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x80000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             72
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x40000000000
-autoclear_features        0x40000000000
+incompatible_features     []
+compatible_features       [42]
+autoclear_features        [42]
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             72
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             72
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x1
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       [0]
+autoclear_features        []
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x10000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x1
-compatible_features       0x1
-autoclear_features        0x0
+incompatible_features     [0]
+compatible_features       [0]
+autoclear_features        []
 refcount_order            4
 header_length             104
 
@@ -XXX,XX +XXX,XX @@ refcount_table_offset     0x80000
 refcount_table_clusters   1
 nb_snapshots              0
 snapshot_offset           0x0
-incompatible_features     0x0
-compatible_features       0x0
-autoclear_features        0x0
+incompatible_features     []
+compatible_features       []
+autoclear_features        []
 refcount_order            4
 header_length             104
 
diff --git a/tests/qemu-iotests/137.out b/tests/qemu-iotests/137.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/137.out
+++ b/tests/qemu-iotests/137.out
@@ -XXX,XX +XXX,XX @@ qemu-io: Unsupported value 'blubb' for qcow2 option 'overlap-check'. Allowed are
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
-incompatible_features     0x0
+incompatible_features     []
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
 wrote 65536/65536 bytes at offset 0
 64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
diff --git a/tests/qemu-iotests/qcow2.py b/tests/qemu-iotests/qcow2.py
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/qcow2.py
+++ b/tests/qemu-iotests/qcow2.py
@@ -XXX,XX +XXX,XX @@ class QcowHeader:
         [ uint64_t, '%#x',  'snapshot_offset' ],
 
         # Version 3 header fields
-        [ uint64_t, '%#x',  'incompatible_features' ],
-        [ uint64_t, '%#x',  'compatible_features' ],
-        [ uint64_t, '%#x',  'autoclear_features' ],
+        [ uint64_t, 'mask', 'incompatible_features' ],
+        [ uint64_t, 'mask', 'compatible_features' ],
+        [ uint64_t, 'mask', 'autoclear_features' ],
         [ uint32_t, '%d',   'refcount_order' ],
         [ uint32_t, '%d',   'header_length' ],
     ];
@@ -XXX,XX +XXX,XX @@ class QcowHeader:
 
     def dump(self):
         for f in QcowHeader.fields:
-            print("%-25s" % f[2], f[1] % self.__dict__[f[2]])
+            value = self.__dict__[f[2]]
+            if f[1] == 'mask':
+                bits = []
+                for bit in range(64):
+                    if value & (1 << bit):
+                        bits.append(bit)
+                value_str = str(bits)
+            else:
+                value_str = f[1] % value
+
+            print("%-25s" % f[2], value_str)
         print("")
 
     def dump_extensions(self):
-- 
2.24.1

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-5-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/common.filter | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/common.filter
+++ b/tests/qemu-iotests/common.filter
@@ -XXX,XX +XXX,XX @@ _filter_qmp_empty_return()
     grep -v '{"return": {}}'
 }
 
+_filter_json_filename()
+{
+    $PYTHON -c 'import sys
+result, *fnames = sys.stdin.read().split("json:{")
+depth = 0
+for fname in fnames:
+    depth += 1 # For the opening brace in the split separator
+    for chr_i, chr in enumerate(fname):
+        if chr == "{":
+            depth += 1
+        elif chr == "}":
+            depth -= 1
+            if depth == 0:
+                break
+
+    # json:{} filenames may be nested; filter out everything from
+    # inside the outermost one
+    if depth == 0:
+        chr_i += 1 # First character past the filename
+        result += "json:{ /* filtered */ }" + fname[chr_i:]
+
+sys.stdout.write(result)'
+}
+
 # make sure this script returns success
 true
-- 
2.24.1

This test can run just fine with other values for refcount_bits, so we
should filter the value from qcow2.py's dump-header.  In fact, we can
filter everything but the feature bits and header extensions, because
that is what the test is about.

(036 currently ignores user-specified image options, but that will be
fixed in the next patch.)

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-6-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/036     |  9 ++++---
 tests/qemu-iotests/036.out | 48 --------------------------------------
 2 files changed, 6 insertions(+), 51 deletions(-)

diff --git a/tests/qemu-iotests/036 b/tests/qemu-iotests/036
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/036
+++ b/tests/qemu-iotests/036
@@ -XXX,XX +XXX,XX @@ $PYTHON qcow2.py "$TEST_IMG" set-feature-bit incompatible 63
 
 # Without feature table
 $PYTHON qcow2.py "$TEST_IMG" del-header-ext 0x6803f857
-$PYTHON qcow2.py "$TEST_IMG" dump-header
+$PYTHON qcow2.py "$TEST_IMG" dump-header | grep features
+$PYTHON qcow2.py "$TEST_IMG" dump-header-exts
 _img_info
 
 # With feature table containing bit 63
@@ -XXX,XX +XXX,XX @@ echo === Create image with unknown autoclear feature bit ===
 echo
 _make_test_img 64M
 $PYTHON qcow2.py "$TEST_IMG" set-feature-bit autoclear 63
-$PYTHON qcow2.py "$TEST_IMG" dump-header
+$PYTHON qcow2.py "$TEST_IMG" dump-header | grep features
+$PYTHON qcow2.py "$TEST_IMG" dump-header-exts
 
 echo
 echo === Repair image ===
 echo
 _check_test_img -r all
 
-$PYTHON qcow2.py "$TEST_IMG" dump-header
+$PYTHON qcow2.py "$TEST_IMG" dump-header | grep features
+$PYTHON qcow2.py "$TEST_IMG" dump-header-exts
 
 # success, all done
 echo "*** done"
diff --git a/tests/qemu-iotests/036.out b/tests/qemu-iotests/036.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/036.out
+++ b/tests/qemu-iotests/036.out
@@ -XXX,XX +XXX,XX @@ QA output created by 036
 === Image with unknown incompatible feature bit ===
 
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
-magic                     0x514649fb
-version                   3
-backing_file_offset       0x0
-backing_file_size         0x0
-cluster_bits              16
-size                      67108864
-crypt_method              0
-l1_size                   1
-l1_table_offset           0x30000
-refcount_table_offset     0x10000
-refcount_table_clusters   1
-nb_snapshots              0
-snapshot_offset           0x0
 incompatible_features     [63]
 compatible_features       []
 autoclear_features        []
-refcount_order            4
-header_length             104
-
 qemu-img: Could not open 'TEST_DIR/t.IMGFMT': Unsupported IMGFMT feature(s): Unknown incompatible feature: 8000000000000000
 qemu-img: Could not open 'TEST_DIR/t.IMGFMT': Unsupported IMGFMT feature(s): Test feature
 
@@ -XXX,XX +XXX,XX @@ qemu-img: Could not open 'TEST_DIR/t.IMGFMT': Unsupported IMGFMT feature(s): tes
 === Create image with unknown autoclear feature bit ===
 
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
-magic                     0x514649fb
-version                   3
-backing_file_offset       0x0
-backing_file_size         0x0
-cluster_bits              16
-size                      67108864
-crypt_method              0
-l1_size                   1
-l1_table_offset           0x30000
-refcount_table_offset     0x10000
-refcount_table_clusters   1
-nb_snapshots              0
-snapshot_offset           0x0
 incompatible_features     []
 compatible_features       []
 autoclear_features        [63]
-refcount_order            4
-header_length             104
-
 Header extension:
 magic                     0x6803f857
 length                    192
@@ -XXX,XX +XXX,XX @@ data                      <binary>
 === Repair image ===
 
 No errors were found on the image.
-magic                     0x514649fb
-version                   3
-backing_file_offset       0x0
-backing_file_size         0x0
-cluster_bits              16
-size                      67108864
-crypt_method              0
-l1_size                   1
-l1_table_offset           0x30000
-refcount_table_offset     0x10000
-refcount_table_clusters   1
-nb_snapshots              0
-snapshot_offset           0x0
 incompatible_features     []
 compatible_features       []
 autoclear_features        []
-refcount_order            4
-header_length             104
-
 Header extension:
 magic                     0x6803f857
 length                    192
-- 
2.24.1

Some tests require compat=1.1 and thus set IMGOPTS='compat=1.1'
globally.  That is not how it should be done; instead, they should
simply set _unsupported_imgopts to compat=0.10 (compat=1.1 is the
default anyway).

This makes the tests heed user-specified $IMGOPTS.  Some do not work
with all image options, though, so we need to disable them accordingly.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsky@redhat.com>
Message-id: 20191107163708.833192-7-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/036 | 3 +--
 tests/qemu-iotests/060 | 4 ++--
 tests/qemu-iotests/062 | 3 ++-
 tests/qemu-iotests/066 | 3 ++-
 tests/qemu-iotests/068 | 3 ++-
 tests/qemu-iotests/098 | 4 ++--
 6 files changed, 11 insertions(+), 9 deletions(-)

diff --git a/tests/qemu-iotests/036 b/tests/qemu-iotests/036
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/036
+++ b/tests/qemu-iotests/036
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
-
 # Only qcow2v3 and later supports feature bits
-IMGOPTS="compat=1.1"
+_unsupported_imgopts 'compat=0.10'
 
 echo
 echo === Image with unknown incompatible feature bit ===
diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/060
+++ b/tests/qemu-iotests/060
@@ -XXX,XX +XXX,XX @@ _filter_io_error()
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# These tests only work for compat=1.1 images with refcount_bits=16
+_unsupported_imgopts 'compat=0.10' 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
 
 # The repair process will create a large file - so check for availability first
 _require_large_file 64G
@@ -XXX,XX +XXX,XX @@ l1_offset=196608 # 0x30000 (XXX: just an assumption)
 l2_offset=262144 # 0x40000 (XXX: just an assumption)
 l2_offset_after_snapshot=524288 # 0x80000 (XXX: just an assumption)
 
-IMGOPTS="compat=1.1"
-
 OPEN_RW="open -o overlap-check=all $TEST_IMG"
 # Overlap checks are done before write operations only, therefore opening an
 # image read-only makes the overlap-check option irrelevant
diff --git a/tests/qemu-iotests/062 b/tests/qemu-iotests/062
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/062
+++ b/tests/qemu-iotests/062
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto generic
+# We need zero clusters and snapshots
+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
 
-IMGOPTS="compat=1.1"
 IMG_SIZE=64M
 
 echo
diff --git a/tests/qemu-iotests/066 b/tests/qemu-iotests/066
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/066
+++ b/tests/qemu-iotests/066
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto generic
+# We need zero clusters and snapshots
+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
 
 # Intentionally create an unaligned image
-IMGOPTS="compat=1.1"
 IMG_SIZE=$((64 * 1024 * 1024 + 512))
 
 echo
diff --git a/tests/qemu-iotests/068 b/tests/qemu-iotests/068
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/068
+++ b/tests/qemu-iotests/068
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto generic
+# Internal snapshots are (currently) impossible with refcount_bits=1
+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
 
-IMGOPTS="compat=1.1"
 IMG_SIZE=128K
 
 case "$QEMU_DEFAULT_MACHINE" in
diff --git a/tests/qemu-iotests/098 b/tests/qemu-iotests/098
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/098
+++ b/tests/qemu-iotests/098
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt qcow2
 _supported_proto file
-
-IMGOPTS="compat=1.1"
+# The code path we want to test here only works for compat=1.1 images
+_unsupported_imgopts 'compat=0.10'
 
 for event in l1_update empty_image_prepare reftable_update refblock_alloc; do
 
-- 
2.24.1

This will allow us to add more options than just -b.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-9-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/common.rc | 28 ++++++++++++++++++++--------
 1 file changed, 20 insertions(+), 8 deletions(-)

diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -XXX,XX +XXX,XX @@ _make_test_img()
     # extra qemu-img options can be added by tests
     # at least one argument (the image size) needs to be added
     local extra_img_options=""
-    local image_size=$*
     local optstr=""
     local img_name=""
     local use_backing=0
     local backing_file=""
     local object_options=""
+    local misc_params=()
 
     if [ -n "$TEST_IMG_FILE" ]; then
         img_name=$TEST_IMG_FILE
@@ -XXX,XX +XXX,XX @@ _make_test_img()
         optstr=$(_optstr_add "$optstr" "key-secret=keysec0")
     fi
 
-    if [ "$1" = "-b" ]; then
-        use_backing=1
-        backing_file=$2
-        image_size=$3
-    fi
+    for param; do
+        if [ "$use_backing" = "1" -a -z "$backing_file" ]; then
+            backing_file=$param
+            continue
+        fi
+
+        case "$param" in
+            -b)
+                use_backing=1
+                ;;
+
+            *)
+                misc_params=("${misc_params[@]}" "$param")
+                ;;
+        esac
+    done
+
     if [ $ "$IMGFMT" = "qcow2" -o "$IMGFMT" = "qed" $ -a -n "$CLUSTER_SIZE" ]; then
         optstr=$(_optstr_add "$optstr" "cluster_size=$CLUSTER_SIZE")
     fi
@@ -XXX,XX +XXX,XX @@ _make_test_img()
     # XXX(hch): have global image options?
     (
      if [ $use_backing = 1 ]; then
-        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options -b "$backing_file" "$img_name" $image_size 2>&1
+        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options -b "$backing_file" "$img_name" "${misc_params[@]}" 2>&1
      else
-        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options "$img_name" $image_size 2>&1
+        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options "$img_name" "${misc_params[@]}" 2>&1
      fi
     ) | _filter_img_create
 
-- 
2.24.1

Blindly overriding IMGOPTS is suboptimal as this discards user-specified
options.  Whatever options the test needs should simply be appended.

Some tests do this (with IMGOPTS=$(_optstr_add "$IMGOPTS" "...")), but
that is cumbersome.  It’s simpler to just give _make_test_img an -o
parameter with which tests can add options.

Some tests actually must override the user-specified options, though,
for example when creating an image in a different format than the test
$IMGFMT.  For such cases, --no-opts allows clearing the current option
list.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-10-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/common.rc | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -XXX,XX +XXX,XX @@ _make_test_img()
     local use_backing=0
     local backing_file=""
     local object_options=""
+    local opts_param=false
     local misc_params=()
 
     if [ -n "$TEST_IMG_FILE" ]; then
@@ -XXX,XX +XXX,XX @@ _make_test_img()
         if [ "$use_backing" = "1" -a -z "$backing_file" ]; then
             backing_file=$param
             continue
+        elif $opts_param; then
+            optstr=$(_optstr_add "$optstr" "$param")
+            opts_param=false
+            continue
         fi
 
         case "$param" in
@@ -XXX,XX +XXX,XX @@ _make_test_img()
                 use_backing=1
                 ;;
 
+            -o)
+                opts_param=true
+                ;;
+
+            --no-opts)
+                optstr=""
+                ;;
+
             *)
                 misc_params=("${misc_params[@]}" "$param")
                 ;;
-- 
2.24.1

Tests should not overwrite all user-supplied image options, but only add
to it (which will effectively overwrite conflicting values).  Accomplish
this by passing options to _make_test_img via -o instead of $IMGOPTS.

For some tests, there is no functional change because they already only
appended options to IMGOPTS.  For these, this patch is just a
simplification.

For others, this is a change, so they now heed user-specified $IMGOPTS.
Some of those tests do not work with all image options, though, so we
need to disable them accordingly.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-12-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/031 |  9 ++++---
 tests/qemu-iotests/039 | 24 ++++++------------
 tests/qemu-iotests/059 | 18 ++++++-------
 tests/qemu-iotests/060 |  6 ++---
 tests/qemu-iotests/061 | 57 ++++++++++++++++++++++--------------------
 tests/qemu-iotests/079 |  3 +--
 tests/qemu-iotests/106 |  2 +-
 tests/qemu-iotests/108 |  2 +-
 tests/qemu-iotests/112 | 32 ++++++++++++------------
 tests/qemu-iotests/115 |  3 +--
 tests/qemu-iotests/121 |  6 ++---
 tests/qemu-iotests/125 |  2 +-
 tests/qemu-iotests/137 |  2 +-
 tests/qemu-iotests/138 |  3 +--
 tests/qemu-iotests/175 |  2 +-
 tests/qemu-iotests/190 |  2 +-
 tests/qemu-iotests/191 |  3 +--
 tests/qemu-iotests/220 |  4 ++-
 tests/qemu-iotests/243 |  6 +++--
 tests/qemu-iotests/244 | 10 +++++---
 tests/qemu-iotests/250 |  3 +--
 tests/qemu-iotests/265 |  2 +-
 22 files changed, 100 insertions(+), 101 deletions(-)

diff --git a/tests/qemu-iotests/031 b/tests/qemu-iotests/031
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/031
+++ b/tests/qemu-iotests/031
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
+# We want to test compat=0.10, which does not support refcount widths
+# other than 16
+_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
 
 CLUSTER_SIZE=65536
 
 # qcow2.py output depends on the exact options used, so override the command
 # line here as an exception
-for IMGOPTS in "compat=0.10" "compat=1.1"; do
+for compat in "compat=0.10" "compat=1.1"; do
 
     echo
-    echo ===== Testing with -o $IMGOPTS =====
+    echo ===== Testing with -o $compat =====
     echo
     echo === Create image with unknown header extension ===
     echo
-    _make_test_img 64M
+    _make_test_img -o $compat 64M
     $PYTHON qcow2.py "$TEST_IMG" add-header-ext 0x12345678 "This is a test header extension"
     $PYTHON qcow2.py "$TEST_IMG" dump-header
     _check_test_img
diff --git a/tests/qemu-iotests/039 b/tests/qemu-iotests/039
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/039
+++ b/tests/qemu-iotests/039
@@ -XXX,XX +XXX,XX @@ size=128M
 echo
 echo "== Checking that image is clean on shutdown =="
 
-IMGOPTS="compat=1.1,lazy_refcounts=on"
-_make_test_img $size
+_make_test_img -o "compat=1.1,lazy_refcounts=on" $size
 
 $QEMU_IO -c "write -P 0x5a 0 512" "$TEST_IMG" | _filter_qemu_io
 
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "== Creating a dirty image file =="
 
-IMGOPTS="compat=1.1,lazy_refcounts=on"
-_make_test_img $size
+_make_test_img -o "compat=1.1,lazy_refcounts=on" $size
 
 _NO_VALGRIND \
 $QEMU_IO -c "write -P 0x5a 0 512" \
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0x5a 0 512" "$TEST_IMG" | _filter_qemu_io
 echo
 echo "== Opening a dirty image read/write should repair it =="
 
-IMGOPTS="compat=1.1,lazy_refcounts=on"
-_make_test_img $size
+_make_test_img -o "compat=1.1,lazy_refcounts=on" $size
 
 _NO_VALGRIND \
 $QEMU_IO -c "write -P 0x5a 0 512" \
@@ -XXX,XX +XXX,XX @@ $PYTHON qcow2.py "$TEST_IMG" dump-header | grep incompatible_features
 echo
 echo "== Creating an image file with lazy_refcounts=off =="
 
-IMGOPTS="compat=1.1,lazy_refcounts=off"
-_make_test_img $size
+_make_test_img -o "compat=1.1,lazy_refcounts=off" $size
 
 _NO_VALGRIND \
 $QEMU_IO -c "write -P 0x5a 0 512" \
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "== Committing to a backing file with lazy_refcounts=on =="
 
-IMGOPTS="compat=1.1,lazy_refcounts=on"
-TEST_IMG="$TEST_IMG".base _make_test_img $size
+TEST_IMG="$TEST_IMG".base _make_test_img -o "compat=1.1,lazy_refcounts=on" $size
 
-IMGOPTS="compat=1.1,lazy_refcounts=on,backing_file=$TEST_IMG.base"
-_make_test_img $size
+_make_test_img -o "compat=1.1,lazy_refcounts=on,backing_file=$TEST_IMG.base" $size
 
 $QEMU_IO -c "write 0 512" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IMG commit "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ TEST_IMG="$TEST_IMG".base _check_test_img
 echo
 echo "== Changing lazy_refcounts setting at runtime =="
 
-IMGOPTS="compat=1.1,lazy_refcounts=off"
-_make_test_img $size
+_make_test_img -o "compat=1.1,lazy_refcounts=off" $size
 
 _NO_VALGRIND \
 $QEMU_IO -c "reopen -o lazy-refcounts=on" \
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "reopen -o lazy-refcounts=on" \
 $PYTHON qcow2.py "$TEST_IMG" dump-header | grep incompatible_features
 _check_test_img
 
-IMGOPTS="compat=1.1,lazy_refcounts=on"
-_make_test_img $size
+_make_test_img -o "compat=1.1,lazy_refcounts=on" $size
 
 _NO_VALGRIND \
 $QEMU_IO -c "reopen -o lazy-refcounts=off" \
diff --git a/tests/qemu-iotests/059 b/tests/qemu-iotests/059
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/059
+++ b/tests/qemu-iotests/059
@@ -XXX,XX +XXX,XX @@ poke_file "$TEST_IMG" "$grain_table_size_offset" "\x01\x00\x00\x00"
 
 echo
 echo "=== Testing monolithicFlat creation and opening ==="
-IMGOPTS="subformat=monolithicFlat" _make_test_img 2G
+_make_test_img -o "subformat=monolithicFlat" 2G
 _img_info
 _cleanup_test_img
 
 echo
 echo "=== Testing monolithicFlat with zeroed_grain ==="
-IMGOPTS="subformat=monolithicFlat,zeroed_grain=on" _make_test_img 2G
+_make_test_img -o "subformat=monolithicFlat,zeroed_grain=on" 2G
 _cleanup_test_img
 
 echo
 echo "=== Testing big twoGbMaxExtentFlat ==="
-IMGOPTS="subformat=twoGbMaxExtentFlat" _make_test_img 1000G
+_make_test_img -o "subformat=twoGbMaxExtentFlat" 1000G
 $QEMU_IMG info $TEST_IMG | _filter_testdir | sed -e 's/cid: [0-9]*/cid: XXXXXXXX/'
 _cleanup_test_img
 
@@ -XXX,XX +XXX,XX @@ _img_info
 
 echo
 echo "=== Testing truncated sparse ==="
-IMGOPTS="subformat=monolithicSparse" _make_test_img 100G
+_make_test_img -o "subformat=monolithicSparse" 100G
 truncate -s 10M $TEST_IMG
 _img_info
 
 echo
 echo "=== Converting to streamOptimized from image with small cluster size==="
-TEST_IMG="$TEST_IMG.qcow2" IMGFMT=qcow2 IMGOPTS="cluster_size=4096" _make_test_img 1G
+TEST_IMG="$TEST_IMG.qcow2" IMGFMT=qcow2 _make_test_img -o "cluster_size=4096" 1G
 $QEMU_IO -f qcow2 -c "write -P 0xa 0 512" "$TEST_IMG.qcow2" | _filter_qemu_io
 $QEMU_IO -f qcow2 -c "write -P 0xb 10240 512" "$TEST_IMG.qcow2" | _filter_qemu_io
 $QEMU_IMG convert -f qcow2 -O vmdk -o subformat=streamOptimized "$TEST_IMG.qcow2" "$TEST_IMG" 2>&1
@@ -XXX,XX +XXX,XX @@ echo "=== Testing monolithicFlat with internally generated JSON file name ==="
 
 echo '--- blkdebug ---'
 # Should work, because bdrv_dirname() works fine with blkdebug
-IMGOPTS="subformat=monolithicFlat" _make_test_img 64M
+_make_test_img -o "subformat=monolithicFlat" 64M
 $QEMU_IO -c "open -o driver=$IMGFMT,file.driver=blkdebug,file.image.filename=$TEST_IMG,file.inject-error.0.event=read_aio" \
          -c info \
     2>&1 \
@@ -XXX,XX +XXX,XX @@ _cleanup_test_img
 
 echo '--- quorum ---'
 # Should not work, because bdrv_dirname() does not work with quorum
-IMGOPTS="subformat=monolithicFlat" _make_test_img 64M
+_make_test_img -o "subformat=monolithicFlat" 64M
 cp "$TEST_IMG" "$TEST_IMG.orig"
 
 filename="json:{
@@ -XXX,XX +XXX,XX @@ _cleanup_test_img
 
 echo
 echo "=== Testing 4TB monolithicFlat creation and IO ==="
-IMGOPTS="subformat=monolithicFlat" _make_test_img 4T
+_make_test_img -o "subformat=monolithicFlat" 4T
 _img_info
 $QEMU_IO -c "write -P 0xa 900G 512" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "read -v 900G 1024" "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _cleanup_test_img
 echo
 echo "=== Testing qemu-img map on extents ==="
 for fmt in monolithicSparse twoGbMaxExtentSparse; do
-    IMGOPTS="subformat=$fmt" _make_test_img 31G
+    _make_test_img -o "subformat=$fmt" 31G
     $QEMU_IO -c "write 65024 1k" "$TEST_IMG" | _filter_qemu_io
     $QEMU_IO -c "write 2147483136 1k" "$TEST_IMG" | _filter_qemu_io
     $QEMU_IO -c "write 5G 1k" "$TEST_IMG" | _filter_qemu_io
diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/060
+++ b/tests/qemu-iotests/060
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c 'write 0k 64k' "$BACKING_IMG" | _filter_qemu_io
 # compat=0.10 is required in order to make the following discard actually
 # unallocate the sector rather than make it a zero sector - we want COW, after
 # all.
-IMGOPTS='compat=0.10' _make_test_img -b "$BACKING_IMG" 1G
+_make_test_img -o 'compat=0.10' -b "$BACKING_IMG" 1G
 # Write two clusters, the second one enforces creation of an L2 table after
 # the first data cluster.
 $QEMU_IO -c 'write 0k 64k' -c 'write 512M 64k' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
 echo "=== Discarding a non-covered in-bounds refblock ==="
 echo
 
-IMGOPTS='refcount_bits=1' _make_test_img 64M
+_make_test_img -o 'refcount_bits=1' 64M
 
 # Pretend there's a refblock somewhere where there is no refblock to
 # cover it (but the covering refblock has a valid index in the
@@ -XXX,XX +XXX,XX @@ echo
 echo "=== Discarding a refblock covered by an unaligned refblock ==="
 echo
 
-IMGOPTS='refcount_bits=1' _make_test_img 64M
+_make_test_img -o 'refcount_bits=1' 64M
 
 # Same as above
 poke_file "$TEST_IMG" "$(($rt_offset+8))" "\x00\x00\x00\x10\x00\x00\x00\x00"
diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/061
+++ b/tests/qemu-iotests/061
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# Conversion between different compat versions can only really work
+# with refcount_bits=16
+_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
 
 echo
 echo "=== Testing version downgrade with zero expansion ==="
 echo
-IMGOPTS="compat=1.1,lazy_refcounts=on" _make_test_img 64M
+_make_test_img -o "compat=1.1,lazy_refcounts=on" 64M
 $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
 $PYTHON qcow2.py "$TEST_IMG" dump-header
 $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing version downgrade with zero expansion and 4K cache entries ==="
 echo
-IMGOPTS="compat=1.1,lazy_refcounts=on" _make_test_img 64M
+_make_test_img -o "compat=1.1,lazy_refcounts=on" 64M
 $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "write -z 32M 128k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c map "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing dirty version downgrade ==="
 echo
-IMGOPTS="compat=1.1,lazy_refcounts=on" _make_test_img 64M
+_make_test_img -o "compat=1.1,lazy_refcounts=on" 64M
 _NO_VALGRIND \
 $QEMU_IO -c "write -P 0x2a 0 128k" -c flush \
          -c "sigraise $(kill -l KILL)" "$TEST_IMG" 2>&1 | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing version downgrade with unknown compat/autoclear flags ==="
 echo
-IMGOPTS="compat=1.1" _make_test_img 64M
+_make_test_img -o "compat=1.1" 64M
 $PYTHON qcow2.py "$TEST_IMG" set-feature-bit compatible 42
 $PYTHON qcow2.py "$TEST_IMG" set-feature-bit autoclear 42
 $PYTHON qcow2.py "$TEST_IMG" dump-header
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing version upgrade and resize ==="
 echo
-IMGOPTS="compat=0.10" _make_test_img 64M
+_make_test_img -o "compat=0.10" 64M
 $QEMU_IO -c "write -P 0x2a 42M 64k" "$TEST_IMG" | _filter_qemu_io
 $PYTHON qcow2.py "$TEST_IMG" dump-header
 $QEMU_IMG amend -o "compat=1.1,lazy_refcounts=on,size=128M" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing dirty lazy_refcounts=off ==="
 echo
-IMGOPTS="compat=1.1,lazy_refcounts=on" _make_test_img 64M
+_make_test_img -o "compat=1.1,lazy_refcounts=on" 64M
 _NO_VALGRIND \
 $QEMU_IO -c "write -P 0x2a 0 128k" -c flush \
          -c "sigraise $(kill -l KILL)" "$TEST_IMG" 2>&1 | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing backing file ==="
 echo
-IMGOPTS="compat=1.1" _make_test_img 64M
-IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 64M
+_make_test_img -o "compat=1.1" 64M
+TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M
 $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io
 $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IMG amend -o "backing_file=$TEST_IMG.base,backing_fmt=qcow2" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing invalid configurations ==="
 echo
-IMGOPTS="compat=0.10" _make_test_img 64M
+_make_test_img -o "compat=0.10" 64M
 $QEMU_IMG amend -o "lazy_refcounts=on" "$TEST_IMG"
 $QEMU_IMG amend -o "compat=1.1" "$TEST_IMG" # actually valid
 $QEMU_IMG amend -o "compat=0.10,lazy_refcounts=on" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ $QEMU_IMG amend -o "preallocation=on" "$TEST_IMG"
 echo
 echo "=== Testing correct handling of unset value ==="
 echo
-IMGOPTS="compat=1.1,cluster_size=1k" _make_test_img 64M
+_make_test_img -o "compat=1.1,cluster_size=1k" 64M
 echo "Should work:"
 $QEMU_IMG amend -o "lazy_refcounts=on" "$TEST_IMG"
 echo "Should not work:" # Just to know which of these tests actually fails
@@ -XXX,XX +XXX,XX @@ $QEMU_IMG amend -o "cluster_size=64k" "$TEST_IMG"
 echo
 echo "=== Testing zero expansion on inactive clusters ==="
 echo
-IMGOPTS="compat=1.1" _make_test_img 64M
+_make_test_img -o "compat=1.1" 64M
 $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IMG snapshot -c foo "$TEST_IMG"
 $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io
 echo
 echo "=== Testing zero expansion on shared L2 table ==="
 echo
-IMGOPTS="compat=1.1" _make_test_img 64M
+_make_test_img -o "compat=1.1" 64M
 $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IMG snapshot -c foo "$TEST_IMG"
 $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io
 echo
 echo "=== Testing zero expansion on backed image ==="
 echo
-IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 64M
+TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M
 $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io
-IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 64M
+_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 64M
 $QEMU_IO -c "read -P 0x2a 0 128k" -c "write -z 0 64k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
 _check_test_img
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 64k" -c "read -P 0x2a 64k 64k" "$TEST_IMG" | _filter_qe
 echo
 echo "=== Testing zero expansion on backed inactive clusters ==="
 echo
-IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 64M
+TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M
 $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io
-IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 64M
+_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 64M
 $QEMU_IO -c "write -z 0 64k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IMG snapshot -c foo "$TEST_IMG"
 $QEMU_IO -c "write -P 0x42 0 128k" "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 64k" -c "read -P 0x2a 64k 64k" "$TEST_IMG" | _filter_qe
 echo
 echo "=== Testing zero expansion on backed image with shared L2 table ==="
 echo
-IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 64M
+TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M
 $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io
-IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 64M
+_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 64M
 $QEMU_IO -c "write -z 0 128k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IMG snapshot -c foo "$TEST_IMG"
 $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io
 echo
 echo "=== Testing preallocated zero expansion on full image ==="
 echo
-IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG" _make_test_img 64M
+TEST_IMG="$TEST_IMG" _make_test_img -o "compat=1.1" 64M
 $QEMU_IO -c "write -P 0x2a 0 64M" "$TEST_IMG" -c "write -z 0 64M" | _filter_qemu_io
 $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
 _check_test_img
@@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "read -P 0 0 64M" "$TEST_IMG" | _filter_qemu_io
 echo
 echo "=== Testing progress report without snapshot ==="
 echo
-IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 4G
-IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 4G
+TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 4G
+_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 4G
 $QEMU_IO -c "write -z 0  64k" \
          -c "write -z 1G 64k" \
          -c "write -z 2G 64k" \
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing progress report with snapshot ==="
 echo
-IMGOPTS="compat=1.1" TEST_IMG="$TEST_IMG.base" _make_test_img 4G
-IMGOPTS="compat=1.1" _make_test_img -b "$TEST_IMG.base" 4G
+TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 4G
+_make_test_img -o "compat=1.1" -b "$TEST_IMG.base" 4G
 $QEMU_IO -c "write -z 0  64k" \
          -c "write -z 1G 64k" \
          -c "write -z 2G 64k" \
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Testing version downgrade with external data file ==="
 echo
-IMGOPTS="compat=1.1,data_file=$TEST_IMG.data" _make_test_img 64M
+_make_test_img -o "compat=1.1,data_file=$TEST_IMG.data" 64M
 $QEMU_IMG amend -o "compat=0.10" "$TEST_IMG"
 _img_info --format-specific
 _check_test_img
@@ -XXX,XX +XXX,XX @@ _check_test_img
 echo
 echo "=== Try changing the external data file ==="
 echo
-IMGOPTS="compat=1.1" _make_test_img 64M
+_make_test_img -o "compat=1.1" 64M
 $QEMU_IMG amend -o "data_file=foo" "$TEST_IMG"
 
 echo
-IMGOPTS="compat=1.1,data_file=$TEST_IMG.data" _make_test_img 64M
+_make_test_img -o "compat=1.1,data_file=$TEST_IMG.data" 64M
 $QEMU_IMG amend -o "data_file=foo" "$TEST_IMG"
 _img_info --format-specific
 TEST_IMG="data-file.filename=$TEST_IMG.data,file.filename=$TEST_IMG" _img_info --format-specific --image-opts
@@ -XXX,XX +XXX,XX @@ TEST_IMG="data-file.filename=$TEST_IMG.data,file.filename=$TEST_IMG" _img_info -
 echo
 echo "=== Clearing and setting data-file-raw ==="
 echo
-IMGOPTS="compat=1.1,data_file=$TEST_IMG.data,data_file_raw=on" _make_test_img 64M
+_make_test_img -o "compat=1.1,data_file=$TEST_IMG.data,data_file_raw=on" 64M
 $QEMU_IMG amend -o "data_file_raw=on" "$TEST_IMG"
 _img_info --format-specific
 _check_test_img
diff --git a/tests/qemu-iotests/079 b/tests/qemu-iotests/079
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/079
+++ b/tests/qemu-iotests/079
@@ -XXX,XX +XXX,XX @@ echo
 cluster_sizes="16384 32768 65536 131072 262144 524288 1048576 2097152 4194304"
 
 for s in $cluster_sizes; do
-    IMGOPTS=$(_optstr_add "$IMGOPTS" "preallocation=metadata,cluster_size=$s") \
-        _make_test_img 4G
+    _make_test_img -o "preallocation=metadata,cluster_size=$s" 4G
 done
 
 # success, all done
diff --git a/tests/qemu-iotests/106 b/tests/qemu-iotests/106
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/106
+++ b/tests/qemu-iotests/106
@@ -XXX,XX +XXX,XX @@ for create_mode in off falloc full; do
         echo
         echo "--- create_mode=$create_mode growth_mode=$growth_mode ---"
 
-        IMGOPTS="preallocation=$create_mode" _make_test_img ${CREATION_SIZE}K
+        _make_test_img -o "preallocation=$create_mode" ${CREATION_SIZE}K
         $QEMU_IMG resize -f "$IMGFMT" --preallocation=$growth_mode "$TEST_IMG" +${GROWTH_SIZE}K
 
         expected_size=0
diff --git a/tests/qemu-iotests/108 b/tests/qemu-iotests/108
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/108
+++ b/tests/qemu-iotests/108
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== Repairing unreferenced data cluster in new refblock area ==='
 echo
 
-IMGOPTS='cluster_size=512' _make_test_img 64M
+_make_test_img -o 'cluster_size=512' 64M
 # Allocate the first 128 kB in the image (first refblock)
 $QEMU_IO -c 'write 0 0x1b200' "$TEST_IMG" | _filter_qemu_io
 # should be 131072 == 0x20000
diff --git a/tests/qemu-iotests/112 b/tests/qemu-iotests/112
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/112
+++ b/tests/qemu-iotests/112
@@ -XXX,XX +XXX,XX @@ echo '=== refcount_bits limits ==='
 echo
 
 # Must be positive (non-zero)
-IMGOPTS="$IMGOPTS,refcount_bits=0" _make_test_img 64M
+_make_test_img -o "refcount_bits=0" 64M
 # Must be positive (non-negative)
-IMGOPTS="$IMGOPTS,refcount_bits=-1" _make_test_img 64M
+_make_test_img -o "refcount_bits=-1" 64M
 # May not exceed 64
-IMGOPTS="$IMGOPTS,refcount_bits=128" _make_test_img 64M
+_make_test_img -o "refcount_bits=128" 64M
 # Must be a power of two
-IMGOPTS="$IMGOPTS,refcount_bits=42" _make_test_img 64M
+_make_test_img -o "refcount_bits=42" 64M
 
 # 1 is the minimum
-IMGOPTS="$IMGOPTS,refcount_bits=1" _make_test_img 64M
+_make_test_img -o "refcount_bits=1" 64M
 print_refcount_bits
 
 # 64 is the maximum
-IMGOPTS="$IMGOPTS,refcount_bits=64" _make_test_img 64M
+_make_test_img -o "refcount_bits=64" 64M
 print_refcount_bits
 
 # 16 is the default
@@ -XXX,XX +XXX,XX @@ echo '=== refcount_bits and compat=0.10 ==='
 echo
 
 # Should work
-IMGOPTS="$IMGOPTS,compat=0.10,refcount_bits=16" _make_test_img 64M
+_make_test_img -o "compat=0.10,refcount_bits=16" 64M
 print_refcount_bits
 
 # Should not work
-IMGOPTS="$IMGOPTS,compat=0.10,refcount_bits=1" _make_test_img 64M
-IMGOPTS="$IMGOPTS,compat=0.10,refcount_bits=64" _make_test_img 64M
+_make_test_img -o "compat=0.10,refcount_bits=1" 64M
+_make_test_img -o "compat=0.10,refcount_bits=64" 64M
 
 
 echo
 echo '=== Snapshot limit on refcount_bits=1 ==='
 echo
 
-IMGOPTS="$IMGOPTS,refcount_bits=1" _make_test_img 64M
+_make_test_img -o "refcount_bits=1" 64M
 print_refcount_bits
 
 $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== Snapshot limit on refcount_bits=2 ==='
 echo
 
-IMGOPTS="$IMGOPTS,refcount_bits=2" _make_test_img 64M
+_make_test_img -o "refcount_bits=2" 64M
 print_refcount_bits
 
 $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== Compressed clusters with refcount_bits=1 ==='
 echo
 
-IMGOPTS="$IMGOPTS,refcount_bits=1" _make_test_img 64M
+_make_test_img -o "refcount_bits=1" 64M
 print_refcount_bits
 
 # Both should fit into a single host cluster; instead of failing to increase the
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== MSb set in 64 bit refcount ==='
 echo
 
-IMGOPTS="$IMGOPTS,refcount_bits=64" _make_test_img 64M
+_make_test_img -o "refcount_bits=64" 64M
 print_refcount_bits
 
 $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== Snapshot on maximum 64 bit refcount value ==='
 echo
 
-IMGOPTS="$IMGOPTS,refcount_bits=64" _make_test_img 64M
+_make_test_img -o "refcount_bits=64" 64M
 print_refcount_bits
 
 $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== Testing too many references for check ==='
 echo
 
-IMGOPTS="$IMGOPTS,refcount_bits=1" _make_test_img 64M
+_make_test_img -o "refcount_bits=1" 64M
 print_refcount_bits
 
 # This cluster should be created at 0x50000
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== Multiple walks necessary during amend ==='
 echo
 
-IMGOPTS="$IMGOPTS,refcount_bits=1,cluster_size=512" _make_test_img 64k
+_make_test_img -o "refcount_bits=1,cluster_size=512" 64k
 
 # Cluster 0 is the image header, clusters 1 to 4 are used by the L1 table, a
 # single L2 table, the reftable and a single refblock. This creates 58 data
diff --git a/tests/qemu-iotests/115 b/tests/qemu-iotests/115
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/115
+++ b/tests/qemu-iotests/115
@@ -XXX,XX +XXX,XX @@ echo
 # least 256 MB. We can achieve that by using preallocation=metadata for an image
 # which has a guest disk size of 256 MB.
 
