Series comparison

-[PULL 0/3] Block patches
+[Qemu-devel] [PULL v2 0/8] Block patches
-The following changes since commit 9cf289af47bcfae5c75de37d8e5d6fd23705322c:
+The following changes since commit 474f3938d79ab36b9231c9ad3b5a9314c2aeacde:
-  Merge tag 'qga-pull-request' of gitlab.com:marcandre.lureau/qemu into staging (2022-05-04 03:42:49 -0700)
+  Merge remote-tracking branch 'remotes/amarkovic/tags/mips-queue-jun-21-2019' into staging (2019-06-21 15:40:50 +0100)
 are available in the Git repository at:
-  https://gitlab.com/stefanha/qemu.git tags/block-pull-request
+  https://github.com/XanClic/qemu.git tags/pull-block-2019-06-24
-for you to fetch changes up to bef2e050d6a7feb865854c65570c496ac5a8cf53:
+for you to fetch changes up to ab5d4a30f7f3803ca5106b370969c1b7b54136f8:
-  util/event-loop-base: Introduce options to set the thread pool size (2022-05-04 17:02:19 +0100)
+  iotests: Fix 205 for concurrent runs (2019-06-24 16:01:40 +0200)
 ----------------------------------------------------------------
-Pull request
+Block patches:
 - The SSH block driver now uses libssh instead of libssh2
 - The VMDK block driver gets read-only support for the seSparse
   subformat
 - Various fixes
-Add new thread-pool-min/thread-pool-max parameters to control the thread pool
+---
-used for async I/O.
 v2:
 - Squashed Pino's fix for pre-0.8 libssh into the libssh patch
 ----------------------------------------------------------------
+Anton Nefedov (1):
+  iotest 134: test cluster-misaligned encrypted write
-Nicolas Saenz Julienne (3):
+Klaus Birkelund Jensen (1):
-  Introduce event-loop-base abstract class
+  nvme: do not advertise support for unsupported arbitration mechanism
   util/main-loop: Introduce the main loop into QOM
   util/event-loop-base: Introduce options to set the thread pool size
- qapi/qom.json                    |  43 ++++++++--
+Max Reitz (1):
- meson.build                      |  26 +++---
+  iotests: Fix 205 for concurrent runs
- include/block/aio.h              |  10 +++
- include/block/thread-pool.h      |   3 +
+Pino Toscano (1):
- include/qemu/main-loop.h         |  10 +++
+  ssh: switch from libssh2 to libssh
- include/sysemu/event-loop-base.h |  41 +++++++++
- include/sysemu/iothread.h        |   6 +-
+Sam Eiderman (3):
- event-loop-base.c                | 140 +++++++++++++++++++++++++++++++
+  vmdk: Fix comment regarding max l1_size coverage
- iothread.c                       |  68 +++++----------
+  vmdk: Reduce the max bound for L1 table size
- util/aio-posix.c                 |   1 +
+  vmdk: Add read-only support for seSparse snapshots
- util/async.c                     |  20 +++++
- util/main-loop.c                 |  65 ++++++++++++++
+Vladimir Sementsov-Ogievskiy (1):
- util/thread-pool.c               |  55 +++++++++++-
+  blockdev: enable non-root nodes for transaction drive-backup source
-files changed, 419 insertions(+), 69 deletions(-)
- create mode 100644 include/sysemu/event-loop-base.h
+ configure                                     |  65 +-
- create mode 100644 event-loop-base.c
+ block/Makefile.objs                           |   6 +-
  block/ssh.c                                   | 652 ++++++++++--------
  block/vmdk.c                                  | 372 +++++++++-
  blockdev.c                                    |   2 +-
  hw/block/nvme.c                               |   1 -
  .travis.yml                                   |   4 +-
  block/trace-events                            |  14 +-
  docs/qemu-block-drivers.texi                  |   2 +-
  .../dockerfiles/debian-win32-cross.docker     |   1 -
  .../dockerfiles/debian-win64-cross.docker     |   1 -
  tests/docker/dockerfiles/fedora.docker        |   4 +-
  tests/docker/dockerfiles/ubuntu.docker        |   2 +-
  tests/docker/dockerfiles/ubuntu1804.docker    |   2 +-
  tests/qemu-iotests/059.out                    |   2 +-
  tests/qemu-iotests/134                        |   9 +
  tests/qemu-iotests/134.out                    |  10 +
  tests/qemu-iotests/205                        |   2 +-
  tests/qemu-iotests/207                        |  54 +-
  tests/qemu-iotests/207.out                    |   2 +-
 files changed, 823 insertions(+), 384 deletions(-)
 --
-.35.1
+.21.0

-New patch
+[Qemu-devel] [PULL v2 1/8] nvme: do not advertise support for unsupported arbitration mechanism
+From: Klaus Birkelund Jensen <klaus@birkelund.eu>
+The device mistakenly reports that the Weighted Round Robin with Urgent
+Priority Class arbitration mechanism is supported.
+It is not.