-IMGOPTS="$IMGOPTS,refcount_bits=64,cluster_size=512,preallocation=metadata" \
-    _make_test_img 256M
+_make_test_img -o "refcount_bits=64,cluster_size=512,preallocation=metadata" 256M
 
 # We know for sure that the L1 and refcount tables do not overlap with any other
 # structure because the metadata overlap checks would have caught that case.
diff --git a/tests/qemu-iotests/121 b/tests/qemu-iotests/121
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/121
+++ b/tests/qemu-iotests/121
@@ -XXX,XX +XXX,XX @@ echo
 # Preallocation speeds up the write operation, but preallocating everything will
 # destroy the purpose of the write; so preallocate one KB less than what would
 # cause a reftable growth...
-IMGOPTS='preallocation=metadata,cluster_size=1k' _make_test_img 64512K
+_make_test_img -o 'preallocation=metadata,cluster_size=1k' 64512K
 # ...and make the image the desired size afterwards.
 $QEMU_IMG resize "$TEST_IMG" 65M
 
@@ -XXX,XX +XXX,XX @@ echo
 echo '--- Test 2 ---'
 echo
 
-IMGOPTS='preallocation=metadata,cluster_size=1k' _make_test_img 64513K
+_make_test_img -o 'preallocation=metadata,cluster_size=1k' 64513K
 # This results in an L1 table growth which in turn results in some clusters at
 # the start of the image becoming free
 $QEMU_IMG resize "$TEST_IMG" 65M
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== Allocating a new refcount block must not leave holes in the image ==='
 echo
 
-IMGOPTS='cluster_size=512,refcount_bits=16' _make_test_img 1M
+_make_test_img -o 'cluster_size=512,refcount_bits=16' 1M
 
 # This results in an image with 256 used clusters: the qcow2 header,
 # the refcount table, one refcount block, the L1 table, four L2 tables
diff --git a/tests/qemu-iotests/125 b/tests/qemu-iotests/125
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/125
+++ b/tests/qemu-iotests/125
@@ -XXX,XX +XXX,XX @@ for GROWTH_SIZE in 16 48 80; do
         for growth_mode in off metadata falloc full; do
             echo "--- cluster_size=$cluster_size growth_size=$GROWTH_SIZE create_mode=$create_mode growth_mode=$growth_mode ---"
 
-            IMGOPTS="preallocation=$create_mode,cluster_size=$cluster_size" _make_test_img ${CREATION_SIZE}
+            _make_test_img -o "preallocation=$create_mode,cluster_size=$cluster_size" ${CREATION_SIZE}
             $QEMU_IMG resize -f "$IMGFMT" --preallocation=$growth_mode "$TEST_IMG" +${GROWTH_SIZE}K
 
             host_size_0=$(get_image_size_on_host)
diff --git a/tests/qemu-iotests/137 b/tests/qemu-iotests/137
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/137
+++ b/tests/qemu-iotests/137
@@ -XXX,XX +XXX,XX @@ $QEMU_IO \
     -c "reopen -o cache-clean-interval=-1" \
     "$TEST_IMG" | _filter_qemu_io
 
-IMGOPTS="cluster_size=256k" _make_test_img 32P
+_make_test_img -o "cluster_size=256k" 32P
 $QEMU_IO \
     -c "reopen -o l2-cache-entry-size=512,l2-cache-size=1T" \
     "$TEST_IMG" | _filter_qemu_io
diff --git a/tests/qemu-iotests/138 b/tests/qemu-iotests/138
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/138
+++ b/tests/qemu-iotests/138
@@ -XXX,XX +XXX,XX @@ echo
 echo '=== Check on an image with a multiple of 2^32 clusters ==='
 echo
 
-IMGOPTS=$(_optstr_add "$IMGOPTS" "cluster_size=512") \
-    _make_test_img 512
+_make_test_img -o "cluster_size=512" 512
 
 # Allocate L2 table
 $QEMU_IO -c 'write 0 512' "$TEST_IMG" | _filter_qemu_io
diff --git a/tests/qemu-iotests/175 b/tests/qemu-iotests/175
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/175
+++ b/tests/qemu-iotests/175
@@ -XXX,XX +XXX,XX @@ stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_block
 for mode in off full falloc; do
     echo
     echo "== creating image with preallocation $mode =="
-    IMGOPTS=preallocation=$mode _make_test_img $size | _filter_imgfmt
+    _make_test_img -o preallocation=$mode $size | _filter_imgfmt
     stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $size
 done
 
diff --git a/tests/qemu-iotests/190 b/tests/qemu-iotests/190
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/190
+++ b/tests/qemu-iotests/190
@@ -XXX,XX +XXX,XX @@ _supported_proto file
 echo "== Huge file =="
 echo
 
-IMGOPTS='cluster_size=2M' _make_test_img 2T
+_make_test_img -o 'cluster_size=2M' 2T
 
 $QEMU_IMG measure -O raw -f qcow2 "$TEST_IMG"
 $QEMU_IMG measure -O qcow2 -o cluster_size=64k -f qcow2 "$TEST_IMG"
diff --git a/tests/qemu-iotests/191 b/tests/qemu-iotests/191
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/191
+++ b/tests/qemu-iotests/191
@@ -XXX,XX +XXX,XX @@ echo === Preparing and starting VM ===
 echo
 
 TEST_IMG="${TEST_IMG}.base" _make_test_img $size
-IMGOPTS=$(_optstr_add "$IMGOPTS" "backing_fmt=$IMGFMT") \
-    TEST_IMG="${TEST_IMG}.mid" _make_test_img -b "${TEST_IMG}.base"
+TEST_IMG="${TEST_IMG}.mid" _make_test_img -o "backing_fmt=$IMGFMT" -b "${TEST_IMG}.base"
 _make_test_img -b "${TEST_IMG}.mid"
 TEST_IMG="${TEST_IMG}.ovl2" _make_test_img -b "${TEST_IMG}.mid"
 
diff --git a/tests/qemu-iotests/220 b/tests/qemu-iotests/220
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/220
+++ b/tests/qemu-iotests/220
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# To use a different refcount width but 16 bits we need compat=1.1
+_unsupported_imgopts 'compat=0.10'
 
 echo "== Creating huge file =="
 
@@ -XXX,XX +XXX,XX @@ echo "== Creating huge file =="
 # of a HUGE (but very sparse) file.  tmpfs works, ext4 does not.
 _require_large_file 513T
 
-IMGOPTS='cluster_size=2M,refcount_bits=1' _make_test_img 513T
+_make_test_img -o 'cluster_size=2M,refcount_bits=1' 513T
 
 echo "== Populating refcounts =="
 # We want an image with 256M refcounts * 2M clusters = 512T referenced.
diff --git a/tests/qemu-iotests/243 b/tests/qemu-iotests/243
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/243
+++ b/tests/qemu-iotests/243
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# External data files do not work with compat=0.10
+_unsupported_imgopts 'compat=0.10'
 
 for mode in off metadata falloc full; do
 
@@ -XXX,XX +XXX,XX @@ for mode in off metadata falloc full; do
     echo "=== preallocation=$mode ==="
     echo
 
-    IMGOPTS="preallocation=$mode" _make_test_img 64M
+    _make_test_img -o "preallocation=$mode" 64M
 
     printf "File size: "
     du -b $TEST_IMG | cut -f1
@@ -XXX,XX +XXX,XX @@ for mode in off metadata falloc full; do
     echo "=== External data file: preallocation=$mode ==="
     echo
 
-    IMGOPTS="data_file=$TEST_IMG.data,preallocation=$mode" _make_test_img 64M
+    _make_test_img -o "data_file=$TEST_IMG.data,preallocation=$mode" 64M
 
     echo -n "qcow2 file size: "
     du -b $TEST_IMG | cut -f1
diff --git a/tests/qemu-iotests/244 b/tests/qemu-iotests/244
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/244
+++ b/tests/qemu-iotests/244
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# External data files do not work with compat=0.10
+_unsupported_imgopts 'compat=0.10'
 
 echo
 echo "=== Create and open image with external data file ==="
 echo
 
 echo "With data file name in the image:"
-IMGOPTS="data_file=$TEST_IMG.data" _make_test_img 64M
+_make_test_img -o "data_file=$TEST_IMG.data" 64M
 _check_test_img
 
 $QEMU_IO -c "open $TEST_IMG" -c "read -P 0 0 64k" 2>&1 | _filter_qemu_io | _filter_testdir
@@ -XXX,XX +XXX,XX @@ echo
 echo "=== Standalone image with external data file (efficient) ==="
 echo
 
-IMGOPTS="data_file=$TEST_IMG.data" _make_test_img 64M
+_make_test_img -o "data_file=$TEST_IMG.data" 64M
 
 echo -n "qcow2 file size before I/O: "
 du -b $TEST_IMG | cut -f1
@@ -XXX,XX +XXX,XX @@ echo
 echo "=== Standalone image with external data file (valid raw) ==="
 echo
 
-IMGOPTS="data_file=$TEST_IMG.data,data_file_raw=on" _make_test_img 64M
+_make_test_img -o "data_file=$TEST_IMG.data,data_file_raw=on" 64M
 
 echo -n "qcow2 file size before I/O: "
 du -b $TEST_IMG | cut -f1
@@ -XXX,XX +XXX,XX @@ echo
 echo "=== bdrv_co_block_status test for file and offset=0 ==="
 echo
 
-IMGOPTS="data_file=$TEST_IMG.data" _make_test_img 64M
+_make_test_img -o "data_file=$TEST_IMG.data" 64M
 
 $QEMU_IO -c 'write -P 0x11 0 1M' -f $IMGFMT "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c 'read -P 0x11 0 1M' -f $IMGFMT "$TEST_IMG" | _filter_qemu_io
diff --git a/tests/qemu-iotests/250 b/tests/qemu-iotests/250
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/250
+++ b/tests/qemu-iotests/250
@@ -XXX,XX +XXX,XX @@ disk_usage()
 }
 
 size=2100M
-IMGOPTS="cluster_size=1M,preallocation=metadata"
 
-_make_test_img $size
+_make_test_img -o "cluster_size=1M,preallocation=metadata" $size
 $QEMU_IO -c 'discard 0 10M' -c 'discard 2090M 10M' \
          -c 'write 2090M 10M' -c 'write 0 10M' "$TEST_IMG" | _filter_qemu_io
 
diff --git a/tests/qemu-iotests/265 b/tests/qemu-iotests/265
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/265
+++ b/tests/qemu-iotests/265
@@ -XXX,XX +XXX,XX @@ _supported_os Linux
 echo '--- Writing to the image ---'
 
 # Reduce cluster size so we get more and quicker I/O
-IMGOPTS='cluster_size=4096' _make_test_img 1M
+_make_test_img -o 'cluster_size=4096' 1M
 (for ((kb = 1024 - 4; kb >= 0; kb -= 4)); do \
      echo "aio_write -P 42 $((kb + 1))k 2k"; \
  done) \
-- 
2.24.1

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-13-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/071 | 4 ++--
 tests/qemu-iotests/174 | 2 +-
 tests/qemu-iotests/178 | 4 ++--
 tests/qemu-iotests/197 | 4 ++--
 tests/qemu-iotests/215 | 4 ++--
 5 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/tests/qemu-iotests/071 b/tests/qemu-iotests/071
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/071
+++ b/tests/qemu-iotests/071
@@ -XXX,XX +XXX,XX @@ echo
 echo "=== Testing blkverify through filename ==="
 echo
 
-TEST_IMG="$TEST_IMG.base" IMGOPTS="" IMGFMT="raw" _make_test_img $IMG_SIZE |\
+TEST_IMG="$TEST_IMG.base" IMGFMT="raw" _make_test_img --no-opts $IMG_SIZE |\
     _filter_imgfmt
 _make_test_img $IMG_SIZE
 $QEMU_IO -c "open -o driver=raw,file.driver=blkverify,file.raw.filename=$TEST_IMG.base $TEST_IMG" \
@@ -XXX,XX +XXX,XX @@ echo
 echo "=== Testing blkverify through file blockref ==="
 echo
 
-TEST_IMG="$TEST_IMG.base" IMGOPTS="" IMGFMT="raw" _make_test_img $IMG_SIZE |\
+TEST_IMG="$TEST_IMG.base" IMGFMT="raw" _make_test_img --no-opts $IMG_SIZE |\
     _filter_imgfmt
 _make_test_img $IMG_SIZE
 $QEMU_IO -c "open -o driver=raw,file.driver=blkverify,file.raw.filename=$TEST_IMG.base,file.test.driver=$IMGFMT,file.test.file.filename=$TEST_IMG" \
diff --git a/tests/qemu-iotests/174 b/tests/qemu-iotests/174
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/174
+++ b/tests/qemu-iotests/174
@@ -XXX,XX +XXX,XX @@ _unsupported_fmt raw
 
 
 size=256K
-IMGFMT=raw IMGKEYSECRET= IMGOPTS= _make_test_img $size | _filter_imgfmt
+IMGFMT=raw IMGKEYSECRET= _make_test_img --no-opts $size | _filter_imgfmt
 
 echo
 echo "== reading wrong format should fail =="
diff --git a/tests/qemu-iotests/178 b/tests/qemu-iotests/178
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/178
+++ b/tests/qemu-iotests/178
@@ -XXX,XX +XXX,XX @@ $QEMU_IMG measure -O foo "$TEST_IMG" # unknown image file format
 
 make_test_img_with_fmt() {
     # Shadow global variables within this function
-    local IMGFMT="$1" IMGOPTS=""
-    _make_test_img "$2"
+    local IMGFMT="$1"
+    _make_test_img --no-opts "$2"
 }
 
 qemu_io_with_fmt() {
diff --git a/tests/qemu-iotests/197 b/tests/qemu-iotests/197
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/197
+++ b/tests/qemu-iotests/197
@@ -XXX,XX +XXX,XX @@ if [ "$IMGFMT" = "vpc" ]; then
 fi
 _make_test_img 4G
 $QEMU_IO -c "write -P 55 3G 1k" "$TEST_IMG" | _filter_qemu_io
-IMGPROTO=file IMGFMT=qcow2 IMGOPTS= TEST_IMG_FILE="$TEST_WRAP" \
-    _make_test_img -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create
+IMGPROTO=file IMGFMT=qcow2 TEST_IMG_FILE="$TEST_WRAP" \
+    _make_test_img --no-opts -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create
 $QEMU_IO -f qcow2 -c "write -z -u 1M 64k" "$TEST_WRAP" | _filter_qemu_io
 
 # Ensure that a read of two clusters, but where one is already allocated,
diff --git a/tests/qemu-iotests/215 b/tests/qemu-iotests/215
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/215
+++ b/tests/qemu-iotests/215
@@ -XXX,XX +XXX,XX @@ if [ "$IMGFMT" = "vpc" ]; then
 fi
 _make_test_img 4G
 $QEMU_IO -c "write -P 55 3G 1k" "$TEST_IMG" | _filter_qemu_io
-IMGPROTO=file IMGFMT=qcow2 IMGOPTS= TEST_IMG_FILE="$TEST_WRAP" \
-    _make_test_img -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create
+IMGPROTO=file IMGFMT=qcow2 TEST_IMG_FILE="$TEST_WRAP" \
+    _make_test_img --no-opts -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create
 $QEMU_IO -f qcow2 -c "write -z -u 1M 64k" "$TEST_WRAP" | _filter_qemu_io
 
 # Ensure that a read of two clusters, but where one is already allocated,
-- 
2.24.1

Overwriting IMGOPTS means ignoring all user-supplied options, which is
not what we want.  Replace the current IMGOPTS use by a new BACKING_FILE
variable.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-14-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/267 | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/tests/qemu-iotests/267 b/tests/qemu-iotests/267
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/267
+++ b/tests/qemu-iotests/267
@@ -XXX,XX +XXX,XX @@ size=128M
 
 run_test()
 {
-    _make_test_img $size
+    if [ -n "$BACKING_FILE" ]; then
+        _make_test_img -b "$BACKING_FILE" $size
+    else
+        _make_test_img $size
+    fi
     printf "savevm snap0\ninfo snapshots\nloadvm snap0\n" | run_qemu "$@" | _filter_date
 }
 
@@ -XXX,XX +XXX,XX @@ echo
 
 TEST_IMG="$TEST_IMG.base" _make_test_img $size
 
-IMGOPTS="backing_file=$TEST_IMG.base" \
+BACKING_FILE="$TEST_IMG.base" \
 run_test -blockdev driver=file,filename="$TEST_IMG.base",node-name=backing-file \
          -blockdev driver=file,filename="$TEST_IMG",node-name=file \
          -blockdev driver=$IMGFMT,file=file,backing=backing-file,node-name=fmt
 
-IMGOPTS="backing_file=$TEST_IMG.base" \
+BACKING_FILE="$TEST_IMG.base" \
 run_test -blockdev driver=file,filename="$TEST_IMG.base",node-name=backing-file \
          -blockdev driver=$IMGFMT,file=backing-file,node-name=backing-fmt \
          -blockdev driver=file,filename="$TEST_IMG",node-name=file \
@@ -XXX,XX +XXX,XX @@ echo
 echo "=== -blockdev with NBD server on the backing file ==="
 echo
 
-IMGOPTS="backing_file=$TEST_IMG.base" _make_test_img $size
+_make_test_img -b "$TEST_IMG.base" $size
 cat <<EOF |
 nbd_server_start unix:$SOCK_DIR/nbd
 nbd_server_add -w backing-fmt
-- 
2.24.1

Use _make_test_img whenever possible.  This way, we will not ignore
user-specified image options.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-15-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/094 | 2 +-
 tests/qemu-iotests/111 | 3 +--
 tests/qemu-iotests/123 | 2 +-
 tests/qemu-iotests/153 | 2 +-
 tests/qemu-iotests/200 | 4 ++--
 5 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/tests/qemu-iotests/094 b/tests/qemu-iotests/094
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/094
+++ b/tests/qemu-iotests/094
@@ -XXX,XX +XXX,XX @@ _supported_proto nbd
 _unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat"
 
 _make_test_img 64M
-$QEMU_IMG create -f $IMGFMT "$TEST_DIR/source.$IMGFMT" 64M | _filter_img_create
+TEST_IMG_FILE="$TEST_DIR/source.$IMGFMT" IMGPROTO=file _make_test_img 64M
 
 _launch_qemu -drive if=none,id=src,file="$TEST_DIR/source.$IMGFMT",format=raw \
              -nodefaults
diff --git a/tests/qemu-iotests/111 b/tests/qemu-iotests/111
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/111
+++ b/tests/qemu-iotests/111
@@ -XXX,XX +XXX,XX @@ _supported_fmt qed qcow qcow2 vmdk
 _supported_proto file
 _unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat"
 
-$QEMU_IMG create -f $IMGFMT -b "$TEST_IMG.inexistent" "$TEST_IMG" 2>&1 \
-    | _filter_testdir | _filter_imgfmt
+_make_test_img -b "$TEST_IMG.inexistent"
 
 # success, all done
 echo '*** done'
diff --git a/tests/qemu-iotests/123 b/tests/qemu-iotests/123
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/123
+++ b/tests/qemu-iotests/123
@@ -XXX,XX +XXX,XX @@ _supported_os Linux
 SRC_IMG="$TEST_DIR/source.$IMGFMT"
 
 _make_test_img 1M
-$QEMU_IMG create -f $IMGFMT "$SRC_IMG" 1M | _filter_img_create
+TEST_IMG_FILE=$SRC_IMG IMGPROTO=file _make_test_img 1M
 
 $QEMU_IO -c 'write -P 42 0 1M' "$SRC_IMG" | _filter_qemu_io
 
diff --git a/tests/qemu-iotests/153 b/tests/qemu-iotests/153
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/153
+++ b/tests/qemu-iotests/153
@@ -XXX,XX +XXX,XX @@ for opts1 in "" "read-only=on" "read-only=on,force-share=on"; do
 
     echo
     echo "== Creating test image =="
-    $QEMU_IMG create -f $IMGFMT "${TEST_IMG}" -b ${TEST_IMG}.base | _filter_img_create
+    _make_test_img -b "${TEST_IMG}.base"
 
     echo
     echo "== Launching QEMU, opts: '$opts1' =="
diff --git a/tests/qemu-iotests/200 b/tests/qemu-iotests/200
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/200
+++ b/tests/qemu-iotests/200
@@ -XXX,XX +XXX,XX @@ _supported_proto file
 BACKING_IMG="${TEST_DIR}/backing.img"
 TEST_IMG="${TEST_DIR}/test.img"
 
-${QEMU_IMG} create -f $IMGFMT "${BACKING_IMG}" 512M | _filter_img_create
-${QEMU_IMG} create -f $IMGFMT -F $IMGFMT "${TEST_IMG}" -b "${BACKING_IMG}" 512M | _filter_img_create
+TEST_IMG="$BACKING_IMG" _make_test_img 512M
+_make_test_img -F $IMGFMT -b "$BACKING_IMG" 512M
 
 ${QEMU_IO} -c "write -P 0xa5 512 300M" "${BACKING_IMG}" | _filter_qemu_io
 
-- 
2.24.1

Just rm will not delete external data files.  Use _rm_test_img every
time we delete a test image.

(In the process, clean up the indentation of every _cleanup() this patch
touches.)

((Also, use quotes consistently.  I am happy to see unquoted instances
like "rm -rf $TEST_DIR/..." go.))

diff --git a/tests/qemu-iotests/019 b/tests/qemu-iotests/019
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/019
+++ b/tests/qemu-iotests/019
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-	_cleanup_test_img
-    rm -f "$TEST_IMG.base"
-    rm -f "$TEST_IMG.orig"
+    _cleanup_test_img
+    _rm_test_img "$TEST_IMG.base"
+    _rm_test_img "$TEST_IMG.orig"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/020 b/tests/qemu-iotests/020
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/020
+++ b/tests/qemu-iotests/020
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-	_cleanup_test_img
-    rm -f "$TEST_IMG.base"
-    rm -f "$TEST_IMG.orig"
+    _cleanup_test_img
+    _rm_test_img "$TEST_IMG.base"
+    _rm_test_img "$TEST_IMG.orig"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/024 b/tests/qemu-iotests/024
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/024
+++ b/tests/qemu-iotests/024
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_DIR/t.$IMGFMT.base_old"
-    rm -f "$TEST_DIR/t.$IMGFMT.base_new"
+    _rm_test_img "$TEST_DIR/t.$IMGFMT.base_old"
+    _rm_test_img "$TEST_DIR/t.$IMGFMT.base_new"
 
-    rm -f "$TEST_DIR/subdir/t.$IMGFMT"
-    rm -f "$TEST_DIR/subdir/t.$IMGFMT.base_old"
-    rm -f "$TEST_DIR/subdir/t.$IMGFMT.base_new"
+    _rm_test_img "$TEST_DIR/subdir/t.$IMGFMT"
+    _rm_test_img "$TEST_DIR/subdir/t.$IMGFMT.base_old"
+    _rm_test_img "$TEST_DIR/subdir/t.$IMGFMT.base_new"
     rmdir "$TEST_DIR/subdir" 2> /dev/null
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/028 b/tests/qemu-iotests/028
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/028
+++ b/tests/qemu-iotests/028
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_qemu
-    rm -f "${TEST_IMG}.copy"
+    _rm_test_img "${TEST_IMG}.copy"
     _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/029 b/tests/qemu-iotests/029
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/029
+++ b/tests/qemu-iotests/029
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-    rm -f $TEST_IMG.snap
+    _rm_test_img "$TEST_IMG.snap"
     _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/043 b/tests/qemu-iotests/043
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/043
+++ b/tests/qemu-iotests/043
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG".[123].base
+    for img in "$TEST_IMG".[123].base; do
+        _rm_test_img "$img"
+    done
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/048 b/tests/qemu-iotests/048
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/048
+++ b/tests/qemu-iotests/048
@@ -XXX,XX +XXX,XX @@ _cleanup()
 {
     echo "Cleanup"
     _cleanup_test_img
-    rm "${TEST_IMG_FILE2}"
+    _rm_test_img "${TEST_IMG_FILE2}"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/050 b/tests/qemu-iotests/050
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/050
+++ b/tests/qemu-iotests/050
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.old"
-    rm -f "$TEST_IMG.new"
+    _rm_test_img "$TEST_IMG.old"
+    _rm_test_img "$TEST_IMG.new"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/053 b/tests/qemu-iotests/053
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/053
+++ b/tests/qemu-iotests/053
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-	rm -f "$TEST_IMG.orig"
-	_cleanup_test_img
+    _rm_test_img "$TEST_IMG.orig"
+    _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/058 b/tests/qemu-iotests/058
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/058
+++ b/tests/qemu-iotests/058
@@ -XXX,XX +XXX,XX @@ _cleanup()
 {
     nbd_server_stop
     _cleanup_test_img
-    rm -f "$converted_image"
+    _rm_test_img "$converted_image"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/059 b/tests/qemu-iotests/059
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/059
+++ b/tests/qemu-iotests/059
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.qcow2"
+    IMGFMT=qcow2 _rm_test_img "$TEST_IMG.qcow2"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/061
+++ b/tests/qemu-iotests/061
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f $TEST_IMG.data
+    _rm_test_img "$TEST_IMG.data"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/063 b/tests/qemu-iotests/063
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/063
+++ b/tests/qemu-iotests/063
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-	_cleanup_test_img
-	rm -f "$TEST_IMG.orig" "$TEST_IMG.raw1" "$TEST_IMG.raw2"
+    _cleanup_test_img
+    for img in "$TEST_IMG".{orig,raw1,raw2,target}; do
+        _rm_test_img "$img"
+    done
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/069 b/tests/qemu-iotests/069
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/069
+++ b/tests/qemu-iotests/069
@@ -XXX,XX +XXX,XX @@ echo "=== Creating an image with a backing file and deleting that file ==="
 echo
 TEST_IMG="$TEST_IMG.base" _make_test_img $IMG_SIZE
 _make_test_img -b "$TEST_IMG.base" $IMG_SIZE
-rm -f "$TEST_IMG.base"
+_rm_test_img "$TEST_IMG.base"
 # Just open the image and close it right again (this should print an error message)
 $QEMU_IO -c quit "$TEST_IMG" 2>&1 | _filter_testdir | _filter_imgfmt
 
diff --git a/tests/qemu-iotests/074 b/tests/qemu-iotests/074
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/074
+++ b/tests/qemu-iotests/074
@@ -XXX,XX +XXX,XX @@ _cleanup()
 {
     echo "Cleanup"
     _cleanup_test_img
-    rm "${TEST_IMG2}"
+    _rm_test_img "${TEST_IMG2}"
     rm -f "$TEST_DIR/blkdebug.conf"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/080 b/tests/qemu-iotests/080
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/080
+++ b/tests/qemu-iotests/080
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-    rm -f $TEST_IMG.snap
+    _rm_test_img "$TEST_IMG.snap"
     _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/081 b/tests/qemu-iotests/081
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/081
+++ b/tests/qemu-iotests/081
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-    rm -rf $TEST_DIR/1.raw
-    rm -rf $TEST_DIR/2.raw
-    rm -rf $TEST_DIR/3.raw
+    _rm_test_img "$TEST_DIR/1.raw"
+    _rm_test_img "$TEST_DIR/2.raw"
+    _rm_test_img "$TEST_DIR/3.raw"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/085 b/tests/qemu-iotests/085
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/085
+++ b/tests/qemu-iotests/085
@@ -XXX,XX +XXX,XX @@ _cleanup()
     _cleanup_qemu
     for i in $(seq 1 ${SNAPSHOTS})
     do
-        rm -f "${TEST_DIR}/${i}-${snapshot_virt0}"
-        rm -f "${TEST_DIR}/${i}-${snapshot_virt1}"
+        _rm_test_img "${TEST_DIR}/${i}-${snapshot_virt0}"
+        _rm_test_img "${TEST_DIR}/${i}-${snapshot_virt1}"
+    done
+    for img in "${TEST_IMG}".{1,2,base}
+    do
+        _rm_test_img "$img"
     done
-    rm -f "${TEST_IMG}" "${TEST_IMG}.1" "${TEST_IMG}.2" "${TEST_IMG}.base"
 
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/088 b/tests/qemu-iotests/088
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/088
+++ b/tests/qemu-iotests/088
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-    rm -f $TEST_IMG.snap
+    _rm_test_img "$TEST_IMG.snap"
     _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/092 b/tests/qemu-iotests/092
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/092
+++ b/tests/qemu-iotests/092
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-    rm -f $TEST_IMG.snap
+    _rm_test_img "$TEST_IMG.snap"
     _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/094 b/tests/qemu-iotests/094
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/094
+++ b/tests/qemu-iotests/094
@@ -XXX,XX +XXX,XX @@ _cleanup()
 {
     _cleanup_qemu
     _cleanup_test_img
-    rm -f "$TEST_DIR/source.$IMGFMT"
+    _rm_test_img "$TEST_DIR/source.$IMGFMT"
 }
 
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/095 b/tests/qemu-iotests/095
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/095
+++ b/tests/qemu-iotests/095
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_qemu
-    rm  -f "${TEST_IMG}.base" "${TEST_IMG}.snp1"
-	_cleanup_test_img
+    _rm_test_img "${TEST_IMG}.base"
+    _rm_test_img "${TEST_IMG}.snp1"
+    _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/099 b/tests/qemu-iotests/099
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/099
+++ b/tests/qemu-iotests/099
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-	_cleanup_test_img
+    _cleanup_test_img
+    _rm_test_img "$TEST_IMG.compare"
+    rm -f "$TEST_DIR/blkdebug.conf"
+
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
@@ -XXX,XX +XXX,XX @@ echo
 test_qemu "file.driver=blkdebug,file.image.filename=$TEST_IMG"
 
 
-rm -f "$TEST_IMG.compare" "$TEST_DIR/blkdebug.conf"
-
 # success, all done
 echo "*** done"
 rm -f $seq.full
diff --git a/tests/qemu-iotests/109 b/tests/qemu-iotests/109
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/109
+++ b/tests/qemu-iotests/109
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_qemu
-    rm -f $TEST_IMG.src
-	_cleanup_test_img
+    _rm_test_img "$TEST_IMG.src"
+    _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/110 b/tests/qemu-iotests/110
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/110
+++ b/tests/qemu-iotests/110
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-	_cleanup_test_img
-        rm -f "$TEST_IMG.copy"
+    _cleanup_test_img
+    _rm_test_img "$TEST_IMG.copy"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/122 b/tests/qemu-iotests/122
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/122
+++ b/tests/qemu-iotests/122
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-    rm -f "$TEST_IMG".[123]
-	_cleanup_test_img
+    for img in "$TEST_IMG".[123]; do
+        _rm_test_img "$img"
+    done
+    _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/123 b/tests/qemu-iotests/123
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/123
+++ b/tests/qemu-iotests/123
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$SRC_IMG"
+    _rm_test_img "$SRC_IMG"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/141 b/tests/qemu-iotests/141
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/141
+++ b/tests/qemu-iotests/141
@@ -XXX,XX +XXX,XX @@ _cleanup()
 {
     _cleanup_qemu
     _cleanup_test_img
-    rm -f "$TEST_DIR"/{b,m,o}.$IMGFMT
+    for img in "$TEST_DIR"/{b,m,o}.$IMGFMT; do
+        _rm_test_img "$img"
+    done
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/142 b/tests/qemu-iotests/142
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/142
+++ b/tests/qemu-iotests/142
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f $TEST_IMG.snap
+    _rm_test_img "$TEST_IMG.snap"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/144 b/tests/qemu-iotests/144
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/144
+++ b/tests/qemu-iotests/144
@@ -XXX,XX +XXX,XX @@ TMP_SNAP2=${TEST_DIR}/tmp2.qcow2
 _cleanup()
 {
     _cleanup_qemu
-    rm -f "${TEST_IMG}" "${TMP_SNAP1}" "${TMP_SNAP2}"
+    for img in "${TEST_IMG}" "${TMP_SNAP1}" "${TMP_SNAP2}"; do
+        _rm_test_img "$img"
+    done
 }
 
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/153 b/tests/qemu-iotests/153
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/153
+++ b/tests/qemu-iotests/153
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "${TEST_IMG}.base"
-    rm -f "${TEST_IMG}.overlay"
-    rm -f "${TEST_IMG}.convert"
-    rm -f "${TEST_IMG}.a"
-    rm -f "${TEST_IMG}.b"
-    rm -f "${TEST_IMG}.c"
-    rm -f "${TEST_IMG}.lnk"
+    for img in "${TEST_IMG}".{base,overlay,convert,a,b,c,lnk}; do
+        _rm_test_img "$img"
+    done
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/156 b/tests/qemu-iotests/156
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/156
+++ b/tests/qemu-iotests/156
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_qemu
-    rm -f "$TEST_IMG"{,.target}{,.backing,.overlay}
+    for img in "$TEST_IMG"{,.target}{,.backing,.overlay}; do
+        _rm_test_img "$img"
+    done
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
@@ -XXX,XX +XXX,XX @@ _send_qemu_cmd $QEMU_HANDLE \
     '"status": "null"'
 
 # Remove the source images
-rm -f "$TEST_IMG{,.backing,.overlay}"
+for img in "$TEST_IMG{,.backing,.overlay}"; do
+    _rm_test_img "$img"
+done
 
 echo
 
diff --git a/tests/qemu-iotests/159 b/tests/qemu-iotests/159
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/159
+++ b/tests/qemu-iotests/159
@@ -XXX,XX +XXX,XX @@ status=1
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.out"
+    _rm_test_img "$TEST_IMG.out"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/160 b/tests/qemu-iotests/160
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/160
+++ b/tests/qemu-iotests/160
@@ -XXX,XX +XXX,XX @@ status=1
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.out" "$TEST_IMG.out.dd"
+    _rm_test_img "$TEST_IMG.out"
+    _rm_test_img "$TEST_IMG.out.dd"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/161 b/tests/qemu-iotests/161
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/161
+++ b/tests/qemu-iotests/161
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.base"
-    rm -f "$TEST_IMG.int"
+    _rm_test_img "$TEST_IMG.base"
+    _rm_test_img "$TEST_IMG.int"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/170 b/tests/qemu-iotests/170
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/170
+++ b/tests/qemu-iotests/170
@@ -XXX,XX +XXX,XX @@ status=1
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.out"
+    _rm_test_img  "$TEST_IMG.out"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/172 b/tests/qemu-iotests/172
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/172
+++ b/tests/qemu-iotests/172
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-	_cleanup_test_img
-    rm -f "$TEST_IMG.2"
-    rm -f "$TEST_IMG.3"
+    _cleanup_test_img
+    _rm_test_img "$TEST_IMG.2"
+    _rm_test_img "$TEST_IMG.3"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/173 b/tests/qemu-iotests/173
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/173
+++ b/tests/qemu-iotests/173
@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
 _cleanup()
 {
     _cleanup_qemu
-    rm  -f "${QEMU_TEST_DIR}/image.base" "${QEMU_TEST_DIR}/image.snp1"
+    _rm_test_img "${TEST_DIR}/image.base"
+    _rm_test_img "${TEST_DIR}/image.snp1"
     _cleanup_test_img
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/178 b/tests/qemu-iotests/178
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/178
+++ b/tests/qemu-iotests/178
@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.converted"
+    _rm_test_img "$TEST_IMG.converted"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/182 b/tests/qemu-iotests/182
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/182
+++ b/tests/qemu-iotests/182
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.overlay"
+    _rm_test_img "$TEST_IMG.overlay"
     rm -f "$SOCK_DIR/nbd.socket"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/183 b/tests/qemu-iotests/183
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/183
+++ b/tests/qemu-iotests/183
@@ -XXX,XX +XXX,XX @@ MIG_SOCKET="${SOCK_DIR}/migrate"
 _cleanup()
 {
     rm -f "${MIG_SOCKET}"
-    rm -f "${TEST_IMG}.dest"
+    _rm_test_img "${TEST_IMG}.dest"
     _cleanup_test_img
     _cleanup_qemu
 }
diff --git a/tests/qemu-iotests/185 b/tests/qemu-iotests/185
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/185
+++ b/tests/qemu-iotests/185
@@ -XXX,XX +XXX,XX @@ status=1 # failure is the default!
 
 _cleanup()
 {
-    rm -f "${TEST_IMG}.mid"
-    rm -f "${TEST_IMG}.copy"
+    _rm_test_img "${TEST_IMG}.mid"
+    _rm_test_img "${TEST_IMG}.copy"
     _cleanup_test_img
     _cleanup_qemu
 }
diff --git a/tests/qemu-iotests/187 b/tests/qemu-iotests/187
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/187
+++ b/tests/qemu-iotests/187
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _cleanup()
 {
-	_cleanup_test_img
-    rm -f "$TEST_IMG.2"
-    rm -f "$TEST_IMG.3"
+    _cleanup_test_img
+    _rm_test_img "$TEST_IMG.2"
+    _rm_test_img "$TEST_IMG.3"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/190 b/tests/qemu-iotests/190
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/190
+++ b/tests/qemu-iotests/190
@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.converted"
+    _rm_test_img "$TEST_IMG.converted"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/191 b/tests/qemu-iotests/191
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/191
+++ b/tests/qemu-iotests/191
@@ -XXX,XX +XXX,XX @@ status=1 # failure is the default!
 
 _cleanup()
 {
-    rm -f "${TEST_IMG}.mid"
-    rm -f "${TEST_IMG}.ovl2"
-    rm -f "${TEST_IMG}.ovl3"
+    _rm_test_img "${TEST_IMG}.mid"
+    _rm_test_img "${TEST_IMG}.ovl2"
+    _rm_test_img "${TEST_IMG}.ovl3"
     _cleanup_test_img
     _cleanup_qemu
 }
diff --git a/tests/qemu-iotests/195 b/tests/qemu-iotests/195
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/195
+++ b/tests/qemu-iotests/195
@@ -XXX,XX +XXX,XX @@ status=1 # failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.mid"
+    _rm_test_img "$TEST_IMG.mid"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/197 b/tests/qemu-iotests/197
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/197
+++ b/tests/qemu-iotests/197
@@ -XXX,XX +XXX,XX @@ esac
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_WRAP"
+    _rm_test_img "$TEST_WRAP"
     rm -f "$BLKDBG_CONF"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/200 b/tests/qemu-iotests/200
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/200
+++ b/tests/qemu-iotests/200
@@ -XXX,XX +XXX,XX @@ status=1    # failure is the default!
 _cleanup()
 {
     _cleanup_qemu
-    rm -f "${TEST_IMG}" "${BACKING_IMG}"
+    _rm_test_img "${TEST_IMG}"
+    _rm_test_img "${BACKING_IMG}"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/215 b/tests/qemu-iotests/215
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/215
+++ b/tests/qemu-iotests/215
@@ -XXX,XX +XXX,XX @@ esac
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_WRAP"
+    _rm_test_img "$TEST_WRAP"
     rm -f "$BLKDBG_CONF"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
diff --git a/tests/qemu-iotests/225 b/tests/qemu-iotests/225
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/225
+++ b/tests/qemu-iotests/225
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.not_base"
+    _rm_test_img "$TEST_IMG.not_base"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/229 b/tests/qemu-iotests/229
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/229
+++ b/tests/qemu-iotests/229
@@ -XXX,XX +XXX,XX @@ _cleanup()
 {
     _cleanup_qemu
     _cleanup_test_img
-    rm -f "$TEST_IMG" "$DEST_IMG"
+    _rm_test_img "$TEST_IMG"
+    _rm_test_img "$DEST_IMG"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/232 b/tests/qemu-iotests/232
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/232
+++ b/tests/qemu-iotests/232
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f $TEST_IMG.[01234]
+    for img in "$TEST_IMG".[01234]; do
+        _rm_test_img "$img"
+    done
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/243 b/tests/qemu-iotests/243
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/243
+++ b/tests/qemu-iotests/243
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f $TEST_IMG.data
+    _rm_test_img "$TEST_IMG.data"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/244 b/tests/qemu-iotests/244
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/244
+++ b/tests/qemu-iotests/244
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f $TEST_IMG.data
-    rm -f $TEST_IMG.src
+    _rm_test_img "$TEST_IMG.data"
+    _rm_test_img "$TEST_IMG.src"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/247 b/tests/qemu-iotests/247
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/247
+++ b/tests/qemu-iotests/247
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f $TEST_IMG.[01234]
+    for img in "$TEST_IMG".[01234]; do
+        _rm_test_img "$img"
+    done
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/249 b/tests/qemu-iotests/249
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/249
+++ b/tests/qemu-iotests/249
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.base"
-    rm -f "$TEST_IMG.int"
+    _rm_test_img "$TEST_IMG.base"
+    _rm_test_img "$TEST_IMG.int"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
diff --git a/tests/qemu-iotests/252 b/tests/qemu-iotests/252
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/252
+++ b/tests/qemu-iotests/252
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 _cleanup()
 {
     _cleanup_test_img
-    rm -f "$TEST_IMG.base_new"
+    _rm_test_img "$TEST_IMG.base_new"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
-- 
2.24.1

This will not work with external data files, so try to get tests working
without it as far as possible.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-17-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/063     | 12 ++++--------
 tests/qemu-iotests/063.out |  3 ++-
 tests/qemu-iotests/085     |  9 +++------
 tests/qemu-iotests/085.out |  8 ++++----
 4 files changed, 13 insertions(+), 19 deletions(-)

diff --git a/tests/qemu-iotests/063 b/tests/qemu-iotests/063
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/063
+++ b/tests/qemu-iotests/063
@@ -XXX,XX +XXX,XX @@ _unsupported_imgopts "subformat=monolithicFlat" \
 _make_test_img 4M
 
 echo "== Testing conversion with -n fails with no target file =="
-# check .orig file does not exist
-rm -f "$TEST_IMG.orig"
 if $QEMU_IMG convert -f $IMGFMT -O $IMGFMT -n "$TEST_IMG" "$TEST_IMG.orig" >/dev/null 2>&1; then
     exit 1
 fi
 
 echo "== Testing conversion with -n succeeds with a target file =="
-rm -f "$TEST_IMG.orig"
-cp "$TEST_IMG" "$TEST_IMG.orig"
+_rm_test_img "$TEST_IMG.orig"
+TEST_IMG="$TEST_IMG.orig" _make_test_img 4M
 if ! $QEMU_IMG convert -f $IMGFMT -O $IMGFMT -n "$TEST_IMG" "$TEST_IMG.orig" ; then
     exit 1
 fi
@@ -XXX,XX +XXX,XX @@ fi
 _check_test_img
 
 echo "== Testing conversion to a smaller file fails =="
-rm -f "$TEST_IMG.orig"
-mv "$TEST_IMG" "$TEST_IMG.orig"
-_make_test_img 2M
-if $QEMU_IMG convert -f $IMGFMT -O $IMGFMT -n "$TEST_IMG.orig" "$TEST_IMG" >/dev/null 2>&1; then
+TEST_IMG="$TEST_IMG.target" _make_test_img 2M
+if $QEMU_IMG convert -f $IMGFMT -O $IMGFMT -n "$TEST_IMG" "$TEST_IMG.target" >/dev/null 2>&1; then
     exit 1
 fi
 
diff --git a/tests/qemu-iotests/063.out b/tests/qemu-iotests/063.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/063.out
+++ b/tests/qemu-iotests/063.out
@@ -XXX,XX +XXX,XX @@ QA output created by 063
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=4194304
 == Testing conversion with -n fails with no target file ==
 == Testing conversion with -n succeeds with a target file ==
+Formatting 'TEST_DIR/t.IMGFMT.orig', fmt=IMGFMT size=4194304
 == Testing conversion to raw is the same after conversion with -n ==
 == Testing conversion back to original format ==
 No errors were found on the image.
 == Testing conversion to a smaller file fails ==
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2097152
+Formatting 'TEST_DIR/t.IMGFMT.target', fmt=IMGFMT size=2097152
 == Regression testing for copy offloading bug ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576
 Formatting 'TEST_DIR/t.IMGFMT.target', fmt=IMGFMT size=1048576
diff --git a/tests/qemu-iotests/085 b/tests/qemu-iotests/085
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/085
+++ b/tests/qemu-iotests/085
@@ -XXX,XX +XXX,XX @@ add_snapshot_image()
 {
     base_image="${TEST_DIR}/$((${1}-1))-${snapshot_virt0}"
     snapshot_file="${TEST_DIR}/${1}-${snapshot_virt0}"
-    _make_test_img -u -b "${base_image}" "$size"
-    mv "${TEST_IMG}" "${snapshot_file}"
+    TEST_IMG=$snapshot_file _make_test_img -u -b "${base_image}" "$size"
     do_blockdev_add "$1" "'backing': null, " "${snapshot_file}"
 }
 
@@ -XXX,XX +XXX,XX @@ blockdev_snapshot()
 
 size=128M
 
-_make_test_img $size
-mv "${TEST_IMG}" "${TEST_IMG}.1"
-_make_test_img $size
-mv "${TEST_IMG}" "${TEST_IMG}.2"
+TEST_IMG="$TEST_IMG.1" _make_test_img $size
+TEST_IMG="$TEST_IMG.2" _make_test_img $size
 
 echo
 echo === Running QEMU ===
diff --git a/tests/qemu-iotests/085.out b/tests/qemu-iotests/085.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/085.out
+++ b/tests/qemu-iotests/085.out
@@ -XXX,XX +XXX,XX @@
 QA output created by 085
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+Formatting 'TEST_DIR/t.IMGFMT.1', fmt=IMGFMT size=134217728
+Formatting 'TEST_DIR/t.IMGFMT.2', fmt=IMGFMT size=134217728
 
 === Running QEMU ===
 
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/10-snapshot-v1.qcow2', fmt=qcow2 size=134217728 backing_fil
 
 === Create a couple of snapshots using blockdev-snapshot ===
 
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/10-snapshot-v0.IMGFMT
+Formatting 'TEST_DIR/11-snapshot-v0.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/10-snapshot-v0.IMGFMT
 { 'execute': 'blockdev-add', 'arguments': { 'driver': 'IMGFMT', 'node-name': 'snap_11', 'backing': null, 'file': { 'driver': 'file', 'filename': 'TEST_DIR/11-snapshot-v0.IMGFMT', 'node-name': 'file_11' } } }
 {"return": {}}
 { 'execute': 'blockdev-snapshot', 'arguments': { 'node': 'virtio0', 'overlay':'snap_11' } }
 {"return": {}}
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/11-snapshot-v0.IMGFMT
+Formatting 'TEST_DIR/12-snapshot-v0.IMGFMT', fmt=IMGFMT size=134217728 backing_file=TEST_DIR/11-snapshot-v0.IMGFMT
 { 'execute': 'blockdev-add', 'arguments': { 'driver': 'IMGFMT', 'node-name': 'snap_12', 'backing': null, 'file': { 'driver': 'file', 'filename': 'TEST_DIR/12-snapshot-v0.IMGFMT', 'node-name': 'file_12' } } }
 {"return": {}}
 { 'execute': 'blockdev-snapshot', 'arguments': { 'node': 'virtio0', 'overlay':'snap_12' } }
-- 
2.24.1

The image end offset as reported by qemu-img check is different when
using an external data file; we do not care about its value here, so we
can just filter it.  Incidentally, common.rc already has _check_test_img
for us which does exactly that.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-18-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/091     | 2 +-
 tests/qemu-iotests/091.out | 2 --
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/tests/qemu-iotests/091 b/tests/qemu-iotests/091
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/091
+++ b/tests/qemu-iotests/091
@@ -XXX,XX +XXX,XX @@ echo "Check image pattern"
 ${QEMU_IO} -c "read -P 0x22 0 4M" "${TEST_IMG}" | _filter_testdir | _filter_qemu_io
 
 echo "Running 'qemu-img check -r all \$TEST_IMG'"
-"${QEMU_IMG}" check -r all "${TEST_IMG}" 2>&1 | _filter_testdir | _filter_qemu
+_check_test_img -r all
 
 echo "*** done"
 rm -f $seq.full
diff --git a/tests/qemu-iotests/091.out b/tests/qemu-iotests/091.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/091.out
+++ b/tests/qemu-iotests/091.out
@@ -XXX,XX +XXX,XX @@ read 4194304/4194304 bytes at offset 0
 4 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 Running 'qemu-img check -r all $TEST_IMG'
 No errors were found on the image.
-80/16384 = 0.49% allocated, 0.00% fragmented, 0.00% compressed clusters
-Image end offset: 5570560
 *** done
-- 
2.24.1

The only difference is that the json:{} filename of the image looks
different.  We actually do not care about that filename in this test, we
are only interested in (1) that there is a json:{} filename, and (2)
whether the backing filename can be constructed.

So just filter out the json:{} data, thus making this test pass both
with and without data_file.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-19-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/110     | 7 +++++--
 tests/qemu-iotests/110.out | 4 ++--
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/tests/qemu-iotests/110 b/tests/qemu-iotests/110
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/110
+++ b/tests/qemu-iotests/110
@@ -XXX,XX +XXX,XX @@ echo
 # Across blkdebug without a config file, you cannot reconstruct filenames, so
 # qemu is incapable of knowing the directory of the top image from the filename
 # alone. However, using bdrv_dirname(), it should still work.
+# (Filter out the json:{} filename so this test works with external data files)
 TEST_IMG="json:{
     'driver': '$IMGFMT',
     'file': {
@@ -XXX,XX +XXX,XX @@ TEST_IMG="json:{
             }
         ]
     }
-}" _img_info | _filter_img_info | grep -v 'backing file format'
+}" _img_info | _filter_img_info | grep -v 'backing file format' \
+    | _filter_json_filename
 
 echo
 echo '=== Backing name is always relative to the backed image ==='
@@ -XXX,XX +XXX,XX @@ TEST_IMG="json:{
             }
         ]
     }
-}" _img_info | _filter_img_info | grep -v 'backing file format'
+}" _img_info | _filter_img_info | grep -v 'backing file format' \
+    | _filter_json_filename
 
 
 # success, all done
diff --git a/tests/qemu-iotests/110.out b/tests/qemu-iotests/110.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/110.out
+++ b/tests/qemu-iotests/110.out
@@ -XXX,XX +XXX,XX @@ backing file: t.IMGFMT.base (actual path: TEST_DIR/t.IMGFMT.base)
 
 === Non-reconstructable filename ===
 
-image: json:{"driver": "IMGFMT", "file": {"set-state.0.event": "read_aio", "image": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}, "driver": "blkdebug", "set-state.0.new_state": 42}}
+image: json:{ /* filtered */ }
 file format: IMGFMT
 virtual size: 64 MiB (67108864 bytes)
 backing file: t.IMGFMT.base (actual path: TEST_DIR/t.IMGFMT.base)
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 backing_file=t.IMGFMT.b
 
 === Nodes without a common directory ===
 
-image: json:{"driver": "IMGFMT", "file": {"children": [{"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}, {"driver": "file", "filename": "TEST_DIR/t.IMGFMT.copy"}], "driver": "quorum", "vote-threshold": 1}}
+image: json:{ /* filtered */ }
 file format: IMGFMT
 virtual size: 64 MiB (67108864 bytes)
 backing file: t.IMGFMT.base (cannot determine actual path)
-- 
2.24.1

When using an external data file, there are no refcounts for data
clusters.  We thus have to adjust the corruption test in this patch to
not be based around a data cluster allocation, but the L2 table
allocation (L2 tables are still refcounted with external data files).

Furthermore, we should not print qcow2.py's list of incompatible
features because it differs depending on whether there is an external
data file or not.

With those two changes, the test will work both with and without
external data files (once that options works with the iotests at all).

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-20-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/137     | 15 +++++++++++----
 tests/qemu-iotests/137.out |  6 ++----
 2 files changed, 13 insertions(+), 8 deletions(-)

diff --git a/tests/qemu-iotests/137 b/tests/qemu-iotests/137
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/137
+++ b/tests/qemu-iotests/137
@@ -XXX,XX +XXX,XX @@ $QEMU_IO \
     "$TEST_IMG" 2>&1 | _filter_qemu_io
 
 # The dirty bit must not be set
-$PYTHON qcow2.py "$TEST_IMG" dump-header | grep incompatible_features
+# (Filter the external data file bit)
+if $PYTHON qcow2.py "$TEST_IMG" dump-header | grep incompatible_features \
+    | grep -q '\<0\>'
+then
+    echo 'ERROR: Dirty bit set'
+else
+    echo 'OK: Dirty bit not set'
+fi
 
 # Similarly we can test whether corruption detection has been enabled:
-# Create L1/L2, overwrite first entry in refcount block, allocate something.
+# Create L1, overwrite refcounts, force allocation of L2 by writing
+# data.
 # Disabling the checks should fail, so the corruption must be detected.
 _make_test_img 64M
-$QEMU_IO -c "write 0 64k" "$TEST_IMG" | _filter_qemu_io
-poke_file "$TEST_IMG" "$((0x20000))" "\x00\x00"
+poke_file "$TEST_IMG" "$((0x20000))" "\x00\x00\x00\x00\x00\x00\x00\x00"
 $QEMU_IO \
     -c "reopen -o overlap-check=none,lazy-refcounts=42" \
     -c "write 64k 64k" \
diff --git a/tests/qemu-iotests/137.out b/tests/qemu-iotests/137.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/137.out
+++ b/tests/qemu-iotests/137.out
@@ -XXX,XX +XXX,XX @@ qemu-io: Unsupported value 'blubb' for qcow2 option 'overlap-check'. Allowed are
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.rc: Killed                  ( VALGRIND_QEMU="${VALGRIND_QEMU_IO}" _qemu_proc_exec "${VALGRIND_LOGFILE}" "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" )
-incompatible_features     []
+OK: Dirty bit not set
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
-wrote 65536/65536 bytes at offset 0
-64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 qemu-io: Parameter 'lazy-refcounts' expects 'on' or 'off'
-qcow2: Marking image as corrupt: Preventing invalid write on metadata (overlaps with qcow2_header); further corruption events will be suppressed
+qcow2: Marking image as corrupt: Preventing invalid allocation of L2 table at offset 0; further corruption events will be suppressed
 write failed: Input/output error
 *** done
-- 
2.24.1

We do not care about the json:{} filenames here, so we can just filter
them out and thus make the test work both with and without external data
files.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-21-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/198     | 6 ++++--
 tests/qemu-iotests/198.out | 4 ++--
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/tests/qemu-iotests/198 b/tests/qemu-iotests/198
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/198
+++ b/tests/qemu-iotests/198
@@ -XXX,XX +XXX,XX @@ echo
 echo "== checking image base =="
 $QEMU_IMG info --image-opts $IMGSPECBASE | _filter_img_info --format-specific \
     | sed -e "/^disk size:/ D" -e '/refcount bits:/ D' -e '/compat:/ D' \
-          -e '/lazy refcounts:/ D' -e '/corrupt:/ D'
+          -e '/lazy refcounts:/ D' -e '/corrupt:/ D' -e '/^\s*data file/ D' \
+    | _filter_json_filename
 
 echo
 echo "== checking image layer =="
 $QEMU_IMG info --image-opts $IMGSPECLAYER | _filter_img_info --format-specific \
     | sed -e "/^disk size:/ D" -e '/refcount bits:/ D' -e '/compat:/ D' \
-          -e '/lazy refcounts:/ D' -e '/corrupt:/ D'
+          -e '/lazy refcounts:/ D' -e '/corrupt:/ D' -e '/^\s*data file/ D' \
+    | _filter_json_filename
 
 
 # success, all done
diff --git a/tests/qemu-iotests/198.out b/tests/qemu-iotests/198.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/198.out
+++ b/tests/qemu-iotests/198.out
@@ -XXX,XX +XXX,XX @@ read 16777216/16777216 bytes at offset 0
 16 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
 == checking image base ==
-image: json:{"encrypt.key-secret": "sec0", "driver": "IMGFMT", "file": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT.base"}}
+image: json:{ /* filtered */ }
 file format: IMGFMT
 virtual size: 16 MiB (16777216 bytes)
 Format specific information:
@@ -XXX,XX +XXX,XX @@ Format specific information:
         master key iters: 1024
 
 == checking image layer ==
-image: json:{"encrypt.key-secret": "sec1", "driver": "IMGFMT", "file": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}}
+image: json:{ /* filtered */ }
 file format: IMGFMT
 virtual size: 16 MiB (16777216 bytes)
 backing file: TEST_DIR/t.IMGFMT.base
-- 
2.24.1

diff --git a/tests/qemu-iotests/007 b/tests/qemu-iotests/007
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/007
+++ b/tests/qemu-iotests/007
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto generic
 # refcount_bits must be at least 4 so we can create ten internal snapshots
-# (1 bit supports none, 2 bits support two, 4 bits support 14)
-_unsupported_imgopts 'refcount_bits=$1\|2$[^0-9]'
+# (1 bit supports none, 2 bits support two, 4 bits support 14);
+# snapshot are generally impossible with external data files
+_unsupported_imgopts 'refcount_bits=$1\|2$[^0-9]' data_file
 
 echo
 echo "creating image"
diff --git a/tests/qemu-iotests/014 b/tests/qemu-iotests/014
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/014
+++ b/tests/qemu-iotests/014
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# Compression and snapshots do not work with external data files
+_unsupported_imgopts data_file
 
 TEST_OFFSETS="0 4294967296"
 TEST_OPS="writev read write readv"
diff --git a/tests/qemu-iotests/015 b/tests/qemu-iotests/015
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/015
+++ b/tests/qemu-iotests/015
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # actually any format that supports snapshots
 _supported_fmt qcow2
 _supported_proto generic
-# Internal snapshots are (currently) impossible with refcount_bits=1
-_unsupported_imgopts 'refcount_bits=1[^0-9]'
+# Internal snapshots are (currently) impossible with refcount_bits=1,
+# and generally impossible with external data files
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
 
 echo
 echo "creating image"
diff --git a/tests/qemu-iotests/026 b/tests/qemu-iotests/026
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/026
+++ b/tests/qemu-iotests/026
@@ -XXX,XX +XXX,XX @@ _supported_cache_modes writethrough none
 # 32 and 64 bits do not work either, however, due to different leaked cluster
 # count on error.
 # Thus, the only remaining option is refcount_bits=16.
-_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
+#
+# As for data_file, none of the refcount tests can work for it.
+_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)' \
+    data_file
 
 echo "Errors while writing 128 kB"
 echo
diff --git a/tests/qemu-iotests/029 b/tests/qemu-iotests/029
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/029
+++ b/tests/qemu-iotests/029
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto generic
 _unsupported_proto vxhs
-# Internal snapshots are (currently) impossible with refcount_bits=1
-_unsupported_imgopts 'refcount_bits=1[^0-9]'
+# Internal snapshots are (currently) impossible with refcount_bits=1,
+# and generally impossible with external data files
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
 
 offset_size=24
 offset_l1_size=36
diff --git a/tests/qemu-iotests/031 b/tests/qemu-iotests/031
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/031
+++ b/tests/qemu-iotests/031
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
-# We want to test compat=0.10, which does not support refcount widths
-# other than 16
-_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
+# We want to test compat=0.10, which does not support external data
+# files or refcount widths other than 16
+_unsupported_imgopts data_file 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
 
 CLUSTER_SIZE=65536
 
diff --git a/tests/qemu-iotests/036 b/tests/qemu-iotests/036
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/036
+++ b/tests/qemu-iotests/036
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto file
-# Only qcow2v3 and later supports feature bits
-_unsupported_imgopts 'compat=0.10'
+# Only qcow2v3 and later supports feature bits;
+# qcow2.py does not support external data files
+_unsupported_imgopts 'compat=0.10' data_file
 
 echo
 echo === Image with unknown incompatible feature bit ===
diff --git a/tests/qemu-iotests/039 b/tests/qemu-iotests/039
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/039
+++ b/tests/qemu-iotests/039
@@ -XXX,XX +XXX,XX @@ _supported_proto file
 _supported_os Linux
 _default_cache_mode writethrough
 _supported_cache_modes writethrough
+# Some of these test cases expect no external data file so that all
+# clusters are part of the qcow2 image and refcounted
+_unsupported_imgopts data_file
 
 size=128M
 
diff --git a/tests/qemu-iotests/046 b/tests/qemu-iotests/046
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/046
+++ b/tests/qemu-iotests/046
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt qcow2
 _supported_proto file
+# data_file does not support compressed clusters
+_unsupported_imgopts data_file
 
 CLUSTER_SIZE=64k
 size=128M
diff --git a/tests/qemu-iotests/048 b/tests/qemu-iotests/048
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/048
+++ b/tests/qemu-iotests/048
@@ -XXX,XX +XXX,XX @@ _compare()
 _supported_fmt raw qcow2 qed luks
 _supported_proto file
 _supported_os Linux
+# Using 'cp' is incompatible with external data files
+_unsupported_imgopts data_file
 
 # Remove once all tests are fixed to use TEST_IMG_FILE
 # correctly and common.rc sets it unconditionally
diff --git a/tests/qemu-iotests/051 b/tests/qemu-iotests/051
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/051
+++ b/tests/qemu-iotests/051
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 # A compat=0.10 image is created in this test which does not support anything
-# other than refcount_bits=16
-_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
+# other than refcount_bits=16;
+# it also will not support an external data file
+_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)' data_file
 _require_drivers nbd
 
 do_run_qemu()
diff --git a/tests/qemu-iotests/058 b/tests/qemu-iotests/058
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/058
+++ b/tests/qemu-iotests/058
@@ -XXX,XX +XXX,XX @@ _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
 _require_command QEMU_NBD
-# Internal snapshots are (currently) impossible with refcount_bits=1
-_unsupported_imgopts 'refcount_bits=1[^0-9]'
+# Internal snapshots are (currently) impossible with refcount_bits=1,
+# and generally impossible with external data files
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
 
 nbd_snapshot_img="nbd:unix:$nbd_unix_socket"
 
diff --git a/tests/qemu-iotests/060 b/tests/qemu-iotests/060
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/060
+++ b/tests/qemu-iotests/060
@@ -XXX,XX +XXX,XX @@ _filter_io_error()
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
-# These tests only work for compat=1.1 images with refcount_bits=16
-_unsupported_imgopts 'compat=0.10' 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
+# These tests only work for compat=1.1 images without an external
+# data file with refcount_bits=16
+_unsupported_imgopts 'compat=0.10' data_file \
+    'refcount_bits=$[^1]\|.\([^6]\|$$\)'
 
 # The repair process will create a large file - so check for availability first
 _require_large_file 64G
diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/061
+++ b/tests/qemu-iotests/061
@@ -XXX,XX +XXX,XX @@ _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
 # Conversion between different compat versions can only really work
-# with refcount_bits=16
-_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
+# with refcount_bits=16;
+# we have explicit tests for data_file here, but the whole test does
+# not work with it
+_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)' data_file
 
 echo
 echo "=== Testing version downgrade with zero expansion ==="
diff --git a/tests/qemu-iotests/062 b/tests/qemu-iotests/062
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/062
+++ b/tests/qemu-iotests/062
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto generic
 # We need zero clusters and snapshots
-_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]' data_file
 
 IMG_SIZE=64M
 
diff --git a/tests/qemu-iotests/066 b/tests/qemu-iotests/066
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/066
+++ b/tests/qemu-iotests/066
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto generic
 # We need zero clusters and snapshots
-_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
+# (TODO: Consider splitting the snapshot part into a separate test
+#        file, so this one runs with refcount_bits=1 and data_file)
+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]' data_file
 
 # Intentionally create an unaligned image
 IMG_SIZE=$((64 * 1024 * 1024 + 512))
diff --git a/tests/qemu-iotests/067 b/tests/qemu-iotests/067
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/067
+++ b/tests/qemu-iotests/067
@@ -XXX,XX +XXX,XX @@ status=1	# failure is the default!
 
 _supported_fmt qcow2
 _supported_proto file
-# Because anything other than 16 would change the output of query-block
-_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
+# Because anything other than 16 would change the output of query-block,
+# and external data files would change the output of
+# query-named-block-nodes
+_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)' data_file
 
 do_run_qemu()
 {
diff --git a/tests/qemu-iotests/068 b/tests/qemu-iotests/068
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/068
+++ b/tests/qemu-iotests/068
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 # This tests qcow2-specific low-level functionality
 _supported_fmt qcow2
 _supported_proto generic
-# Internal snapshots are (currently) impossible with refcount_bits=1
-_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]'
+# Internal snapshots are (currently) impossible with refcount_bits=1,
+# and generally impossible with external data files
+_unsupported_imgopts 'compat=0.10' 'refcount_bits=1[^0-9]' data_file
 
 IMG_SIZE=128K
 
diff --git a/tests/qemu-iotests/071 b/tests/qemu-iotests/071
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/071
+++ b/tests/qemu-iotests/071
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _require_drivers blkdebug blkverify
+# blkdebug can only inject errors on bs->file, not on the data_file,
+# so thie test does not work with external data files
+_unsupported_imgopts data_file
 
 do_run_qemu()
 {
diff --git a/tests/qemu-iotests/073 b/tests/qemu-iotests/073
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/073
+++ b/tests/qemu-iotests/073
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto generic
 _unsupported_proto vxhs
+# External data files do not support compressed clusters
+# (TODO: Consider writing a version for external data files that does
+#        not test compressed clusters)
+_unsupported_imgopts data_file
 
 CLUSTER_SIZE=64k
 size=128M
diff --git a/tests/qemu-iotests/074 b/tests/qemu-iotests/074
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/074
+++ b/tests/qemu-iotests/074
@@ -XXX,XX +XXX,XX @@ _compare()
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# blkdebug can only inject errors on bs->file
+_unsupported_imgopts data_file
 
 # Setup test basic parameters
 TEST_IMG2=$TEST_IMG.2
diff --git a/tests/qemu-iotests/080 b/tests/qemu-iotests/080
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/080
+++ b/tests/qemu-iotests/080
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
-# - Internal snapshots are (currently) impossible with refcount_bits=1
+# - Internal snapshots are (currently) impossible with refcount_bits=1,
+#   and generally impossible with external data files
 # - This is generally a test for compat=1.1 images
-_unsupported_imgopts 'refcount_bits=1[^0-9]' 'compat=0.10'
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file 'compat=0.10'
 
 header_size=104
 
diff --git a/tests/qemu-iotests/090 b/tests/qemu-iotests/090
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/090
+++ b/tests/qemu-iotests/090
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt qcow2
 _supported_proto file nfs
+# External data files do not support compressed clusters
+_unsupported_imgopts data_file
 
 IMG_SIZE=128K
 
diff --git a/tests/qemu-iotests/098 b/tests/qemu-iotests/098
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/098
+++ b/tests/qemu-iotests/098
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt qcow2
 _supported_proto file
-# The code path we want to test here only works for compat=1.1 images
-_unsupported_imgopts 'compat=0.10'
+# The code path we want to test here only works for compat=1.1 images;
+# blkdebug can only inject errors on bs->file, so external data files
+# do not work with this test
+_unsupported_imgopts 'compat=0.10' data_file
 
 for event in l1_update empty_image_prepare reftable_update refblock_alloc; do
 
diff --git a/tests/qemu-iotests/099 b/tests/qemu-iotests/099
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/099
+++ b/tests/qemu-iotests/099
@@ -XXX,XX +XXX,XX @@ _supported_fmt qcow qcow2 qed vdi vhdx vmdk vpc
 _supported_proto file
 _supported_os Linux
 _require_drivers blkdebug blkverify
+# data_file would change the json:{} filenames
 _unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" \
-    "subformat=twoGbMaxExtentSparse"
+    "subformat=twoGbMaxExtentSparse" data_file
 
 do_run_qemu()
 {
diff --git a/tests/qemu-iotests/103 b/tests/qemu-iotests/103
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/103
+++ b/tests/qemu-iotests/103
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt qcow2
 _supported_proto file nfs
-# Internal snapshots are (currently) impossible with refcount_bits=1
-_unsupported_imgopts 'refcount_bits=1[^0-9]'
+# Internal snapshots are (currently) impossible with refcount_bits=1,
+# and generally impossible with external data files
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
 
 IMG_SIZE=64K
 
diff --git a/tests/qemu-iotests/108 b/tests/qemu-iotests/108
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/108
+++ b/tests/qemu-iotests/108
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
-# This test directly modifies a refblock so it relies on refcount_bits being 16
-_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
+# This test directly modifies a refblock so it relies on refcount_bits being 16;
+# and the low-level modification it performs are not tuned for external data
+# files
+_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)' data_file
 
 echo
 echo '=== Repairing an image without any refcount table ==='
diff --git a/tests/qemu-iotests/112 b/tests/qemu-iotests/112
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/112
+++ b/tests/qemu-iotests/112
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 # This test will set refcount_bits on its own which would conflict with the
-# manual setting; compat will be overridden as well
-_unsupported_imgopts refcount_bits 'compat=0.10'
+# manual setting; compat will be overridden as well;
+# and external data files do not work well with our refcount testing
+_unsupported_imgopts refcount_bits 'compat=0.10' data_file
 
 print_refcount_bits()
 {
diff --git a/tests/qemu-iotests/114 b/tests/qemu-iotests/114
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/114
+++ b/tests/qemu-iotests/114
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto generic
 _unsupported_proto vxhs
+# qcow2.py does not work too well with external data files
+_unsupported_imgopts data_file
 
 
 TEST_IMG="$TEST_IMG.base" _make_test_img 64M
diff --git a/tests/qemu-iotests/121 b/tests/qemu-iotests/121
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/121
+++ b/tests/qemu-iotests/121
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# Refcount structures are used much differently with external data
+# files
+_unsupported_imgopts data_file
 
 echo
 echo '=== New refcount structures may not conflict with existing structures ==='
diff --git a/tests/qemu-iotests/138 b/tests/qemu-iotests/138
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/138
+++ b/tests/qemu-iotests/138
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# With an external data file, data clusters are not refcounted
+# (and so qemu-img check does not check their refcount)
+_unsupported_imgopts data_file
 
 echo
 echo '=== Check on an image with a multiple of 2^32 clusters ==='
diff --git a/tests/qemu-iotests/156 b/tests/qemu-iotests/156
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/156
+++ b/tests/qemu-iotests/156
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2 qed
 _supported_proto generic
 _unsupported_proto vxhs
+# Copying files around with cp does not work with external data files
+_unsupported_imgopts data_file
 
 # Create source disk
 TEST_IMG="$TEST_IMG.backing" _make_test_img 1M
diff --git a/tests/qemu-iotests/176 b/tests/qemu-iotests/176
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/176
+++ b/tests/qemu-iotests/176
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
-# Persistent dirty bitmaps require compat=1.1
-_unsupported_imgopts 'compat=0.10'
+# Persistent dirty bitmaps require compat=1.1;
+# Internal snapshots forbid using an external data file
+# (they work with refcount_bits=1 here, though, because there actually
+# is no data when creating the snapshot)
+_unsupported_imgopts 'compat=0.10' data_file
 
 run_qemu()
 {
diff --git a/tests/qemu-iotests/191 b/tests/qemu-iotests/191
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/191
+++ b/tests/qemu-iotests/191
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _supported_fmt qcow2
 _supported_proto file
+# An external data file would change the query-named-block-nodes output
+_unsupported_imgopts data_file
 
 size=64M
 
diff --git a/tests/qemu-iotests/201 b/tests/qemu-iotests/201
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/201
+++ b/tests/qemu-iotests/201
@@ -XXX,XX +XXX,XX @@ _supported_fmt qcow2
 _supported_proto generic
 _supported_os Linux
 
-# Internal snapshots are (currently) impossible with refcount_bits=1
-# This was taken from test 080
-_unsupported_imgopts 'refcount_bits=1[^0-9]'
+# Internal snapshots are (currently) impossible with refcount_bits=1,
+# and generally impossible with external data files
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
 
 size=64M
 _make_test_img $size
diff --git a/tests/qemu-iotests/214 b/tests/qemu-iotests/214
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/214
+++ b/tests/qemu-iotests/214
@@ -XXX,XX +XXX,XX @@ _supported_proto file
 
 # Repairing the corrupted image requires qemu-img check to store a
 # refcount up to 3, which requires at least two refcount bits.
-_unsupported_imgopts 'refcount_bits=1[^0-9]'
+# External data files do not support compressed clusters.
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
 
 
 echo
diff --git a/tests/qemu-iotests/217 b/tests/qemu-iotests/217
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/217
+++ b/tests/qemu-iotests/217
@@ -XXX,XX +XXX,XX @@ _supported_proto file
 
 # This test needs clusters with at least a refcount of 2 so that
 # OFLAG_COPIED is not set.  refcount_bits=1 is therefore unsupported.
-_unsupported_imgopts 'refcount_bits=1[^0-9]'
+# (As are external data files.)
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
 
 echo
 echo '=== Simulating an I/O error during snapshot deletion ==='
diff --git a/tests/qemu-iotests/220 b/tests/qemu-iotests/220
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/220
+++ b/tests/qemu-iotests/220
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
-# To use a different refcount width but 16 bits we need compat=1.1
-_unsupported_imgopts 'compat=0.10'
+# To use a different refcount width but 16 bits we need compat=1.1,
+# and external data files do not support compressed clusters.
+_unsupported_imgopts 'compat=0.10' data_file
 
 echo "== Creating huge file =="
 
diff --git a/tests/qemu-iotests/243 b/tests/qemu-iotests/243
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/243
+++ b/tests/qemu-iotests/243
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
-# External data files do not work with compat=0.10
-_unsupported_imgopts 'compat=0.10'
+# External data files do not work with compat=0.10, and because there
+# is an explicit case for external data files here, we cannot allow
+# the user to specify whether to use one
+_unsupported_imgopts 'compat=0.10' data_file
 
 for mode in off metadata falloc full; do
 
diff --git a/tests/qemu-iotests/244 b/tests/qemu-iotests/244
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/244
+++ b/tests/qemu-iotests/244
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
-# External data files do not work with compat=0.10
-_unsupported_imgopts 'compat=0.10'
+# External data files do not work with compat=0.10, and because we use
+# our own external data file, we cannot let the user specify one
+_unsupported_imgopts 'compat=0.10' data_file
 
 echo
 echo "=== Create and open image with external data file ==="
diff --git a/tests/qemu-iotests/250 b/tests/qemu-iotests/250
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/250
+++ b/tests/qemu-iotests/250
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# This test does not make much sense with external data files
+_unsupported_imgopts data_file
 
 # This test checks that qcow2_process_discards does not truncate a discard
 # request > 2G.
diff --git a/tests/qemu-iotests/261 b/tests/qemu-iotests/261
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/261
+++ b/tests/qemu-iotests/261
@@ -XXX,XX +XXX,XX @@ _supported_os Linux
 # (1) We create a v2 image that supports nothing but refcount_bits=16
 # (2) We do some refcount management on our own which expects
 #     refcount_bits=16
-_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)'
+# As for data files, they do not support snapshots at all.
+_unsupported_imgopts 'refcount_bits=$[^1]\|.\([^6]\|$$\)' data_file
 
 # Parameters:
 #   $1: image filename
diff --git a/tests/qemu-iotests/267 b/tests/qemu-iotests/267
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/267
+++ b/tests/qemu-iotests/267
@@ -XXX,XX +XXX,XX @@ _supported_proto file
 _supported_os Linux
 _require_drivers copy-on-read
 
-# Internal snapshots are (currently) impossible with refcount_bits=1
-_unsupported_imgopts 'refcount_bits=1[^0-9]'
+# Internal snapshots are (currently) impossible with refcount_bits=1,
+# and generally impossible with external data files
+_unsupported_imgopts 'refcount_bits=1[^0-9]' data_file
 
 do_run_qemu()
 {
diff --git a/tests/qemu-iotests/273 b/tests/qemu-iotests/273
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/273
+++ b/tests/qemu-iotests/273
@@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 _supported_fmt qcow2
 _supported_proto file
 _supported_os Linux
+# External data files would add nodes to the block graph, so it would
+# not match the reference output
+_unsupported_imgopts data_file
 
 do_run_qemu()
 {
-- 
2.24.1

The problem with allowing the data_file option is that you want to use a
different data file per image used in the test.  Therefore, we need to
allow patterns like -o data_file='$TEST_IMG.data_file'.

Then, we need to filter it out from qemu-img map, qemu-img create, and
remove the data file in _rm_test_img.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-id: 20191107163708.833192-23-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/common.filter | 23 +++++++++++++++++++++--
 tests/qemu-iotests/common.rc     | 22 +++++++++++++++++++++-
 2 files changed, 42 insertions(+), 3 deletions(-)

diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/common.filter
+++ b/tests/qemu-iotests/common.filter
@@ -XXX,XX +XXX,XX @@ _filter_actual_image_size()
 # replace driver-specific options in the "Formatting..." line
 _filter_img_create()
 {
-    $SED -e "s#$REMOTE_TEST_DIR#TEST_DIR#g" \
+    data_file_filter=()
+    if data_file=$(_get_data_file "$TEST_IMG"); then
+        data_file_filter=(-e "s# data_file=$data_file##")
+    fi
+
+    $SED "${data_file_filter[@]}" \
+        -e "s#$REMOTE_TEST_DIR#TEST_DIR#g" \
         -e "s#$IMGPROTO:$TEST_DIR#TEST_DIR#g" \
         -e "s#$TEST_DIR#TEST_DIR#g" \
         -e "s#$SOCK_DIR#SOCK_DIR#g" \
@@ -XXX,XX +XXX,XX @@ _filter_img_info()
 # human and json output
 _filter_qemu_img_map()
 {
+    # Assuming the data_file value in $IMGOPTS contains a '$TEST_IMG',
+    # create a filter that replaces the data file name by $TEST_IMG.
+    # Example:
+    #   In $IMGOPTS: 'data_file=$TEST_IMG.data_file'
+    #   Then data_file_pattern == '$.*$.data_file'
+    #   And  data_file_filter  == -e 's#$.*$.data_file#\1#
+    data_file_filter=()
+    if data_file_pattern=$(_get_data_file '\$.*\$'); then
+        data_file_filter=(-e "s#$data_file_pattern#\\1#")
+    fi
+
     $SED -e 's/$[0-9a-fx]* *[0-9a-fx]* *$[0-9a-fx]* */\1/g' \
         -e 's/"offset": [0-9]\+/"offset": OFFSET/g' \
-        -e 's/Mapped to *//' | _filter_testdir | _filter_imgfmt
+        -e 's/Mapped to *//' \
+        "${data_file_filter[@]}" \
+        | _filter_testdir | _filter_imgfmt
 }
 
 _filter_nbd()
diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -XXX,XX +XXX,XX @@ _stop_nbd_server()
     fi
 }
 
+# Gets the data_file value from IMGOPTS and replaces the '$TEST_IMG'
+# pattern by '$1'
+# Caution: The replacement is done with sed, so $1 must be escaped
+#          properly.  (The delimiter is '#'.)
+_get_data_file()
+{
+    if ! echo "$IMGOPTS" | grep -q 'data_file='; then
+        return 1
+    fi
+
+    echo "$IMGOPTS" | sed -e 's/.*data_file=$[^,]*$.*/\1/' \
+                    | sed -e "s#\\\$TEST_IMG#$1#"
+}
+
 _make_test_img()
 {
     # extra qemu-img options can be added by tests
@@ -XXX,XX +XXX,XX @@ _make_test_img()
     fi
 
     if [ -n "$IMGOPTS" ]; then
-        optstr=$(_optstr_add "$optstr" "$IMGOPTS")
+        imgopts_expanded=$(echo "$IMGOPTS" | sed -e "s#\\\$TEST_IMG#$img_name#")
+        optstr=$(_optstr_add "$optstr" "$imgopts_expanded")
     fi
     if [ -n "$IMGKEYSECRET" ]; then
         object_options="--object secret,id=keysec0,data=$IMGKEYSECRET"
@@ -XXX,XX +XXX,XX @@ _rm_test_img()
         # Remove all the extents for vmdk
         "$QEMU_IMG" info "$img" 2>/dev/null | grep 'filename:' | cut -f 2 -d: \
             | xargs -I {} rm -f "{}"
+    elif [ "$IMGFMT" = "qcow2" ]; then
+        # Remove external data file
+        if data_file=$(_get_data_file "$img"); then
+            rm -f "$data_file"
+        fi
     fi
     rm -f "$img"
 }
-- 
2.24.1

From: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>

Allow writing all the data compressed through the filter driver.
The written data will be aligned by the cluster size.
Based on the QEMU current implementation, that data can be written to
unallocated clusters only. May be used for a backup job.

Suggested-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 1575288906-551879-2-git-send-email-andrey.shinkevich@virtuozzo.com
[mreitz: Replace NULL bdrv_get_format_name() by "(no format)"]
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/Makefile.objs     |   1 +
 block/filter-compress.c | 168 ++++++++++++++++++++++++++++++++++++++++
 qapi/block-core.json    |  10 ++-
 3 files changed, 175 insertions(+), 4 deletions(-)
 create mode 100644 block/filter-compress.c

diff --git a/block/Makefile.objs b/block/Makefile.objs
index XXXXXXX..XXXXXXX 100644
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -XXX,XX +XXX,XX @@ block-obj-y += crypto.o
 
 block-obj-y += aio_task.o
 block-obj-y += backup-top.o
+block-obj-y += filter-compress.o
 
 common-obj-y += stream.o
 
diff --git a/block/filter-compress.c b/block/filter-compress.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/block/filter-compress.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * Compress filter block driver
+ *
+ * Copyright (c) 2019 Virtuozzo International GmbH
+ *
+ * Author:
+ *   Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
+ *   (based on block/copy-on-read.c by Max Reitz)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 or
+ * (at your option) any later version of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include "block/block_int.h"
+#include "qemu/module.h"
+#include "qapi/error.h"
+
+
+static int compress_open(BlockDriverState *bs, QDict *options, int flags,
+                         Error **errp)
+{
+    bs->file = bdrv_open_child(NULL, options, "file", bs, &child_file, false,
+                               errp);
+    if (!bs->file) {
+        return -EINVAL;
+    }
+
+    if (!bs->file->bs->drv || !block_driver_can_compress(bs->file->bs->drv)) {
+        error_setg(errp,
+                   "Compression is not supported for underlying format: %s",
+                   bdrv_get_format_name(bs->file->bs) ?: "(no format)");
+
+        return -ENOTSUP;
+    }
+
+    bs->supported_write_flags = BDRV_REQ_WRITE_UNCHANGED |
+        (BDRV_REQ_FUA & bs->file->bs->supported_write_flags);
+
+    bs->supported_zero_flags = BDRV_REQ_WRITE_UNCHANGED |
+        ((BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK) &
+            bs->file->bs->supported_zero_flags);
+
+    return 0;
+}
+
+
+static int64_t compress_getlength(BlockDriverState *bs)
+{
+    return bdrv_getlength(bs->file->bs);
+}
+
+
+static int coroutine_fn compress_co_preadv_part(BlockDriverState *bs,
+                                                uint64_t offset, uint64_t bytes,
+                                                QEMUIOVector *qiov,
+                                                size_t qiov_offset,
+                                                int flags)
+{
+    return bdrv_co_preadv_part(bs->file, offset, bytes, qiov, qiov_offset,
+                               flags);
+}
+
+
+static int coroutine_fn compress_co_pwritev_part(BlockDriverState *bs,
+                                                 uint64_t offset,
+                                                 uint64_t bytes,
+                                                 QEMUIOVector *qiov,
+                                                 size_t qiov_offset, int flags)
+{
+    return bdrv_co_pwritev_part(bs->file, offset, bytes, qiov, qiov_offset,
+                                flags | BDRV_REQ_WRITE_COMPRESSED);
+}
+
+
+static int coroutine_fn compress_co_pwrite_zeroes(BlockDriverState *bs,
+                                                  int64_t offset, int bytes,
+                                                  BdrvRequestFlags flags)
+{
+    return bdrv_co_pwrite_zeroes(bs->file, offset, bytes, flags);
+}
+
+
+static int coroutine_fn compress_co_pdiscard(BlockDriverState *bs,
+                                             int64_t offset, int bytes)
+{
+    return bdrv_co_pdiscard(bs->file, offset, bytes);
+}
+
+
+static void compress_refresh_limits(BlockDriverState *bs, Error **errp)
+{
+    BlockDriverInfo bdi;
+    int ret;
+
+    if (!bs->file) {
+        return;
+    }
+
+    ret = bdrv_get_info(bs->file->bs, &bdi);
+    if (ret < 0 || bdi.cluster_size == 0) {
+        return;
+    }
+
+    bs->bl.request_alignment = bdi.cluster_size;
+}
+
+
+static void compress_eject(BlockDriverState *bs, bool eject_flag)
+{
+    bdrv_eject(bs->file->bs, eject_flag);
+}
+
+
+static void compress_lock_medium(BlockDriverState *bs, bool locked)
+{
+    bdrv_lock_medium(bs->file->bs, locked);
+}
+
+
+static bool compress_recurse_is_first_non_filter(BlockDriverState *bs,
+                                                 BlockDriverState *candidate)
+{
+    return bdrv_recurse_is_first_non_filter(bs->file->bs, candidate);
+}
+
+
+static BlockDriver bdrv_compress = {
+    .format_name                        = "compress",
+
+    .bdrv_open                          = compress_open,
+    .bdrv_child_perm                    = bdrv_filter_default_perms,
+
+    .bdrv_getlength                     = compress_getlength,
+
+    .bdrv_co_preadv_part                = compress_co_preadv_part,
+    .bdrv_co_pwritev_part               = compress_co_pwritev_part,
+    .bdrv_co_pwrite_zeroes              = compress_co_pwrite_zeroes,
+    .bdrv_co_pdiscard                   = compress_co_pdiscard,
+    .bdrv_refresh_limits                = compress_refresh_limits,
+
+    .bdrv_eject                         = compress_eject,
+    .bdrv_lock_medium                   = compress_lock_medium,
+
+    .bdrv_co_block_status               = bdrv_co_block_status_from_file,
+
+    .bdrv_recurse_is_first_non_filter   = compress_recurse_is_first_non_filter,
+
+    .has_variable_length                = true,
+    .is_filter                          = true,
+};
+
+static void bdrv_compress_init(void)
+{
+    bdrv_register(&bdrv_compress);
+}
+
+block_init(bdrv_compress_init);
diff --git a/qapi/block-core.json b/qapi/block-core.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/block-core.json
+++ b/qapi/block-core.json
@@ -XXX,XX +XXX,XX @@
 # @copy-on-read: Since 3.0
 # @blklogwrites: Since 3.0
 # @blkreplay: Since 4.2
+# @compress: Since 5.0
 #
 # Since: 2.9
 ##
 { 'enum': 'BlockdevDriver',
   'data': [ 'blkdebug', 'blklogwrites', 'blkreplay', 'blkverify', 'bochs',
-            'cloop', 'copy-on-read', 'dmg', 'file', 'ftp', 'ftps', 'gluster',
-            'host_cdrom', 'host_device', 'http', 'https', 'iscsi', 'luks',
-            'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels', 'qcow',
-            'qcow2', 'qed', 'quorum', 'raw', 'rbd',
+            'cloop', 'compress', 'copy-on-read', 'dmg', 'file', 'ftp', 'ftps',
+            'gluster', 'host_cdrom', 'host_device', 'http', 'https', 'iscsi',
+            'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
+            'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
             { 'name': 'replication', 'if': 'defined(CONFIG_REPLICATION)' },
             'sheepdog',
             'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat', 'vxhs' ] }
@@ -XXX,XX +XXX,XX @@
       'blkreplay':  'BlockdevOptionsBlkreplay',
       'bochs':      'BlockdevOptionsGenericFormat',
       'cloop':      'BlockdevOptionsGenericFormat',
+      'compress':   'BlockdevOptionsGenericFormat',
       'copy-on-read':'BlockdevOptionsGenericFormat',
       'dmg':        'BlockdevOptionsGenericFormat',
       'file':       'BlockdevOptionsFile',
-- 
2.24.1

From: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>

QEMU currently supports writing compressed data of the size equal to
one cluster. This patch allows writing QCOW2 compressed data that
exceed one cluster. Now, we split buffered data into separate clusters
and write them compressed using the block/aio_task API.