+Signed-off-by: Klaus Birkelund Jensen <klaus.jensen@cnexlabs.com>
+Message-id: 20190606092530.14206-1-klaus@birkelund.eu
+Acked-by: Maxim Levitsky <mlevitsk@redhat.com>
+Signed-off-by: Max Reitz <mreitz@redhat.com>
+---
+ hw/block/nvme.c | 1 -
+file changed, 1 deletion(-)
+diff --git a/hw/block/nvme.c b/hw/block/nvme.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/block/nvme.c
++++ b/hw/block/nvme.c
+@@ -XXX,XX +XXX,XX @@ static void nvme_realize(PCIDevice *pci_dev, Error **errp)
+     n->bar.cap = 0;
+     NVME_CAP_SET_MQES(n->bar.cap, 0x7ff);
+     NVME_CAP_SET_CQR(n->bar.cap, 1);
+-    NVME_CAP_SET_AMS(n->bar.cap, 1);
+     NVME_CAP_SET_TO(n->bar.cap, 0xf);
+     NVME_CAP_SET_CSS(n->bar.cap, 1);
+     NVME_CAP_SET_MPSMAX(n->bar.cap, 4);
+--
+.21.0

-[PULL 3/3] util/event-loop-base: Introduce options to set the thread pool size
+[Qemu-devel] [PULL v2 2/8] blockdev: enable non-root nodes for transaction drive-backup source
-From: Nicolas Saenz Julienne <nsaenzju@redhat.com>
+From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
-The thread pool regulates itself: when idle, it kills threads until
+We forget to enable it for transaction .prepare, while it is already
-empty, when in demand, it creates new threads until full. This behaviour
+enabled in do_drive_backup since commit a2d665c1bc362
-doesn't play well with latency sensitive workloads where the price of
+    "blockdev: loosen restrictions on drive-backup source node"
 creating a new thread is too high. For example, when paired with qemu's
 '-mlock', or using safety features like SafeStack, creating a new thread
 has been measured take multiple milliseconds.
-In order to mitigate this let's introduce a new 'EventLoopBase'
+Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
-property to set the thread pool size. The threads will be created during
+Message-id: 20190618140804.59214-1-vsementsov@virtuozzo.com
-the pool's initialization or upon updating the property's value, remain
+Reviewed-by: John Snow <jsnow@redhat.com>
-available during its lifetime regardless of demand, and destroyed upon
+Signed-off-by: Max Reitz <mreitz@redhat.com>
-freeing it. A properly characterized workload will then be able to
+---
-configure the pool to avoid any latency spikes.
+ blockdev.c | 2 +-
 file changed, 1 insertion(+), 1 deletion(-)
-Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
+diff --git a/blockdev.c b/blockdev.c
 Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
 Acked-by: Markus Armbruster <armbru@redhat.com>
 Message-id: 20220425075723.20019-4-nsaenzju@redhat.com
 Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
 ---
  qapi/qom.json                    | 10 +++++-
  include/block/aio.h              | 10 ++++++
  include/block/thread-pool.h      |  3 ++
  include/sysemu/event-loop-base.h |  4 +++
  event-loop-base.c                | 23 +++++++++++++
  iothread.c                       |  3 ++
  util/aio-posix.c                 |  1 +
  util/async.c                     | 20 ++++++++++++
  util/main-loop.c                 |  9 ++++++
  util/thread-pool.c               | 55 +++++++++++++++++++++++++++++---
 files changed, 133 insertions(+), 5 deletions(-)
 diff --git a/qapi/qom.json b/qapi/qom.json
 index XXXXXXX..XXXXXXX 100644
---- a/qapi/qom.json
+--- a/blockdev.c
-+++ b/qapi/qom.json
++++ b/blockdev.c
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ static void drive_backup_prepare(BlkActionState *common, Error **errp)
- #                 0 means that the engine will use its default.
+     assert(common->action->type == TRANSACTION_ACTION_KIND_DRIVE_BACKUP);
- #                 (default: 0)
+     backup = common->action->u.drive_backup.data;
- #
-+# @thread-pool-min: minimum number of threads reserved in the thread pool
+-    bs = qmp_get_root_bs(backup->device, errp);
-+#                   (default:0)
++    bs = bdrv_lookup_bs(backup->device, backup->device, errp);
-+#
+     if (!bs) {
 +# @thread-pool-max: maximum number of threads the thread pool can contain
 +#                   (default:64)
 +#
  # Since: 7.1
  ##
  { 'struct': 'EventLoopBaseProperties',
 -  'data': { '*aio-max-batch': 'int' } }
 +  'data': { '*aio-max-batch': 'int',
 +            '*thread-pool-min': 'int',
 +            '*thread-pool-max': 'int' } }
  ##
  # @IothreadProperties:
 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 @@ struct AioContext {
      QSLIST_HEAD(, Coroutine) scheduled_coroutines;
      QEMUBH *co_schedule_bh;
 +    int thread_pool_min;
 +    int thread_pool_max;
      /* Thread pool for performing work and receiving completion callbacks.
       * Has its own locking.
       */
@@ -XXX,XX +XXX,XX @@ void aio_context_set_poll_params(AioContext *ctx, int64_t max_ns,
  void aio_context_set_aio_params(AioContext *ctx, int64_t max_batch,
                                  Error **errp);
 +/**
 + * aio_context_set_thread_pool_params:
 + * @ctx: the aio context
 + * @min: min number of threads to have readily available in the thread pool
 + * @min: max number of threads the thread pool can contain
 + */
 +void aio_context_set_thread_pool_params(AioContext *ctx, int64_t min,
 +                                        int64_t max, Error **errp);
  #endif
 diff --git a/include/block/thread-pool.h b/include/block/thread-pool.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/block/thread-pool.h
 +++ b/include/block/thread-pool.h
@@ -XXX,XX +XXX,XX @@
  #include "block/block.h"
 +#define THREAD_POOL_MAX_THREADS_DEFAULT         64
 +
  typedef int ThreadPoolFunc(void *opaque);
  typedef struct ThreadPool ThreadPool;
@@ -XXX,XX +XXX,XX @@ BlockAIOCB *thread_pool_submit_aio(ThreadPool *pool,
  int coroutine_fn thread_pool_submit_co(ThreadPool *pool,
          ThreadPoolFunc *func, void *arg);
  void thread_pool_submit(ThreadPool *pool, ThreadPoolFunc *func, void *arg);
 +void thread_pool_update_params(ThreadPool *pool, struct AioContext *ctx);
  #endif
 diff --git a/include/sysemu/event-loop-base.h b/include/sysemu/event-loop-base.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/sysemu/event-loop-base.h
 +++ b/include/sysemu/event-loop-base.h
@@ -XXX,XX +XXX,XX @@ struct EventLoopBase {
      /* AioContext AIO engine parameters */
      int64_t aio_max_batch;
 +
 +    /* AioContext thread pool parameters */
 +    int64_t thread_pool_min;
 +    int64_t thread_pool_max;
  };
  #endif
 diff --git a/event-loop-base.c b/event-loop-base.c
 index XXXXXXX..XXXXXXX 100644
 --- a/event-loop-base.c
 +++ b/event-loop-base.c
@@ -XXX,XX +XXX,XX @@
  #include "qemu/osdep.h"
  #include "qom/object_interfaces.h"
  #include "qapi/error.h"
 +#include "block/thread-pool.h"
  #include "sysemu/event-loop-base.h"
  typedef struct {
@@ -XXX,XX +XXX,XX @@ typedef struct {
      ptrdiff_t offset; /* field's byte offset in EventLoopBase struct */
  } EventLoopBaseParamInfo;
 +static void event_loop_base_instance_init(Object *obj)
 +{
 +    EventLoopBase *base = EVENT_LOOP_BASE(obj);
 +
 +    base->thread_pool_max = THREAD_POOL_MAX_THREADS_DEFAULT;
 +}
 +
  static EventLoopBaseParamInfo aio_max_batch_info = {
      "aio-max-batch", offsetof(EventLoopBase, aio_max_batch),
  };
 +static EventLoopBaseParamInfo thread_pool_min_info = {
 +    "thread-pool-min", offsetof(EventLoopBase, thread_pool_min),
 +};
 +static EventLoopBaseParamInfo thread_pool_max_info = {
 +    "thread-pool-max", offsetof(EventLoopBase, thread_pool_max),
 +};
  static void event_loop_base_get_param(Object *obj, Visitor *v,
          const char *name, void *opaque, Error **errp)
@@ -XXX,XX +XXX,XX @@ static void event_loop_base_class_init(ObjectClass *klass, void *class_data)
                                event_loop_base_get_param,
                                event_loop_base_set_param,
                                NULL, &aio_max_batch_info);
 +    object_class_property_add(klass, "thread-pool-min", "int",
 +                              event_loop_base_get_param,
 +                              event_loop_base_set_param,
 +                              NULL, &thread_pool_min_info);
 +    object_class_property_add(klass, "thread-pool-max", "int",
 +                              event_loop_base_get_param,
 +                              event_loop_base_set_param,
 +                              NULL, &thread_pool_max_info);
  }
  static const TypeInfo event_loop_base_info = {
      .name = TYPE_EVENT_LOOP_BASE,
      .parent = TYPE_OBJECT,
      .instance_size = sizeof(EventLoopBase),
 +    .instance_init = event_loop_base_instance_init,
      .class_size = sizeof(EventLoopBaseClass),
      .class_init = event_loop_base_class_init,
      .abstract = true,
 diff --git a/iothread.c b/iothread.c
 index XXXXXXX..XXXXXXX 100644
 --- a/iothread.c
 +++ b/iothread.c
@@ -XXX,XX +XXX,XX @@ static void iothread_set_aio_context_params(EventLoopBase *base, Error **errp)
      aio_context_set_aio_params(iothread->ctx,
                                 iothread->parent_obj.aio_max_batch,
                                 errp);
 +
 +    aio_context_set_thread_pool_params(iothread->ctx, base->thread_pool_min,
 +                                       base->thread_pool_max, errp);
  }