Suggested-by: Pavel Butsykin <pbutsykin@virtuozzo.com>
Suggested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 1575288906-551879-3-git-send-email-andrey.shinkevich@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/qcow2.c | 102 +++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 75 insertions(+), 27 deletions(-)

diff --git a/block/qcow2.c b/block/qcow2.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -XXX,XX +XXX,XX @@ fail:
     return ret;
 }
 
-/* XXX: put compressed sectors first, then all the cluster aligned
-   tables to avoid losing bytes in alignment */
 static coroutine_fn int
-qcow2_co_pwritev_compressed_part(BlockDriverState *bs,
+qcow2_co_pwritev_compressed_task(BlockDriverState *bs,
                                  uint64_t offset, uint64_t bytes,
                                  QEMUIOVector *qiov, size_t qiov_offset)
 {
@@ -XXX,XX +XXX,XX @@ qcow2_co_pwritev_compressed_part(BlockDriverState *bs,
     uint8_t *buf, *out_buf;
     uint64_t cluster_offset;
 
-    if (has_data_file(bs)) {
-        return -ENOTSUP;
-    }
-
-    if (bytes == 0) {
-        /* align end of file to a sector boundary to ease reading with
-           sector based I/Os */
-        int64_t len = bdrv_getlength(bs->file->bs);
-        if (len < 0) {
-            return len;
-        }
-        return bdrv_co_truncate(bs->file, len, false, PREALLOC_MODE_OFF, NULL);
-    }
-
-    if (offset_into_cluster(s, offset)) {
-        return -EINVAL;
-    }
+    assert(bytes == s->cluster_size || (bytes < s->cluster_size &&
+           (offset + bytes == bs->total_sectors << BDRV_SECTOR_BITS)));
 
     buf = qemu_blockalign(bs, s->cluster_size);
-    if (bytes != s->cluster_size) {
-        if (bytes > s->cluster_size ||
-            offset + bytes != bs->total_sectors << BDRV_SECTOR_BITS)
-        {
-            qemu_vfree(buf);
-            return -EINVAL;
-        }
+    if (bytes < s->cluster_size) {
         /* Zero-pad last write if image size is not cluster aligned */
         memset(buf + bytes, 0, s->cluster_size - bytes);
     }
@@ -XXX,XX +XXX,XX @@ fail:
     return ret;
 }
 
+static coroutine_fn int qcow2_co_pwritev_compressed_task_entry(AioTask *task)
+{
+    Qcow2AioTask *t = container_of(task, Qcow2AioTask, task);
+
+    assert(!t->cluster_type && !t->l2meta);
+
+    return qcow2_co_pwritev_compressed_task(t->bs, t->offset, t->bytes, t->qiov,
+                                            t->qiov_offset);
+}
+
+/*
+ * XXX: put compressed sectors first, then all the cluster aligned
+ * tables to avoid losing bytes in alignment
+ */
+static coroutine_fn int
+qcow2_co_pwritev_compressed_part(BlockDriverState *bs,
+                                 uint64_t offset, uint64_t bytes,
+                                 QEMUIOVector *qiov, size_t qiov_offset)
+{
+    BDRVQcow2State *s = bs->opaque;
+    AioTaskPool *aio = NULL;
+    int ret = 0;
+
+    if (has_data_file(bs)) {
+        return -ENOTSUP;
+    }
+
+    if (bytes == 0) {
+        /*
+         * align end of file to a sector boundary to ease reading with
+         * sector based I/Os
+         */
+        int64_t len = bdrv_getlength(bs->file->bs);
+        if (len < 0) {
+            return len;
+        }
+        return bdrv_co_truncate(bs->file, len, false, PREALLOC_MODE_OFF, NULL);
+    }
+
+    if (offset_into_cluster(s, offset)) {
+        return -EINVAL;
+    }
+
+    while (bytes && aio_task_pool_status(aio) == 0) {
+        uint64_t chunk_size = MIN(bytes, s->cluster_size);
+
+        if (!aio && chunk_size != bytes) {
+            aio = aio_task_pool_new(QCOW2_MAX_WORKERS);
+        }
+
+        ret = qcow2_add_task(bs, aio, qcow2_co_pwritev_compressed_task_entry,
+                             0, 0, offset, chunk_size, qiov, qiov_offset, NULL);
+        if (ret < 0) {
+            break;
+        }
+        qiov_offset += chunk_size;
+        offset += chunk_size;
+        bytes -= chunk_size;
+    }
+
+    if (aio) {
+        aio_task_pool_wait_all(aio);
+        if (ret == 0) {
+            ret = aio_task_pool_status(aio);
+        }
+        g_free(aio);
+    }
+
+    return ret;
+}
+
 static int coroutine_fn
 qcow2_co_preadv_compressed(BlockDriverState *bs,
                            uint64_t file_cluster_offset,
-- 
2.24.1

From: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>

Add the case to the iotest #214 that checks possibility of writing
compressed data of more than one cluster size. The test case involves
the compress filter driver showing a sample usage of that.

Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 1575288906-551879-4-git-send-email-andrey.shinkevich@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/214     | 43 ++++++++++++++++++++++++++++++++++++++
 tests/qemu-iotests/214.out | 14 +++++++++++++
 2 files changed, 57 insertions(+)

diff --git a/tests/qemu-iotests/214 b/tests/qemu-iotests/214
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/214
+++ b/tests/qemu-iotests/214
@@ -XXX,XX +XXX,XX @@ _check_test_img -r all
 $QEMU_IO -c "read  -P 0x11  0 4M" "$TEST_IMG" 2>&1 | _filter_qemu_io | _filter_testdir
 $QEMU_IO -c "read  -P 0x22 4M 4M" "$TEST_IMG" 2>&1 | _filter_qemu_io | _filter_testdir
 
+echo
+echo "=== Write compressed data of multiple clusters ==="
+echo
+cluster_size=0x10000
+_make_test_img 2M -o cluster_size=$cluster_size
+
+echo "Write uncompressed data:"
+let data_size="8 * $cluster_size"
+$QEMU_IO -c "write -P 0xaa 0 $data_size" "$TEST_IMG" \
+         2>&1 | _filter_qemu_io | _filter_testdir
+sizeA=$($QEMU_IMG info --output=json "$TEST_IMG" |
+        sed -n '/"actual-size":/ s/[^0-9]//gp')
+
+_make_test_img 2M -o cluster_size=$cluster_size
+echo "Write compressed data:"
+let data_size="3 * $cluster_size + $cluster_size / 2"
+# Set compress on. That will align the written data
+# by the cluster size and will write them compressed.
+QEMU_IO_OPTIONS=$QEMU_IO_OPTIONS_NO_FMT \
+$QEMU_IO -c "write -P 0xbb 0 $data_size" --image-opts \
+         "driver=compress,file.driver=$IMGFMT,file.file.driver=file,file.file.filename=$TEST_IMG" \
+         2>&1 | _filter_qemu_io | _filter_testdir
+
+let offset="4 * $cluster_size + $cluster_size / 4"
+QEMU_IO_OPTIONS=$QEMU_IO_OPTIONS_NO_FMT \
+$QEMU_IO -c "write -P 0xcc $offset $data_size" "json:{\
+    'driver': 'compress',
+    'file': {'driver': '$IMGFMT',
+             'file': {'driver': 'file',
+                      'filename': '$TEST_IMG'}}}" | \
+                          _filter_qemu_io | _filter_testdir
+
+sizeB=$($QEMU_IMG info --output=json "$TEST_IMG" |
+        sed -n '/"actual-size":/ s/[^0-9]//gp')
+
+if [ $sizeA -le $sizeB ]
+then
+    echo "Compression ERROR"
+fi
+
+$QEMU_IMG check --output=json "$TEST_IMG" |
+          sed -n 's/,$//; /"compressed-clusters":/ s/^ *//p'
+
 # success, all done
 echo '*** done'
 rm -f $seq.full
diff --git a/tests/qemu-iotests/214.out b/tests/qemu-iotests/214.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/214.out
+++ b/tests/qemu-iotests/214.out
@@ -XXX,XX +XXX,XX @@ read 4194304/4194304 bytes at offset 0
 4 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 read 4194304/4194304 bytes at offset 4194304
 4 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+=== Write compressed data of multiple clusters ===
+
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2097152
+Write uncompressed data:
+wrote 524288/524288 bytes at offset 0
+512 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2097152
+Write compressed data:
+wrote 229376/229376 bytes at offset 0
+224 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 229376/229376 bytes at offset 278528
+224 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+"compressed-clusters": 8
 *** done
-- 
2.24.1

From: Philippe Mathieu-Daudé <philmd@redhat.com>

Commit 6f6e1698a6 desugarized "-machine accel=" to a list
of "-accel" options. Since now "-machine accel" and "-accel"
became incompatible, update the iotests to the new format.

Error reported here:
https://gitlab.com/qemu-project/qemu/-/jobs/385801004#L3400

Reported-by: GitLab CI
Fixes: 6f6e1698a6 (vl: configure accelerators from -accel options)
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200106130951.29873-1-philmd@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/235   | 2 +-
 tests/qemu-iotests/check | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/tests/qemu-iotests/235 b/tests/qemu-iotests/235
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/235
+++ b/tests/qemu-iotests/235
@@ -XXX,XX +XXX,XX @@ qemu_img_create('-f', iotests.imgfmt, '-o', 'preallocation=metadata', disk,
                 str(size))
 
 vm = QEMUMachine(iotests.qemu_prog)
-vm.add_args('-machine', 'accel=kvm:tcg')
+vm.add_args('-accel', 'kvm', '-accel', 'tcg')
 if iotests.qemu_default_machine == 's390-ccw-virtio':
         vm.add_args('-no-shutdown')
 vm.add_args('-drive', 'id=src,file=' + disk)
diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/check
+++ b/tests/qemu-iotests/check
@@ -XXX,XX +XXX,XX @@ export QEMU_PROG="$(type -p "$QEMU_PROG")"
 
 case "$QEMU_PROG" in
     *qemu-system-arm|*qemu-system-aarch64)
-        export QEMU_OPTIONS="-nodefaults -display none -machine virt,accel=qtest"
+        export QEMU_OPTIONS="-nodefaults -display none -machine virt -accel qtest"
         ;;
     *qemu-system-tricore)
-        export QEMU_OPTIONS="-nodefaults -display none -machine tricore_testboard,accel=qtest"
+        export QEMU_OPTIONS="-nodefaults -display none -machine tricore_testboard -accel qtest"
         ;;
     *)
-        export QEMU_OPTIONS="-nodefaults -display none -machine accel=qtest"
+        export QEMU_OPTIONS="-nodefaults -display none -accel qtest"
         ;;
 esac
 
-- 
2.24.1

When dropping backup-top, we need to drain the node before freeing the
BlockCopyState.  Otherwise, requests may still be in flight and then the
assertion in shres_destroy() will fail.

(This becomes visible in intermittent failure of 056.)

Cc: qemu-stable@nongnu.org
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20191219182638.104621-1-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/backup-top.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/block/backup-top.c b/block/backup-top.c
index XXXXXXX..XXXXXXX 100644
--- a/block/backup-top.c
+++ b/block/backup-top.c
@@ -XXX,XX +XXX,XX @@ void bdrv_backup_top_drop(BlockDriverState *bs)
     BDRVBackupTopState *s = bs->opaque;
     AioContext *aio_context = bdrv_get_aio_context(bs);
 
-    block_copy_state_free(s->bcs);
-
     aio_context_acquire(aio_context);
 
     bdrv_drained_begin(bs);
 
+    block_copy_state_free(s->bcs);
+
     s->active = false;
     bdrv_child_refresh_perms(bs, bs->backing, &error_abort);
     bdrv_replace_node(bs, backing_bs(bs), &error_abort);
-- 
2.24.1

The following changes since commit 56f9e46b841c7be478ca038d8d4085d776ab4b0d:

Merge remote-tracking branch 'remotes/armbru/tags/pull-qapi-2017-02-20' into staging (2017-02-20 17:42:47 +0000)

are available in the git repository at:

git://github.com/stefanha/qemu.git tags/block-pull-request

for you to fetch changes up to a7b91d35bab97a2d3e779d0c64c9b837b52a6cf7:

coroutine-lock: make CoRwlock thread-safe and fair (2017-02-21 11:39:40 +0000)

----------------------------------------------------------------
Pull request

v2:
 * Rebased to resolve scsi conflicts

----------------------------------------------------------------

Paolo Bonzini (24):
  block: move AioContext, QEMUTimer, main-loop to libqemuutil
  aio: introduce aio_co_schedule and aio_co_wake
  block-backend: allow blk_prw from coroutine context
  test-thread-pool: use generic AioContext infrastructure
  io: add methods to set I/O handlers on AioContext
  io: make qio_channel_yield aware of AioContexts
  nbd: convert to use qio_channel_yield
  coroutine-lock: reschedule coroutine on the AioContext it was running
    on
  blkdebug: reschedule coroutine on the AioContext it is running on
  qed: introduce qed_aio_start_io and qed_aio_next_io_cb
  aio: push aio_context_acquire/release down to dispatching
  block: explicitly acquire aiocontext in timers that need it
  block: explicitly acquire aiocontext in callbacks that need it
  block: explicitly acquire aiocontext in bottom halves that need it
  block: explicitly acquire aiocontext in aio callbacks that need it
  aio-posix: partially inline aio_dispatch into aio_poll
  async: remove unnecessary inc/dec pairs
  block: document fields protected by AioContext lock
  coroutine-lock: make CoMutex thread-safe
  coroutine-lock: add limited spinning to CoMutex
  test-aio-multithread: add performance comparison with thread-based
    mutexes
  coroutine-lock: place CoMutex before CoQueue in header
  coroutine-lock: add mutex argument to CoQueue APIs
  coroutine-lock: make CoRwlock thread-safe and fair

-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

AioContext is fairly self contained, the only dependency is QEMUTimer but
that in turn doesn't need anything else.  So move them out of block-obj-y
to avoid introducing a dependency from io/ to block-obj-y.

main-loop and its dependency iohandler also need to be moved, because
later in this series io/ will call iohandler_get_aio_context.

[Changed copyright "the QEMU team" to "other QEMU contributors" as
suggested by Daniel Berrange and agreed by Paolo.
--Stefan]

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213135235.12274-2-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 Makefile.objs                       |  4 ---
 stubs/Makefile.objs                 |  1 +
 tests/Makefile.include              | 11 ++++----
 util/Makefile.objs                  |  6 +++-
 block/io.c                          | 29 -------------------
 stubs/linux-aio.c                   | 32 +++++++++++++++++++++
 stubs/set-fd-handler.c              | 11 --------
 aio-posix.c => util/aio-posix.c     |  2 +-
 aio-win32.c => util/aio-win32.c     |  0
 util/aiocb.c                        | 55 +++++++++++++++++++++++++++++++++++++
 async.c => util/async.c             |  3 +-
 iohandler.c => util/iohandler.c     |  0
 main-loop.c => util/main-loop.c     |  0
 qemu-timer.c => util/qemu-timer.c   |  0
 thread-pool.c => util/thread-pool.c |  2 +-
 trace-events                        | 11 --------
 util/trace-events                   | 11 ++++++++
 17 files changed, 114 insertions(+), 64 deletions(-)
 create mode 100644 stubs/linux-aio.c
 rename aio-posix.c => util/aio-posix.c (99%)
 rename aio-win32.c => util/aio-win32.c (100%)
 create mode 100644 util/aiocb.c
 rename async.c => util/async.c (99%)
 rename iohandler.c => util/iohandler.c (100%)
 rename main-loop.c => util/main-loop.c (100%)
 rename qemu-timer.c => util/qemu-timer.c (100%)
 rename thread-pool.c => util/thread-pool.c (99%)

diff --git a/Makefile.objs b/Makefile.objs
index XXXXXXX..XXXXXXX 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -XXX,XX +XXX,XX @@ chardev-obj-y = chardev/
 #######################################################################
 # block-obj-y is code used by both qemu system emulation and qemu-img
 
-block-obj-y = async.o thread-pool.o
 block-obj-y += nbd/
 block-obj-y += block.o blockjob.o
-block-obj-y += main-loop.o iohandler.o qemu-timer.o
-block-obj-$(CONFIG_POSIX) += aio-posix.o
-block-obj-$(CONFIG_WIN32) += aio-win32.o
 block-obj-y += block/
 block-obj-y += qemu-io-cmds.o
 block-obj-$(CONFIG_REPLICATION) += replication.o
diff --git a/stubs/Makefile.objs b/stubs/Makefile.objs
index XXXXXXX..XXXXXXX 100644
--- a/stubs/Makefile.objs
+++ b/stubs/Makefile.objs
@@ -XXX,XX +XXX,XX @@ stub-obj-y += get-vm-name.o
 stub-obj-y += iothread.o
 stub-obj-y += iothread-lock.o
 stub-obj-y += is-daemonized.o
+stub-obj-$(CONFIG_LINUX_AIO) += linux-aio.o
 stub-obj-y += machine-init-done.o
 stub-obj-y += migr-blocker.o
 stub-obj-y += monitor.o
diff --git a/tests/Makefile.include b/tests/Makefile.include
index XXXXXXX..XXXXXXX 100644
--- a/tests/Makefile.include
+++ b/tests/Makefile.include
@@ -XXX,XX +XXX,XX @@ check-unit-y += tests/test-visitor-serialization$(EXESUF)
 check-unit-y += tests/test-iov$(EXESUF)
 gcov-files-test-iov-y = util/iov.c
 check-unit-y += tests/test-aio$(EXESUF)
+gcov-files-test-aio-y = util/async.c util/qemu-timer.o
+gcov-files-test-aio-$(CONFIG_WIN32) += util/aio-win32.c
+gcov-files-test-aio-$(CONFIG_POSIX) += util/aio-posix.c
 check-unit-y += tests/test-throttle$(EXESUF)
 gcov-files-test-aio-$(CONFIG_WIN32) = aio-win32.c
 gcov-files-test-aio-$(CONFIG_POSIX) = aio-posix.c
@@ -XXX,XX +XXX,XX @@ tests/check-qjson$(EXESUF): tests/check-qjson.o $(test-util-obj-y)
 tests/check-qom-interface$(EXESUF): tests/check-qom-interface.o $(test-qom-obj-y)
 tests/check-qom-proplist$(EXESUF): tests/check-qom-proplist.o $(test-qom-obj-y)
 
-tests/test-char$(EXESUF): tests/test-char.o qemu-timer.o \
-	$(test-util-obj-y) $(qtest-obj-y) $(test-block-obj-y) $(chardev-obj-y)
+tests/test-char$(EXESUF): tests/test-char.o $(test-util-obj-y) $(qtest-obj-y) $(test-io-obj-y) $(chardev-obj-y)
 tests/test-coroutine$(EXESUF): tests/test-coroutine.o $(test-block-obj-y)
 tests/test-aio$(EXESUF): tests/test-aio.o $(test-block-obj-y)
 tests/test-throttle$(EXESUF): tests/test-throttle.o $(test-block-obj-y)
@@ -XXX,XX +XXX,XX @@ tests/test-vmstate$(EXESUF): tests/test-vmstate.o \
 	migration/vmstate.o migration/qemu-file.o \
         migration/qemu-file-channel.o migration/qjson.o \
 	$(test-io-obj-y)
-tests/test-timed-average$(EXESUF): tests/test-timed-average.o qemu-timer.o \
-	$(test-util-obj-y)
+tests/test-timed-average$(EXESUF): tests/test-timed-average.o $(test-util-obj-y)
 tests/test-base64$(EXESUF): tests/test-base64.o \
 	libqemuutil.a libqemustub.a
 tests/ptimer-test$(EXESUF): tests/ptimer-test.o tests/ptimer-test-stubs.o hw/core/ptimer.o libqemustub.a
@@ -XXX,XX +XXX,XX @@ tests/usb-hcd-ehci-test$(EXESUF): tests/usb-hcd-ehci-test.o $(libqos-usb-obj-y)
 tests/usb-hcd-xhci-test$(EXESUF): tests/usb-hcd-xhci-test.o $(libqos-usb-obj-y)
 tests/pc-cpu-test$(EXESUF): tests/pc-cpu-test.o
 tests/postcopy-test$(EXESUF): tests/postcopy-test.o
-tests/vhost-user-test$(EXESUF): tests/vhost-user-test.o qemu-timer.o \
+tests/vhost-user-test$(EXESUF): tests/vhost-user-test.o $(test-util-obj-y) \
 	$(qtest-obj-y) $(test-io-obj-y) $(libqos-virtio-obj-y) $(libqos-pc-obj-y) \
 	$(chardev-obj-y)
 tests/qemu-iotests/socket_scm_helper$(EXESUF): tests/qemu-iotests/socket_scm_helper.o
diff --git a/util/Makefile.objs b/util/Makefile.objs
index XXXXXXX..XXXXXXX 100644
--- a/util/Makefile.objs
+++ b/util/Makefile.objs
@@ -XXX,XX +XXX,XX @@
 util-obj-y = osdep.o cutils.o unicode.o qemu-timer-common.o
 util-obj-y += bufferiszero.o
 util-obj-y += lockcnt.o
+util-obj-y += aiocb.o async.o thread-pool.o qemu-timer.o
+util-obj-y += main-loop.o iohandler.o
+util-obj-$(CONFIG_POSIX) += aio-posix.o
 util-obj-$(CONFIG_POSIX) += compatfd.o
 util-obj-$(CONFIG_POSIX) += event_notifier-posix.o
 util-obj-$(CONFIG_POSIX) += mmap-alloc.o
 util-obj-$(CONFIG_POSIX) += oslib-posix.o
 util-obj-$(CONFIG_POSIX) += qemu-openpty.o
 util-obj-$(CONFIG_POSIX) += qemu-thread-posix.o
-util-obj-$(CONFIG_WIN32) += event_notifier-win32.o
 util-obj-$(CONFIG_POSIX) += memfd.o
+util-obj-$(CONFIG_WIN32) += aio-win32.o
+util-obj-$(CONFIG_WIN32) += event_notifier-win32.o
 util-obj-$(CONFIG_WIN32) += oslib-win32.o
 util-obj-$(CONFIG_WIN32) += qemu-thread-win32.o
 util-obj-y += envlist.o path.o module.o
diff --git a/block/io.c b/block/io.c
index XXXXXXX..XXXXXXX 100644
--- a/block/io.c
+++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ BlockAIOCB *bdrv_aio_flush(BlockDriverState *bs,
     return &acb->common;
 }
 
-void *qemu_aio_get(const AIOCBInfo *aiocb_info, BlockDriverState *bs,
-                   BlockCompletionFunc *cb, void *opaque)
-{
-    BlockAIOCB *acb;
-
-    acb = g_malloc(aiocb_info->aiocb_size);
-    acb->aiocb_info = aiocb_info;
-    acb->bs = bs;
-    acb->cb = cb;
-    acb->opaque = opaque;
-    acb->refcnt = 1;
-    return acb;
-}
-
-void qemu_aio_ref(void *p)
-{
-    BlockAIOCB *acb = p;
-    acb->refcnt++;
-}
-
-void qemu_aio_unref(void *p)
-{
-    BlockAIOCB *acb = p;
-    assert(acb->refcnt > 0);
-    if (--acb->refcnt == 0) {
-        g_free(acb);
-    }
-}
-
 /**************************************************************/
 /* Coroutine block device emulation */
 
diff --git a/stubs/linux-aio.c b/stubs/linux-aio.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/stubs/linux-aio.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * Linux native AIO support.
+ *
+ * Copyright (C) 2009 IBM, Corp.
+ * Copyright (C) 2009 Red Hat, Inc.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+#include "qemu/osdep.h"
+#include "block/aio.h"
+#include "block/raw-aio.h"
+
+void laio_detach_aio_context(LinuxAioState *s, AioContext *old_context)
+{
+    abort();
+}
+
+void laio_attach_aio_context(LinuxAioState *s, AioContext *new_context)
+{
+    abort();
+}
+
+LinuxAioState *laio_init(void)
+{
+    abort();
+}
+
+void laio_cleanup(LinuxAioState *s)
+{
+    abort();
+}
diff --git a/stubs/set-fd-handler.c b/stubs/set-fd-handler.c
index XXXXXXX..XXXXXXX 100644
--- a/stubs/set-fd-handler.c
+++ b/stubs/set-fd-handler.c
@@ -XXX,XX +XXX,XX @@ void qemu_set_fd_handler(int fd,
 {
     abort();
 }
-
-void aio_set_fd_handler(AioContext *ctx,
-                        int fd,
-                        bool is_external,
-                        IOHandler *io_read,
-                        IOHandler *io_write,
-                        AioPollFn *io_poll,
-                        void *opaque)
-{
-    abort();
-}
diff --git a/aio-posix.c b/util/aio-posix.c
similarity index 99%
rename from aio-posix.c
rename to util/aio-posix.c
index XXXXXXX..XXXXXXX 100644
--- a/aio-posix.c
+++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/rcu_queue.h"
 #include "qemu/sockets.h"
 #include "qemu/cutils.h"
-#include "trace-root.h"
+#include "trace.h"
 #ifdef CONFIG_EPOLL_CREATE1
 #include <sys/epoll.h>
 #endif
diff --git a/aio-win32.c b/util/aio-win32.c
similarity index 100%
rename from aio-win32.c
rename to util/aio-win32.c
diff --git a/util/aiocb.c b/util/aiocb.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/util/aiocb.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * BlockAIOCB allocation
+ *
+ * Copyright (c) 2003-2017 Fabrice Bellard and other QEMU contributors
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "qemu/osdep.h"
+#include "block/aio.h"
+
+void *qemu_aio_get(const AIOCBInfo *aiocb_info, BlockDriverState *bs,
+                   BlockCompletionFunc *cb, void *opaque)
+{
+    BlockAIOCB *acb;
+
+    acb = g_malloc(aiocb_info->aiocb_size);
+    acb->aiocb_info = aiocb_info;
+    acb->bs = bs;
+    acb->cb = cb;
+    acb->opaque = opaque;
+    acb->refcnt = 1;
+    return acb;
+}
+
+void qemu_aio_ref(void *p)
+{
+    BlockAIOCB *acb = p;
+    acb->refcnt++;
+}
+
+void qemu_aio_unref(void *p)
+{
+    BlockAIOCB *acb = p;
+    assert(acb->refcnt > 0);
+    if (--acb->refcnt == 0) {
+        g_free(acb);
+    }
+}
diff --git a/async.c b/util/async.c
similarity index 99%
rename from async.c
rename to util/async.c
index XXXXXXX..XXXXXXX 100644
--- a/async.c
+++ b/util/async.c
@@ -XXX,XX +XXX,XX @@
 /*
- * QEMU System Emulator
+ * Data plane event loop
  *
  * Copyright (c) 2003-2008 Fabrice Bellard
+ * Copyright (c) 2009-2017 QEMU contributors
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
diff --git a/iohandler.c b/util/iohandler.c
similarity index 100%
rename from iohandler.c
rename to util/iohandler.c
diff --git a/main-loop.c b/util/main-loop.c
similarity index 100%
rename from main-loop.c
rename to util/main-loop.c
diff --git a/qemu-timer.c b/util/qemu-timer.c
similarity index 100%
rename from qemu-timer.c
rename to util/qemu-timer.c
diff --git a/thread-pool.c b/util/thread-pool.c
similarity index 99%
rename from thread-pool.c
rename to util/thread-pool.c
index XXXXXXX..XXXXXXX 100644
--- a/thread-pool.c
+++ b/util/thread-pool.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/queue.h"
 #include "qemu/thread.h"
 #include "qemu/coroutine.h"
-#include "trace-root.h"
+#include "trace.h"
 #include "block/thread-pool.h"
 #include "qemu/main-loop.h"
 
diff --git a/trace-events b/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/trace-events
+++ b/trace-events
@@ -XXX,XX +XXX,XX @@
 #
 # The <format-string> should be a sprintf()-compatible format string.
 
-# aio-posix.c
-run_poll_handlers_begin(void *ctx, int64_t max_ns) "ctx %p max_ns %"PRId64
-run_poll_handlers_end(void *ctx, bool progress) "ctx %p progress %d"
-poll_shrink(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
-poll_grow(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
-
-# thread-pool.c
-thread_pool_submit(void *pool, void *req, void *opaque) "pool %p req %p opaque %p"
-thread_pool_complete(void *pool, void *req, void *opaque, int ret) "pool %p req %p opaque %p ret %d"
-thread_pool_cancel(void *req, void *opaque) "req %p opaque %p"
-
 # ioport.c
 cpu_in(unsigned int addr, char size, unsigned int val) "addr %#x(%c) value %u"
 cpu_out(unsigned int addr, char size, unsigned int val) "addr %#x(%c) value %u"
diff --git a/util/trace-events b/util/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/util/trace-events
+++ b/util/trace-events
@@ -XXX,XX +XXX,XX @@
 # See docs/tracing.txt for syntax documentation.
 
+# util/aio-posix.c
+run_poll_handlers_begin(void *ctx, int64_t max_ns) "ctx %p max_ns %"PRId64
+run_poll_handlers_end(void *ctx, bool progress) "ctx %p progress %d"
+poll_shrink(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
+poll_grow(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
+
+# util/thread-pool.c
+thread_pool_submit(void *pool, void *req, void *opaque) "pool %p req %p opaque %p"
+thread_pool_complete(void *pool, void *req, void *opaque, int ret) "pool %p req %p opaque %p ret %d"
+thread_pool_cancel(void *req, void *opaque) "req %p opaque %p"
+
 # util/buffer.c
 buffer_resize(const char *buf, size_t olen, size_t len) "%s: old %zd, new %zd"
 buffer_move_empty(const char *buf, size_t len, const char *from) "%s: %zd bytes from %s"
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

aio_co_wake provides the infrastructure to start a coroutine on a "home"
AioContext.  It will be used by CoMutex and CoQueue, so that coroutines
don't jump from one context to another when they go to sleep on a
mutex or waitqueue.  However, it can also be used as a more efficient
alternative to one-shot bottom halves, and saves the effort of tracking
which AioContext a coroutine is running on.

aio_co_schedule is the part of aio_co_wake that starts a coroutine
on a remove AioContext, but it is also useful to implement e.g.
bdrv_set_aio_context callbacks.

The implementation of aio_co_schedule is based on a lock-free
multiple-producer, single-consumer queue.  The multiple producers use
cmpxchg to add to a LIFO stack.  The consumer (a per-AioContext bottom
half) grabs all items added so far, inverts the list to make it FIFO,
and goes through it one item at a time until it's empty.  The data
structure was inspired by OSv, which uses it in the very code we'll
"port" to QEMU for the thread-safe CoMutex.

Most of the new code is really tests.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213135235.12274-3-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 tests/Makefile.include       |   8 +-
 include/block/aio.h          |  32 +++++++
 include/qemu/coroutine_int.h |  11 ++-
 tests/iothread.h             |  25 +++++
 tests/iothread.c             |  91 ++++++++++++++++++
 tests/test-aio-multithread.c | 213 +++++++++++++++++++++++++++++++++++++++++++
 util/async.c                 |  65 +++++++++++++
 util/qemu-coroutine.c        |   8 ++
 util/trace-events            |   4 +
 9 files changed, 453 insertions(+), 4 deletions(-)
 create mode 100644 tests/iothread.h
 create mode 100644 tests/iothread.c
 create mode 100644 tests/test-aio-multithread.c

diff --git a/tests/Makefile.include b/tests/Makefile.include
index XXXXXXX..XXXXXXX 100644
--- a/tests/Makefile.include
+++ b/tests/Makefile.include
@@ -XXX,XX +XXX,XX @@ check-unit-y += tests/test-aio$(EXESUF)
 gcov-files-test-aio-y = util/async.c util/qemu-timer.o
 gcov-files-test-aio-$(CONFIG_WIN32) += util/aio-win32.c
 gcov-files-test-aio-$(CONFIG_POSIX) += util/aio-posix.c
+check-unit-y += tests/test-aio-multithread$(EXESUF)
+gcov-files-test-aio-multithread-y = $(gcov-files-test-aio-y)
+gcov-files-test-aio-multithread-y += util/qemu-coroutine.c tests/iothread.c
 check-unit-y += tests/test-throttle$(EXESUF)
-gcov-files-test-aio-$(CONFIG_WIN32) = aio-win32.c
-gcov-files-test-aio-$(CONFIG_POSIX) = aio-posix.c
 check-unit-y += tests/test-thread-pool$(EXESUF)
 gcov-files-test-thread-pool-y = thread-pool.c
 gcov-files-test-hbitmap-y = util/hbitmap.c
@@ -XXX,XX +XXX,XX @@ test-qapi-obj-y = tests/test-qapi-visit.o tests/test-qapi-types.o \
 	$(test-qom-obj-y)
 test-crypto-obj-y = $(crypto-obj-y) $(test-qom-obj-y)
 test-io-obj-y = $(io-obj-y) $(test-crypto-obj-y)
-test-block-obj-y = $(block-obj-y) $(test-io-obj-y)
+test-block-obj-y = $(block-obj-y) $(test-io-obj-y) tests/iothread.o
 
 tests/check-qint$(EXESUF): tests/check-qint.o $(test-util-obj-y)
 tests/check-qstring$(EXESUF): tests/check-qstring.o $(test-util-obj-y)
@@ -XXX,XX +XXX,XX @@ tests/check-qom-proplist$(EXESUF): tests/check-qom-proplist.o $(test-qom-obj-y)
 tests/test-char$(EXESUF): tests/test-char.o $(test-util-obj-y) $(qtest-obj-y) $(test-io-obj-y) $(chardev-obj-y)
 tests/test-coroutine$(EXESUF): tests/test-coroutine.o $(test-block-obj-y)
 tests/test-aio$(EXESUF): tests/test-aio.o $(test-block-obj-y)
+tests/test-aio-multithread$(EXESUF): tests/test-aio-multithread.o $(test-block-obj-y)
 tests/test-throttle$(EXESUF): tests/test-throttle.o $(test-block-obj-y)
 tests/test-blockjob$(EXESUF): tests/test-blockjob.o $(test-block-obj-y) $(test-util-obj-y)
 tests/test-blockjob-txn$(EXESUF): tests/test-blockjob-txn.o $(test-block-obj-y) $(test-util-obj-y)
diff --git a/include/block/aio.h b/include/block/aio.h
index XXXXXXX..XXXXXXX 100644
--- a/include/block/aio.h
+++ b/include/block/aio.h
@@ -XXX,XX +XXX,XX @@ typedef void QEMUBHFunc(void *opaque);
 typedef bool AioPollFn(void *opaque);
 typedef void IOHandler(void *opaque);
 
+struct Coroutine;
 struct ThreadPool;
 struct LinuxAioState;
 
@@ -XXX,XX +XXX,XX @@ struct AioContext {
     bool notified;
     EventNotifier notifier;
 
+    QSLIST_HEAD(, Coroutine) scheduled_coroutines;
+    QEMUBH *co_schedule_bh;
+
     /* Thread pool for performing work and receiving completion callbacks.
      * Has its own locking.
      */
@@ -XXX,XX +XXX,XX @@ static inline bool aio_node_check(AioContext *ctx, bool is_external)
 }
 
 /**
+ * aio_co_schedule:
+ * @ctx: the aio context
+ * @co: the coroutine
+ *
+ * Start a coroutine on a remote AioContext.
+ *
+ * The coroutine must not be entered by anyone else while aio_co_schedule()
+ * is active.  In addition the coroutine must have yielded unless ctx
+ * is the context in which the coroutine is running (i.e. the value of
+ * qemu_get_current_aio_context() from the coroutine itself).
+ */
+void aio_co_schedule(AioContext *ctx, struct Coroutine *co);
+
+/**
+ * aio_co_wake:
+ * @co: the coroutine
+ *
+ * Restart a coroutine on the AioContext where it was running last, thus
+ * preventing coroutines from jumping from one context to another when they
+ * go to sleep.
+ *
+ * aio_co_wake may be executed either in coroutine or non-coroutine
+ * context.  The coroutine must not be entered by anyone else while
+ * aio_co_wake() is active.
+ */
+void aio_co_wake(struct Coroutine *co);
+
+/**
  * Return the AioContext whose event loop runs in the current thread.
  *
  * If called from an IOThread this will be the IOThread's AioContext.  If
diff --git a/include/qemu/coroutine_int.h b/include/qemu/coroutine_int.h
index XXXXXXX..XXXXXXX 100644
--- a/include/qemu/coroutine_int.h
+++ b/include/qemu/coroutine_int.h
@@ -XXX,XX +XXX,XX @@ struct Coroutine {
     CoroutineEntry *entry;
     void *entry_arg;
     Coroutine *caller;
+
+    /* Only used when the coroutine has terminated.  */
     QSLIST_ENTRY(Coroutine) pool_next;
+
     size_t locks_held;
 
-    /* Coroutines that should be woken up when we yield or terminate */
+    /* Coroutines that should be woken up when we yield or terminate.
+     * Only used when the coroutine is running.
+     */
     QSIMPLEQ_HEAD(, Coroutine) co_queue_wakeup;
+
+    /* Only used when the coroutine has yielded.  */
+    AioContext *ctx;
     QSIMPLEQ_ENTRY(Coroutine) co_queue_next;
+    QSLIST_ENTRY(Coroutine) co_scheduled_next;
 };
 
 Coroutine *qemu_coroutine_new(void);
diff --git a/tests/iothread.h b/tests/iothread.h
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/tests/iothread.h
@@ -XXX,XX +XXX,XX @@
+/*
+ * Event loop thread implementation for unit tests
+ *
+ * Copyright Red Hat Inc., 2013, 2016
+ *
+ * Authors:
+ *  Stefan Hajnoczi   <stefanha@redhat.com>
+ *  Paolo Bonzini     <pbonzini@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+#ifndef TEST_IOTHREAD_H
+#define TEST_IOTHREAD_H
+
+#include "block/aio.h"
+#include "qemu/thread.h"
+
+typedef struct IOThread IOThread;
+
+IOThread *iothread_new(void);
+void iothread_join(IOThread *iothread);
+AioContext *iothread_get_aio_context(IOThread *iothread);
+
+#endif
diff --git a/tests/iothread.c b/tests/iothread.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/tests/iothread.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * Event loop thread implementation for unit tests
+ *
+ * Copyright Red Hat Inc., 2013, 2016
+ *
+ * Authors:
+ *  Stefan Hajnoczi   <stefanha@redhat.com>
+ *  Paolo Bonzini     <pbonzini@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qemu/osdep.h"
+#include "qapi/error.h"
+#include "block/aio.h"
+#include "qemu/main-loop.h"
+#include "qemu/rcu.h"
+#include "iothread.h"
+
+struct IOThread {
+    AioContext *ctx;
+
+    QemuThread thread;
+    QemuMutex init_done_lock;
+    QemuCond init_done_cond;    /* is thread initialization done? */
+    bool stopping;
+};
+
+static __thread IOThread *my_iothread;
+
+AioContext *qemu_get_current_aio_context(void)
+{
+    return my_iothread ? my_iothread->ctx : qemu_get_aio_context();
+}
+
+static void *iothread_run(void *opaque)
+{
+    IOThread *iothread = opaque;
+
+    rcu_register_thread();
+
+    my_iothread = iothread;
+    qemu_mutex_lock(&iothread->init_done_lock);
+    iothread->ctx = aio_context_new(&error_abort);
+    qemu_cond_signal(&iothread->init_done_cond);
+    qemu_mutex_unlock(&iothread->init_done_lock);
+
+    while (!atomic_read(&iothread->stopping)) {
+        aio_poll(iothread->ctx, true);
+    }
+
+    rcu_unregister_thread();
+    return NULL;
+}
+
+void iothread_join(IOThread *iothread)
+{
+    iothread->stopping = true;
+    aio_notify(iothread->ctx);
+    qemu_thread_join(&iothread->thread);
+    qemu_cond_destroy(&iothread->init_done_cond);
+    qemu_mutex_destroy(&iothread->init_done_lock);
+    aio_context_unref(iothread->ctx);
+    g_free(iothread);
+}
+
+IOThread *iothread_new(void)
+{
+    IOThread *iothread = g_new0(IOThread, 1);
+
+    qemu_mutex_init(&iothread->init_done_lock);
+    qemu_cond_init(&iothread->init_done_cond);
+    qemu_thread_create(&iothread->thread, NULL, iothread_run,
+                       iothread, QEMU_THREAD_JOINABLE);
+
+    /* Wait for initialization to complete */
+    qemu_mutex_lock(&iothread->init_done_lock);
+    while (iothread->ctx == NULL) {
+        qemu_cond_wait(&iothread->init_done_cond,
+                       &iothread->init_done_lock);
+    }
+    qemu_mutex_unlock(&iothread->init_done_lock);
+    return iothread;
+}
+
+AioContext *iothread_get_aio_context(IOThread *iothread)
+{
+    return iothread->ctx;
+}
diff --git a/tests/test-aio-multithread.c b/tests/test-aio-multithread.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/tests/test-aio-multithread.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * AioContext multithreading tests
+ *
+ * Copyright Red Hat, Inc. 2016
+ *
+ * Authors:
+ *  Paolo Bonzini    <pbonzini@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2 or later.
+ * See the COPYING.LIB file in the top-level directory.
+ */
+
+#include "qemu/osdep.h"
+#include <glib.h>
+#include "block/aio.h"
+#include "qapi/error.h"
+#include "qemu/coroutine.h"
+#include "qemu/thread.h"
+#include "qemu/error-report.h"
+#include "iothread.h"
+
+/* AioContext management */
+
+#define NUM_CONTEXTS 5
+
+static IOThread *threads[NUM_CONTEXTS];
+static AioContext *ctx[NUM_CONTEXTS];
+static __thread int id = -1;
+
+static QemuEvent done_event;
+
+/* Run a function synchronously on a remote iothread. */
+
+typedef struct CtxRunData {
+    QEMUBHFunc *cb;
+    void *arg;
+} CtxRunData;
+
+static void ctx_run_bh_cb(void *opaque)
+{
+    CtxRunData *data = opaque;
+
+    data->cb(data->arg);
+    qemu_event_set(&done_event);
+}
+
+static void ctx_run(int i, QEMUBHFunc *cb, void *opaque)
+{
+    CtxRunData data = {
+        .cb = cb,
+        .arg = opaque
+    };
+
+    qemu_event_reset(&done_event);
+    aio_bh_schedule_oneshot(ctx[i], ctx_run_bh_cb, &data);
+    qemu_event_wait(&done_event);
+}
+
+/* Starting the iothreads. */
+
+static void set_id_cb(void *opaque)
+{
+    int *i = opaque;
+
+    id = *i;
+}
+
+static void create_aio_contexts(void)
+{
+    int i;
+
+    for (i = 0; i < NUM_CONTEXTS; i++) {
+        threads[i] = iothread_new();
+        ctx[i] = iothread_get_aio_context(threads[i]);
+    }
+
+    qemu_event_init(&done_event, false);
+    for (i = 0; i < NUM_CONTEXTS; i++) {
+        ctx_run(i, set_id_cb, &i);
+    }
+}
+
+/* Stopping the iothreads. */
+
+static void join_aio_contexts(void)
+{
+    int i;
+
+    for (i = 0; i < NUM_CONTEXTS; i++) {
+        aio_context_ref(ctx[i]);
+    }
+    for (i = 0; i < NUM_CONTEXTS; i++) {
+        iothread_join(threads[i]);
+    }
+    for (i = 0; i < NUM_CONTEXTS; i++) {
+        aio_context_unref(ctx[i]);
+    }
+    qemu_event_destroy(&done_event);
+}
+
+/* Basic test for the stuff above. */
+
+static void test_lifecycle(void)
+{
+    create_aio_contexts();
+    join_aio_contexts();
+}
+
+/* aio_co_schedule test.  */
+
+static Coroutine *to_schedule[NUM_CONTEXTS];
+
+static bool now_stopping;
+
+static int count_retry;
+static int count_here;
+static int count_other;
+
+static bool schedule_next(int n)
+{
+    Coroutine *co;
+
+    co = atomic_xchg(&to_schedule[n], NULL);
+    if (!co) {
+        atomic_inc(&count_retry);
+        return false;
+    }
+
+    if (n == id) {
+        atomic_inc(&count_here);
+    } else {
+        atomic_inc(&count_other);
+    }
+
+    aio_co_schedule(ctx[n], co);
+    return true;
+}
+
+static void finish_cb(void *opaque)
+{
+    schedule_next(id);
+}
+
+static coroutine_fn void test_multi_co_schedule_entry(void *opaque)
+{
+    g_assert(to_schedule[id] == NULL);
+    atomic_mb_set(&to_schedule[id], qemu_coroutine_self());
+
+    while (!atomic_mb_read(&now_stopping)) {
+        int n;
+
+        n = g_test_rand_int_range(0, NUM_CONTEXTS);
+        schedule_next(n);
+        qemu_coroutine_yield();
+
+        g_assert(to_schedule[id] == NULL);
+        atomic_mb_set(&to_schedule[id], qemu_coroutine_self());
+    }
+}
+
+
+static void test_multi_co_schedule(int seconds)
+{
+    int i;
+
+    count_here = count_other = count_retry = 0;
+    now_stopping = false;
+
+    create_aio_contexts();
+    for (i = 0; i < NUM_CONTEXTS; i++) {
+        Coroutine *co1 = qemu_coroutine_create(test_multi_co_schedule_entry, NULL);
+        aio_co_schedule(ctx[i], co1);
+    }
+
+    g_usleep(seconds * 1000000);
+
+    atomic_mb_set(&now_stopping, true);
+    for (i = 0; i < NUM_CONTEXTS; i++) {
+        ctx_run(i, finish_cb, NULL);
+        to_schedule[i] = NULL;
+    }
+
+    join_aio_contexts();
+    g_test_message("scheduled %d, queued %d, retry %d, total %d\n",
+                  count_other, count_here, count_retry,
+                  count_here + count_other + count_retry);
+}
+
+static void test_multi_co_schedule_1(void)
+{
+    test_multi_co_schedule(1);
+}
+
+static void test_multi_co_schedule_10(void)
+{
+    test_multi_co_schedule(10);
+}
+
+/* End of tests.  */
+
+int main(int argc, char **argv)
+{
+    init_clocks();
+
+    g_test_init(&argc, &argv, NULL);
+    g_test_add_func("/aio/multi/lifecycle", test_lifecycle);
+    if (g_test_quick()) {
+        g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_1);
+    } else {
+        g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_10);
+    }
+    return g_test_run();
+}
diff --git a/util/async.c b/util/async.c
index XXXXXXX..XXXXXXX 100644
--- a/util/async.c
+++ b/util/async.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/main-loop.h"
 #include "qemu/atomic.h"
 #include "block/raw-aio.h"
+#include "qemu/coroutine_int.h"
+#include "trace.h"
 
 /***********************************************************/
 /* bottom halves (can be seen as timers which expire ASAP) */
@@ -XXX,XX +XXX,XX @@ aio_ctx_finalize(GSource     *source)
     }
 #endif
 
+    assert(QSLIST_EMPTY(&ctx->scheduled_coroutines));
+    qemu_bh_delete(ctx->co_schedule_bh);
+
     qemu_lockcnt_lock(&ctx->list_lock);
     assert(!qemu_lockcnt_count(&ctx->list_lock));
     while (ctx->first_bh) {
@@ -XXX,XX +XXX,XX @@ static bool event_notifier_poll(void *opaque)
     return atomic_read(&ctx->notified);
 }
 
+static void co_schedule_bh_cb(void *opaque)
+{
+    AioContext *ctx = opaque;
+    QSLIST_HEAD(, Coroutine) straight, reversed;
+
+    QSLIST_MOVE_ATOMIC(&reversed, &ctx->scheduled_coroutines);
+    QSLIST_INIT(&straight);
+
+    while (!QSLIST_EMPTY(&reversed)) {
+        Coroutine *co = QSLIST_FIRST(&reversed);
+        QSLIST_REMOVE_HEAD(&reversed, co_scheduled_next);
+        QSLIST_INSERT_HEAD(&straight, co, co_scheduled_next);
+    }
+
+    while (!QSLIST_EMPTY(&straight)) {
+        Coroutine *co = QSLIST_FIRST(&straight);
+        QSLIST_REMOVE_HEAD(&straight, co_scheduled_next);
+        trace_aio_co_schedule_bh_cb(ctx, co);
+        qemu_coroutine_enter(co);
+    }
+}
+
 AioContext *aio_context_new(Error **errp)
 {
     int ret;
@@ -XXX,XX +XXX,XX @@ AioContext *aio_context_new(Error **errp)
     }
     g_source_set_can_recurse(&ctx->source, true);
     qemu_lockcnt_init(&ctx->list_lock);
+
+    ctx->co_schedule_bh = aio_bh_new(ctx, co_schedule_bh_cb, ctx);
+    QSLIST_INIT(&ctx->scheduled_coroutines);
+
     aio_set_event_notifier(ctx, &ctx->notifier,
                            false,
                            (EventNotifierHandler *)
@@ -XXX,XX +XXX,XX @@ fail:
     return NULL;
 }
 
+void aio_co_schedule(AioContext *ctx, Coroutine *co)
+{
+    trace_aio_co_schedule(ctx, co);
+    QSLIST_INSERT_HEAD_ATOMIC(&ctx->scheduled_coroutines,
+                              co, co_scheduled_next);
+    qemu_bh_schedule(ctx->co_schedule_bh);
+}
+
+void aio_co_wake(struct Coroutine *co)
+{
+    AioContext *ctx;
+
+    /* Read coroutine before co->ctx.  Matches smp_wmb in
+     * qemu_coroutine_enter.
+     */
+    smp_read_barrier_depends();
+    ctx = atomic_read(&co->ctx);
+
+    if (ctx != qemu_get_current_aio_context()) {
+        aio_co_schedule(ctx, co);
+        return;
+    }
+
+    if (qemu_in_coroutine()) {
+        Coroutine *self = qemu_coroutine_self();
+        assert(self != co);
+        QSIMPLEQ_INSERT_TAIL(&self->co_queue_wakeup, co, co_queue_next);
+    } else {
+        aio_context_acquire(ctx);
+        qemu_coroutine_enter(co);
+        aio_context_release(ctx);
+    }
+}
+
 void aio_context_ref(AioContext *ctx)
 {
     g_source_ref(&ctx->source);
diff --git a/util/qemu-coroutine.c b/util/qemu-coroutine.c
index XXXXXXX..XXXXXXX 100644
--- a/util/qemu-coroutine.c
+++ b/util/qemu-coroutine.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/atomic.h"
 #include "qemu/coroutine.h"
 #include "qemu/coroutine_int.h"
+#include "block/aio.h"
 
 enum {
     POOL_BATCH_SIZE = 64,
@@ -XXX,XX +XXX,XX @@ void qemu_coroutine_enter(Coroutine *co)
     }
 
     co->caller = self;
+    co->ctx = qemu_get_current_aio_context();
+
+    /* Store co->ctx before anything that stores co.  Matches
+     * barrier in aio_co_wake.
+     */
+    smp_wmb();
+
     ret = qemu_coroutine_switch(self, co, COROUTINE_ENTER);
 
     qemu_co_queue_run_restart(co);
diff --git a/util/trace-events b/util/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/util/trace-events
+++ b/util/trace-events
@@ -XXX,XX +XXX,XX @@ run_poll_handlers_end(void *ctx, bool progress) "ctx %p progress %d"
 poll_shrink(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
 poll_grow(void *ctx, int64_t old, int64_t new) "ctx %p old %"PRId64" new %"PRId64
 
+# util/async.c
+aio_co_schedule(void *ctx, void *co) "ctx %p co %p"
+aio_co_schedule_bh_cb(void *ctx, void *co) "ctx %p co %p"
+
 # util/thread-pool.c
 thread_pool_submit(void *pool, void *req, void *opaque) "pool %p req %p opaque %p"
 thread_pool_complete(void *pool, void *req, void *opaque, int ret) "pool %p req %p opaque %p ret %d"
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

qcow2_create2 calls this.  Do not run a nested event loop, as that
breaks when aio_co_wake tries to queue the coroutine on the co_queue_wakeup
list of the currently running one.