 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 @@
  #include "qemu/osdep.h"
  #include "block/block.h"
 +#include "block/thread-pool.h"
  #include "qemu/main-loop.h"
  #include "qemu/rcu.h"
  #include "qemu/rcu_queue.h"
 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 @@ AioContext *aio_context_new(Error **errp)
      ctx->aio_max_batch = 0;
 +    ctx->thread_pool_min = 0;
 +    ctx->thread_pool_max = THREAD_POOL_MAX_THREADS_DEFAULT;
 +
      return ctx;
  fail:
      g_source_destroy(&ctx->source);
@@ -XXX,XX +XXX,XX @@ void qemu_set_current_aio_context(AioContext *ctx)
      assert(!get_my_aiocontext());
      set_my_aiocontext(ctx);
  }
 +
 +void aio_context_set_thread_pool_params(AioContext *ctx, int64_t min,
 +                                        int64_t max, Error **errp)
 +{
 +
 +    if (min > max || !max || min > INT_MAX || max > INT_MAX) {
 +        error_setg(errp, "bad thread-pool-min/thread-pool-max values");
 +        return;
 +    }
 +
 +    ctx->thread_pool_min = min;
 +    ctx->thread_pool_max = max;
 +
 +    if (ctx->thread_pool) {
 +        thread_pool_update_params(ctx->thread_pool, ctx);
 +    }
 +}
 diff --git a/util/main-loop.c b/util/main-loop.c
 index XXXXXXX..XXXXXXX 100644
 --- a/util/main-loop.c
 +++ b/util/main-loop.c
@@ -XXX,XX +XXX,XX @@
  #include "sysemu/replay.h"
  #include "qemu/main-loop.h"
  #include "block/aio.h"
 +#include "block/thread-pool.h"
  #include "qemu/error-report.h"
  #include "qemu/queue.h"
  #include "qemu/compiler.h"
@@ -XXX,XX +XXX,XX @@ int qemu_init_main_loop(Error **errp)
  static void main_loop_update_params(EventLoopBase *base, Error **errp)
  {
 +    ERRP_GUARD();
 +
      if (!qemu_aio_context) {
          error_setg(errp, "qemu aio context not ready");
          return;
      }
-     aio_context_set_aio_params(qemu_aio_context, base->aio_max_batch, errp);
-+    if (*errp) {
-+        return;
-+    }
-+
-+    aio_context_set_thread_pool_params(qemu_aio_context, base->thread_pool_min,
-+                                       base->thread_pool_max, errp);
- }
- MainLoop *mloop;
-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 @@ struct ThreadPool {
-     QemuMutex lock;
-     QemuCond worker_stopped;
-     QemuSemaphore sem;
--    int max_threads;
-     QEMUBH *new_thread_bh;
-     /* The following variables are only accessed from one AioContext. */
-@@ -XXX,XX +XXX,XX @@ struct ThreadPool {
-     int new_threads;     /* backlog of threads we need to create */
-     int pending_threads; /* threads created but not running yet */
-     bool stopping;
-+    int min_threads;
-+    int max_threads;
- };
-+static inline bool back_to_sleep(ThreadPool *pool, int ret)
-+{
-+    /*
-+     * The semaphore timed out, we should exit the loop except when:
-+     *  - There is work to do, we raced with the signal.
-+     *  - The max threads threshold just changed, we raced with the signal.
-+     *  - The thread pool forces a minimum number of readily available threads.
-+     */
-+    if (ret == -1 && (!QTAILQ_EMPTY(&pool->request_list) ||
-+            pool->cur_threads > pool->max_threads ||
-+            pool->cur_threads <= pool->min_threads)) {
-+            return true;
-+    }
-+
-+    return false;
-+}
-+
- static void *worker_thread(void *opaque)
- {
-     ThreadPool *pool = opaque;
-@@ -XXX,XX +XXX,XX @@ static void *worker_thread(void *opaque)
-             ret = qemu_sem_timedwait(&pool->sem, 10000);
-             qemu_mutex_lock(&pool->lock);
-             pool->idle_threads--;
--        } while (ret == -1 && !QTAILQ_EMPTY(&pool->request_list));
--        if (ret == -1 || pool->stopping) {
-+        } while (back_to_sleep(pool, ret));
-+        if (ret == -1 || pool->stopping ||
-+            pool->cur_threads > pool->max_threads) {
-             break;
-         }
-@@ -XXX,XX +XXX,XX @@ void thread_pool_submit(ThreadPool *pool, ThreadPoolFunc *func, void *arg)
-     thread_pool_submit_aio(pool, func, arg, NULL, NULL);
- }
-+void thread_pool_update_params(ThreadPool *pool, AioContext *ctx)
-+{
-+    qemu_mutex_lock(&pool->lock);
-+
-+    pool->min_threads = ctx->thread_pool_min;
-+    pool->max_threads = ctx->thread_pool_max;
-+
-+    /*
-+     * We either have to:
-+     *  - Increase the number available of threads until over the min_threads
-+     *    threshold.
-+     *  - Decrease the number of available threads until under the max_threads
-+     *    threshold.
-+     *  - Do nothing. The current number of threads fall in between the min and
-+     *    max thresholds. We'll let the pool manage itself.
-+     */
-+    for (int i = pool->cur_threads; i < pool->min_threads; i++) {
-+        spawn_thread(pool);
-+    }
-+
-+    for (int i = pool->cur_threads; i > pool->max_threads; i--) {
-+        qemu_sem_post(&pool->sem);
-+    }
-+
-+    qemu_mutex_unlock(&pool->lock);
-+}
-+
- static void thread_pool_init_one(ThreadPool *pool, AioContext *ctx)
- {
-     if (!ctx) {
-@@ -XXX,XX +XXX,XX @@ static void thread_pool_init_one(ThreadPool *pool, AioContext *ctx)
-     qemu_mutex_init(&pool->lock);
-     qemu_cond_init(&pool->worker_stopped);
-     qemu_sem_init(&pool->sem, 0);
--    pool->max_threads = 64;
-     pool->new_thread_bh = aio_bh_new(ctx, spawn_thread_bh_fn, pool);
-     QLIST_INIT(&pool->head);
-     QTAILQ_INIT(&pool->request_list);
-+
-+    thread_pool_update_params(pool, ctx);
- }
- ThreadPool *thread_pool_new(AioContext *ctx)
 --
-.35.1
+.21.0

-New patch
+[Qemu-devel] [PULL v2 3/8] iotest 134: test cluster-misaligned encrypted write
+From: Anton Nefedov <anton.nefedov@virtuozzo.com>
+COW (even empty/zero) areas require encryption too
+Signed-off-by: Anton Nefedov <anton.nefedov@virtuozzo.com>
+Reviewed-by: Eric Blake <eblake@redhat.com>
+Reviewed-by: Max Reitz <mreitz@redhat.com>
+Reviewed-by: Alberto Garcia <berto@igalia.com>
+Message-id: 20190516143028.81155-1-anton.nefedov@virtuozzo.com
+Signed-off-by: Max Reitz <mreitz@redhat.com>
+---
+ tests/qemu-iotests/134     |  9 +++++++++
+ tests/qemu-iotests/134.out | 10 ++++++++++
+files changed, 19 insertions(+)
+diff --git a/tests/qemu-iotests/134 b/tests/qemu-iotests/134
+index XXXXXXX..XXXXXXX 100755
+--- a/tests/qemu-iotests/134
++++ b/tests/qemu-iotests/134
+@@ -XXX,XX +XXX,XX @@ echo
+ echo "== reading whole image =="
+ $QEMU_IO --object $SECRET -c "read 0 $size" --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
++echo
++echo "== rewriting cluster part =="
++$QEMU_IO --object $SECRET -c "write -P 0xb 512 512" --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
++
++echo
++echo "== verify pattern =="
++$QEMU_IO --object $SECRET -c "read -P 0 0 512"  --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
++$QEMU_IO --object $SECRET -c "read -P 0xb 512 512"  --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
++
+ echo
+ echo "== rewriting whole image =="
+ $QEMU_IO --object $SECRET -c "write -P 0xa 0 $size" --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
+diff --git a/tests/qemu-iotests/134.out b/tests/qemu-iotests/134.out
+index XXXXXXX..XXXXXXX 100644
+--- a/tests/qemu-iotests/134.out
++++ b/tests/qemu-iotests/134.out
+@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 encryption=on encrypt.
+ read 134217728/134217728 bytes at offset 0
+MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
++== rewriting cluster part ==
++wrote 512/512 bytes at offset 512
++512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
++
++== verify pattern ==
++read 512/512 bytes at offset 0
++512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
++read 512/512 bytes at offset 512
++512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
++
+ == rewriting whole image ==
+ wrote 134217728/134217728 bytes at offset 0
+MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+--
+.21.0

-New patch
+[Qemu-devel] [PULL v2 4/8] vmdk: Fix comment regarding max l1_size coverage
+From: Sam Eiderman <shmuel.eiderman@oracle.com>
+Commit b0651b8c246d ("vmdk: Move l1_size check into vmdk_add_extent")
+extended the l1_size check from VMDK4 to VMDK3 but did not update the
+default coverage in the moved comment.
+The previous vmdk4 calculation:
+    (512 * 1024 * 1024) * 512(l2 entries) * 65536(grain) = 16PB
+The added vmdk3 calculation:
+    (512 * 1024 * 1024) * 4096(l2 entries) * 512(grain) = 1PB
+Adding the calculation of vmdk3 to the comment.
+In any case, VMware does not offer virtual disks more than 2TB for
+vmdk4/vmdk3 or 64TB for the new undocumented seSparse format which is
+not implemented yet in qemu.
+Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
+Reviewed-by: Eyal Moscovici <eyal.moscovici@oracle.com>
+Reviewed-by: Liran Alon <liran.alon@oracle.com>
+Reviewed-by: Arbel Moshe <arbel.moshe@oracle.com>
+Signed-off-by: Sam Eiderman <shmuel.eiderman@oracle.com>
+Message-id: 20190620091057.47441-2-shmuel.eiderman@oracle.com
+Reviewed-by: yuchenlin <yuchenlin@synology.com>
+Reviewed-by: Max Reitz <mreitz@redhat.com>
+Signed-off-by: Max Reitz <mreitz@redhat.com>
+---
+ block/vmdk.c | 11 ++++++++---
+file changed, 8 insertions(+), 3 deletions(-)
+diff --git a/block/vmdk.c b/block/vmdk.c
+index XXXXXXX..XXXXXXX 100644
+--- a/block/vmdk.c
++++ b/block/vmdk.c
+@@ -XXX,XX +XXX,XX @@ static int vmdk_add_extent(BlockDriverState *bs,
+         return -EFBIG;
+     }
+     if (l1_size > 512 * 1024 * 1024) {
+-        /* Although with big capacity and small l1_entry_sectors, we can get a
++        /*
++         * Although with big capacity and small l1_entry_sectors, we can get a
+          * big l1_size, we don't want unbounded value to allocate the table.