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213135235.12274-4-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 block/block-backend.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/block/block-backend.c b/block/block-backend.c
index XXXXXXX..XXXXXXX 100644
--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -XXX,XX +XXX,XX @@ static int blk_prw(BlockBackend *blk, int64_t offset, uint8_t *buf,
 {
     QEMUIOVector qiov;
     struct iovec iov;
-    Coroutine *co;
     BlkRwCo rwco;
 
     iov = (struct iovec) {
@@ -XXX,XX +XXX,XX @@ static int blk_prw(BlockBackend *blk, int64_t offset, uint8_t *buf,
         .ret    = NOT_DONE,
     };
 
-    co = qemu_coroutine_create(co_entry, &rwco);
-    qemu_coroutine_enter(co);
-    BDRV_POLL_WHILE(blk_bs(blk), rwco.ret == NOT_DONE);
+    if (qemu_in_coroutine()) {
+        /* Fast-path if already in coroutine context */
+        co_entry(&rwco);
+    } else {
+        Coroutine *co = qemu_coroutine_create(co_entry, &rwco);
+        qemu_coroutine_enter(co);
+        BDRV_POLL_WHILE(blk_bs(blk), rwco.ret == NOT_DONE);
+    }
 
     return rwco.ret;
 }
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Once the thread pool starts using aio_co_wake, it will also need
qemu_get_current_aio_context().  Make test-thread-pool create
an AioContext with qemu_init_main_loop, so that stubs/iothread.c
and tests/iothread.c can provide the rest.

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213135235.12274-5-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 tests/test-thread-pool.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/tests/test-thread-pool.c b/tests/test-thread-pool.c
index XXXXXXX..XXXXXXX 100644
--- a/tests/test-thread-pool.c
+++ b/tests/test-thread-pool.c
@@ -XXX,XX +XXX,XX @@
 #include "qapi/error.h"
 #include "qemu/timer.h"
 #include "qemu/error-report.h"
+#include "qemu/main-loop.h"
 
 static AioContext *ctx;
 static ThreadPool *pool;
@@ -XXX,XX +XXX,XX @@ static void test_cancel_async(void)
 int main(int argc, char **argv)
 {
     int ret;
-    Error *local_error = NULL;
 
-    init_clocks();
-
-    ctx = aio_context_new(&local_error);
-    if (!ctx) {
-        error_reportf_err(local_error, "Failed to create AIO Context: ");
-        exit(1);
-    }
+    qemu_init_main_loop(&error_abort);
+    ctx = qemu_get_current_aio_context();
     pool = aio_get_thread_pool(ctx);
 
     g_test_init(&argc, &argv, NULL);
@@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv)
 
     ret = g_test_run();
 
-    aio_context_unref(ctx);
     return ret;
 }
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

This is in preparation for making qio_channel_yield work on
AioContexts other than the main one.

Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213135235.12274-6-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 include/io/channel.h | 25 +++++++++++++++++++++++++
 io/channel-command.c | 13 +++++++++++++
 io/channel-file.c    | 11 +++++++++++
 io/channel-socket.c  | 16 +++++++++++-----
 io/channel-tls.c     | 12 ++++++++++++
 io/channel-watch.c   |  6 ++++++
 io/channel.c         | 11 +++++++++++
 7 files changed, 89 insertions(+), 5 deletions(-)

diff --git a/include/io/channel.h b/include/io/channel.h
index XXXXXXX..XXXXXXX 100644
--- a/include/io/channel.h
+++ b/include/io/channel.h
@@ -XXX,XX +XXX,XX @@
 
 #include "qemu-common.h"
 #include "qom/object.h"
+#include "block/aio.h"
 
 #define TYPE_QIO_CHANNEL "qio-channel"
 #define QIO_CHANNEL(obj)                                    \
@@ -XXX,XX +XXX,XX @@ struct QIOChannelClass {
                      off_t offset,
                      int whence,
                      Error **errp);
+    void (*io_set_aio_fd_handler)(QIOChannel *ioc,
+                                  AioContext *ctx,
+                                  IOHandler *io_read,
+                                  IOHandler *io_write,
+                                  void *opaque);
 };
 
 /* General I/O handling functions */
@@ -XXX,XX +XXX,XX @@ void qio_channel_yield(QIOChannel *ioc,
 void qio_channel_wait(QIOChannel *ioc,
                       GIOCondition condition);
 
+/**
+ * qio_channel_set_aio_fd_handler:
+ * @ioc: the channel object
+ * @ctx: the AioContext to set the handlers on
+ * @io_read: the read handler
+ * @io_write: the write handler
+ * @opaque: the opaque value passed to the handler
+ *
+ * This is used internally by qio_channel_yield().  It can
+ * be used by channel implementations to forward the handlers
+ * to another channel (e.g. from #QIOChannelTLS to the
+ * underlying socket).
+ */
+void qio_channel_set_aio_fd_handler(QIOChannel *ioc,
+                                    AioContext *ctx,
+                                    IOHandler *io_read,
+                                    IOHandler *io_write,
+                                    void *opaque);
+
 #endif /* QIO_CHANNEL_H */
diff --git a/io/channel-command.c b/io/channel-command.c
index XXXXXXX..XXXXXXX 100644
--- a/io/channel-command.c
+++ b/io/channel-command.c
@@ -XXX,XX +XXX,XX @@ static int qio_channel_command_close(QIOChannel *ioc,
 }
 
 
+static void qio_channel_command_set_aio_fd_handler(QIOChannel *ioc,
+                                                   AioContext *ctx,
+                                                   IOHandler *io_read,
+                                                   IOHandler *io_write,
+                                                   void *opaque)
+{
+    QIOChannelCommand *cioc = QIO_CHANNEL_COMMAND(ioc);
+    aio_set_fd_handler(ctx, cioc->readfd, false, io_read, NULL, NULL, opaque);
+    aio_set_fd_handler(ctx, cioc->writefd, false, NULL, io_write, NULL, opaque);
+}
+
+
 static GSource *qio_channel_command_create_watch(QIOChannel *ioc,
                                                  GIOCondition condition)
 {
@@ -XXX,XX +XXX,XX @@ static void qio_channel_command_class_init(ObjectClass *klass,
     ioc_klass->io_set_blocking = qio_channel_command_set_blocking;
     ioc_klass->io_close = qio_channel_command_close;
     ioc_klass->io_create_watch = qio_channel_command_create_watch;
+    ioc_klass->io_set_aio_fd_handler = qio_channel_command_set_aio_fd_handler;
 }
 
 static const TypeInfo qio_channel_command_info = {
diff --git a/io/channel-file.c b/io/channel-file.c
index XXXXXXX..XXXXXXX 100644
--- a/io/channel-file.c
+++ b/io/channel-file.c
@@ -XXX,XX +XXX,XX @@ static int qio_channel_file_close(QIOChannel *ioc,
 }
 
 
+static void qio_channel_file_set_aio_fd_handler(QIOChannel *ioc,
+                                                AioContext *ctx,
+                                                IOHandler *io_read,
+                                                IOHandler *io_write,
+                                                void *opaque)
+{
+    QIOChannelFile *fioc = QIO_CHANNEL_FILE(ioc);
+    aio_set_fd_handler(ctx, fioc->fd, false, io_read, io_write, NULL, opaque);
+}
+
 static GSource *qio_channel_file_create_watch(QIOChannel *ioc,
                                               GIOCondition condition)
 {
@@ -XXX,XX +XXX,XX @@ static void qio_channel_file_class_init(ObjectClass *klass,
     ioc_klass->io_seek = qio_channel_file_seek;
     ioc_klass->io_close = qio_channel_file_close;
     ioc_klass->io_create_watch = qio_channel_file_create_watch;
+    ioc_klass->io_set_aio_fd_handler = qio_channel_file_set_aio_fd_handler;
 }
 
 static const TypeInfo qio_channel_file_info = {
diff --git a/io/channel-socket.c b/io/channel-socket.c
index XXXXXXX..XXXXXXX 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -XXX,XX +XXX,XX @@ qio_channel_socket_set_blocking(QIOChannel *ioc,
         qemu_set_block(sioc->fd);
     } else {
         qemu_set_nonblock(sioc->fd);
-#ifdef WIN32
-        WSAEventSelect(sioc->fd, ioc->event,
-                       FD_READ | FD_ACCEPT | FD_CLOSE |
-                       FD_CONNECT | FD_WRITE | FD_OOB);
-#endif
     }
     return 0;
 }
@@ -XXX,XX +XXX,XX @@ qio_channel_socket_shutdown(QIOChannel *ioc,
     return 0;
 }
 
+static void qio_channel_socket_set_aio_fd_handler(QIOChannel *ioc,
+                                                  AioContext *ctx,
+                                                  IOHandler *io_read,
+                                                  IOHandler *io_write,
+                                                  void *opaque)
+{
+    QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
+    aio_set_fd_handler(ctx, sioc->fd, false, io_read, io_write, NULL, opaque);
+}
+
 static GSource *qio_channel_socket_create_watch(QIOChannel *ioc,
                                                 GIOCondition condition)
 {
@@ -XXX,XX +XXX,XX @@ static void qio_channel_socket_class_init(ObjectClass *klass,
     ioc_klass->io_set_cork = qio_channel_socket_set_cork;
     ioc_klass->io_set_delay = qio_channel_socket_set_delay;
     ioc_klass->io_create_watch = qio_channel_socket_create_watch;
+    ioc_klass->io_set_aio_fd_handler = qio_channel_socket_set_aio_fd_handler;
 }
 
 static const TypeInfo qio_channel_socket_info = {
diff --git a/io/channel-tls.c b/io/channel-tls.c
index XXXXXXX..XXXXXXX 100644
--- a/io/channel-tls.c
+++ b/io/channel-tls.c
@@ -XXX,XX +XXX,XX @@ static int qio_channel_tls_close(QIOChannel *ioc,
     return qio_channel_close(tioc->master, errp);
 }
 
+static void qio_channel_tls_set_aio_fd_handler(QIOChannel *ioc,
+                                               AioContext *ctx,
+                                               IOHandler *io_read,
+                                               IOHandler *io_write,
+                                               void *opaque)
+{
+    QIOChannelTLS *tioc = QIO_CHANNEL_TLS(ioc);
+
+    qio_channel_set_aio_fd_handler(tioc->master, ctx, io_read, io_write, opaque);
+}
+
 static GSource *qio_channel_tls_create_watch(QIOChannel *ioc,
                                              GIOCondition condition)
 {
@@ -XXX,XX +XXX,XX @@ static void qio_channel_tls_class_init(ObjectClass *klass,
     ioc_klass->io_close = qio_channel_tls_close;
     ioc_klass->io_shutdown = qio_channel_tls_shutdown;
     ioc_klass->io_create_watch = qio_channel_tls_create_watch;
+    ioc_klass->io_set_aio_fd_handler = qio_channel_tls_set_aio_fd_handler;
 }
 
 static const TypeInfo qio_channel_tls_info = {
diff --git a/io/channel-watch.c b/io/channel-watch.c
index XXXXXXX..XXXXXXX 100644
--- a/io/channel-watch.c
+++ b/io/channel-watch.c
@@ -XXX,XX +XXX,XX @@ GSource *qio_channel_create_socket_watch(QIOChannel *ioc,
     GSource *source;
     QIOChannelSocketSource *ssource;
 
+#ifdef WIN32
+    WSAEventSelect(socket, ioc->event,
+                   FD_READ | FD_ACCEPT | FD_CLOSE |
+                   FD_CONNECT | FD_WRITE | FD_OOB);
+#endif
+
     source = g_source_new(&qio_channel_socket_source_funcs,
                           sizeof(QIOChannelSocketSource));
     ssource = (QIOChannelSocketSource *)source;
diff --git a/io/channel.c b/io/channel.c
index XXXXXXX..XXXXXXX 100644
--- a/io/channel.c
+++ b/io/channel.c
@@ -XXX,XX +XXX,XX @@ GSource *qio_channel_create_watch(QIOChannel *ioc,
 }
 
 
+void qio_channel_set_aio_fd_handler(QIOChannel *ioc,
+                                    AioContext *ctx,
+                                    IOHandler *io_read,
+                                    IOHandler *io_write,
+                                    void *opaque)
+{
+    QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
+
+    klass->io_set_aio_fd_handler(ioc, ctx, io_read, io_write, opaque);
+}
+
 guint qio_channel_add_watch(QIOChannel *ioc,
                             GIOCondition condition,
                             QIOChannelFunc func,
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Support separate coroutines for reading and writing, and place the
read/write handlers on the AioContext that the QIOChannel is registered
with.

Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213135235.12274-7-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 include/io/channel.h | 47 ++++++++++++++++++++++++++--
 io/channel.c         | 86 +++++++++++++++++++++++++++++++++++++++-------------
 2 files changed, 109 insertions(+), 24 deletions(-)

diff --git a/include/io/channel.h b/include/io/channel.h
index XXXXXXX..XXXXXXX 100644
--- a/include/io/channel.h
+++ b/include/io/channel.h
@@ -XXX,XX +XXX,XX @@
 
 #include "qemu-common.h"
 #include "qom/object.h"
+#include "qemu/coroutine.h"
 #include "block/aio.h"
 
 #define TYPE_QIO_CHANNEL "qio-channel"
@@ -XXX,XX +XXX,XX @@ struct QIOChannel {
     Object parent;
     unsigned int features; /* bitmask of QIOChannelFeatures */
     char *name;
+    AioContext *ctx;
+    Coroutine *read_coroutine;
+    Coroutine *write_coroutine;
 #ifdef _WIN32
     HANDLE event; /* For use with GSource on Win32 */
 #endif
@@ -XXX,XX +XXX,XX @@ guint qio_channel_add_watch(QIOChannel *ioc,
 
 
 /**
+ * qio_channel_attach_aio_context:
+ * @ioc: the channel object
+ * @ctx: the #AioContext to set the handlers on
+ *
+ * Request that qio_channel_yield() sets I/O handlers on
+ * the given #AioContext.  If @ctx is %NULL, qio_channel_yield()
+ * uses QEMU's main thread event loop.
+ *
+ * You can move a #QIOChannel from one #AioContext to another even if
+ * I/O handlers are set for a coroutine.  However, #QIOChannel provides
+ * no synchronization between the calls to qio_channel_yield() and
+ * qio_channel_attach_aio_context().
+ *
+ * Therefore you should first call qio_channel_detach_aio_context()
+ * to ensure that the coroutine is not entered concurrently.  Then,
+ * while the coroutine has yielded, call qio_channel_attach_aio_context(),
+ * and then aio_co_schedule() to place the coroutine on the new
+ * #AioContext.  The calls to qio_channel_detach_aio_context()
+ * and qio_channel_attach_aio_context() should be protected with
+ * aio_context_acquire() and aio_context_release().
+ */
+void qio_channel_attach_aio_context(QIOChannel *ioc,
+                                    AioContext *ctx);
+
+/**
+ * qio_channel_detach_aio_context:
+ * @ioc: the channel object
+ *
+ * Disable any I/O handlers set by qio_channel_yield().  With the
+ * help of aio_co_schedule(), this allows moving a coroutine that was
+ * paused by qio_channel_yield() to another context.
+ */
+void qio_channel_detach_aio_context(QIOChannel *ioc);
+
+/**
  * qio_channel_yield:
  * @ioc: the channel object
  * @condition: the I/O condition to wait for
  *
- * Yields execution from the current coroutine until
- * the condition indicated by @condition becomes
- * available.
+ * Yields execution from the current coroutine until the condition
+ * indicated by @condition becomes available.  @condition must
+ * be either %G_IO_IN or %G_IO_OUT; it cannot contain both.  In
+ * addition, no two coroutine can be waiting on the same condition
+ * and channel at the same time.
  *
  * This must only be called from coroutine context
  */
diff --git a/io/channel.c b/io/channel.c
index XXXXXXX..XXXXXXX 100644
--- a/io/channel.c
+++ b/io/channel.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/osdep.h"
 #include "io/channel.h"
 #include "qapi/error.h"
-#include "qemu/coroutine.h"
+#include "qemu/main-loop.h"
 
 bool qio_channel_has_feature(QIOChannel *ioc,
                              QIOChannelFeature feature)
@@ -XXX,XX +XXX,XX @@ off_t qio_channel_io_seek(QIOChannel *ioc,
 }
 
 
-typedef struct QIOChannelYieldData QIOChannelYieldData;
-struct QIOChannelYieldData {
-    QIOChannel *ioc;
-    Coroutine *co;
-};
+static void qio_channel_set_aio_fd_handlers(QIOChannel *ioc);
 
+static void qio_channel_restart_read(void *opaque)
+{
+    QIOChannel *ioc = opaque;
+    Coroutine *co = ioc->read_coroutine;
+
+    ioc->read_coroutine = NULL;
+    qio_channel_set_aio_fd_handlers(ioc);
+    aio_co_wake(co);
+}
 
-static gboolean qio_channel_yield_enter(QIOChannel *ioc,
-                                        GIOCondition condition,
-                                        gpointer opaque)
+static void qio_channel_restart_write(void *opaque)
 {
-    QIOChannelYieldData *data = opaque;
-    qemu_coroutine_enter(data->co);
-    return FALSE;
+    QIOChannel *ioc = opaque;
+    Coroutine *co = ioc->write_coroutine;
+
+    ioc->write_coroutine = NULL;
+    qio_channel_set_aio_fd_handlers(ioc);
+    aio_co_wake(co);
 }
 
+static void qio_channel_set_aio_fd_handlers(QIOChannel *ioc)
+{
+    IOHandler *rd_handler = NULL, *wr_handler = NULL;
+    AioContext *ctx;
+
+    if (ioc->read_coroutine) {
+        rd_handler = qio_channel_restart_read;
+    }
+    if (ioc->write_coroutine) {
+        wr_handler = qio_channel_restart_write;
+    }
+
+    ctx = ioc->ctx ? ioc->ctx : iohandler_get_aio_context();
+    qio_channel_set_aio_fd_handler(ioc, ctx, rd_handler, wr_handler, ioc);
+}
+
+void qio_channel_attach_aio_context(QIOChannel *ioc,
+                                    AioContext *ctx)
+{
+    AioContext *old_ctx;
+    if (ioc->ctx == ctx) {
+        return;
+    }
+
+    old_ctx = ioc->ctx ? ioc->ctx : iohandler_get_aio_context();
+    qio_channel_set_aio_fd_handler(ioc, old_ctx, NULL, NULL, NULL);
+    ioc->ctx = ctx;
+    qio_channel_set_aio_fd_handlers(ioc);
+}
+
+void qio_channel_detach_aio_context(QIOChannel *ioc)
+{
+    ioc->read_coroutine = NULL;
+    ioc->write_coroutine = NULL;
+    qio_channel_set_aio_fd_handlers(ioc);
+    ioc->ctx = NULL;
+}
 
 void coroutine_fn qio_channel_yield(QIOChannel *ioc,
                                     GIOCondition condition)
 {
-    QIOChannelYieldData data;
-
     assert(qemu_in_coroutine());
-    data.ioc = ioc;
-    data.co = qemu_coroutine_self();
-    qio_channel_add_watch(ioc,
-                          condition,
-                          qio_channel_yield_enter,
-                          &data,
-                          NULL);
+    if (condition == G_IO_IN) {
+        assert(!ioc->read_coroutine);
+        ioc->read_coroutine = qemu_coroutine_self();
+    } else if (condition == G_IO_OUT) {
+        assert(!ioc->write_coroutine);
+        ioc->write_coroutine = qemu_coroutine_self();
+    } else {
+        abort();
+    }
+    qio_channel_set_aio_fd_handlers(ioc);
     qemu_coroutine_yield();
 }
 
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

In the client, read the reply headers from a coroutine, switching the
read side between the "read header" coroutine and the I/O coroutine that
reads the body of the reply.

In the server, if the server can read more requests it will create a new
"read request" coroutine as soon as a request has been read.  Otherwise,
the new coroutine is created in nbd_request_put.

diff --git a/block/nbd-client.h b/block/nbd-client.h
index XXXXXXX..XXXXXXX 100644
--- a/block/nbd-client.h
+++ b/block/nbd-client.h
@@ -XXX,XX +XXX,XX @@ typedef struct NBDClientSession {
 
     CoMutex send_mutex;
     CoQueue free_sema;
-    Coroutine *send_coroutine;
+    Coroutine *read_reply_co;
     int in_flight;
 
     Coroutine *recv_coroutine[MAX_NBD_REQUESTS];
diff --git a/block/nbd-client.c b/block/nbd-client.c
index XXXXXXX..XXXXXXX 100644
--- a/block/nbd-client.c
+++ b/block/nbd-client.c
@@ -XXX,XX +XXX,XX @@
 #define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs))
 #define INDEX_TO_HANDLE(bs, index)  ((index)  ^ ((uint64_t)(intptr_t)bs))
 
-static void nbd_recv_coroutines_enter_all(NBDClientSession *s)
+static void nbd_recv_coroutines_enter_all(BlockDriverState *bs)
 {
+    NBDClientSession *s = nbd_get_client_session(bs);
     int i;
 
     for (i = 0; i < MAX_NBD_REQUESTS; i++) {
@@ -XXX,XX +XXX,XX @@ static void nbd_recv_coroutines_enter_all(NBDClientSession *s)
             qemu_coroutine_enter(s->recv_coroutine[i]);
         }
     }
+    BDRV_POLL_WHILE(bs, s->read_reply_co);
 }
 
 static void nbd_teardown_connection(BlockDriverState *bs)
@@ -XXX,XX +XXX,XX @@ static void nbd_teardown_connection(BlockDriverState *bs)
     qio_channel_shutdown(client->ioc,
                          QIO_CHANNEL_SHUTDOWN_BOTH,
                          NULL);
-    nbd_recv_coroutines_enter_all(client);
+    nbd_recv_coroutines_enter_all(bs);
 
     nbd_client_detach_aio_context(bs);
     object_unref(OBJECT(client->sioc));
@@ -XXX,XX +XXX,XX @@ static void nbd_teardown_connection(BlockDriverState *bs)
     client->ioc = NULL;
 }
 
-static void nbd_reply_ready(void *opaque)
+static coroutine_fn void nbd_read_reply_entry(void *opaque)
 {
-    BlockDriverState *bs = opaque;
-    NBDClientSession *s = nbd_get_client_session(bs);
+    NBDClientSession *s = opaque;
     uint64_t i;
     int ret;
 
-    if (!s->ioc) { /* Already closed */
-        return;
-    }
-
-    if (s->reply.handle == 0) {
-        /* No reply already in flight.  Fetch a header.  It is possible
-         * that another thread has done the same thing in parallel, so
-         * the socket is not readable anymore.
-         */
+    for (;;) {
+        assert(s->reply.handle == 0);
         ret = nbd_receive_reply(s->ioc, &s->reply);
-        if (ret == -EAGAIN) {
-            return;
-        }
         if (ret < 0) {
-            s->reply.handle = 0;
-            goto fail;
+            break;
         }
-    }
 
-    /* There's no need for a mutex on the receive side, because the
-     * handler acts as a synchronization point and ensures that only
-     * one coroutine is called until the reply finishes.  */
-    i = HANDLE_TO_INDEX(s, s->reply.handle);
-    if (i >= MAX_NBD_REQUESTS) {
-        goto fail;
-    }
+        /* There's no need for a mutex on the receive side, because the
+         * handler acts as a synchronization point and ensures that only
+         * one coroutine is called until the reply finishes.
+         */
+        i = HANDLE_TO_INDEX(s, s->reply.handle);
+        if (i >= MAX_NBD_REQUESTS || !s->recv_coroutine[i]) {
+            break;
+        }
 
-    if (s->recv_coroutine[i]) {
-        qemu_coroutine_enter(s->recv_coroutine[i]);
-        return;
+        /* We're woken up by the recv_coroutine itself.  Note that there
+         * is no race between yielding and reentering read_reply_co.  This
+         * is because:
+         *
+         * - if recv_coroutine[i] runs on the same AioContext, it is only
+         *   entered after we yield
+         *
+         * - if recv_coroutine[i] runs on a different AioContext, reentering
+         *   read_reply_co happens through a bottom half, which can only
+         *   run after we yield.
+         */
+        aio_co_wake(s->recv_coroutine[i]);
+        qemu_coroutine_yield();
     }
-
-fail:
-    nbd_teardown_connection(bs);
-}
-
-static void nbd_restart_write(void *opaque)
-{
-    BlockDriverState *bs = opaque;
-
-    qemu_coroutine_enter(nbd_get_client_session(bs)->send_coroutine);
+    s->read_reply_co = NULL;
 }
 
 static int nbd_co_send_request(BlockDriverState *bs,
@@ -XXX,XX +XXX,XX @@ static int nbd_co_send_request(BlockDriverState *bs,
                                QEMUIOVector *qiov)
 {
     NBDClientSession *s = nbd_get_client_session(bs);
-    AioContext *aio_context;
     int rc, ret, i;
 
     qemu_co_mutex_lock(&s->send_mutex);
@@ -XXX,XX +XXX,XX @@ static int nbd_co_send_request(BlockDriverState *bs,
         return -EPIPE;
     }
 
-    s->send_coroutine = qemu_coroutine_self();
-    aio_context = bdrv_get_aio_context(bs);
-
-    aio_set_fd_handler(aio_context, s->sioc->fd, false,
-                       nbd_reply_ready, nbd_restart_write, NULL, bs);
     if (qiov) {
         qio_channel_set_cork(s->ioc, true);
         rc = nbd_send_request(s->ioc, request);
@@ -XXX,XX +XXX,XX @@ static int nbd_co_send_request(BlockDriverState *bs,
     } else {
         rc = nbd_send_request(s->ioc, request);
     }
-    aio_set_fd_handler(aio_context, s->sioc->fd, false,
-                       nbd_reply_ready, NULL, NULL, bs);
-    s->send_coroutine = NULL;
     qemu_co_mutex_unlock(&s->send_mutex);
     return rc;
 }
@@ -XXX,XX +XXX,XX @@ static void nbd_co_receive_reply(NBDClientSession *s,
 {
     int ret;
 
-    /* Wait until we're woken up by the read handler.  TODO: perhaps
-     * peek at the next reply and avoid yielding if it's ours?  */
+    /* Wait until we're woken up by nbd_read_reply_entry.  */
     qemu_coroutine_yield();
     *reply = s->reply;
     if (reply->handle != request->handle ||
@@ -XXX,XX +XXX,XX @@ static void nbd_coroutine_start(NBDClientSession *s,
     /* s->recv_coroutine[i] is set as soon as we get the send_lock.  */
 }
 
-static void nbd_coroutine_end(NBDClientSession *s,
+static void nbd_coroutine_end(BlockDriverState *bs,
                               NBDRequest *request)
 {
+    NBDClientSession *s = nbd_get_client_session(bs);
     int i = HANDLE_TO_INDEX(s, request->handle);
+
     s->recv_coroutine[i] = NULL;
-    if (s->in_flight-- == MAX_NBD_REQUESTS) {
-        qemu_co_queue_next(&s->free_sema);
+    s->in_flight--;
+    qemu_co_queue_next(&s->free_sema);
+
+    /* Kick the read_reply_co to get the next reply.  */
+    if (s->read_reply_co) {
+        aio_co_wake(s->read_reply_co);
     }
 }
 
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_preadv(BlockDriverState *bs, uint64_t offset,
     } else {
         nbd_co_receive_reply(client, &request, &reply, qiov);
     }
-    nbd_coroutine_end(client, &request);
+    nbd_coroutine_end(bs, &request);
     return -reply.error;
 }
 
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_pwritev(BlockDriverState *bs, uint64_t offset,
     } else {
         nbd_co_receive_reply(client, &request, &reply, NULL);
     }
-    nbd_coroutine_end(client, &request);
+    nbd_coroutine_end(bs, &request);
     return -reply.error;
 }
 
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,
     } else {
         nbd_co_receive_reply(client, &request, &reply, NULL);
     }
-    nbd_coroutine_end(client, &request);
+    nbd_coroutine_end(bs, &request);
     return -reply.error;
 }
 
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_flush(BlockDriverState *bs)
     } else {
         nbd_co_receive_reply(client, &request, &reply, NULL);
     }
-    nbd_coroutine_end(client, &request);
+    nbd_coroutine_end(bs, &request);
     return -reply.error;
 }
 
@@ -XXX,XX +XXX,XX @@ int nbd_client_co_pdiscard(BlockDriverState *bs, int64_t offset, int count)
     } else {
         nbd_co_receive_reply(client, &request, &reply, NULL);
     }
-    nbd_coroutine_end(client, &request);
+    nbd_coroutine_end(bs, &request);
     return -reply.error;
 
 }
 
 void nbd_client_detach_aio_context(BlockDriverState *bs)
 {
-    aio_set_fd_handler(bdrv_get_aio_context(bs),
-                       nbd_get_client_session(bs)->sioc->fd,
-                       false, NULL, NULL, NULL, NULL);
+    NBDClientSession *client = nbd_get_client_session(bs);
+    qio_channel_detach_aio_context(QIO_CHANNEL(client->sioc));
 }
 
 void nbd_client_attach_aio_context(BlockDriverState *bs,
                                    AioContext *new_context)
 {
-    aio_set_fd_handler(new_context, nbd_get_client_session(bs)->sioc->fd,
-                       false, nbd_reply_ready, NULL, NULL, bs);
+    NBDClientSession *client = nbd_get_client_session(bs);
+    qio_channel_attach_aio_context(QIO_CHANNEL(client->sioc), new_context);
+    aio_co_schedule(new_context, client->read_reply_co);
 }
 
 void nbd_client_close(BlockDriverState *bs)
@@ -XXX,XX +XXX,XX @@ int nbd_client_init(BlockDriverState *bs,
     /* Now that we're connected, set the socket to be non-blocking and
      * kick the reply mechanism.  */
     qio_channel_set_blocking(QIO_CHANNEL(sioc), false, NULL);
-
+    client->read_reply_co = qemu_coroutine_create(nbd_read_reply_entry, client);
     nbd_client_attach_aio_context(bs, bdrv_get_aio_context(bs));
 
     logout("Established connection with NBD server\n");
diff --git a/nbd/client.c b/nbd/client.c
index XXXXXXX..XXXXXXX 100644
--- a/nbd/client.c
+++ b/nbd/client.c
@@ -XXX,XX +XXX,XX @@ ssize_t nbd_receive_reply(QIOChannel *ioc, NBDReply *reply)
     ssize_t ret;
 
     ret = read_sync(ioc, buf, sizeof(buf));
-    if (ret < 0) {
+    if (ret <= 0) {
         return ret;
     }
 
diff --git a/nbd/common.c b/nbd/common.c
index XXXXXXX..XXXXXXX 100644
--- a/nbd/common.c
+++ b/nbd/common.c
@@ -XXX,XX +XXX,XX @@ ssize_t nbd_wr_syncv(QIOChannel *ioc,
         }
         if (len == QIO_CHANNEL_ERR_BLOCK) {
             if (qemu_in_coroutine()) {
-                /* XXX figure out if we can create a variant on
-                 * qio_channel_yield() that works with AIO contexts
-                 * and consider using that in this branch */
-                qemu_coroutine_yield();
-            } else if (done) {
-                /* XXX this is needed by nbd_reply_ready.  */
-                qio_channel_wait(ioc,
-                                 do_read ? G_IO_IN : G_IO_OUT);
+                qio_channel_yield(ioc, do_read ? G_IO_IN : G_IO_OUT);
             } else {
                 return -EAGAIN;
             }
diff --git a/nbd/server.c b/nbd/server.c
index XXXXXXX..XXXXXXX 100644
--- a/nbd/server.c
+++ b/nbd/server.c
@@ -XXX,XX +XXX,XX @@ struct NBDClient {
     CoMutex send_lock;
     Coroutine *send_coroutine;
 
-    bool can_read;
-
     QTAILQ_ENTRY(NBDClient) next;
     int nb_requests;
     bool closing;
@@ -XXX,XX +XXX,XX @@ struct NBDClient {
 
 /* That's all folks */
 
-static void nbd_set_handlers(NBDClient *client);
-static void nbd_unset_handlers(NBDClient *client);
-static void nbd_update_can_read(NBDClient *client);
+static void nbd_client_receive_next_request(NBDClient *client);
 
 static gboolean nbd_negotiate_continue(QIOChannel *ioc,
                                        GIOCondition condition,
@@ -XXX,XX +XXX,XX @@ void nbd_client_put(NBDClient *client)
          */
         assert(client->closing);
 
-        nbd_unset_handlers(client);
+        qio_channel_detach_aio_context(client->ioc);
         object_unref(OBJECT(client->sioc));
         object_unref(OBJECT(client->ioc));
         if (client->tlscreds) {
@@ -XXX,XX +XXX,XX @@ static NBDRequestData *nbd_request_get(NBDClient *client)
 
     assert(client->nb_requests <= MAX_NBD_REQUESTS - 1);
     client->nb_requests++;
-    nbd_update_can_read(client);
 
     req = g_new0(NBDRequestData, 1);
     nbd_client_get(client);
@@ -XXX,XX +XXX,XX @@ static void nbd_request_put(NBDRequestData *req)
     g_free(req);
 
     client->nb_requests--;
-    nbd_update_can_read(client);
+    nbd_client_receive_next_request(client);
+
     nbd_client_put(client);
 }
 
@@ -XXX,XX +XXX,XX @@ static void blk_aio_attached(AioContext *ctx, void *opaque)
     exp->ctx = ctx;
 
     QTAILQ_FOREACH(client, &exp->clients, next) {
-        nbd_set_handlers(client);
+        qio_channel_attach_aio_context(client->ioc, ctx);
+        if (client->recv_coroutine) {
+            aio_co_schedule(ctx, client->recv_coroutine);
+        }
+        if (client->send_coroutine) {
+            aio_co_schedule(ctx, client->send_coroutine);
+        }
     }
 }
 
@@ -XXX,XX +XXX,XX @@ static void blk_aio_detach(void *opaque)
     TRACE("Export %s: Detaching clients from AIO context %p\n", exp->name, exp->ctx);
 
     QTAILQ_FOREACH(client, &exp->clients, next) {
-        nbd_unset_handlers(client);
+        qio_channel_detach_aio_context(client->ioc);
     }
 
     exp->ctx = NULL;
@@ -XXX,XX +XXX,XX @@ static ssize_t nbd_co_send_reply(NBDRequestData *req, NBDReply *reply,
     g_assert(qemu_in_coroutine());
     qemu_co_mutex_lock(&client->send_lock);
     client->send_coroutine = qemu_coroutine_self();
-    nbd_set_handlers(client);
 
     if (!len) {
         rc = nbd_send_reply(client->ioc, reply);
@@ -XXX,XX +XXX,XX @@ static ssize_t nbd_co_send_reply(NBDRequestData *req, NBDReply *reply,
     }
 
     client->send_coroutine = NULL;
-    nbd_set_handlers(client);
     qemu_co_mutex_unlock(&client->send_lock);
     return rc;
 }
@@ -XXX,XX +XXX,XX @@ static ssize_t nbd_co_receive_request(NBDRequestData *req,
     ssize_t rc;
 
     g_assert(qemu_in_coroutine());
-    client->recv_coroutine = qemu_coroutine_self();
-    nbd_update_can_read(client);
-
+    assert(client->recv_coroutine == qemu_coroutine_self());
     rc = nbd_receive_request(client->ioc, request);
     if (rc < 0) {
         if (rc != -EAGAIN) {
@@ -XXX,XX +XXX,XX @@ static ssize_t nbd_co_receive_request(NBDRequestData *req,
 
 out:
     client->recv_coroutine = NULL;
-    nbd_update_can_read(client);
+    nbd_client_receive_next_request(client);
 
     return rc;
 }
 
-static void nbd_trip(void *opaque)
+/* Owns a reference to the NBDClient passed as opaque.  */
+static coroutine_fn void nbd_trip(void *opaque)
 {
     NBDClient *client = opaque;
     NBDExport *exp = client->exp;
     NBDRequestData *req;
-    NBDRequest request;
+    NBDRequest request = { 0 };    /* GCC thinks it can be used uninitialized */
     NBDReply reply;
     ssize_t ret;
     int flags;
 
     TRACE("Reading request.");
     if (client->closing) {
+        nbd_client_put(client);
         return;
     }
 
@@ -XXX,XX +XXX,XX @@ static void nbd_trip(void *opaque)
 
 done:
     nbd_request_put(req);
+    nbd_client_put(client);
     return;
 
 out:
     nbd_request_put(req);
     client_close(client);
+    nbd_client_put(client);
 }
 
-static void nbd_read(void *opaque)
+static void nbd_client_receive_next_request(NBDClient *client)
 {
-    NBDClient *client = opaque;
-
-    if (client->recv_coroutine) {
-        qemu_coroutine_enter(client->recv_coroutine);
-    } else {
-        qemu_coroutine_enter(qemu_coroutine_create(nbd_trip, client));
-    }
-}
-
-static void nbd_restart_write(void *opaque)
-{
-    NBDClient *client = opaque;
-
-    qemu_coroutine_enter(client->send_coroutine);
-}
-
-static void nbd_set_handlers(NBDClient *client)
-{
-    if (client->exp && client->exp->ctx) {
-        aio_set_fd_handler(client->exp->ctx, client->sioc->fd, true,
-                           client->can_read ? nbd_read : NULL,
-                           client->send_coroutine ? nbd_restart_write : NULL,
-                           NULL, client);
-    }
-}
-
-static void nbd_unset_handlers(NBDClient *client)
-{
-    if (client->exp && client->exp->ctx) {
-        aio_set_fd_handler(client->exp->ctx, client->sioc->fd, true, NULL,
-                           NULL, NULL, NULL);
-    }
-}
-
-static void nbd_update_can_read(NBDClient *client)
-{
-    bool can_read = client->recv_coroutine ||
-                    client->nb_requests < MAX_NBD_REQUESTS;
-
-    if (can_read != client->can_read) {
-        client->can_read = can_read;
-        nbd_set_handlers(client);
-
-        /* There is no need to invoke aio_notify(), since aio_set_fd_handler()
-         * in nbd_set_handlers() will have taken care of that */
+    if (!client->recv_coroutine && client->nb_requests < MAX_NBD_REQUESTS) {
+        nbd_client_get(client);
+        client->recv_coroutine = qemu_coroutine_create(nbd_trip, client);
+        aio_co_schedule(client->exp->ctx, client->recv_coroutine);
     }
 }
 
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void nbd_co_client_start(void *opaque)
         goto out;
     }
     qemu_co_mutex_init(&client->send_lock);
-    nbd_set_handlers(client);
 
     if (exp) {
         QTAILQ_INSERT_TAIL(&exp->clients, client, next);
     }
+
+    nbd_client_receive_next_request(client);
+
 out:
     g_free(data);
 }
@@ -XXX,XX +XXX,XX @@ void nbd_client_new(NBDExport *exp,
     object_ref(OBJECT(client->sioc));
     client->ioc = QIO_CHANNEL(sioc);
     object_ref(OBJECT(client->ioc));
-    client->can_read = true;
     client->close = close_fn;
 
     data->client = client;
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

As a small step towards the introduction of multiqueue, we want
coroutines to remain on the same AioContext that started them,
unless they are moved explicitly with e.g. aio_co_schedule.  This patch
avoids that coroutines switch AioContext when they use a CoMutex.
For now it does not make much of a difference, because the CoMutex
is not thread-safe and the AioContext itself is used to protect the
CoMutex from concurrent access.  However, this is going to change.

diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
index XXXXXXX..XXXXXXX 100644
--- a/util/qemu-coroutine-lock.c
+++ b/util/qemu-coroutine-lock.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/coroutine.h"
 #include "qemu/coroutine_int.h"
 #include "qemu/queue.h"
+#include "block/aio.h"
 #include "trace.h"
 
 void qemu_co_queue_init(CoQueue *queue)
@@ -XXX,XX +XXX,XX @@ void qemu_co_queue_run_restart(Coroutine *co)
 
 static bool qemu_co_queue_do_restart(CoQueue *queue, bool single)
 {
-    Coroutine *self = qemu_coroutine_self();
     Coroutine *next;
 
     if (QSIMPLEQ_EMPTY(&queue->entries)) {
@@ -XXX,XX +XXX,XX @@ static bool qemu_co_queue_do_restart(CoQueue *queue, bool single)
 
     while ((next = QSIMPLEQ_FIRST(&queue->entries)) != NULL) {
         QSIMPLEQ_REMOVE_HEAD(&queue->entries, co_queue_next);
-        QSIMPLEQ_INSERT_TAIL(&self->co_queue_wakeup, next, co_queue_next);
-        trace_qemu_co_queue_next(next);
+        aio_co_wake(next);
         if (single) {
             break;
         }
diff --git a/util/trace-events b/util/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/util/trace-events
+++ b/util/trace-events
@@ -XXX,XX +XXX,XX @@ qemu_coroutine_terminate(void *co) "self %p"
 
 # util/qemu-coroutine-lock.c
 qemu_co_queue_run_restart(void *co) "co %p"
-qemu_co_queue_next(void *nxt) "next %p"
 qemu_co_mutex_lock_entry(void *mutex, void *self) "mutex %p self %p"
 qemu_co_mutex_lock_return(void *mutex, void *self) "mutex %p self %p"
 qemu_co_mutex_unlock_entry(void *mutex, void *self) "mutex %p self %p"
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Keep the coroutine on the same AioContext.  Without this change,
there would be a race between yielding the coroutine and reentering it.
While the race cannot happen now, because the code only runs from a single
AioContext, this will change with multiqueue support in the block layer.

While doing the change, replace custom bottom half with aio_co_schedule.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 20170213135235.12274-10-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 block/blkdebug.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/block/blkdebug.c b/block/blkdebug.c
index XXXXXXX..XXXXXXX 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -XXX,XX +XXX,XX @@ out:
     return ret;
 }
 
-static void error_callback_bh(void *opaque)
-{
-    Coroutine *co = opaque;
-    qemu_coroutine_enter(co);
-}
-
 static int inject_error(BlockDriverState *bs, BlkdebugRule *rule)
 {
     BDRVBlkdebugState *s = bs->opaque;
@@ -XXX,XX +XXX,XX @@ static int inject_error(BlockDriverState *bs, BlkdebugRule *rule)
     }
 
     if (!immediately) {
-        aio_bh_schedule_oneshot(bdrv_get_aio_context(bs), error_callback_bh,
-                                qemu_coroutine_self());
+        aio_co_schedule(qemu_get_current_aio_context(), qemu_coroutine_self());
         qemu_coroutine_yield();
     }
 
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

qed_aio_start_io and qed_aio_next_io will not have to acquire/release
the AioContext, while qed_aio_next_io_cb will.  Split the functionality
and gain a little type-safety in the process.

diff --git a/block/qed.c b/block/qed.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -XXX,XX +XXX,XX @@ static CachedL2Table *qed_new_l2_table(BDRVQEDState *s)
     return l2_table;
 }
 
-static void qed_aio_next_io(void *opaque, int ret);
+static void qed_aio_next_io(QEDAIOCB *acb, int ret);
+
+static void qed_aio_start_io(QEDAIOCB *acb)
+{
+    qed_aio_next_io(acb, 0);
+}
+
+static void qed_aio_next_io_cb(void *opaque, int ret)
+{
+    QEDAIOCB *acb = opaque;
+
+    qed_aio_next_io(acb, ret);
+}
 
 static void qed_plug_allocating_write_reqs(BDRVQEDState *s)
 {
@@ -XXX,XX +XXX,XX @@ static void qed_unplug_allocating_write_reqs(BDRVQEDState *s)
 
     acb = QSIMPLEQ_FIRST(&s->allocating_write_reqs);
     if (acb) {
-        qed_aio_next_io(acb, 0);
+        qed_aio_start_io(acb);
     }
 }
 
@@ -XXX,XX +XXX,XX @@ static void qed_aio_complete(QEDAIOCB *acb, int ret)
         QSIMPLEQ_REMOVE_HEAD(&s->allocating_write_reqs, next);
         acb = QSIMPLEQ_FIRST(&s->allocating_write_reqs);
         if (acb) {
-            qed_aio_next_io(acb, 0);
+            qed_aio_start_io(acb);
         } else if (s->header.features & QED_F_NEED_CHECK) {
             qed_start_need_check_timer(s);
         }
@@ -XXX,XX +XXX,XX @@ static void qed_commit_l2_update(void *opaque, int ret)
     acb->request.l2_table = qed_find_l2_cache_entry(&s->l2_cache, l2_offset);
     assert(acb->request.l2_table != NULL);
 
-    qed_aio_next_io(opaque, ret);
+    qed_aio_next_io(acb, ret);
 }
 
 /**
@@ -XXX,XX +XXX,XX @@ static void qed_aio_write_l2_update(QEDAIOCB *acb, int ret, uint64_t offset)
     if (need_alloc) {
         /* Write out the whole new L2 table */
         qed_write_l2_table(s, &acb->request, 0, s->table_nelems, true,
-                            qed_aio_write_l1_update, acb);
+                           qed_aio_write_l1_update, acb);
     } else {
         /* Write out only the updated part of the L2 table */
         qed_write_l2_table(s, &acb->request, index, acb->cur_nclusters, false,
-                            qed_aio_next_io, acb);
+                           qed_aio_next_io_cb, acb);
     }
     return;
 
@@ -XXX,XX +XXX,XX @@ static void qed_aio_write_main(void *opaque, int ret)
     }
 
     if (acb->find_cluster_ret == QED_CLUSTER_FOUND) {
-        next_fn = qed_aio_next_io;
+        next_fn = qed_aio_next_io_cb;
     } else {
         if (s->bs->backing) {
             next_fn = qed_aio_write_flush_before_l2_update;
@@ -XXX,XX +XXX,XX @@ static void qed_aio_write_alloc(QEDAIOCB *acb, size_t len)
     if (acb->flags & QED_AIOCB_ZERO) {
         /* Skip ahead if the clusters are already zero */
         if (acb->find_cluster_ret == QED_CLUSTER_ZERO) {
-            qed_aio_next_io(acb, 0);
+            qed_aio_start_io(acb);
             return;
         }
 
@@ -XXX,XX +XXX,XX @@ static void qed_aio_read_data(void *opaque, int ret,
     /* Handle zero cluster and backing file reads */
     if (ret == QED_CLUSTER_ZERO) {
         qemu_iovec_memset(&acb->cur_qiov, 0, 0, acb->cur_qiov.size);
-        qed_aio_next_io(acb, 0);
+        qed_aio_start_io(acb);
         return;
     } else if (ret != QED_CLUSTER_FOUND) {
         qed_read_backing_file(s, acb->cur_pos, &acb->cur_qiov,
-                              &acb->backing_qiov, qed_aio_next_io, acb);
+                              &acb->backing_qiov, qed_aio_next_io_cb, acb);
         return;
     }
 
     BLKDBG_EVENT(bs->file, BLKDBG_READ_AIO);
     bdrv_aio_readv(bs->file, offset / BDRV_SECTOR_SIZE,
                    &acb->cur_qiov, acb->cur_qiov.size / BDRV_SECTOR_SIZE,
-                   qed_aio_next_io, acb);
+                   qed_aio_next_io_cb, acb);
     return;
 
 err:
@@ -XXX,XX +XXX,XX @@ err:
 /**
  * Begin next I/O or complete the request
  */
-static void qed_aio_next_io(void *opaque, int ret)
+static void qed_aio_next_io(QEDAIOCB *acb, int ret)
 {
-    QEDAIOCB *acb = opaque;
     BDRVQEDState *s = acb_to_s(acb);
     QEDFindClusterFunc *io_fn = (acb->flags & QED_AIOCB_WRITE) ?
                                 qed_aio_write_data : qed_aio_read_data;
@@ -XXX,XX +XXX,XX @@ static BlockAIOCB *qed_aio_setup(BlockDriverState *bs,
     qemu_iovec_init(&acb->cur_qiov, qiov->niov);
 
     /* Start request */
-    qed_aio_next_io(acb, 0);
+    qed_aio_start_io(acb);
     return &acb->common;
 }
 
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

The AioContext data structures are now protected by list_lock and/or
they are walked with FOREACH_RCU primitives.  There is no need anymore
to acquire the AioContext for the entire duration of aio_dispatch.
Instead, just acquire it before and after invoking the callbacks.
The next step is then to push it further down.

diff --git a/util/aio-posix.c b/util/aio-posix.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-posix.c
+++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
             (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR)) &&
             aio_node_check(ctx, node->is_external) &&
             node->io_read) {
+            aio_context_acquire(ctx);
             node->io_read(node->opaque);
+            aio_context_release(ctx);
 
             /* aio_notify() does not count as progress */
             if (node->opaque != &ctx->notifier) {
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
             (revents & (G_IO_OUT | G_IO_ERR)) &&
             aio_node_check(ctx, node->is_external) &&
             node->io_write) {
+            aio_context_acquire(ctx);
             node->io_write(node->opaque);
+            aio_context_release(ctx);
             progress = true;
         }
 
@@ -XXX,XX +XXX,XX @@ bool aio_dispatch(AioContext *ctx, bool dispatch_fds)
     }
 
     /* Run our timers */
+    aio_context_acquire(ctx);
     progress |= timerlistgroup_run_timers(&ctx->tlg);
+    aio_context_release(ctx);
 
     return progress;
 }
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
     int64_t timeout;
     int64_t start = 0;
 
-    aio_context_acquire(ctx);
-    progress = false;
-
     /* aio_notify can avoid the expensive event_notifier_set if
      * everything (file descriptors, bottom halves, timers) will
      * be re-evaluated before the next blocking poll().  This is
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         start = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
     }
 
-    if (try_poll_mode(ctx, blocking)) {
-        progress = true;
-    } else {
+    aio_context_acquire(ctx);
+    progress = try_poll_mode(ctx, blocking);
+    aio_context_release(ctx);
+
+    if (!progress) {
         assert(npfd == 0);
 
         /* fill pollfds */
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         timeout = blocking ? aio_compute_timeout(ctx) : 0;
 
         /* wait until next event */
-        if (timeout) {
-            aio_context_release(ctx);
-        }
         if (aio_epoll_check_poll(ctx, pollfds, npfd, timeout)) {
             AioHandler epoll_handler;
 
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         } else  {
             ret = qemu_poll_ns(pollfds, npfd, timeout);
         }
-        if (timeout) {
-            aio_context_acquire(ctx);
-        }
     }
 
     if (blocking) {
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         progress = true;
     }
 
-    aio_context_release(ctx);
-
     return progress;
 }
 
diff --git a/util/aio-win32.c b/util/aio-win32.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-win32.c
+++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
             (revents || event_notifier_get_handle(node->e) == event) &&
             node->io_notify) {
             node->pfd.revents = 0;
+            aio_context_acquire(ctx);
             node->io_notify(node->e);
+            aio_context_release(ctx);
 
             /* aio_notify() does not count as progress */
             if (node->e != &ctx->notifier) {
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
             (node->io_read || node->io_write)) {
             node->pfd.revents = 0;
             if ((revents & G_IO_IN) && node->io_read) {
+                aio_context_acquire(ctx);
                 node->io_read(node->opaque);
+                aio_context_release(ctx);
                 progress = true;
             }
             if ((revents & G_IO_OUT) && node->io_write) {
+                aio_context_acquire(ctx);
                 node->io_write(node->opaque);
+                aio_context_release(ctx);
                 progress = true;
             }
 
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
     int count;
     int timeout;
 
-    aio_context_acquire(ctx);
     progress = false;
 
     /* aio_notify can avoid the expensive event_notifier_set if
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
 
         timeout = blocking && !have_select_revents
             ? qemu_timeout_ns_to_ms(aio_compute_timeout(ctx)) : 0;
-        if (timeout) {
-            aio_context_release(ctx);
-        }
         ret = WaitForMultipleObjects(count, events, FALSE, timeout);
         if (blocking) {
             assert(first);
             atomic_sub(&ctx->notify_me, 2);
         }
-        if (timeout) {
-            aio_context_acquire(ctx);
-        }
 
         if (first) {
             aio_notify_accept(ctx);
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         progress |= aio_dispatch_handlers(ctx, event);
     } while (count > 0);
 
+    aio_context_acquire(ctx);
     progress |= timerlistgroup_run_timers(&ctx->tlg);
-
     aio_context_release(ctx);
     return progress;
 }
diff --git a/util/async.c b/util/async.c
index XXXXXXX..XXXXXXX 100644
--- a/util/async.c
+++ b/util/async.c
@@ -XXX,XX +XXX,XX @@ int aio_bh_poll(AioContext *ctx)
                 ret = 1;
             }
             bh->idle = 0;
+            aio_context_acquire(ctx);
             aio_bh_call(bh);
+            aio_context_release(ctx);
         }
         if (bh->deleted) {
             deleted = true;
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 20170213135235.12274-13-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 block/qed.h                 |  3 +++
 block/curl.c                |  2 ++
 block/io.c                  |  5 +++++
 block/iscsi.c               |  8 ++++++--
 block/null.c                |  4 ++++
 block/qed.c                 | 12 ++++++++++++
 block/throttle-groups.c     |  2 ++
 util/aio-posix.c            |  2 --
 util/aio-win32.c            |  2 --
 util/qemu-coroutine-sleep.c |  2 +-
 10 files changed, 35 insertions(+), 7 deletions(-)

diff --git a/block/qed.h b/block/qed.h
index XXXXXXX..XXXXXXX 100644
--- a/block/qed.h
+++ b/block/qed.h
@@ -XXX,XX +XXX,XX @@ enum {
  */
 typedef void QEDFindClusterFunc(void *opaque, int ret, uint64_t offset, size_t len);
 
+void qed_acquire(BDRVQEDState *s);
+void qed_release(BDRVQEDState *s);
+
 /**
  * Generic callback for chaining async callbacks
  */
diff --git a/block/curl.c b/block/curl.c
index XXXXXXX..XXXXXXX 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -XXX,XX +XXX,XX @@ static void curl_multi_timeout_do(void *arg)
         return;
     }
 
+    aio_context_acquire(s->aio_context);
     curl_multi_socket_action(s->multi, CURL_SOCKET_TIMEOUT, 0, &running);
 
     curl_multi_check_completion(s);
+    aio_context_release(s->aio_context);
 #else
     abort();
 #endif
diff --git a/block/io.c b/block/io.c
index XXXXXXX..XXXXXXX 100644
--- a/block/io.c
+++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ void bdrv_aio_cancel(BlockAIOCB *acb)
         if (acb->aiocb_info->get_aio_context) {
             aio_poll(acb->aiocb_info->get_aio_context(acb), true);
         } else if (acb->bs) {
+            /* qemu_aio_ref and qemu_aio_unref are not thread-safe, so
+             * assert that we're not using an I/O thread.  Thread-safe
+             * code should use bdrv_aio_cancel_async exclusively.
+             */
+            assert(bdrv_get_aio_context(acb->bs) == qemu_get_aio_context());
             aio_poll(bdrv_get_aio_context(acb->bs), true);
         } else {
             abort();
diff --git a/block/iscsi.c b/block/iscsi.c
index XXXXXXX..XXXXXXX 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -XXX,XX +XXX,XX @@ static void iscsi_retry_timer_expired(void *opaque)
     struct IscsiTask *iTask = opaque;
     iTask->complete = 1;
     if (iTask->co) {
-        qemu_coroutine_enter(iTask->co);
+        aio_co_wake(iTask->co);
     }
 }
 
@@ -XXX,XX +XXX,XX @@ static void iscsi_nop_timed_event(void *opaque)
 {
     IscsiLun *iscsilun = opaque;
 
+    aio_context_acquire(iscsilun->aio_context);
     if (iscsi_get_nops_in_flight(iscsilun->iscsi) >= MAX_NOP_FAILURES) {
         error_report("iSCSI: NOP timeout. Reconnecting...");
         iscsilun->request_timed_out = true;
     } else if (iscsi_nop_out_async(iscsilun->iscsi, NULL, NULL, 0, NULL) != 0) {
         error_report("iSCSI: failed to sent NOP-Out. Disabling NOP messages.");
-        return;
+        goto out;
     }
 
     timer_mod(iscsilun->nop_timer, qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + NOP_INTERVAL);
     iscsi_set_events(iscsilun);
+
+out:
+    aio_context_release(iscsilun->aio_context);
 }
 
 static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp)
diff --git a/block/null.c b/block/null.c
index XXXXXXX..XXXXXXX 100644
--- a/block/null.c
+++ b/block/null.c
@@ -XXX,XX +XXX,XX @@ static void null_bh_cb(void *opaque)
 static void null_timer_cb(void *opaque)
 {
     NullAIOCB *acb = opaque;
+    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
+
+    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, 0);
+    aio_context_release(ctx);
     timer_deinit(&acb->timer);
     qemu_aio_unref(acb);
 }
diff --git a/block/qed.c b/block/qed.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -XXX,XX +XXX,XX @@ static void qed_need_check_timer_cb(void *opaque)
 
     trace_qed_need_check_timer_cb(s);
 
+    qed_acquire(s);
     qed_plug_allocating_write_reqs(s);
 
     /* Ensure writes are on disk before clearing flag */
     bdrv_aio_flush(s->bs->file->bs, qed_clear_need_check, s);
+    qed_release(s);
+}
+
+void qed_acquire(BDRVQEDState *s)
+{
+    aio_context_acquire(bdrv_get_aio_context(s->bs));
+}
+
+void qed_release(BDRVQEDState *s)
+{
+    aio_context_release(bdrv_get_aio_context(s->bs));
 }
 
 static void qed_start_need_check_timer(BDRVQEDState *s)
diff --git a/block/throttle-groups.c b/block/throttle-groups.c
index XXXXXXX..XXXXXXX 100644
--- a/block/throttle-groups.c
+++ b/block/throttle-groups.c
@@ -XXX,XX +XXX,XX @@ static void timer_cb(BlockBackend *blk, bool is_write)
     qemu_mutex_unlock(&tg->lock);
 
     /* Run the request that was waiting for this timer */
+    aio_context_acquire(blk_get_aio_context(blk));
     empty_queue = !qemu_co_enter_next(&blkp->throttled_reqs[is_write]);
+    aio_context_release(blk_get_aio_context(blk));
 
     /* If the request queue was empty then we have to take care of
      * scheduling the next one */
diff --git a/util/aio-posix.c b/util/aio-posix.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-posix.c
+++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@ bool aio_dispatch(AioContext *ctx, bool dispatch_fds)
     }
 
     /* Run our timers */
-    aio_context_acquire(ctx);
     progress |= timerlistgroup_run_timers(&ctx->tlg);
-    aio_context_release(ctx);
 
     return progress;
 }
diff --git a/util/aio-win32.c b/util/aio-win32.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-win32.c
+++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         progress |= aio_dispatch_handlers(ctx, event);
     } while (count > 0);
 
-    aio_context_acquire(ctx);
     progress |= timerlistgroup_run_timers(&ctx->tlg);
-    aio_context_release(ctx);
     return progress;
 }
 
diff --git a/util/qemu-coroutine-sleep.c b/util/qemu-coroutine-sleep.c
index XXXXXXX..XXXXXXX 100644
--- a/util/qemu-coroutine-sleep.c
+++ b/util/qemu-coroutine-sleep.c
@@ -XXX,XX +XXX,XX @@ static void co_sleep_cb(void *opaque)
 {
     CoSleepCB *sleep_cb = opaque;
 
-    qemu_coroutine_enter(sleep_cb->co);
+    aio_co_wake(sleep_cb->co);
 }
 
 void coroutine_fn co_aio_sleep_ns(AioContext *ctx, QEMUClockType type,
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

This covers both file descriptor callbacks and polling callbacks,
since they execute related code.

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 20170213135235.12274-14-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 block/curl.c          | 16 +++++++++++++---
 block/iscsi.c         |  4 ++++
 block/linux-aio.c     |  4 ++++
 block/nfs.c           |  6 ++++++
 block/sheepdog.c      | 29 +++++++++++++++--------------
 block/ssh.c           | 29 +++++++++--------------------
 block/win32-aio.c     | 10 ++++++----
 hw/block/virtio-blk.c |  5 ++++-
 hw/scsi/virtio-scsi.c |  7 +++++++
 util/aio-posix.c      |  7 -------
 util/aio-win32.c      |  6 ------
 11 files changed, 68 insertions(+), 55 deletions(-)

diff --git a/block/curl.c b/block/curl.c
index XXXXXXX..XXXXXXX 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -XXX,XX +XXX,XX @@ static void curl_multi_check_completion(BDRVCURLState *s)
     }
 }
 
-static void curl_multi_do(void *arg)
+static void curl_multi_do_locked(CURLState *s)
 {
-    CURLState *s = (CURLState *)arg;
     CURLSocket *socket, *next_socket;
     int running;
     int r;
@@ -XXX,XX +XXX,XX @@ static void curl_multi_do(void *arg)
     }
 }
 
+static void curl_multi_do(void *arg)
+{
+    CURLState *s = (CURLState *)arg;
+
+    aio_context_acquire(s->s->aio_context);
+    curl_multi_do_locked(s);
+    aio_context_release(s->s->aio_context);
+}
+
 static void curl_multi_read(void *arg)
 {
     CURLState *s = (CURLState *)arg;
 
-    curl_multi_do(arg);
+    aio_context_acquire(s->s->aio_context);
+    curl_multi_do_locked(s);
     curl_multi_check_completion(s->s);
+    aio_context_release(s->s->aio_context);
 }
 
 static void curl_multi_timeout_do(void *arg)
diff --git a/block/iscsi.c b/block/iscsi.c
index XXXXXXX..XXXXXXX 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -XXX,XX +XXX,XX @@ iscsi_process_read(void *arg)
     IscsiLun *iscsilun = arg;
     struct iscsi_context *iscsi = iscsilun->iscsi;
 
+    aio_context_acquire(iscsilun->aio_context);
     iscsi_service(iscsi, POLLIN);
     iscsi_set_events(iscsilun);
+    aio_context_release(iscsilun->aio_context);
 }
 
 static void
@@ -XXX,XX +XXX,XX @@ iscsi_process_write(void *arg)
     IscsiLun *iscsilun = arg;
     struct iscsi_context *iscsi = iscsilun->iscsi;
 
+    aio_context_acquire(iscsilun->aio_context);
     iscsi_service(iscsi, POLLOUT);
     iscsi_set_events(iscsilun);
+    aio_context_release(iscsilun->aio_context);
 }
 
 static int64_t sector_lun2qemu(int64_t sector, IscsiLun *iscsilun)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index XXXXXXX..XXXXXXX 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_completion_cb(EventNotifier *e)
     LinuxAioState *s = container_of(e, LinuxAioState, e);
 
     if (event_notifier_test_and_clear(&s->e)) {
+        aio_context_acquire(s->aio_context);
         qemu_laio_process_completions_and_submit(s);
+        aio_context_release(s->aio_context);
     }
 }
 
@@ -XXX,XX +XXX,XX @@ static bool qemu_laio_poll_cb(void *opaque)
         return false;
     }
 
+    aio_context_acquire(s->aio_context);
     qemu_laio_process_completions_and_submit(s);
+    aio_context_release(s->aio_context);
     return true;
 }
 
diff --git a/block/nfs.c b/block/nfs.c
index XXXXXXX..XXXXXXX 100644
--- a/block/nfs.c
+++ b/block/nfs.c
@@ -XXX,XX +XXX,XX @@ static void nfs_set_events(NFSClient *client)
 static void nfs_process_read(void *arg)
 {
     NFSClient *client = arg;
+
+    aio_context_acquire(client->aio_context);
     nfs_service(client->context, POLLIN);
     nfs_set_events(client);
+    aio_context_release(client->aio_context);
 }
 
 static void nfs_process_write(void *arg)
 {
     NFSClient *client = arg;
+
+    aio_context_acquire(client->aio_context);
     nfs_service(client->context, POLLOUT);
     nfs_set_events(client);
+    aio_context_release(client->aio_context);
 }
 
 static void nfs_co_init_task(BlockDriverState *bs, NFSRPC *task)
diff --git a/block/sheepdog.c b/block/sheepdog.c
index XXXXXXX..XXXXXXX 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -XXX,XX +XXX,XX @@ static coroutine_fn int send_co_req(int sockfd, SheepdogReq *hdr, void *data,
     return ret;
 }
 
-static void restart_co_req(void *opaque)
-{
-    Coroutine *co = opaque;
-
-    qemu_coroutine_enter(co);
-}
-
 typedef struct SheepdogReqCo {
     int sockfd;
     BlockDriverState *bs;
@@ -XXX,XX +XXX,XX @@ typedef struct SheepdogReqCo {
     unsigned int *rlen;
     int ret;
     bool finished;
+    Coroutine *co;
 } SheepdogReqCo;
 
+static void restart_co_req(void *opaque)
+{
+    SheepdogReqCo *srco = opaque;
+
+    aio_co_wake(srco->co);
+}
+
 static coroutine_fn void do_co_req(void *opaque)
 {
     int ret;
-    Coroutine *co;
     SheepdogReqCo *srco = opaque;
     int sockfd = srco->sockfd;
     SheepdogReq *hdr = srco->hdr;
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void do_co_req(void *opaque)
     unsigned int *wlen = srco->wlen;
     unsigned int *rlen = srco->rlen;
 
-    co = qemu_coroutine_self();
+    srco->co = qemu_coroutine_self();
     aio_set_fd_handler(srco->aio_context, sockfd, false,
-                       NULL, restart_co_req, NULL, co);
+                       NULL, restart_co_req, NULL, srco);
 
     ret = send_co_req(sockfd, hdr, data, wlen);
     if (ret < 0) {
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void do_co_req(void *opaque)
     }
 
     aio_set_fd_handler(srco->aio_context, sockfd, false,
-                       restart_co_req, NULL, NULL, co);
+                       restart_co_req, NULL, NULL, srco);
 
     ret = qemu_co_recv(sockfd, hdr, sizeof(*hdr));
     if (ret != sizeof(*hdr)) {
@@ -XXX,XX +XXX,XX @@ out:
     aio_set_fd_handler(srco->aio_context, sockfd, false,
                        NULL, NULL, NULL, NULL);
 
+    srco->co = NULL;
     srco->ret = ret;
     srco->finished = true;
     if (srco->bs) {
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn aio_read_response(void *opaque)
          * We've finished all requests which belong to the AIOCB, so
          * we can switch back to sd_co_readv/writev now.
          */
-        qemu_coroutine_enter(acb->coroutine);
+        aio_co_wake(acb->coroutine);
     }
 
     return;
@@ -XXX,XX +XXX,XX @@ static void co_read_response(void *opaque)
         s->co_recv = qemu_coroutine_create(aio_read_response, opaque);
     }
 
-    qemu_coroutine_enter(s->co_recv);
+    aio_co_wake(s->co_recv);
 }
 
 static void co_write_request(void *opaque)
 {
     BDRVSheepdogState *s = opaque;
 
-    qemu_coroutine_enter(s->co_send);
+    aio_co_wake(s->co_send);
 }
 
 /*
diff --git a/block/ssh.c b/block/ssh.c
index XXXXXXX..XXXXXXX 100644
--- a/block/ssh.c
+++ b/block/ssh.c
@@ -XXX,XX +XXX,XX @@ static void restart_coroutine(void *opaque)
 
     DPRINTF("co=%p", co);
 
-    qemu_coroutine_enter(co);
+    aio_co_wake(co);
 }
 
-static coroutine_fn void set_fd_handler(BDRVSSHState *s, BlockDriverState *bs)
+/* A non-blocking call returned EAGAIN, so yield, ensuring the
+ * handlers are set up so that we'll be rescheduled when there is an
+ * interesting event on the socket.
+ */
+static coroutine_fn void co_yield(BDRVSSHState *s, BlockDriverState *bs)
 {
     int r;
     IOHandler *rd_handler = NULL, *wr_handler = NULL;
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void set_fd_handler(BDRVSSHState *s, BlockDriverState *bs)
 
     aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock,
                        false, rd_handler, wr_handler, NULL, co);
-}
-
-static coroutine_fn void clear_fd_handler(BDRVSSHState *s,
-                                          BlockDriverState *bs)
-{
-    DPRINTF("s->sock=%d", s->sock);
-    aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock,
-                       false, NULL, NULL, NULL, NULL);
-}
-
-/* A non-blocking call returned EAGAIN, so yield, ensuring the
- * handlers are set up so that we'll be rescheduled when there is an
- * interesting event on the socket.
- */
-static coroutine_fn void co_yield(BDRVSSHState *s, BlockDriverState *bs)
-{
-    set_fd_handler(s, bs);
     qemu_coroutine_yield();
-    clear_fd_handler(s, bs);
+    DPRINTF("s->sock=%d - back", s->sock);
+    aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock, false,
+                       NULL, NULL, NULL, NULL);
 }
 
 /* SFTP has a function `libssh2_sftp_seek64' which seeks to a position
diff --git a/block/win32-aio.c b/block/win32-aio.c
index XXXXXXX..XXXXXXX 100644
--- a/block/win32-aio.c
+++ b/block/win32-aio.c
@@ -XXX,XX +XXX,XX @@ struct QEMUWin32AIOState {
     HANDLE hIOCP;
     EventNotifier e;
     int count;
-    bool is_aio_context_attached;
+    AioContext *aio_ctx;
 };
 
 typedef struct QEMUWin32AIOCB {
@@ -XXX,XX +XXX,XX @@ static void win32_aio_process_completion(QEMUWin32AIOState *s,
     }
 
 
+    aio_context_acquire(s->aio_ctx);
     waiocb->common.cb(waiocb->common.opaque, ret);
+    aio_context_release(s->aio_ctx);
     qemu_aio_unref(waiocb);
 }
 
@@ -XXX,XX +XXX,XX @@ void win32_aio_detach_aio_context(QEMUWin32AIOState *aio,
                                   AioContext *old_context)
 {
     aio_set_event_notifier(old_context, &aio->e, false, NULL, NULL);
-    aio->is_aio_context_attached = false;
+    aio->aio_ctx = NULL;
 }
 
 void win32_aio_attach_aio_context(QEMUWin32AIOState *aio,
                                   AioContext *new_context)
 {
-    aio->is_aio_context_attached = true;
+    aio->aio_ctx = new_context;
     aio_set_event_notifier(new_context, &aio->e, false,
                            win32_aio_completion_cb, NULL);
 }
@@ -XXX,XX +XXX,XX @@ out_free_state:
 
 void win32_aio_cleanup(QEMUWin32AIOState *aio)
 {
-    assert(!aio->is_aio_context_attached);
+    assert(!aio->aio_ctx);
     CloseHandle(aio->hIOCP);
     event_notifier_cleanup(&aio->e);
     g_free(aio);
diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_ioctl_complete(void *opaque, int status)
 {
     VirtIOBlockIoctlReq *ioctl_req = opaque;
     VirtIOBlockReq *req = ioctl_req->req;
-    VirtIODevice *vdev = VIRTIO_DEVICE(req->dev);
+    VirtIOBlock *s = req->dev;
+    VirtIODevice *vdev = VIRTIO_DEVICE(s);
     struct virtio_scsi_inhdr *scsi;
     struct sg_io_hdr *hdr;
 
@@ -XXX,XX +XXX,XX @@ bool virtio_blk_handle_vq(VirtIOBlock *s, VirtQueue *vq)
     MultiReqBuffer mrb = {};
     bool progress = false;
 
+    aio_context_acquire(blk_get_aio_context(s->blk));
     blk_io_plug(s->blk);
 
     do {
@@ -XXX,XX +XXX,XX @@ bool virtio_blk_handle_vq(VirtIOBlock *s, VirtQueue *vq)
     }
 
     blk_io_unplug(s->blk);
+    aio_context_release(blk_get_aio_context(s->blk));
     return progress;
 }
 
diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -XXX,XX +XXX,XX @@ bool virtio_scsi_handle_ctrl_vq(VirtIOSCSI *s, VirtQueue *vq)
     VirtIOSCSIReq *req;
     bool progress = false;
 
+    virtio_scsi_acquire(s);
     while ((req = virtio_scsi_pop_req(s, vq))) {
         progress = true;
         virtio_scsi_handle_ctrl_req(s, req);
     }
+    virtio_scsi_release(s);
     return progress;
 }
 
@@ -XXX,XX +XXX,XX @@ bool virtio_scsi_handle_cmd_vq(VirtIOSCSI *s, VirtQueue *vq)
 
     QTAILQ_HEAD(, VirtIOSCSIReq) reqs = QTAILQ_HEAD_INITIALIZER(reqs);
 
+    virtio_scsi_acquire(s);
     do {
         virtio_queue_set_notification(vq, 0);
 
@@ -XXX,XX +XXX,XX @@ bool virtio_scsi_handle_cmd_vq(VirtIOSCSI *s, VirtQueue *vq)
     QTAILQ_FOREACH_SAFE(req, &reqs, next, next) {
         virtio_scsi_handle_cmd_req_submit(s, req);
     }
+    virtio_scsi_release(s);
     return progress;
 }
 
@@ -XXX,XX +XXX,XX @@ out:
 
 bool virtio_scsi_handle_event_vq(VirtIOSCSI *s, VirtQueue *vq)
 {
+    virtio_scsi_acquire(s);
     if (s->events_dropped) {
         virtio_scsi_push_event(s, NULL, VIRTIO_SCSI_T_NO_EVENT, 0);
+        virtio_scsi_release(s);
         return true;
     }
+    virtio_scsi_release(s);
     return false;
 }
 
diff --git a/util/aio-posix.c b/util/aio-posix.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-posix.c
+++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
             (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR)) &&
             aio_node_check(ctx, node->is_external) &&
             node->io_read) {
-            aio_context_acquire(ctx);
             node->io_read(node->opaque);
-            aio_context_release(ctx);
 
             /* aio_notify() does not count as progress */
             if (node->opaque != &ctx->notifier) {
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
             (revents & (G_IO_OUT | G_IO_ERR)) &&
             aio_node_check(ctx, node->is_external) &&
             node->io_write) {
-            aio_context_acquire(ctx);
             node->io_write(node->opaque);
-            aio_context_release(ctx);
             progress = true;
         }
 
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         start = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
     }
 
-    aio_context_acquire(ctx);
     progress = try_poll_mode(ctx, blocking);
-    aio_context_release(ctx);
-
     if (!progress) {
         assert(npfd == 0);
 
diff --git a/util/aio-win32.c b/util/aio-win32.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-win32.c
+++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
             (revents || event_notifier_get_handle(node->e) == event) &&
             node->io_notify) {
             node->pfd.revents = 0;
-            aio_context_acquire(ctx);
             node->io_notify(node->e);
-            aio_context_release(ctx);
 
             /* aio_notify() does not count as progress */
             if (node->e != &ctx->notifier) {
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
             (node->io_read || node->io_write)) {
             node->pfd.revents = 0;
             if ((revents & G_IO_IN) && node->io_read) {
-                aio_context_acquire(ctx);
                 node->io_read(node->opaque);
-                aio_context_release(ctx);
                 progress = true;
             }
             if ((revents & G_IO_OUT) && node->io_write) {
-                aio_context_acquire(ctx);
                 node->io_write(node->opaque);
-                aio_context_release(ctx);
                 progress = true;
             }
 
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 20170213135235.12274-15-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 block/archipelago.c   |  3 +++
 block/blkreplay.c     |  2 +-
 block/block-backend.c |  6 ++++++
 block/curl.c          | 26 ++++++++++++++++++--------
 block/gluster.c       |  9 +--------
 block/io.c            |  6 +++++-
 block/iscsi.c         |  6 +++++-
 block/linux-aio.c     | 15 +++++++++------
 block/nfs.c           |  3 ++-
 block/null.c          |  4 ++++
 block/qed.c           |  3 +++
 block/rbd.c           |  4 ++++
 dma-helpers.c         |  2 ++
 hw/block/virtio-blk.c |  2 ++
 hw/scsi/scsi-bus.c    |  2 ++
 util/async.c          |  4 ++--
 util/thread-pool.c    |  2 ++
 17 files changed, 71 insertions(+), 28 deletions(-)

diff --git a/block/archipelago.c b/block/archipelago.c
index XXXXXXX..XXXXXXX 100644
--- a/block/archipelago.c
+++ b/block/archipelago.c
@@ -XXX,XX +XXX,XX @@ static void qemu_archipelago_complete_aio(void *opaque)
 {
     AIORequestData *reqdata = (AIORequestData *) opaque;
     ArchipelagoAIOCB *aio_cb = (ArchipelagoAIOCB *) reqdata->aio_cb;
+    AioContext *ctx = bdrv_get_aio_context(aio_cb->common.bs);
 
+    aio_context_acquire(ctx);
     aio_cb->common.cb(aio_cb->common.opaque, aio_cb->ret);
+    aio_context_release(ctx);
     aio_cb->status = 0;
 
     qemu_aio_unref(aio_cb);
diff --git a/block/blkreplay.c b/block/blkreplay.c
index XXXXXXX..XXXXXXX 100755
--- a/block/blkreplay.c
+++ b/block/blkreplay.c
@@ -XXX,XX +XXX,XX @@ static int64_t blkreplay_getlength(BlockDriverState *bs)
 static void blkreplay_bh_cb(void *opaque)
 {
     Request *req = opaque;
-    qemu_coroutine_enter(req->co);
+    aio_co_wake(req->co);
     qemu_bh_delete(req->bh);
     g_free(req);
 }
diff --git a/block/block-backend.c b/block/block-backend.c
index XXXXXXX..XXXXXXX 100644
--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -XXX,XX +XXX,XX @@ int blk_make_zero(BlockBackend *blk, BdrvRequestFlags flags)
 static void error_callback_bh(void *opaque)
 {
     struct BlockBackendAIOCB *acb = opaque;
+    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 
     bdrv_dec_in_flight(acb->common.bs);
+    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, acb->ret);
+    aio_context_release(ctx);
     qemu_aio_unref(acb);
 }
 
@@ -XXX,XX +XXX,XX @@ static void blk_aio_complete(BlkAioEmAIOCB *acb)
 static void blk_aio_complete_bh(void *opaque)
 {
     BlkAioEmAIOCB *acb = opaque;
+    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 
     assert(acb->has_returned);
+    aio_context_acquire(ctx);
     blk_aio_complete(acb);
+    aio_context_release(ctx);
 }
 
 static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes,
diff --git a/block/curl.c b/block/curl.c
index XXXXXXX..XXXXXXX 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
 {
     CURLState *state;
     int running;
+    int ret = -EINPROGRESS;
 
     CURLAIOCB *acb = p;
-    BDRVCURLState *s = acb->common.bs->opaque;
+    BlockDriverState *bs = acb->common.bs;
+    BDRVCURLState *s = bs->opaque;
+    AioContext *ctx = bdrv_get_aio_context(bs);
 
     size_t start = acb->sector_num * BDRV_SECTOR_SIZE;
     size_t end;
 
+    aio_context_acquire(ctx);
+
     // In case we have the requested data already (e.g. read-ahead),
     // we can just call the callback and be done.
     switch (curl_find_buf(s, start, acb->nb_sectors * BDRV_SECTOR_SIZE, acb)) {
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
             qemu_aio_unref(acb);
             // fall through
         case FIND_RET_WAIT:
-            return;
+            goto out;
         default:
             break;
     }
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
     // No cache found, so let's start a new request
     state = curl_init_state(acb->common.bs, s);
     if (!state) {
-        acb->common.cb(acb->common.opaque, -EIO);
-        qemu_aio_unref(acb);
-        return;
+        ret = -EIO;
+        goto out;
     }
 
     acb->start = 0;
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
     state->orig_buf = g_try_malloc(state->buf_len);
     if (state->buf_len && state->orig_buf == NULL) {
         curl_clean_state(state);
-        acb->common.cb(acb->common.opaque, -ENOMEM);
-        qemu_aio_unref(acb);
-        return;
+        ret = -ENOMEM;
+        goto out;
     }
     state->acb[0] = acb;
 
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
 
     /* Tell curl it needs to kick things off */
     curl_multi_socket_action(s->multi, CURL_SOCKET_TIMEOUT, 0, &running);
+
+out:
+    if (ret != -EINPROGRESS) {
+        acb->common.cb(acb->common.opaque, ret);
+        qemu_aio_unref(acb);
+    }
+    aio_context_release(ctx);
 }
 
 static BlockAIOCB *curl_aio_readv(BlockDriverState *bs,
diff --git a/block/gluster.c b/block/gluster.c
index XXXXXXX..XXXXXXX 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -XXX,XX +XXX,XX @@ static struct glfs *qemu_gluster_init(BlockdevOptionsGluster *gconf,
     return qemu_gluster_glfs_init(gconf, errp);
 }
 
-static void qemu_gluster_complete_aio(void *opaque)
-{
-    GlusterAIOCB *acb = (GlusterAIOCB *)opaque;
-
-    qemu_coroutine_enter(acb->coroutine);
-}
-
 /*
  * AIO callback routine called from GlusterFS thread.
  */
@@ -XXX,XX +XXX,XX @@ static void gluster_finish_aiocb(struct glfs_fd *fd, ssize_t ret, void *arg)
         acb->ret = -EIO; /* Partial read/write - fail it */
     }
 