+-         * Limit it to 512M, which is 16PB for default cluster and L2 table
+-         * size */
++         * Limit it to 512M, which is:
++         *     16PB - for default "Hosted Sparse Extent" (VMDK4)
++         *            cluster size: 64KB, L2 table size: 512 entries
++         *     1PB  - for default "ESXi Host Sparse Extent" (VMDK3/vmfsSparse)
++         *            cluster size: 512B, L2 table size: 4096 entries
++         */
+         error_setg(errp, "L1 size too big");
+         return -EFBIG;
+     }
+--
+.21.0

-New patch
+[Qemu-devel] [PULL v2 5/8] vmdk: Reduce the max bound for L1 table size
+From: Sam Eiderman <shmuel.eiderman@oracle.com>
+M of L1 entries is a very loose bound, only 32M are required to store
+the maximal supported VMDK file size of 2TB.
+Fixed qemu-iotest 59# - now failure occures before on impossible L1
+table size.
+Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
+Reviewed-by: Eyal Moscovici <eyal.moscovici@oracle.com>
+Reviewed-by: Liran Alon <liran.alon@oracle.com>
+Reviewed-by: Arbel Moshe <arbel.moshe@oracle.com>
+Signed-off-by: Sam Eiderman <shmuel.eiderman@oracle.com>
+Message-id: 20190620091057.47441-3-shmuel.eiderman@oracle.com
+Reviewed-by: Max Reitz <mreitz@redhat.com>
+Signed-off-by: Max Reitz <mreitz@redhat.com>
+---
+ block/vmdk.c               | 13 +++++++------
+ tests/qemu-iotests/059.out |  2 +-
+files changed, 8 insertions(+), 7 deletions(-)
+diff --git a/block/vmdk.c b/block/vmdk.c
+index XXXXXXX..XXXXXXX 100644
+--- a/block/vmdk.c
++++ b/block/vmdk.c
+@@ -XXX,XX +XXX,XX @@ static int vmdk_add_extent(BlockDriverState *bs,
+         error_setg(errp, "Invalid granularity, image may be corrupt");
+         return -EFBIG;
+     }
+-    if (l1_size > 512 * 1024 * 1024) {
++    if (l1_size > 32 * 1024 * 1024) {
+         /*
+          * Although with big capacity and small l1_entry_sectors, we can get a
+          * big l1_size, we don't want unbounded value to allocate the table.
+-         * Limit it to 512M, which is:
+-         *     16PB - for default "Hosted Sparse Extent" (VMDK4)
+-         *            cluster size: 64KB, L2 table size: 512 entries
+-         *     1PB  - for default "ESXi Host Sparse Extent" (VMDK3/vmfsSparse)
+-         *            cluster size: 512B, L2 table size: 4096 entries
++         * Limit it to 32M, which is enough to store:
++         *     8TB  - for both VMDK3 & VMDK4 with
++         *            minimal cluster size: 512B
++         *            minimal L2 table size: 512 entries
++         *            8 TB is still more than the maximal value supported for
++         *            VMDK3 & VMDK4 which is 2TB.
+          */
+         error_setg(errp, "L1 size too big");
+         return -EFBIG;
+diff --git a/tests/qemu-iotests/059.out b/tests/qemu-iotests/059.out
+index XXXXXXX..XXXXXXX 100644
+--- a/tests/qemu-iotests/059.out
++++ b/tests/qemu-iotests/059.out
+@@ -XXX,XX +XXX,XX @@ Offset          Length          Mapped to       File
+x140000000     0x10000         0x50000         TEST_DIR/t-s003.vmdk
+ === Testing afl image with a very large capacity ===
+-qemu-img: Can't get image size 'TEST_DIR/afl9.IMGFMT': File too large
++qemu-img: Could not open 'TEST_DIR/afl9.IMGFMT': L1 size too big
+ *** done
+--
+.21.0

-[PULL 1/3] Introduce event-loop-base abstract class
+[Qemu-devel] [PULL v2 6/8] vmdk: Add read-only support for seSparse snapshots
-From: Nicolas Saenz Julienne <nsaenzju@redhat.com>
+From: Sam Eiderman <shmuel.eiderman@oracle.com>
-Introduce the 'event-loop-base' abstract class, it'll hold the
+Until ESXi 6.5 VMware used the vmfsSparse format for snapshots (VMDK3 in
-properties common to all event loops and provide the necessary hooks for
+QEMU).
-their creation and maintenance. Then have iothread inherit from it.
+This format was lacking in the following:
-EventLoopBaseClass is defined as user creatable and provides a hook for
-its children to attach themselves to the user creatable class 'complete'
+    * Grain directory (L1) and grain table (L2) entries were 32-bit,
-function. It also provides an update_params() callback to propagate
+      allowing access to only 2TB (slightly less) of data.
-property changes onto its children.
+    * The grain size (default) was 512 bytes - leading to data
+      fragmentation and many grain tables.
-The new 'event-loop-base' class will live in the root directory. It is
+    * For space reclamation purposes, it was necessary to find all the
-built on its own using the 'link_whole' option (there are no direct
+      grains which are not pointed to by any grain table - so a reverse
-function dependencies between the class and its children, it all happens
+      mapping of "offset of grain in vmdk" to "grain table" must be
-trough 'constructor' magic). And also imposes new compilation
+      constructed - which takes large amounts of CPU/RAM.