-    aio_bh_schedule_oneshot(acb->aio_context, qemu_gluster_complete_aio, acb);
+    aio_co_schedule(acb->aio_context, acb->coroutine);
 }
 
 static void qemu_gluster_parse_flags(int bdrv_flags, int *open_flags)
diff --git a/block/io.c b/block/io.c
index XXXXXXX..XXXXXXX 100644
--- a/block/io.c
+++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ static void bdrv_co_drain_bh_cb(void *opaque)
     bdrv_dec_in_flight(bs);
     bdrv_drained_begin(bs);
     data->done = true;
-    qemu_coroutine_enter(co);
+    aio_co_wake(co);
 }
 
 static void coroutine_fn bdrv_co_yield_to_drain(BlockDriverState *bs)
@@ -XXX,XX +XXX,XX @@ static void bdrv_co_complete(BlockAIOCBCoroutine *acb)
 static void bdrv_co_em_bh(void *opaque)
 {
     BlockAIOCBCoroutine *acb = opaque;
+    BlockDriverState *bs = acb->common.bs;
+    AioContext *ctx = bdrv_get_aio_context(bs);
 
     assert(!acb->need_bh);
+    aio_context_acquire(ctx);
     bdrv_co_complete(acb);
+    aio_context_release(ctx);
 }
 
 static void bdrv_co_maybe_schedule_bh(BlockAIOCBCoroutine *acb)
diff --git a/block/iscsi.c b/block/iscsi.c
index XXXXXXX..XXXXXXX 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -XXX,XX +XXX,XX @@ static void
 iscsi_bh_cb(void *p)
 {
     IscsiAIOCB *acb = p;
+    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 
     qemu_bh_delete(acb->bh);
 
     g_free(acb->buf);
     acb->buf = NULL;
 
+    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, acb->status);
+    aio_context_release(ctx);
 
     if (acb->task != NULL) {
         scsi_free_scsi_task(acb->task);
@@ -XXX,XX +XXX,XX @@ iscsi_schedule_bh(IscsiAIOCB *acb)
 static void iscsi_co_generic_bh_cb(void *opaque)
 {
     struct IscsiTask *iTask = opaque;
+
     iTask->complete = 1;
-    qemu_coroutine_enter(iTask->co);
+    aio_co_wake(iTask->co);
 }
 
 static void iscsi_retry_timer_expired(void *opaque)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index XXXXXXX..XXXXXXX 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -XXX,XX +XXX,XX @@ struct LinuxAioState {
     io_context_t ctx;
     EventNotifier e;
 
-    /* io queue for submit at batch */
+    /* io queue for submit at batch.  Protected by AioContext lock. */
     LaioQueue io_q;
 
-    /* I/O completion processing */
+    /* I/O completion processing.  Only runs in I/O thread.  */
     QEMUBH *completion_bh;
     int event_idx;
     int event_max;
@@ -XXX,XX +XXX,XX @@ static inline ssize_t io_event_ret(struct io_event *ev)
  */
 static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
 {
+    LinuxAioState *s = laiocb->ctx;
     int ret;
 
     ret = laiocb->ret;
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
     }
 
     laiocb->ret = ret;
+    aio_context_acquire(s->aio_context);
     if (laiocb->co) {
         /* If the coroutine is already entered it must be in ioq_submit() and
          * will notice laio->ret has been filled in when it eventually runs
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
         laiocb->common.cb(laiocb->common.opaque, ret);
         qemu_aio_unref(laiocb);
     }
+    aio_context_release(s->aio_context);
 }
 
 /**
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completions(LinuxAioState *s)
 static void qemu_laio_process_completions_and_submit(LinuxAioState *s)
 {
     qemu_laio_process_completions(s);
+
+    aio_context_acquire(s->aio_context);
     if (!s->io_q.plugged && !QSIMPLEQ_EMPTY(&s->io_q.pending)) {
         ioq_submit(s);
     }
+    aio_context_release(s->aio_context);
 }
 
 static void qemu_laio_completion_bh(void *opaque)
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_completion_cb(EventNotifier *e)
     LinuxAioState *s = container_of(e, LinuxAioState, e);
 
     if (event_notifier_test_and_clear(&s->e)) {
-        aio_context_acquire(s->aio_context);
         qemu_laio_process_completions_and_submit(s);
-        aio_context_release(s->aio_context);
     }
 }
 
@@ -XXX,XX +XXX,XX @@ static bool qemu_laio_poll_cb(void *opaque)
         return false;
     }
 
-    aio_context_acquire(s->aio_context);
     qemu_laio_process_completions_and_submit(s);
-    aio_context_release(s->aio_context);
     return true;
 }
 
@@ -XXX,XX +XXX,XX @@ void laio_detach_aio_context(LinuxAioState *s, AioContext *old_context)
 {
     aio_set_event_notifier(old_context, &s->e, false, NULL, NULL);
     qemu_bh_delete(s->completion_bh);
+    s->aio_context = NULL;
 }
 
 void laio_attach_aio_context(LinuxAioState *s, AioContext *new_context)
diff --git a/block/nfs.c b/block/nfs.c
index XXXXXXX..XXXXXXX 100644
--- a/block/nfs.c
+++ b/block/nfs.c
@@ -XXX,XX +XXX,XX @@ static void nfs_co_init_task(BlockDriverState *bs, NFSRPC *task)
 static void nfs_co_generic_bh_cb(void *opaque)
 {
     NFSRPC *task = opaque;
+
     task->complete = 1;
-    qemu_coroutine_enter(task->co);
+    aio_co_wake(task->co);
 }
 
 static void
diff --git a/block/null.c b/block/null.c
index XXXXXXX..XXXXXXX 100644
--- a/block/null.c
+++ b/block/null.c
@@ -XXX,XX +XXX,XX @@ static const AIOCBInfo null_aiocb_info = {
 static void null_bh_cb(void *opaque)
 {
     NullAIOCB *acb = opaque;
+    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
+
+    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, 0);
+    aio_context_release(ctx);
     qemu_aio_unref(acb);
 }
 
diff --git a/block/qed.c b/block/qed.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -XXX,XX +XXX,XX @@ static void qed_update_l2_table(BDRVQEDState *s, QEDTable *table, int index,
 static void qed_aio_complete_bh(void *opaque)
 {
     QEDAIOCB *acb = opaque;
+    BDRVQEDState *s = acb_to_s(acb);
     BlockCompletionFunc *cb = acb->common.cb;
     void *user_opaque = acb->common.opaque;
     int ret = acb->bh_ret;
@@ -XXX,XX +XXX,XX @@ static void qed_aio_complete_bh(void *opaque)
     qemu_aio_unref(acb);
 
     /* Invoke callback */
+    qed_acquire(s);
     cb(user_opaque, ret);
+    qed_release(s);
 }
 
 static void qed_aio_complete(QEDAIOCB *acb, int ret)
diff --git a/block/rbd.c b/block/rbd.c
index XXXXXXX..XXXXXXX 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -XXX,XX +XXX,XX @@ shutdown:
 static void qemu_rbd_complete_aio(RADOSCB *rcb)
 {
     RBDAIOCB *acb = rcb->acb;
+    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
     int64_t r;
 
     r = rcb->ret;
@@ -XXX,XX +XXX,XX @@ static void qemu_rbd_complete_aio(RADOSCB *rcb)
         qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
     }
     qemu_vfree(acb->bounce);
+
+    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
+    aio_context_release(ctx);
 
     qemu_aio_unref(acb);
 }
diff --git a/dma-helpers.c b/dma-helpers.c
index XXXXXXX..XXXXXXX 100644
--- a/dma-helpers.c
+++ b/dma-helpers.c
@@ -XXX,XX +XXX,XX @@ static void dma_blk_cb(void *opaque, int ret)
                                 QEMU_ALIGN_DOWN(dbs->iov.size, dbs->align));
     }
 
+    aio_context_acquire(dbs->ctx);
     dbs->acb = dbs->io_func(dbs->offset, &dbs->iov,
                             dma_blk_cb, dbs, dbs->io_func_opaque);
+    aio_context_release(dbs->ctx);
     assert(dbs->acb);
 }
 
diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_dma_restart_bh(void *opaque)
 
     s->rq = NULL;
 
+    aio_context_acquire(blk_get_aio_context(s->conf.conf.blk));
     while (req) {
         VirtIOBlockReq *next = req->next;
         if (virtio_blk_handle_request(req, &mrb)) {
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_dma_restart_bh(void *opaque)
     if (mrb.num_reqs) {
         virtio_blk_submit_multireq(s->blk, &mrb);
     }
+    aio_context_release(blk_get_aio_context(s->conf.conf.blk));
 }
 
 static void virtio_blk_dma_restart_cb(void *opaque, int running,
diff --git a/hw/scsi/scsi-bus.c b/hw/scsi/scsi-bus.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/scsi/scsi-bus.c
+++ b/hw/scsi/scsi-bus.c
@@ -XXX,XX +XXX,XX @@ static void scsi_dma_restart_bh(void *opaque)
     qemu_bh_delete(s->bh);
     s->bh = NULL;
 
+    aio_context_acquire(blk_get_aio_context(s->conf.blk));
     QTAILQ_FOREACH_SAFE(req, &s->requests, next, next) {
         scsi_req_ref(req);
         if (req->retry) {
@@ -XXX,XX +XXX,XX @@ static void scsi_dma_restart_bh(void *opaque)
         }
         scsi_req_unref(req);
     }
+    aio_context_release(blk_get_aio_context(s->conf.blk));
 }
 
 void scsi_req_retry(SCSIRequest *req)
diff --git a/util/async.c b/util/async.c
index XXXXXXX..XXXXXXX 100644
--- a/util/async.c
+++ b/util/async.c
@@ -XXX,XX +XXX,XX @@ int aio_bh_poll(AioContext *ctx)
                 ret = 1;
             }
             bh->idle = 0;
-            aio_context_acquire(ctx);
             aio_bh_call(bh);
-            aio_context_release(ctx);
         }
         if (bh->deleted) {
             deleted = true;
@@ -XXX,XX +XXX,XX @@ static void co_schedule_bh_cb(void *opaque)
         Coroutine *co = QSLIST_FIRST(&straight);
         QSLIST_REMOVE_HEAD(&straight, co_scheduled_next);
         trace_aio_co_schedule_bh_cb(ctx, co);
+        aio_context_acquire(ctx);
         qemu_coroutine_enter(co);
+        aio_context_release(ctx);
     }
 }
 
diff --git a/util/thread-pool.c b/util/thread-pool.c
index XXXXXXX..XXXXXXX 100644
--- a/util/thread-pool.c
+++ b/util/thread-pool.c
@@ -XXX,XX +XXX,XX @@ static void thread_pool_completion_bh(void *opaque)
     ThreadPool *pool = opaque;
     ThreadPoolElement *elem, *next;
 
+    aio_context_acquire(pool->ctx);
 restart:
     QLIST_FOREACH_SAFE(elem, &pool->head, all, next) {
         if (elem->state != THREAD_DONE) {
@@ -XXX,XX +XXX,XX @@ restart:
             qemu_aio_unref(elem);
         }
     }
+    aio_context_release(pool->ctx);
 }
 
 static void thread_pool_cancel(BlockAIOCB *acb)
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 20170213135235.12274-16-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 block/archipelago.c    |  3 ---
 block/block-backend.c  |  7 -------
 block/curl.c           |  2 +-
 block/io.c             |  6 +-----
 block/iscsi.c          |  3 ---
 block/linux-aio.c      |  5 +----
 block/mirror.c         | 12 +++++++++---
 block/null.c           |  8 --------
 block/qed-cluster.c    |  2 ++
 block/qed-table.c      | 12 ++++++++++--
 block/qed.c            |  4 ++--
 block/rbd.c            |  4 ----
 block/win32-aio.c      |  3 ---
 hw/block/virtio-blk.c  | 12 +++++++++++-
 hw/scsi/scsi-disk.c    | 15 +++++++++++++++
 hw/scsi/scsi-generic.c | 20 +++++++++++++++++---
 util/thread-pool.c     |  4 +++-
 17 files changed, 72 insertions(+), 50 deletions(-)

diff --git a/block/archipelago.c b/block/archipelago.c
index XXXXXXX..XXXXXXX 100644
--- a/block/archipelago.c
+++ b/block/archipelago.c
@@ -XXX,XX +XXX,XX @@ static void qemu_archipelago_complete_aio(void *opaque)
 {
     AIORequestData *reqdata = (AIORequestData *) opaque;
     ArchipelagoAIOCB *aio_cb = (ArchipelagoAIOCB *) reqdata->aio_cb;
-    AioContext *ctx = bdrv_get_aio_context(aio_cb->common.bs);
 
-    aio_context_acquire(ctx);
     aio_cb->common.cb(aio_cb->common.opaque, aio_cb->ret);
-    aio_context_release(ctx);
     aio_cb->status = 0;
 
     qemu_aio_unref(aio_cb);
diff --git a/block/block-backend.c b/block/block-backend.c
index XXXXXXX..XXXXXXX 100644
--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -XXX,XX +XXX,XX @@ int blk_make_zero(BlockBackend *blk, BdrvRequestFlags flags)
 static void error_callback_bh(void *opaque)
 {
     struct BlockBackendAIOCB *acb = opaque;
-    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 
     bdrv_dec_in_flight(acb->common.bs);
-    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, acb->ret);
-    aio_context_release(ctx);
     qemu_aio_unref(acb);
 }
 
@@ -XXX,XX +XXX,XX @@ static void blk_aio_complete(BlkAioEmAIOCB *acb)
 static void blk_aio_complete_bh(void *opaque)
 {
     BlkAioEmAIOCB *acb = opaque;
-    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
-
     assert(acb->has_returned);
-    aio_context_acquire(ctx);
     blk_aio_complete(acb);
-    aio_context_release(ctx);
 }
 
 static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes,
diff --git a/block/curl.c b/block/curl.c
index XXXXXXX..XXXXXXX 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -XXX,XX +XXX,XX @@ static void curl_readv_bh_cb(void *p)
     curl_multi_socket_action(s->multi, CURL_SOCKET_TIMEOUT, 0, &running);
 
 out:
+    aio_context_release(ctx);
     if (ret != -EINPROGRESS) {
         acb->common.cb(acb->common.opaque, ret);
         qemu_aio_unref(acb);
     }
-    aio_context_release(ctx);
 }
 
 static BlockAIOCB *curl_aio_readv(BlockDriverState *bs,
diff --git a/block/io.c b/block/io.c
index XXXXXXX..XXXXXXX 100644
--- a/block/io.c
+++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ static void bdrv_co_io_em_complete(void *opaque, int ret)
     CoroutineIOCompletion *co = opaque;
 
     co->ret = ret;
-    qemu_coroutine_enter(co->coroutine);
+    aio_co_wake(co->coroutine);
 }
 
 static int coroutine_fn bdrv_driver_preadv(BlockDriverState *bs,
@@ -XXX,XX +XXX,XX @@ static void bdrv_co_complete(BlockAIOCBCoroutine *acb)
 static void bdrv_co_em_bh(void *opaque)
 {
     BlockAIOCBCoroutine *acb = opaque;
-    BlockDriverState *bs = acb->common.bs;
-    AioContext *ctx = bdrv_get_aio_context(bs);
 
     assert(!acb->need_bh);
-    aio_context_acquire(ctx);
     bdrv_co_complete(acb);
-    aio_context_release(ctx);
 }
 
 static void bdrv_co_maybe_schedule_bh(BlockAIOCBCoroutine *acb)
diff --git a/block/iscsi.c b/block/iscsi.c
index XXXXXXX..XXXXXXX 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -XXX,XX +XXX,XX @@ static void
 iscsi_bh_cb(void *p)
 {
     IscsiAIOCB *acb = p;
-    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
 
     qemu_bh_delete(acb->bh);
 
     g_free(acb->buf);
     acb->buf = NULL;
 
-    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, acb->status);
-    aio_context_release(ctx);
 
     if (acb->task != NULL) {
         scsi_free_scsi_task(acb->task);
diff --git a/block/linux-aio.c b/block/linux-aio.c
index XXXXXXX..XXXXXXX 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -XXX,XX +XXX,XX @@ static inline ssize_t io_event_ret(struct io_event *ev)
  */
 static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
 {
-    LinuxAioState *s = laiocb->ctx;
     int ret;
 
     ret = laiocb->ret;
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
     }
 
     laiocb->ret = ret;
-    aio_context_acquire(s->aio_context);
     if (laiocb->co) {
         /* If the coroutine is already entered it must be in ioq_submit() and
          * will notice laio->ret has been filled in when it eventually runs
@@ -XXX,XX +XXX,XX @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
          * that!
          */
         if (!qemu_coroutine_entered(laiocb->co)) {
-            qemu_coroutine_enter(laiocb->co);
+            aio_co_wake(laiocb->co);
         }
     } else {
         laiocb->common.cb(laiocb->common.opaque, ret);
         qemu_aio_unref(laiocb);
     }
-    aio_context_release(s->aio_context);
 }
 
 /**
diff --git a/block/mirror.c b/block/mirror.c
index XXXXXXX..XXXXXXX 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -XXX,XX +XXX,XX @@ static void mirror_write_complete(void *opaque, int ret)
 {
     MirrorOp *op = opaque;
     MirrorBlockJob *s = op->s;
+
+    aio_context_acquire(blk_get_aio_context(s->common.blk));
     if (ret < 0) {
         BlockErrorAction action;
 
@@ -XXX,XX +XXX,XX @@ static void mirror_write_complete(void *opaque, int ret)
         }
     }
     mirror_iteration_done(op, ret);
+    aio_context_release(blk_get_aio_context(s->common.blk));
 }
 
 static void mirror_read_complete(void *opaque, int ret)
 {
     MirrorOp *op = opaque;
     MirrorBlockJob *s = op->s;
+
+    aio_context_acquire(blk_get_aio_context(s->common.blk));
     if (ret < 0) {
         BlockErrorAction action;
 
@@ -XXX,XX +XXX,XX @@ static void mirror_read_complete(void *opaque, int ret)
         }
 
         mirror_iteration_done(op, ret);
-        return;
+    } else {
+        blk_aio_pwritev(s->target, op->sector_num * BDRV_SECTOR_SIZE, &op->qiov,
+                        0, mirror_write_complete, op);
     }
-    blk_aio_pwritev(s->target, op->sector_num * BDRV_SECTOR_SIZE, &op->qiov,
-                    0, mirror_write_complete, op);
+    aio_context_release(blk_get_aio_context(s->common.blk));
 }
 
 static inline void mirror_clip_sectors(MirrorBlockJob *s,
diff --git a/block/null.c b/block/null.c
index XXXXXXX..XXXXXXX 100644
--- a/block/null.c
+++ b/block/null.c
@@ -XXX,XX +XXX,XX @@ static const AIOCBInfo null_aiocb_info = {
 static void null_bh_cb(void *opaque)
 {
     NullAIOCB *acb = opaque;
-    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
-
-    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, 0);
-    aio_context_release(ctx);
     qemu_aio_unref(acb);
 }
 
 static void null_timer_cb(void *opaque)
 {
     NullAIOCB *acb = opaque;
-    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
-
-    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, 0);
-    aio_context_release(ctx);
     timer_deinit(&acb->timer);
     qemu_aio_unref(acb);
 }
diff --git a/block/qed-cluster.c b/block/qed-cluster.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qed-cluster.c
+++ b/block/qed-cluster.c
@@ -XXX,XX +XXX,XX @@ static void qed_find_cluster_cb(void *opaque, int ret)
     unsigned int index;
     unsigned int n;
 
+    qed_acquire(s);
     if (ret) {
         goto out;
     }
@@ -XXX,XX +XXX,XX @@ static void qed_find_cluster_cb(void *opaque, int ret)
 
 out:
     find_cluster_cb->cb(find_cluster_cb->opaque, ret, offset, len);
+    qed_release(s);
     g_free(find_cluster_cb);
 }
 
diff --git a/block/qed-table.c b/block/qed-table.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qed-table.c
+++ b/block/qed-table.c
@@ -XXX,XX +XXX,XX @@ static void qed_read_table_cb(void *opaque, int ret)
 {
     QEDReadTableCB *read_table_cb = opaque;
     QEDTable *table = read_table_cb->table;
+    BDRVQEDState *s = read_table_cb->s;
     int noffsets = read_table_cb->qiov.size / sizeof(uint64_t);
     int i;
 
@@ -XXX,XX +XXX,XX @@ static void qed_read_table_cb(void *opaque, int ret)
     }
 
     /* Byteswap offsets */
+    qed_acquire(s);
     for (i = 0; i < noffsets; i++) {
         table->offsets[i] = le64_to_cpu(table->offsets[i]);
     }
+    qed_release(s);
 
 out:
     /* Completion */
-    trace_qed_read_table_cb(read_table_cb->s, read_table_cb->table, ret);
+    trace_qed_read_table_cb(s, read_table_cb->table, ret);
     gencb_complete(&read_table_cb->gencb, ret);
 }
 
@@ -XXX,XX +XXX,XX @@ typedef struct {
 static void qed_write_table_cb(void *opaque, int ret)
 {
     QEDWriteTableCB *write_table_cb = opaque;
+    BDRVQEDState *s = write_table_cb->s;
 
-    trace_qed_write_table_cb(write_table_cb->s,
+    trace_qed_write_table_cb(s,
                              write_table_cb->orig_table,
                              write_table_cb->flush,
                              ret);
@@ -XXX,XX +XXX,XX @@ static void qed_write_table_cb(void *opaque, int ret)
     if (write_table_cb->flush) {
         /* We still need to flush first */
         write_table_cb->flush = false;
+        qed_acquire(s);
         bdrv_aio_flush(write_table_cb->s->bs, qed_write_table_cb,
                        write_table_cb);
+        qed_release(s);
         return;
     }
 
@@ -XXX,XX +XXX,XX @@ static void qed_read_l2_table_cb(void *opaque, int ret)
     CachedL2Table *l2_table = request->l2_table;
     uint64_t l2_offset = read_l2_table_cb->l2_offset;
 
+    qed_acquire(s);
     if (ret) {
         /* can't trust loaded L2 table anymore */
         qed_unref_l2_cache_entry(l2_table);
@@ -XXX,XX +XXX,XX @@ static void qed_read_l2_table_cb(void *opaque, int ret)
         request->l2_table = qed_find_l2_cache_entry(&s->l2_cache, l2_offset);
         assert(request->l2_table != NULL);
     }
+    qed_release(s);
 
     gencb_complete(&read_l2_table_cb->gencb, ret);
 }
diff --git a/block/qed.c b/block/qed.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -XXX,XX +XXX,XX @@ static void qed_is_allocated_cb(void *opaque, int ret, uint64_t offset, size_t l
     }
 
     if (cb->co) {
-        qemu_coroutine_enter(cb->co);
+        aio_co_wake(cb->co);
     }
 }
 
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn qed_co_pwrite_zeroes_cb(void *opaque, int ret)
     cb->done = true;
     cb->ret = ret;
     if (cb->co) {
-        qemu_coroutine_enter(cb->co);
+        aio_co_wake(cb->co);
     }
 }
 
diff --git a/block/rbd.c b/block/rbd.c
index XXXXXXX..XXXXXXX 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -XXX,XX +XXX,XX @@ shutdown:
 static void qemu_rbd_complete_aio(RADOSCB *rcb)
 {
     RBDAIOCB *acb = rcb->acb;
-    AioContext *ctx = bdrv_get_aio_context(acb->common.bs);
     int64_t r;
 
     r = rcb->ret;
@@ -XXX,XX +XXX,XX @@ static void qemu_rbd_complete_aio(RADOSCB *rcb)
         qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
     }
     qemu_vfree(acb->bounce);
-
-    aio_context_acquire(ctx);
     acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
-    aio_context_release(ctx);
 
     qemu_aio_unref(acb);
 }
diff --git a/block/win32-aio.c b/block/win32-aio.c
index XXXXXXX..XXXXXXX 100644
--- a/block/win32-aio.c
+++ b/block/win32-aio.c
@@ -XXX,XX +XXX,XX @@ static void win32_aio_process_completion(QEMUWin32AIOState *s,
         qemu_vfree(waiocb->buf);
     }
 
-
-    aio_context_acquire(s->aio_ctx);
     waiocb->common.cb(waiocb->common.opaque, ret);
-    aio_context_release(s->aio_ctx);
     qemu_aio_unref(waiocb);
 }
 
diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -XXX,XX +XXX,XX @@ static int virtio_blk_handle_rw_error(VirtIOBlockReq *req, int error,
 static void virtio_blk_rw_complete(void *opaque, int ret)
 {
     VirtIOBlockReq *next = opaque;
+    VirtIOBlock *s = next->dev;
 
+    aio_context_acquire(blk_get_aio_context(s->conf.conf.blk));
     while (next) {
         VirtIOBlockReq *req = next;
         next = req->mr_next;
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_rw_complete(void *opaque, int ret)
         block_acct_done(blk_get_stats(req->dev->blk), &req->acct);
         virtio_blk_free_request(req);
     }
+    aio_context_release(blk_get_aio_context(s->conf.conf.blk));
 }
 
 static void virtio_blk_flush_complete(void *opaque, int ret)
 {
     VirtIOBlockReq *req = opaque;
+    VirtIOBlock *s = req->dev;
 
+    aio_context_acquire(blk_get_aio_context(s->conf.conf.blk));
     if (ret) {
         if (virtio_blk_handle_rw_error(req, -ret, 0)) {
-            return;
+            goto out;
         }
     }
 
     virtio_blk_req_complete(req, VIRTIO_BLK_S_OK);
     block_acct_done(blk_get_stats(req->dev->blk), &req->acct);
     virtio_blk_free_request(req);
+
+out:
+    aio_context_release(blk_get_aio_context(s->conf.conf.blk));
 }
 
 #ifdef __linux__
@@ -XXX,XX +XXX,XX @@ static void virtio_blk_ioctl_complete(void *opaque, int status)
     virtio_stl_p(vdev, &scsi->data_len, hdr->dxfer_len);
 
 out:
+    aio_context_acquire(blk_get_aio_context(s->conf.conf.blk));
     virtio_blk_req_complete(req, status);
     virtio_blk_free_request(req);
+    aio_context_release(blk_get_aio_context(s->conf.conf.blk));
     g_free(ioctl_req);
 }
 
diff --git a/hw/scsi/scsi-disk.c b/hw/scsi/scsi-disk.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/scsi/scsi-disk.c
+++ b/hw/scsi/scsi-disk.c
@@ -XXX,XX +XXX,XX @@ static void scsi_aio_complete(void *opaque, int ret)
 
     assert(r->req.aiocb != NULL);
     r->req.aiocb = NULL;
+    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
     if (scsi_disk_req_check_error(r, ret, true)) {
         goto done;
     }
@@ -XXX,XX +XXX,XX @@ static void scsi_aio_complete(void *opaque, int ret)
     scsi_req_complete(&r->req, GOOD);
 
 done:
+    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
     scsi_req_unref(&r->req);
 }
 
@@ -XXX,XX +XXX,XX @@ static void scsi_dma_complete(void *opaque, int ret)
     assert(r->req.aiocb != NULL);
     r->req.aiocb = NULL;
 
+    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
     if (ret < 0) {
         block_acct_failed(blk_get_stats(s->qdev.conf.blk), &r->acct);
     } else {
         block_acct_done(blk_get_stats(s->qdev.conf.blk), &r->acct);
     }
     scsi_dma_complete_noio(r, ret);
+    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
 }
 
 static void scsi_read_complete(void * opaque, int ret)
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
 
     assert(r->req.aiocb != NULL);
     r->req.aiocb = NULL;
+    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
     if (scsi_disk_req_check_error(r, ret, true)) {
         goto done;
     }
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
 
 done:
     scsi_req_unref(&r->req);
+    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
 }
 
 /* Actually issue a read to the block device.  */
@@ -XXX,XX +XXX,XX @@ static void scsi_do_read_cb(void *opaque, int ret)
     assert (r->req.aiocb != NULL);
     r->req.aiocb = NULL;
 
+    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
     if (ret < 0) {
         block_acct_failed(blk_get_stats(s->qdev.conf.blk), &r->acct);
     } else {
         block_acct_done(blk_get_stats(s->qdev.conf.blk), &r->acct);
     }
     scsi_do_read(opaque, ret);
+    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
 }
 
 /* Read more data from scsi device into buffer.  */
@@ -XXX,XX +XXX,XX @@ static void scsi_write_complete(void * opaque, int ret)
     assert (r->req.aiocb != NULL);
     r->req.aiocb = NULL;
 
+    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
     if (ret < 0) {
         block_acct_failed(blk_get_stats(s->qdev.conf.blk), &r->acct);
     } else {
         block_acct_done(blk_get_stats(s->qdev.conf.blk), &r->acct);
     }
     scsi_write_complete_noio(r, ret);
+    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
 }
 
 static void scsi_write_data(SCSIRequest *req)
@@ -XXX,XX +XXX,XX @@ static void scsi_unmap_complete(void *opaque, int ret)
 {
     UnmapCBData *data = opaque;
     SCSIDiskReq *r = data->r;
+    SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev);
 
     assert(r->req.aiocb != NULL);
     r->req.aiocb = NULL;
 
+    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
     scsi_unmap_complete_noio(data, ret);
+    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
 }
 
 static void scsi_disk_emulate_unmap(SCSIDiskReq *r, uint8_t *inbuf)
@@ -XXX,XX +XXX,XX @@ static void scsi_write_same_complete(void *opaque, int ret)
 
     assert(r->req.aiocb != NULL);
     r->req.aiocb = NULL;
+    aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
     if (scsi_disk_req_check_error(r, ret, true)) {
         goto done;
     }
@@ -XXX,XX +XXX,XX @@ done:
     scsi_req_unref(&r->req);
     qemu_vfree(data->iov.iov_base);
     g_free(data);
+    aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
 }
 
 static void scsi_disk_emulate_write_same(SCSIDiskReq *r, uint8_t *inbuf)
diff --git a/hw/scsi/scsi-generic.c b/hw/scsi/scsi-generic.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/scsi/scsi-generic.c
+++ b/hw/scsi/scsi-generic.c
@@ -XXX,XX +XXX,XX @@ done:
 static void scsi_command_complete(void *opaque, int ret)
 {
     SCSIGenericReq *r = (SCSIGenericReq *)opaque;
+    SCSIDevice *s = r->req.dev;
 
     assert(r->req.aiocb != NULL);
     r->req.aiocb = NULL;
+
+    aio_context_acquire(blk_get_aio_context(s->conf.blk));
     scsi_command_complete_noio(r, ret);
+    aio_context_release(blk_get_aio_context(s->conf.blk));
 }
 
 static int execute_command(BlockBackend *blk,
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
     assert(r->req.aiocb != NULL);
     r->req.aiocb = NULL;
 
+    aio_context_acquire(blk_get_aio_context(s->conf.blk));
+
     if (ret || r->req.io_canceled) {
         scsi_command_complete_noio(r, ret);
-        return;
+        goto done;
     }
 
     len = r->io_header.dxfer_len - r->io_header.resid;
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
     r->len = -1;
     if (len == 0) {
         scsi_command_complete_noio(r, 0);
-        return;
+        goto done;
     }
 
     /* Snoop READ CAPACITY output to set the blocksize.  */
@@ -XXX,XX +XXX,XX @@ static void scsi_read_complete(void * opaque, int ret)
     }
     scsi_req_data(&r->req, len);
     scsi_req_unref(&r->req);
+
+done:
+    aio_context_release(blk_get_aio_context(s->conf.blk));
 }
 
 /* Read more data from scsi device into buffer.  */
@@ -XXX,XX +XXX,XX @@ static void scsi_write_complete(void * opaque, int ret)
     assert(r->req.aiocb != NULL);
     r->req.aiocb = NULL;
 
+    aio_context_acquire(blk_get_aio_context(s->conf.blk));
+
     if (ret || r->req.io_canceled) {
         scsi_command_complete_noio(r, ret);
-        return;
+        goto done;
     }
 
     if (r->req.cmd.buf[0] == MODE_SELECT && r->req.cmd.buf[4] == 12 &&
@@ -XXX,XX +XXX,XX @@ static void scsi_write_complete(void * opaque, int ret)
     }
 
     scsi_command_complete_noio(r, ret);
+
+done:
+    aio_context_release(blk_get_aio_context(s->conf.blk));
 }
 
 /* Write data to a scsi device.  Returns nonzero on failure.
diff --git a/util/thread-pool.c b/util/thread-pool.c
index XXXXXXX..XXXXXXX 100644
--- a/util/thread-pool.c
+++ b/util/thread-pool.c
@@ -XXX,XX +XXX,XX @@ restart:
              */
             qemu_bh_schedule(pool->completion_bh);
 
+            aio_context_release(pool->ctx);
             elem->common.cb(elem->common.opaque, elem->ret);
+            aio_context_acquire(pool->ctx);
             qemu_aio_unref(elem);
             goto restart;
         } else {
@@ -XXX,XX +XXX,XX @@ static void thread_pool_co_cb(void *opaque, int ret)
     ThreadPoolCo *co = opaque;
 
     co->ret = ret;
-    qemu_coroutine_enter(co->co);
+    aio_co_wake(co->co);
 }
 
 int coroutine_fn thread_pool_submit_co(ThreadPool *pool, ThreadPoolFunc *func,
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

This patch prepares for the removal of unnecessary lockcnt inc/dec pairs.
Extract the dispatching loop for file descriptor handlers into a new
function aio_dispatch_handlers, and then inline aio_dispatch into
aio_poll.

aio_dispatch can now become void.

diff --git a/include/block/aio.h b/include/block/aio.h
index XXXXXXX..XXXXXXX 100644
--- a/include/block/aio.h
+++ b/include/block/aio.h
@@ -XXX,XX +XXX,XX @@ bool aio_pending(AioContext *ctx);
 /* Dispatch any pending callbacks from the GSource attached to the AioContext.
  *
  * This is used internally in the implementation of the GSource.
- *
- * @dispatch_fds: true to process fds, false to skip them
- *                (can be used as an optimization by callers that know there
- *                are no fds ready)
  */
-bool aio_dispatch(AioContext *ctx, bool dispatch_fds);
+void aio_dispatch(AioContext *ctx);
 
 /* Progress in completing AIO work to occur.  This can issue new pending
  * aio as a result of executing I/O completion or bh callbacks.
diff --git a/util/aio-posix.c b/util/aio-posix.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-posix.c
+++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
     AioHandler *node, *tmp;
     bool progress = false;
 
-    /*
-     * We have to walk very carefully in case aio_set_fd_handler is
-     * called while we're walking.
-     */
-    qemu_lockcnt_inc(&ctx->list_lock);
-
     QLIST_FOREACH_SAFE_RCU(node, &ctx->aio_handlers, node, tmp) {
         int revents;
 
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
         }
     }
 
-    qemu_lockcnt_dec(&ctx->list_lock);
     return progress;
 }
 
-/*
- * Note that dispatch_fds == false has the side-effect of post-poning the
- * freeing of deleted handlers.
- */
-bool aio_dispatch(AioContext *ctx, bool dispatch_fds)
+void aio_dispatch(AioContext *ctx)
 {
-    bool progress;
+    aio_bh_poll(ctx);
 
-    /*
-     * If there are callbacks left that have been queued, we need to call them.
-     * Do not call select in this case, because it is possible that the caller
-     * does not need a complete flush (as is the case for aio_poll loops).
-     */
-    progress = aio_bh_poll(ctx);
+    qemu_lockcnt_inc(&ctx->list_lock);
+    aio_dispatch_handlers(ctx);
+    qemu_lockcnt_dec(&ctx->list_lock);
 
-    if (dispatch_fds) {
-        progress |= aio_dispatch_handlers(ctx);
-    }
-
-    /* Run our timers */
-    progress |= timerlistgroup_run_timers(&ctx->tlg);
-
-    return progress;
+    timerlistgroup_run_timers(&ctx->tlg);
 }
 
 /* These thread-local variables are used only in a small part of aio_poll
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
     npfd = 0;
     qemu_lockcnt_dec(&ctx->list_lock);
 
-    /* Run dispatch even if there were no readable fds to run timers */
-    if (aio_dispatch(ctx, ret > 0)) {
-        progress = true;
+    progress |= aio_bh_poll(ctx);
+
+    if (ret > 0) {
+        qemu_lockcnt_inc(&ctx->list_lock);
+        progress |= aio_dispatch_handlers(ctx);
+        qemu_lockcnt_dec(&ctx->list_lock);
     }
 
+    progress |= timerlistgroup_run_timers(&ctx->tlg);
+
     return progress;
 }
 
diff --git a/util/aio-win32.c b/util/aio-win32.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-win32.c
+++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
     return progress;
 }
 
-bool aio_dispatch(AioContext *ctx, bool dispatch_fds)
+void aio_dispatch(AioContext *ctx)
 {
-    bool progress;
-
-    progress = aio_bh_poll(ctx);
-    if (dispatch_fds) {
-        progress |= aio_dispatch_handlers(ctx, INVALID_HANDLE_VALUE);
-    }
-    progress |= timerlistgroup_run_timers(&ctx->tlg);
-    return progress;
+    aio_bh_poll(ctx);
+    aio_dispatch_handlers(ctx, INVALID_HANDLE_VALUE);
+    timerlistgroup_run_timers(&ctx->tlg);
 }
 
 bool aio_poll(AioContext *ctx, bool blocking)
diff --git a/util/async.c b/util/async.c
index XXXXXXX..XXXXXXX 100644
--- a/util/async.c
+++ b/util/async.c
@@ -XXX,XX +XXX,XX @@ aio_ctx_dispatch(GSource     *source,
     AioContext *ctx = (AioContext *) source;
 
     assert(callback == NULL);
-    aio_dispatch(ctx, true);
+    aio_dispatch(ctx);
     return true;
 }
 
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Pull the increment/decrement pair out of aio_bh_poll and into the
callers.

diff --git a/util/aio-posix.c b/util/aio-posix.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-posix.c
+++ b/util/aio-posix.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx)
 
 void aio_dispatch(AioContext *ctx)
 {
+    qemu_lockcnt_inc(&ctx->list_lock);
     aio_bh_poll(ctx);
-
-    qemu_lockcnt_inc(&ctx->list_lock);
     aio_dispatch_handlers(ctx);
     qemu_lockcnt_dec(&ctx->list_lock);
 
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
     }
 
     npfd = 0;
-    qemu_lockcnt_dec(&ctx->list_lock);
 
     progress |= aio_bh_poll(ctx);
 
     if (ret > 0) {
-        qemu_lockcnt_inc(&ctx->list_lock);
         progress |= aio_dispatch_handlers(ctx);
-        qemu_lockcnt_dec(&ctx->list_lock);
     }
 
+    qemu_lockcnt_dec(&ctx->list_lock);
+
     progress |= timerlistgroup_run_timers(&ctx->tlg);
 
     return progress;
diff --git a/util/aio-win32.c b/util/aio-win32.c
index XXXXXXX..XXXXXXX 100644
--- a/util/aio-win32.c
+++ b/util/aio-win32.c
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
     bool progress = false;
     AioHandler *tmp;
 
-    qemu_lockcnt_inc(&ctx->list_lock);
-
     /*
      * We have to walk very carefully in case aio_set_fd_handler is
      * called while we're walking.
@@ -XXX,XX +XXX,XX @@ static bool aio_dispatch_handlers(AioContext *ctx, HANDLE event)
         }
     }
 
-    qemu_lockcnt_dec(&ctx->list_lock);
     return progress;
 }
 
 void aio_dispatch(AioContext *ctx)
 {
+    qemu_lockcnt_inc(&ctx->list_lock);
     aio_bh_poll(ctx);
     aio_dispatch_handlers(ctx, INVALID_HANDLE_VALUE);
+    qemu_lockcnt_dec(&ctx->list_lock);
     timerlistgroup_run_timers(&ctx->tlg);
 }
 
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         }
     }
 
-    qemu_lockcnt_dec(&ctx->list_lock);
     first = true;
 
     /* ctx->notifier is always registered.  */
@@ -XXX,XX +XXX,XX @@ bool aio_poll(AioContext *ctx, bool blocking)
         progress |= aio_dispatch_handlers(ctx, event);
     } while (count > 0);
 
+    qemu_lockcnt_dec(&ctx->list_lock);
+
     progress |= timerlistgroup_run_timers(&ctx->tlg);
     return progress;
 }
diff --git a/util/async.c b/util/async.c
index XXXXXXX..XXXXXXX 100644
--- a/util/async.c
+++ b/util/async.c
@@ -XXX,XX +XXX,XX @@ void aio_bh_call(QEMUBH *bh)
     bh->cb(bh->opaque);
 }
 
-/* Multiple occurrences of aio_bh_poll cannot be called concurrently */
+/* Multiple occurrences of aio_bh_poll cannot be called concurrently.
+ * The count in ctx->list_lock is incremented before the call, and is
+ * not affected by the call.
+ */
 int aio_bh_poll(AioContext *ctx)
 {
     QEMUBH *bh, **bhp, *next;
     int ret;
     bool deleted = false;
 
-    qemu_lockcnt_inc(&ctx->list_lock);
-
     ret = 0;
     for (bh = atomic_rcu_read(&ctx->first_bh); bh; bh = next) {
         next = atomic_rcu_read(&bh->next);
@@ -XXX,XX +XXX,XX @@ int aio_bh_poll(AioContext *ctx)
 
     /* remove deleted bhs */
     if (!deleted) {
-        qemu_lockcnt_dec(&ctx->list_lock);
         return ret;
     }
 
-    if (qemu_lockcnt_dec_and_lock(&ctx->list_lock)) {
+    if (qemu_lockcnt_dec_if_lock(&ctx->list_lock)) {
         bhp = &ctx->first_bh;
         while (*bhp) {
             bh = *bhp;
@@ -XXX,XX +XXX,XX @@ int aio_bh_poll(AioContext *ctx)
                 bhp = &bh->next;
             }
         }
-        qemu_lockcnt_unlock(&ctx->list_lock);
+        qemu_lockcnt_inc_and_unlock(&ctx->list_lock);
     }
     return ret;
 }
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/include/block/block_int.h b/include/block/block_int.h
index XXXXXXX..XXXXXXX 100644
--- a/include/block/block_int.h
+++ b/include/block/block_int.h
@@ -XXX,XX +XXX,XX @@ struct BdrvChild {
  * copied as well.
  */
 struct BlockDriverState {
-    int64_t total_sectors; /* if we are reading a disk image, give its
-                              size in sectors */
+    /* Protected by big QEMU lock or read-only after opening.  No special
+     * locking needed during I/O...
+     */
     int open_flags; /* flags used to open the file, re-used for re-open */
     bool read_only; /* if true, the media is read only */
     bool encrypted; /* if true, the media is encrypted */
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
     bool sg;        /* if true, the device is a /dev/sg* */
     bool probed;    /* if true, format was probed rather than specified */
 
-    int copy_on_read; /* if nonzero, copy read backing sectors into image.
-                         note this is a reference count */
-
-    CoQueue flush_queue;            /* Serializing flush queue */
-    bool active_flush_req;          /* Flush request in flight? */
-    unsigned int write_gen;         /* Current data generation */
-    unsigned int flushed_gen;       /* Flushed write generation */
-
     BlockDriver *drv; /* NULL means no media */
     void *opaque;
 
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
     BdrvChild *backing;
     BdrvChild *file;
 
-    /* Callback before write request is processed */
-    NotifierWithReturnList before_write_notifiers;
-
-    /* number of in-flight requests; overall and serialising */
-    unsigned int in_flight;
-    unsigned int serialising_in_flight;
-
-    bool wakeup;
-
-    /* Offset after the highest byte written to */
-    uint64_t wr_highest_offset;
-
     /* I/O Limits */
     BlockLimits bl;
 
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
     QTAILQ_ENTRY(BlockDriverState) bs_list;
     /* element of the list of monitor-owned BDS */
     QTAILQ_ENTRY(BlockDriverState) monitor_list;
-    QLIST_HEAD(, BdrvDirtyBitmap) dirty_bitmaps;
     int refcnt;
 
-    QLIST_HEAD(, BdrvTrackedRequest) tracked_requests;
-
     /* operation blockers */
     QLIST_HEAD(, BdrvOpBlocker) op_blockers[BLOCK_OP_TYPE_MAX];
 
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
     /* The error object in use for blocking operations on backing_hd */
     Error *backing_blocker;
 
+    /* Protected by AioContext lock */
+
+    /* If true, copy read backing sectors into image.  Can be >1 if more
+     * than one client has requested copy-on-read.
+     */
+    int copy_on_read;
+
+    /* If we are reading a disk image, give its size in sectors.
+     * Generally read-only; it is written to by load_vmstate and save_vmstate,
+     * but the block layer is quiescent during those.
+     */
+    int64_t total_sectors;
+
+    /* Callback before write request is processed */
+    NotifierWithReturnList before_write_notifiers;
+
+    /* number of in-flight requests; overall and serialising */
+    unsigned int in_flight;
+    unsigned int serialising_in_flight;
+
+    bool wakeup;
+
+    /* Offset after the highest byte written to */
+    uint64_t wr_highest_offset;
+
     /* threshold limit for writes, in bytes. "High water mark". */
     uint64_t write_threshold_offset;
     NotifierWithReturn write_threshold_notifier;
@@ -XXX,XX +XXX,XX @@ struct BlockDriverState {
     /* counter for nested bdrv_io_plug */
     unsigned io_plugged;
 
+    QLIST_HEAD(, BdrvTrackedRequest) tracked_requests;
+    CoQueue flush_queue;                  /* Serializing flush queue */
+    bool active_flush_req;                /* Flush request in flight? */
+    unsigned int write_gen;               /* Current data generation */
+    unsigned int flushed_gen;             /* Flushed write generation */
+
+    QLIST_HEAD(, BdrvDirtyBitmap) dirty_bitmaps;
+
+    /* do we need to tell the quest if we have a volatile write cache? */
+    int enable_write_cache;
+
     int quiesce_counter;
 };
 
diff --git a/include/sysemu/block-backend.h b/include/sysemu/block-backend.h
index XXXXXXX..XXXXXXX 100644
--- a/include/sysemu/block-backend.h
+++ b/include/sysemu/block-backend.h
@@ -XXX,XX +XXX,XX @@ typedef struct BlockDevOps {
  * fields that must be public. This is in particular for QLIST_ENTRY() and
  * friends so that BlockBackends can be kept in lists outside block-backend.c */
 typedef struct BlockBackendPublic {
-    /* I/O throttling.
-     * throttle_state tells us if this BlockBackend has I/O limits configured.
-     * io_limits_disabled tells us if they are currently being enforced */
+    /* I/O throttling has its own locking, but also some fields are
+     * protected by the AioContext lock.
+     */
+
+    /* Protected by AioContext lock.  */
     CoQueue      throttled_reqs[2];
+
+    /* Nonzero if the I/O limits are currently being ignored; generally
+     * it is zero.  */
     unsigned int io_limits_disabled;
 
     /* The following fields are protected by the ThrottleGroup lock.
-     * See the ThrottleGroup documentation for details. */
+     * See the ThrottleGroup documentation for details.
+     * throttle_state tells us if I/O limits are configured. */
     ThrottleState *throttle_state;
     ThrottleTimers throttle_timers;
     unsigned       pending_reqs[2];
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

This uses the lock-free mutex described in the paper '"Blocking without
Locking", or LFTHREADS: A lock-free thread library' by Gidenstam and
Papatriantafilou.  The same technique is used in OSv, and in fact
the code is essentially a conversion to C of OSv's code.

[Added missing coroutine_fn in tests/test-aio-multithread.c.
--Stefan]

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213181244.16297-2-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 include/qemu/coroutine.h     |  17 ++++-
 tests/test-aio-multithread.c |  86 ++++++++++++++++++++++++
 util/qemu-coroutine-lock.c   | 155 ++++++++++++++++++++++++++++++++++++++++---
 util/trace-events            |   1 +
 4 files changed, 246 insertions(+), 13 deletions(-)

diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
index XXXXXXX..XXXXXXX 100644
--- a/include/qemu/coroutine.h
+++ b/include/qemu/coroutine.h
@@ -XXX,XX +XXX,XX @@ bool qemu_co_queue_empty(CoQueue *queue);
 /**
  * Provides a mutex that can be used to synchronise coroutines
  */
+struct CoWaitRecord;
 typedef struct CoMutex {
-    bool locked;
+    /* Count of pending lockers; 0 for a free mutex, 1 for an
+     * uncontended mutex.
+     */
+    unsigned locked;
+
+    /* A queue of waiters.  Elements are added atomically in front of
+     * from_push.  to_pop is only populated, and popped from, by whoever
+     * is in charge of the next wakeup.  This can be an unlocker or,
+     * through the handoff protocol, a locker that is about to go to sleep.
+     */
+    QSLIST_HEAD(, CoWaitRecord) from_push, to_pop;
+
+    unsigned handoff, sequence;
+
     Coroutine *holder;
-    CoQueue queue;
 } CoMutex;
 
 /**
diff --git a/tests/test-aio-multithread.c b/tests/test-aio-multithread.c
index XXXXXXX..XXXXXXX 100644
--- a/tests/test-aio-multithread.c
+++ b/tests/test-aio-multithread.c
@@ -XXX,XX +XXX,XX @@ static void test_multi_co_schedule_10(void)
     test_multi_co_schedule(10);
 }
 
+/* CoMutex thread-safety.  */
+
+static uint32_t atomic_counter;
+static uint32_t running;
+static uint32_t counter;
+static CoMutex comutex;
+
+static void coroutine_fn test_multi_co_mutex_entry(void *opaque)
+{
+    while (!atomic_mb_read(&now_stopping)) {
+        qemu_co_mutex_lock(&comutex);
+        counter++;
+        qemu_co_mutex_unlock(&comutex);
+
+        /* Increase atomic_counter *after* releasing the mutex.  Otherwise
+         * there is a chance (it happens about 1 in 3 runs) that the iothread
+         * exits before the coroutine is woken up, causing a spurious
+         * assertion failure.
+         */
+        atomic_inc(&atomic_counter);
+    }
+    atomic_dec(&running);
+}
+
+static void test_multi_co_mutex(int threads, int seconds)
+{
+    int i;
+
+    qemu_co_mutex_init(&comutex);
+    counter = 0;
+    atomic_counter = 0;
+    now_stopping = false;
+
+    create_aio_contexts();
+    assert(threads <= NUM_CONTEXTS);
+    running = threads;
+    for (i = 0; i < threads; i++) {
+        Coroutine *co1 = qemu_coroutine_create(test_multi_co_mutex_entry, NULL);
+        aio_co_schedule(ctx[i], co1);
+    }
+
+    g_usleep(seconds * 1000000);
+
+    atomic_mb_set(&now_stopping, true);
+    while (running > 0) {
+        g_usleep(100000);
+    }
+
+    join_aio_contexts();
+    g_test_message("%d iterations/second\n", counter / seconds);
+    g_assert_cmpint(counter, ==, atomic_counter);
+}
+
+/* Testing with NUM_CONTEXTS threads focuses on the queue.  The mutex however
+ * is too contended (and the threads spend too much time in aio_poll)
+ * to actually stress the handoff protocol.
+ */
+static void test_multi_co_mutex_1(void)
+{
+    test_multi_co_mutex(NUM_CONTEXTS, 1);
+}
+
+static void test_multi_co_mutex_10(void)
+{
+    test_multi_co_mutex(NUM_CONTEXTS, 10);
+}
+
+/* Testing with fewer threads stresses the handoff protocol too.  Still, the
+ * case where the locker _can_ pick up a handoff is very rare, happening
+ * about 10 times in 1 million, so increase the runtime a bit compared to
+ * other "quick" testcases that only run for 1 second.
+ */
+static void test_multi_co_mutex_2_3(void)
+{
+    test_multi_co_mutex(2, 3);
+}
+
+static void test_multi_co_mutex_2_30(void)
+{
+    test_multi_co_mutex(2, 30);
+}
+
 /* End of tests.  */
 
 int main(int argc, char **argv)
@@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv)
     g_test_add_func("/aio/multi/lifecycle", test_lifecycle);
     if (g_test_quick()) {
         g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_1);
+        g_test_add_func("/aio/multi/mutex/contended", test_multi_co_mutex_1);
+        g_test_add_func("/aio/multi/mutex/handoff", test_multi_co_mutex_2_3);
     } else {
         g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_10);
+        g_test_add_func("/aio/multi/mutex/contended", test_multi_co_mutex_10);
+        g_test_add_func("/aio/multi/mutex/handoff", test_multi_co_mutex_2_30);
     }
     return g_test_run();
 }
diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
index XXXXXXX..XXXXXXX 100644
--- a/util/qemu-coroutine-lock.c
+++ b/util/qemu-coroutine-lock.c
@@ -XXX,XX +XXX,XX @@
  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
  * THE SOFTWARE.
+ *
+ * The lock-free mutex implementation is based on OSv
+ * (core/lfmutex.cc, include/lockfree/mutex.hh).
+ * Copyright (C) 2013 Cloudius Systems, Ltd.
  */
 
 #include "qemu/osdep.h"
@@ -XXX,XX +XXX,XX @@ bool qemu_co_queue_empty(CoQueue *queue)
     return QSIMPLEQ_FIRST(&queue->entries) == NULL;
 }
 
+/* The wait records are handled with a multiple-producer, single-consumer
+ * lock-free queue.  There cannot be two concurrent pop_waiter() calls
+ * because pop_waiter() can only be called while mutex->handoff is zero.
+ * This can happen in three cases:
+ * - in qemu_co_mutex_unlock, before the hand-off protocol has started.
+ *   In this case, qemu_co_mutex_lock will see mutex->handoff == 0 and
+ *   not take part in the handoff.
+ * - in qemu_co_mutex_lock, if it steals the hand-off responsibility from
+ *   qemu_co_mutex_unlock.  In this case, qemu_co_mutex_unlock will fail
+ *   the cmpxchg (it will see either 0 or the next sequence value) and
+ *   exit.  The next hand-off cannot begin until qemu_co_mutex_lock has
+ *   woken up someone.
+ * - in qemu_co_mutex_unlock, if it takes the hand-off token itself.
+ *   In this case another iteration starts with mutex->handoff == 0;
+ *   a concurrent qemu_co_mutex_lock will fail the cmpxchg, and
+ *   qemu_co_mutex_unlock will go back to case (1).
+ *
+ * The following functions manage this queue.
+ */
+typedef struct CoWaitRecord {
+    Coroutine *co;
+    QSLIST_ENTRY(CoWaitRecord) next;
+} CoWaitRecord;
+
+static void push_waiter(CoMutex *mutex, CoWaitRecord *w)
+{
+    w->co = qemu_coroutine_self();
+    QSLIST_INSERT_HEAD_ATOMIC(&mutex->from_push, w, next);
+}
+
+static void move_waiters(CoMutex *mutex)
+{
+    QSLIST_HEAD(, CoWaitRecord) reversed;
+    QSLIST_MOVE_ATOMIC(&reversed, &mutex->from_push);
+    while (!QSLIST_EMPTY(&reversed)) {
+        CoWaitRecord *w = QSLIST_FIRST(&reversed);
+        QSLIST_REMOVE_HEAD(&reversed, next);
+        QSLIST_INSERT_HEAD(&mutex->to_pop, w, next);
+    }
+}
+
+static CoWaitRecord *pop_waiter(CoMutex *mutex)
+{
+    CoWaitRecord *w;
+
+    if (QSLIST_EMPTY(&mutex->to_pop)) {
+        move_waiters(mutex);
+        if (QSLIST_EMPTY(&mutex->to_pop)) {
+            return NULL;
+        }
+    }
+    w = QSLIST_FIRST(&mutex->to_pop);
+    QSLIST_REMOVE_HEAD(&mutex->to_pop, next);
+    return w;
+}
+
+static bool has_waiters(CoMutex *mutex)
+{
+    return QSLIST_EMPTY(&mutex->to_pop) || QSLIST_EMPTY(&mutex->from_push);
+}
+
 void qemu_co_mutex_init(CoMutex *mutex)
 {
     memset(mutex, 0, sizeof(*mutex));
-    qemu_co_queue_init(&mutex->queue);
 }
 
-void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex)
+static void coroutine_fn qemu_co_mutex_lock_slowpath(CoMutex *mutex)
 {
     Coroutine *self = qemu_coroutine_self();
+    CoWaitRecord w;
+    unsigned old_handoff;
 
     trace_qemu_co_mutex_lock_entry(mutex, self);
+    w.co = self;
+    push_waiter(mutex, &w);
 
-    while (mutex->locked) {
-        qemu_co_queue_wait(&mutex->queue);
+    /* This is the "Responsibility Hand-Off" protocol; a lock() picks from
+     * a concurrent unlock() the responsibility of waking somebody up.
+     */
+    old_handoff = atomic_mb_read(&mutex->handoff);
+    if (old_handoff &&
+        has_waiters(mutex) &&
+        atomic_cmpxchg(&mutex->handoff, old_handoff, 0) == old_handoff) {
+        /* There can be no concurrent pops, because there can be only
+         * one active handoff at a time.
+         */
+        CoWaitRecord *to_wake = pop_waiter(mutex);
+        Coroutine *co = to_wake->co;
+        if (co == self) {
+            /* We got the lock ourselves!  */
+            assert(to_wake == &w);
+            return;
+        }
+
+        aio_co_wake(co);
     }
 
-    mutex->locked = true;
-    mutex->holder = self;
-    self->locks_held++;
-
+    qemu_coroutine_yield();
     trace_qemu_co_mutex_lock_return(mutex, self);
 }
 
+void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex)
+{
+    Coroutine *self = qemu_coroutine_self();
+
+    if (atomic_fetch_inc(&mutex->locked) == 0) {
+        /* Uncontended.  */
+        trace_qemu_co_mutex_lock_uncontended(mutex, self);
+    } else {
+        qemu_co_mutex_lock_slowpath(mutex);
+    }
+    mutex->holder = self;
+    self->locks_held++;
+}
+
 void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex)
 {
     Coroutine *self = qemu_coroutine_self();
 
     trace_qemu_co_mutex_unlock_entry(mutex, self);
 
-    assert(mutex->locked == true);
+    assert(mutex->locked);
     assert(mutex->holder == self);
     assert(qemu_in_coroutine());
 
-    mutex->locked = false;
     mutex->holder = NULL;
     self->locks_held--;
-    qemu_co_queue_next(&mutex->queue);
+    if (atomic_fetch_dec(&mutex->locked) == 1) {
+        /* No waiting qemu_co_mutex_lock().  Pfew, that was easy!  */
+        return;
+    }
+
+    for (;;) {
+        CoWaitRecord *to_wake = pop_waiter(mutex);
+        unsigned our_handoff;
+
+        if (to_wake) {
+            Coroutine *co = to_wake->co;
+            aio_co_wake(co);
+            break;
+        }
+
+        /* Some concurrent lock() is in progress (we know this because
+         * mutex->locked was >1) but it hasn't yet put itself on the wait
+         * queue.  Pick a sequence number for the handoff protocol (not 0).
+         */
+        if (++mutex->sequence == 0) {
+            mutex->sequence = 1;
+        }
+
+        our_handoff = mutex->sequence;
+        atomic_mb_set(&mutex->handoff, our_handoff);
+        if (!has_waiters(mutex)) {
+            /* The concurrent lock has not added itself yet, so it
+             * will be able to pick our handoff.
+             */
+            break;
+        }
+
+        /* Try to do the handoff protocol ourselves; if somebody else has
+         * already taken it, however, we're done and they're responsible.
+         */
+        if (atomic_cmpxchg(&mutex->handoff, our_handoff, 0) != our_handoff) {
+            break;
+        }
+    }
 
     trace_qemu_co_mutex_unlock_return(mutex, self);
 }
diff --git a/util/trace-events b/util/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/util/trace-events
+++ b/util/trace-events
@@ -XXX,XX +XXX,XX @@ qemu_coroutine_terminate(void *co) "self %p"
 
 # util/qemu-coroutine-lock.c
 qemu_co_queue_run_restart(void *co) "co %p"
+qemu_co_mutex_lock_uncontended(void *mutex, void *self) "mutex %p self %p"
 qemu_co_mutex_lock_entry(void *mutex, void *self) "mutex %p self %p"
 qemu_co_mutex_lock_return(void *mutex, void *self) "mutex %p self %p"
 qemu_co_mutex_unlock_entry(void *mutex, void *self) "mutex %p self %p"
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Running a very small critical section on pthread_mutex_t and CoMutex
shows that pthread_mutex_t is much faster because it doesn't actually
go to sleep.  What happens is that the critical section is shorter
than the latency of entering the kernel and thus FUTEX_WAIT always
fails.  With CoMutex there is no such latency but you still want to
avoid wait and wakeup.  So introduce it artificially.

This only works with one waiters; because CoMutex is fair, it will
always have more waits and wakeups than a pthread_mutex_t.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213181244.16297-3-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 include/qemu/coroutine.h   |  5 +++++
 util/qemu-coroutine-lock.c | 51 ++++++++++++++++++++++++++++++++++++++++------
 util/qemu-coroutine.c      |  2 +-
 3 files changed, 51 insertions(+), 7 deletions(-)

diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
index XXXXXXX..XXXXXXX 100644
--- a/include/qemu/coroutine.h
+++ b/include/qemu/coroutine.h
@@ -XXX,XX +XXX,XX @@ typedef struct CoMutex {
      */
     unsigned locked;
 
+    /* Context that is holding the lock.  Useful to avoid spinning
+     * when two coroutines on the same AioContext try to get the lock. :)
+     */
+    AioContext *ctx;
+
     /* A queue of waiters.  Elements are added atomically in front of
      * from_push.  to_pop is only populated, and popped from, by whoever
      * is in charge of the next wakeup.  This can be an unlocker or,
diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
index XXXXXXX..XXXXXXX 100644
--- a/util/qemu-coroutine-lock.c
+++ b/util/qemu-coroutine-lock.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu-common.h"
 #include "qemu/coroutine.h"
 #include "qemu/coroutine_int.h"
+#include "qemu/processor.h"
 #include "qemu/queue.h"
 #include "block/aio.h"
 #include "trace.h"
@@ -XXX,XX +XXX,XX @@ void qemu_co_mutex_init(CoMutex *mutex)
     memset(mutex, 0, sizeof(*mutex));
 }
 
-static void coroutine_fn qemu_co_mutex_lock_slowpath(CoMutex *mutex)
+static void coroutine_fn qemu_co_mutex_wake(CoMutex *mutex, Coroutine *co)
+{
+    /* Read co before co->ctx; pairs with smp_wmb() in
+     * qemu_coroutine_enter().
+     */
+    smp_read_barrier_depends();
+    mutex->ctx = co->ctx;
+    aio_co_wake(co);
+}
+
+static void coroutine_fn qemu_co_mutex_lock_slowpath(AioContext *ctx,
+                                                     CoMutex *mutex)
 {
     Coroutine *self = qemu_coroutine_self();
     CoWaitRecord w;
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn qemu_co_mutex_lock_slowpath(CoMutex *mutex)
         if (co == self) {
             /* We got the lock ourselves!  */
             assert(to_wake == &w);
+            mutex->ctx = ctx;
             return;
         }
 
-        aio_co_wake(co);
+        qemu_co_mutex_wake(mutex, co);
     }
 
     qemu_coroutine_yield();
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn qemu_co_mutex_lock_slowpath(CoMutex *mutex)
 
 void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex)
 {
+    AioContext *ctx = qemu_get_current_aio_context();
     Coroutine *self = qemu_coroutine_self();
+    int waiters, i;
 
-    if (atomic_fetch_inc(&mutex->locked) == 0) {
+    /* Running a very small critical section on pthread_mutex_t and CoMutex
+     * shows that pthread_mutex_t is much faster because it doesn't actually
+     * go to sleep.  What happens is that the critical section is shorter
+     * than the latency of entering the kernel and thus FUTEX_WAIT always
+     * fails.  With CoMutex there is no such latency but you still want to
+     * avoid wait and wakeup.  So introduce it artificially.
+     */
+    i = 0;
+retry_fast_path:
+    waiters = atomic_cmpxchg(&mutex->locked, 0, 1);
+    if (waiters != 0) {
+        while (waiters == 1 && ++i < 1000) {
+            if (atomic_read(&mutex->ctx) == ctx) {
+                break;
+            }
+            if (atomic_read(&mutex->locked) == 0) {
+                goto retry_fast_path;
+            }
+            cpu_relax();
+        }
+        waiters = atomic_fetch_inc(&mutex->locked);
+    }
+
+    if (waiters == 0) {
         /* Uncontended.  */
         trace_qemu_co_mutex_lock_uncontended(mutex, self);
+        mutex->ctx = ctx;
     } else {
-        qemu_co_mutex_lock_slowpath(mutex);
+        qemu_co_mutex_lock_slowpath(ctx, mutex);
     }
     mutex->holder = self;
     self->locks_held++;
@@ -XXX,XX +XXX,XX @@ void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex)
     assert(mutex->holder == self);
     assert(qemu_in_coroutine());
 
+    mutex->ctx = NULL;
     mutex->holder = NULL;
     self->locks_held--;
     if (atomic_fetch_dec(&mutex->locked) == 1) {
@@ -XXX,XX +XXX,XX @@ void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex)
         unsigned our_handoff;
 
         if (to_wake) {
-            Coroutine *co = to_wake->co;
-            aio_co_wake(co);
+            qemu_co_mutex_wake(mutex, to_wake->co);
             break;
         }
 
diff --git a/util/qemu-coroutine.c b/util/qemu-coroutine.c
index XXXXXXX..XXXXXXX 100644
--- a/util/qemu-coroutine.c
+++ b/util/qemu-coroutine.c
@@ -XXX,XX +XXX,XX @@ void qemu_coroutine_enter(Coroutine *co)
     co->ctx = qemu_get_current_aio_context();
 
     /* Store co->ctx before anything that stores co.  Matches
-     * barrier in aio_co_wake.
+     * barrier in aio_co_wake and qemu_co_mutex_wake.
      */
     smp_wmb();
 
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

Add two implementations of the same benchmark as the previous patch,
but using pthreads.  One uses a normal QemuMutex, the other is Linux
only and implements a fair mutex based on MCS locks and futexes.
This shows that the slower performance of the 5-thread case is due to
the fairness of CoMutex, rather than to coroutines.  If fairness does
not matter, as is the case with two threads, CoMutex can actually be
faster than pthreads.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213181244.16297-4-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 tests/test-aio-multithread.c | 164 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 164 insertions(+)

diff --git a/tests/test-aio-multithread.c b/tests/test-aio-multithread.c
index XXXXXXX..XXXXXXX 100644
--- a/tests/test-aio-multithread.c
+++ b/tests/test-aio-multithread.c
@@ -XXX,XX +XXX,XX @@ static void test_multi_co_mutex_2_30(void)
     test_multi_co_mutex(2, 30);
 }
 
+/* Same test with fair mutexes, for performance comparison.  */
+
+#ifdef CONFIG_LINUX
+#include "qemu/futex.h"
+
+/* The nodes for the mutex reside in this structure (on which we try to avoid
+ * false sharing).  The head of the mutex is in the "mutex_head" variable.
+ */
+static struct {
+    int next, locked;
+    int padding[14];
+} nodes[NUM_CONTEXTS] __attribute__((__aligned__(64)));
+
+static int mutex_head = -1;
+
+static void mcs_mutex_lock(void)
+{
+    int prev;
+
+    nodes[id].next = -1;
+    nodes[id].locked = 1;
+    prev = atomic_xchg(&mutex_head, id);
+    if (prev != -1) {
+        atomic_set(&nodes[prev].next, id);
+        qemu_futex_wait(&nodes[id].locked, 1);
+    }
+}
+
+static void mcs_mutex_unlock(void)
+{
+    int next;
+    if (nodes[id].next == -1) {
+        if (atomic_read(&mutex_head) == id &&
+            atomic_cmpxchg(&mutex_head, id, -1) == id) {
+            /* Last item in the list, exit.  */
+            return;
+        }
+        while (atomic_read(&nodes[id].next) == -1) {
+            /* mcs_mutex_lock did the xchg, but has not updated
+             * nodes[prev].next yet.
+             */
+        }
+    }
+
+    /* Wake up the next in line.  */
+    next = nodes[id].next;
+    nodes[next].locked = 0;
+    qemu_futex_wake(&nodes[next].locked, 1);
+}
+
+static void test_multi_fair_mutex_entry(void *opaque)
+{
+    while (!atomic_mb_read(&now_stopping)) {
+        mcs_mutex_lock();
+        counter++;
+        mcs_mutex_unlock();
+        atomic_inc(&atomic_counter);
+    }
+    atomic_dec(&running);
+}
+
+static void test_multi_fair_mutex(int threads, int seconds)
+{
+    int i;
+
+    assert(mutex_head == -1);
+    counter = 0;
+    atomic_counter = 0;
+    now_stopping = false;
+
+    create_aio_contexts();
+    assert(threads <= NUM_CONTEXTS);
+    running = threads;
+    for (i = 0; i < threads; i++) {
+        Coroutine *co1 = qemu_coroutine_create(test_multi_fair_mutex_entry, NULL);
+        aio_co_schedule(ctx[i], co1);
+    }
+
+    g_usleep(seconds * 1000000);
+
+    atomic_mb_set(&now_stopping, true);
+    while (running > 0) {
+        g_usleep(100000);
+    }
+
+    join_aio_contexts();
+    g_test_message("%d iterations/second\n", counter / seconds);
+    g_assert_cmpint(counter, ==, atomic_counter);
+}
+
+static void test_multi_fair_mutex_1(void)
+{
+    test_multi_fair_mutex(NUM_CONTEXTS, 1);
+}
+
+static void test_multi_fair_mutex_10(void)
+{
+    test_multi_fair_mutex(NUM_CONTEXTS, 10);
+}
+#endif
+
+/* Same test with pthread mutexes, for performance comparison and
+ * portability.  */
+
+static QemuMutex mutex;
+
+static void test_multi_mutex_entry(void *opaque)
+{
+    while (!atomic_mb_read(&now_stopping)) {
+        qemu_mutex_lock(&mutex);
+        counter++;
+        qemu_mutex_unlock(&mutex);
+        atomic_inc(&atomic_counter);
+    }
+    atomic_dec(&running);
+}
+
+static void test_multi_mutex(int threads, int seconds)
+{
+    int i;
+
+    qemu_mutex_init(&mutex);
+    counter = 0;
+    atomic_counter = 0;
+    now_stopping = false;
+
+    create_aio_contexts();
+    assert(threads <= NUM_CONTEXTS);
+    running = threads;
+    for (i = 0; i < threads; i++) {
+        Coroutine *co1 = qemu_coroutine_create(test_multi_mutex_entry, NULL);
+        aio_co_schedule(ctx[i], co1);
+    }
+
+    g_usleep(seconds * 1000000);
+
+    atomic_mb_set(&now_stopping, true);
+    while (running > 0) {
+        g_usleep(100000);
+    }
+
+    join_aio_contexts();
+    g_test_message("%d iterations/second\n", counter / seconds);
+    g_assert_cmpint(counter, ==, atomic_counter);
+}
+
+static void test_multi_mutex_1(void)
+{
+    test_multi_mutex(NUM_CONTEXTS, 1);
+}
+
+static void test_multi_mutex_10(void)
+{
+    test_multi_mutex(NUM_CONTEXTS, 10);
+}
+
 /* End of tests.  */
 
 int main(int argc, char **argv)
@@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv)
         g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_1);
         g_test_add_func("/aio/multi/mutex/contended", test_multi_co_mutex_1);
         g_test_add_func("/aio/multi/mutex/handoff", test_multi_co_mutex_2_3);
+#ifdef CONFIG_LINUX
+        g_test_add_func("/aio/multi/mutex/mcs", test_multi_fair_mutex_1);
+#endif
+        g_test_add_func("/aio/multi/mutex/pthread", test_multi_mutex_1);
     } else {
         g_test_add_func("/aio/multi/schedule", test_multi_co_schedule_10);
         g_test_add_func("/aio/multi/mutex/contended", test_multi_co_mutex_10);
         g_test_add_func("/aio/multi/mutex/handoff", test_multi_co_mutex_2_30);
+#ifdef CONFIG_LINUX
+        g_test_add_func("/aio/multi/mutex/mcs", test_multi_fair_mutex_10);
+#endif
+        g_test_add_func("/aio/multi/mutex/pthread", test_multi_mutex_10);
     }
     return g_test_run();
 }
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

This will avoid forward references in the next patch.  It is also
more logical because CoQueue is not anymore the basic primitive.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213181244.16297-5-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 include/qemu/coroutine.h | 89 ++++++++++++++++++++++++------------------------
 1 file changed, 44 insertions(+), 45 deletions(-)

diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
index XXXXXXX..XXXXXXX 100644
--- a/include/qemu/coroutine.h
+++ b/include/qemu/coroutine.h
@@ -XXX,XX +XXX,XX @@ bool qemu_in_coroutine(void);
  */
 bool qemu_coroutine_entered(Coroutine *co);
 
-
-/**
- * CoQueues are a mechanism to queue coroutines in order to continue executing
- * them later. They provide the fundamental primitives on which coroutine locks
- * are built.
- */
-typedef struct CoQueue {
-    QSIMPLEQ_HEAD(, Coroutine) entries;
-} CoQueue;
-
-/**
- * Initialise a CoQueue. This must be called before any other operation is used
- * on the CoQueue.
- */
-void qemu_co_queue_init(CoQueue *queue);
-
-/**
- * Adds the current coroutine to the CoQueue and transfers control to the
- * caller of the coroutine.
- */
-void coroutine_fn qemu_co_queue_wait(CoQueue *queue);
-
-/**
- * Restarts the next coroutine in the CoQueue and removes it from the queue.
- *
- * Returns true if a coroutine was restarted, false if the queue is empty.
- */
-bool coroutine_fn qemu_co_queue_next(CoQueue *queue);
-
-/**
- * Restarts all coroutines in the CoQueue and leaves the queue empty.
- */
-void coroutine_fn qemu_co_queue_restart_all(CoQueue *queue);
-
-/**
- * Enter the next coroutine in the queue
- */
-bool qemu_co_enter_next(CoQueue *queue);
-
-/**
- * Checks if the CoQueue is empty.
- */
-bool qemu_co_queue_empty(CoQueue *queue);
-
-
 /**
  * Provides a mutex that can be used to synchronise coroutines
  */
@@ -XXX,XX +XXX,XX @@ void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex);
  */
 void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex);
 
+
+/**
+ * CoQueues are a mechanism to queue coroutines in order to continue executing
+ * them later.
+ */
+typedef struct CoQueue {
+    QSIMPLEQ_HEAD(, Coroutine) entries;
+} CoQueue;
+
+/**
+ * Initialise a CoQueue. This must be called before any other operation is used
+ * on the CoQueue.
+ */
+void qemu_co_queue_init(CoQueue *queue);
+
+/**
+ * Adds the current coroutine to the CoQueue and transfers control to the
+ * caller of the coroutine.
+ */
+void coroutine_fn qemu_co_queue_wait(CoQueue *queue);
+
+/**
+ * Restarts the next coroutine in the CoQueue and removes it from the queue.
+ *
+ * Returns true if a coroutine was restarted, false if the queue is empty.
+ */
+bool coroutine_fn qemu_co_queue_next(CoQueue *queue);
+
+/**
+ * Restarts all coroutines in the CoQueue and leaves the queue empty.
+ */
+void coroutine_fn qemu_co_queue_restart_all(CoQueue *queue);
+
+/**
+ * Enter the next coroutine in the queue
+ */
+bool qemu_co_enter_next(CoQueue *queue);
+
+/**
+ * Checks if the CoQueue is empty.
+ */
+bool qemu_co_queue_empty(CoQueue *queue);
+
+
 typedef struct CoRwlock {
     bool writer;
     int reader;
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

All that CoQueue needs in order to become thread-safe is help
from an external mutex.  Add this to the API.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213181244.16297-6-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 include/qemu/coroutine.h   |  8 +++++---
 block/backup.c             |  2 +-
 block/io.c                 |  4 ++--
 block/nbd-client.c         |  2 +-
 block/qcow2-cluster.c      |  4 +---
 block/sheepdog.c           |  2 +-
 block/throttle-groups.c    |  2 +-
 hw/9pfs/9p.c               |  2 +-
 util/qemu-coroutine-lock.c | 24 +++++++++++++++++++++---
 9 files changed, 34 insertions(+), 16 deletions(-)

diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
index XXXXXXX..XXXXXXX 100644
--- a/include/qemu/coroutine.h
+++ b/include/qemu/coroutine.h
@@ -XXX,XX +XXX,XX @@ void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex);
 
 /**
  * CoQueues are a mechanism to queue coroutines in order to continue executing
- * them later.
+ * them later.  They are similar to condition variables, but they need help
+ * from an external mutex in order to maintain thread-safety.
  */
 typedef struct CoQueue {
     QSIMPLEQ_HEAD(, Coroutine) entries;
@@ -XXX,XX +XXX,XX @@ void qemu_co_queue_init(CoQueue *queue);
 
 /**
  * Adds the current coroutine to the CoQueue and transfers control to the
- * caller of the coroutine.
+ * caller of the coroutine.  The mutex is unlocked during the wait and
+ * locked again afterwards.
  */
-void coroutine_fn qemu_co_queue_wait(CoQueue *queue);
+void coroutine_fn qemu_co_queue_wait(CoQueue *queue, CoMutex *mutex);
 
 /**
  * Restarts the next coroutine in the CoQueue and removes it from the queue.
diff --git a/block/backup.c b/block/backup.c
index XXXXXXX..XXXXXXX 100644
--- a/block/backup.c
+++ b/block/backup.c
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn wait_for_overlapping_requests(BackupBlockJob *job,
         retry = false;
         QLIST_FOREACH(req, &job->inflight_reqs, list) {
             if (end > req->start && start < req->end) {
-                qemu_co_queue_wait(&req->wait_queue);
+                qemu_co_queue_wait(&req->wait_queue, NULL);
                 retry = true;
                 break;
             }
diff --git a/block/io.c b/block/io.c
index XXXXXXX..XXXXXXX 100644
--- a/block/io.c
+++ b/block/io.c
@@ -XXX,XX +XXX,XX @@ static bool coroutine_fn wait_serialising_requests(BdrvTrackedRequest *self)
                  * (instead of producing a deadlock in the former case). */
                 if (!req->waiting_for) {
                     self->waiting_for = req;
-                    qemu_co_queue_wait(&req->wait_queue);
+                    qemu_co_queue_wait(&req->wait_queue, NULL);
                     self->waiting_for = NULL;
                     retry = true;
                     waited = true;
@@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_flush(BlockDriverState *bs)
 
     /* Wait until any previous flushes are completed */
     while (bs->active_flush_req) {
-        qemu_co_queue_wait(&bs->flush_queue);
+        qemu_co_queue_wait(&bs->flush_queue, NULL);
     }
 
     bs->active_flush_req = true;
diff --git a/block/nbd-client.c b/block/nbd-client.c
index XXXXXXX..XXXXXXX 100644
--- a/block/nbd-client.c
+++ b/block/nbd-client.c
@@ -XXX,XX +XXX,XX @@ static void nbd_coroutine_start(NBDClientSession *s,
     /* Poor man semaphore.  The free_sema is locked when no other request
      * can be accepted, and unlocked after receiving one reply.  */
     if (s->in_flight == MAX_NBD_REQUESTS) {
-        qemu_co_queue_wait(&s->free_sema);
+        qemu_co_queue_wait(&s->free_sema, NULL);
         assert(s->in_flight < MAX_NBD_REQUESTS);
     }
     s->in_flight++;
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index XXXXXXX..XXXXXXX 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -XXX,XX +XXX,XX @@ static int handle_dependencies(BlockDriverState *bs, uint64_t guest_offset,
             if (bytes == 0) {
                 /* Wait for the dependency to complete. We need to recheck
                  * the free/allocated clusters when we continue. */
-                qemu_co_mutex_unlock(&s->lock);
-                qemu_co_queue_wait(&old_alloc->dependent_requests);
-                qemu_co_mutex_lock(&s->lock);
+                qemu_co_queue_wait(&old_alloc->dependent_requests, &s->lock);
                 return -EAGAIN;
             }
         }
diff --git a/block/sheepdog.c b/block/sheepdog.c
index XXXXXXX..XXXXXXX 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -XXX,XX +XXX,XX @@ static void wait_for_overlapping_aiocb(BDRVSheepdogState *s, SheepdogAIOCB *acb)
 retry:
     QLIST_FOREACH(cb, &s->inflight_aiocb_head, aiocb_siblings) {
         if (AIOCBOverlapping(acb, cb)) {
-            qemu_co_queue_wait(&s->overlapping_queue);
+            qemu_co_queue_wait(&s->overlapping_queue, NULL);
             goto retry;
         }
     }
diff --git a/block/throttle-groups.c b/block/throttle-groups.c
index XXXXXXX..XXXXXXX 100644
--- a/block/throttle-groups.c
+++ b/block/throttle-groups.c
@@ -XXX,XX +XXX,XX @@ void coroutine_fn throttle_group_co_io_limits_intercept(BlockBackend *blk,
     if (must_wait || blkp->pending_reqs[is_write]) {
         blkp->pending_reqs[is_write]++;
         qemu_mutex_unlock(&tg->lock);
-        qemu_co_queue_wait(&blkp->throttled_reqs[is_write]);
+        qemu_co_queue_wait(&blkp->throttled_reqs[is_write], NULL);
         qemu_mutex_lock(&tg->lock);
         blkp->pending_reqs[is_write]--;
     }
diff --git a/hw/9pfs/9p.c b/hw/9pfs/9p.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/9pfs/9p.c
+++ b/hw/9pfs/9p.c
@@ -XXX,XX +XXX,XX @@ static void coroutine_fn v9fs_flush(void *opaque)
         /*
          * Wait for pdu to complete.
          */
-        qemu_co_queue_wait(&cancel_pdu->complete);
+        qemu_co_queue_wait(&cancel_pdu->complete, NULL);
         cancel_pdu->cancelled = 0;
         pdu_free(cancel_pdu);
     }
diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
index XXXXXXX..XXXXXXX 100644
--- a/util/qemu-coroutine-lock.c
+++ b/util/qemu-coroutine-lock.c
@@ -XXX,XX +XXX,XX @@ void qemu_co_queue_init(CoQueue *queue)
     QSIMPLEQ_INIT(&queue->entries);
 }
 
-void coroutine_fn qemu_co_queue_wait(CoQueue *queue)
+void coroutine_fn qemu_co_queue_wait(CoQueue *queue, CoMutex *mutex)
 {
     Coroutine *self = qemu_coroutine_self();
     QSIMPLEQ_INSERT_TAIL(&queue->entries, self, co_queue_next);
+
+    if (mutex) {
+        qemu_co_mutex_unlock(mutex);
+    }
+
+    /* There is no race condition here.  Other threads will call
+     * aio_co_schedule on our AioContext, which can reenter this
+     * coroutine but only after this yield and after the main loop
+     * has gone through the next iteration.
+     */
     qemu_coroutine_yield();
     assert(qemu_in_coroutine());
+
+    /* TODO: OSv implements wait morphing here, where the wakeup
+     * primitive automatically places the woken coroutine on the
+     * mutex's queue.  This avoids the thundering herd effect.
+     */
+    if (mutex) {
+        qemu_co_mutex_lock(mutex);
+    }
 }
 
 /**
@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_rdlock(CoRwlock *lock)
     Coroutine *self = qemu_coroutine_self();
 
     while (lock->writer) {
-        qemu_co_queue_wait(&lock->queue);
+        qemu_co_queue_wait(&lock->queue, NULL);
     }
     lock->reader++;
     self->locks_held++;
@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_wrlock(CoRwlock *lock)
     Coroutine *self = qemu_coroutine_self();
 
     while (lock->writer || lock->reader) {
-        qemu_co_queue_wait(&lock->queue);
+        qemu_co_queue_wait(&lock->queue, NULL);
     }
     lock->writer = true;
     self->locks_held++;
-- 
2.9.3

From: Paolo Bonzini <pbonzini@redhat.com>

This adds a CoMutex around the existing CoQueue.  Because the write-side
can just take CoMutex, the old "writer" field is not necessary anymore.
Instead of removing it altogether, count the number of pending writers
during a read-side critical section and forbid further readers from
entering.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20170213181244.16297-7-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 include/qemu/coroutine.h   |  3 ++-
 util/qemu-coroutine-lock.c | 35 ++++++++++++++++++++++++-----------
 2 files changed, 26 insertions(+), 12 deletions(-)

diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
index XXXXXXX..XXXXXXX 100644
--- a/include/qemu/coroutine.h
+++ b/include/qemu/coroutine.h
@@ -XXX,XX +XXX,XX @@ bool qemu_co_queue_empty(CoQueue *queue);
 
 
 typedef struct CoRwlock {
-    bool writer;
+    int pending_writer;
     int reader;
+    CoMutex mutex;
     CoQueue queue;
 } CoRwlock;
 
diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c
index XXXXXXX..XXXXXXX 100644
--- a/util/qemu-coroutine-lock.c
+++ b/util/qemu-coroutine-lock.c
@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_init(CoRwlock *lock)
 {
     memset(lock, 0, sizeof(*lock));
     qemu_co_queue_init(&lock->queue);
+    qemu_co_mutex_init(&lock->mutex);
 }
 
 void qemu_co_rwlock_rdlock(CoRwlock *lock)
 {
     Coroutine *self = qemu_coroutine_self();
 
-    while (lock->writer) {
-        qemu_co_queue_wait(&lock->queue, NULL);
+    qemu_co_mutex_lock(&lock->mutex);
+    /* For fairness, wait if a writer is in line.  */
+    while (lock->pending_writer) {
+        qemu_co_queue_wait(&lock->queue, &lock->mutex);
     }
     lock->reader++;
+    qemu_co_mutex_unlock(&lock->mutex);
+
+    /* The rest of the read-side critical section is run without the mutex.  */
     self->locks_held++;
 }
 
@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_unlock(CoRwlock *lock)
     Coroutine *self = qemu_coroutine_self();
 
     assert(qemu_in_coroutine());
-    if (lock->writer) {
-        lock->writer = false;
+    if (!lock->reader) {
+        /* The critical section started in qemu_co_rwlock_wrlock.  */
         qemu_co_queue_restart_all(&lock->queue);
     } else {
+        self->locks_held--;
+
+        qemu_co_mutex_lock(&lock->mutex);
         lock->reader--;
         assert(lock->reader >= 0);
         /* Wakeup only one waiting writer */
@@ -XXX,XX +XXX,XX @@ void qemu_co_rwlock_unlock(CoRwlock *lock)
             qemu_co_queue_next(&lock->queue);
         }
     }
-    self->locks_held--;
+    qemu_co_mutex_unlock(&lock->mutex);
 }
 
 void qemu_co_rwlock_wrlock(CoRwlock *lock)
 {
-    Coroutine *self = qemu_coroutine_self();
-
-    while (lock->writer || lock->reader) {
-        qemu_co_queue_wait(&lock->queue, NULL);
+    qemu_co_mutex_lock(&lock->mutex);
+    lock->pending_writer++;
+    while (lock->reader) {
+        qemu_co_queue_wait(&lock->queue, &lock->mutex);
     }
-    lock->writer = true;
-    self->locks_held++;
+    lock->pending_writer--;
+
+    /* The rest of the write-side critical section is run with
+     * the mutex taken, so that lock->reader remains zero.
+     * There is no need to update self->locks_held.
+     */
 }
-- 
2.9.3