-dependencies:
+The format specification can be found in VMware's documentation:
-    qom <- event-loop-base <- blockdev (iothread.c)
+https://www.vmware.com/support/developer/vddk/vmdk_50_technote.pdf
-And in subsequent patches:
+In ESXi 6.5, to support snapshot files larger than 2TB, a new format was
+introduced: SESparse (Space Efficient).
-    qom <- event-loop-base <- qemuutil (util/main-loop.c)
+This format fixes the above issues:
-All this forced some amount of reordering in meson.build:
+    * All entries are now 64-bit.
- - Moved qom build definition before qemuutil. Doing it the other way
+    * The grain size (default) is 4KB.
-   around (i.e. moving qemuutil after qom) isn't possible as a lot of
+    * Grain directory and grain tables are now located at the beginning
-   core libraries that live in between the two depend on it.
+      of the file.
+      + seSparse format reserves space for all grain tables.
- - Process the 'hw' subdir earlier, as it introduces files into the
+      + Grain tables can be addressed using an index.
-   'qom' source set.
+      + Grains are located in the end of the file and can also be
+        addressed with an index.
-No functional changes intended.
+      - seSparse vmdks of large disks (64TB) have huge preallocated
+        headers - mainly due to L2 tables, even for empty snapshots.
-Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
+    * The header contains a reverse mapping ("backmap") of "offset of
-Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
+      grain in vmdk" to "grain table" and a bitmap ("free bitmap") which
-Acked-by: Markus Armbruster <armbru@redhat.com>
+      specifies for each grain - whether it is allocated or not.
-Message-id: 20220425075723.20019-2-nsaenzju@redhat.com
+      Using these data structures we can implement space reclamation
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
+      efficiently.
     * Due to the fact that the header now maintains two mappings:
         * The regular one (grain directory & grain tables)
         * A reverse one (backmap and free bitmap)
       These data structures can lose consistency upon crash and result
       in a corrupted VMDK.
       Therefore, a journal is also added to the VMDK and is replayed
       when the VMware reopens the file after a crash.
 Since ESXi 6.7 - SESparse is the only snapshot format available.
 Unfortunately, VMware does not provide documentation regarding the new
 seSparse format.
 This commit is based on black-box research of the seSparse format.
 Various in-guest block operations and their effect on the snapshot file
 were tested.
 The only VMware provided source of information (regarding the underlying
 implementation) was a log file on the ESXi:
     /var/log/hostd.log
 Whenever an seSparse snapshot is created - the log is being populated
 with seSparse records.
 Relevant log records are of the form:
 [...] Const Header:
 [...]  constMagic     = 0xcafebabe
 [...]  version        = 2.1
 [...]  capacity       = 204800
 [...]  grainSize      = 8
 [...]  grainTableSize = 64
 [...]  flags          = 0
 [...] Extents:
 [...]  Header         : <1 : 1>
 [...]  JournalHdr     : <2 : 2>
 [...]  Journal        : <2048 : 2048>
 [...]  GrainDirectory : <4096 : 2048>
 [...]  GrainTables    : <6144 : 2048>
 [...]  FreeBitmap     : <8192 : 2048>
 [...]  BackMap        : <10240 : 2048>
 [...]  Grain          : <12288 : 204800>
 [...] Volatile Header:
 [...] volatileMagic     = 0xcafecafe
 [...] FreeGTNumber      = 0
 [...] nextTxnSeqNumber  = 0
 [...] replayJournal     = 0
 The sizes that are seen in the log file are in sectors.
 Extents are of the following format: <offset : size>
 This commit is a strict implementation which enforces:
     * magics
     * version number 2.1
     * grain size of 8 sectors  (4KB)
     * grain table size of 64 sectors
     * zero flags
     * extent locations
 Additionally, this commit proivdes only a subset of the functionality
 offered by seSparse's format:
     * Read-only
     * No journal replay
     * No space reclamation
     * No unmap support
 Hence, journal header, journal, free bitmap and backmap extents are
 unused, only the "classic" (L1 -> L2 -> data) grain access is
 implemented.
 However there are several differences in the grain access itself.
 Grain directory (L1):
     * Grain directory entries are indexes (not offsets) to grain
       tables.
     * Valid grain directory entries have their highest nibble set to
 x1.
     * Since grain tables are always located in the beginning of the
       file - the index can fit into 32 bits - so we can use its low
       part if it's valid.
 Grain table (L2):
     * Grain table entries are indexes (not offsets) to grains.
     * If the highest nibble of the entry is:
 x0:
             The grain in not allocated.
             The rest of the bytes are 0.
 x1:
             The grain is unmapped - guest sees a zero grain.
             The rest of the bits point to the previously mapped grain,
             see 0x3 case.
 x2:
             The grain is zero.
 x3:
             The grain is allocated - to get the index calculate:
             ((entry & 0x0fff000000000000) >> 48) |
             ((entry & 0x0000ffffffffffff) << 12)
     * The difference between 0x1 and 0x2 is that 0x1 is an unallocated
       grain which results from the guest using sg_unmap to unmap the
       grain - but the grain itself still exists in the grain extent - a
       space reclamation procedure should delete it.
       Unmapping a zero grain has no effect (0x2 will not change to 0x1)
       but unmapping an unallocated grain will (0x0 to 0x1) - naturally.
 In order to implement seSparse some fields had to be changed to support
 both 32-bit and 64-bit entry sizes.
 Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
 Reviewed-by: Eyal Moscovici <eyal.moscovici@oracle.com>
 Reviewed-by: Arbel Moshe <arbel.moshe@oracle.com>
 Signed-off-by: Sam Eiderman <shmuel.eiderman@oracle.com>
 Message-id: 20190620091057.47441-4-shmuel.eiderman@oracle.com
 Signed-off-by: Max Reitz <mreitz@redhat.com>
 ---
- qapi/qom.json                    |  22 +++++--
+ block/vmdk.c | 358 ++++++++++++++++++++++++++++++++++++++++++++++++---
- meson.build                      |  23 ++++---
+file changed, 342 insertions(+), 16 deletions(-)
- include/sysemu/event-loop-base.h |  36 +++++++++++
- include/sysemu/iothread.h        |   6 +-
+diff --git a/block/vmdk.c b/block/vmdk.c
  event-loop-base.c                | 104 +++++++++++++++++++++++++++++++
  iothread.c                       |  65 ++++++-------------
 files changed, 192 insertions(+), 64 deletions(-)
  create mode 100644 include/sysemu/event-loop-base.h
  create mode 100644 event-loop-base.c
 diff --git a/qapi/qom.json b/qapi/qom.json
 index XXXXXXX..XXXXXXX 100644
---- a/qapi/qom.json
+--- a/block/vmdk.c
-+++ b/qapi/qom.json
++++ b/block/vmdk.c
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ typedef struct {
-             '*repeat': 'bool',
+     uint16_t compressAlgorithm;
-             '*grab-toggle': 'GrabToggleKeys' } }
+ } QEMU_PACKED VMDK4Header;
-+##
++typedef struct VMDKSESparseConstHeader {
-+# @EventLoopBaseProperties:
++    uint64_t magic;
-+#
++    uint64_t version;
-+# Common properties for event loops
++    uint64_t capacity;
-+#
++    uint64_t grain_size;
-+# @aio-max-batch: maximum number of requests in a batch for the AIO engine,
++    uint64_t grain_table_size;
-+#                 0 means that the engine will use its default.
++    uint64_t flags;
-+#                 (default: 0)
++    uint64_t reserved1;
-+#
++    uint64_t reserved2;
-+# Since: 7.1
++    uint64_t reserved3;
-+##
++    uint64_t reserved4;
-+{ 'struct': 'EventLoopBaseProperties',
++    uint64_t volatile_header_offset;
-+  'data': { '*aio-max-batch': 'int' } }
++    uint64_t volatile_header_size;
-+
++    uint64_t journal_header_offset;
- ##
++    uint64_t journal_header_size;
- # @IothreadProperties:
++    uint64_t journal_offset;
- #
++    uint64_t journal_size;
-@@ -XXX,XX +XXX,XX @@
++    uint64_t grain_dir_offset;
- #               algorithm detects it is spending too long polling without
++    uint64_t grain_dir_size;
- #               encountering events. 0 selects a default behaviour (default: 0)
++    uint64_t grain_tables_offset;
- #
++    uint64_t grain_tables_size;
--# @aio-max-batch: maximum number of requests in a batch for the AIO engine,
++    uint64_t free_bitmap_offset;
--#                 0 means that the engine will use its default
++    uint64_t free_bitmap_size;
--#                 (default:0, since 6.1)
++    uint64_t backmap_offset;
-+# The @aio-max-batch option is available since 6.1.
++    uint64_t backmap_size;
- #
++    uint64_t grains_offset;
- # Since: 2.0
++    uint64_t grains_size;
- ##
++    uint8_t pad[304];
- { 'struct': 'IothreadProperties',
++} QEMU_PACKED VMDKSESparseConstHeader;
-+  'base': 'EventLoopBaseProperties',
++
-   'data': { '*poll-max-ns': 'int',
++typedef struct VMDKSESparseVolatileHeader {
-             '*poll-grow': 'int',
++    uint64_t magic;
--            '*poll-shrink': 'int',
++    uint64_t free_gt_number;
--            '*aio-max-batch': 'int' } }
++    uint64_t next_txn_seq_number;
-+            '*poll-shrink': 'int' } }
++    uint64_t replay_journal;
++    uint8_t pad[480];
- ##
++} QEMU_PACKED VMDKSESparseVolatileHeader;
- # @MemoryBackendProperties:
++
-diff --git a/meson.build b/meson.build
+ #define L2_CACHE_SIZE 16
-index XXXXXXX..XXXXXXX 100644
---- a/meson.build
+ typedef struct VmdkExtent {
-+++ b/meson.build
+@@ -XXX,XX +XXX,XX @@ typedef struct VmdkExtent {
-@@ -XXX,XX +XXX,XX @@ subdir('qom')
+     bool compressed;
- subdir('authz')
+     bool has_marker;
- subdir('crypto')
+     bool has_zero_grain;
- subdir('ui')
++    bool sesparse;
-+subdir('hw')
++    uint64_t sesparse_l2_tables_offset;
++    uint64_t sesparse_clusters_offset;
++    int32_t entry_size;
- if enable_modules
+     int version;
-@@ -XXX,XX +XXX,XX @@ if enable_modules
+     int64_t sectors;
-   modulecommon = declare_dependency(link_whole: libmodulecommon, compile_args: '-DBUILD_DSO')
+     int64_t end_sector;
- endif
+     int64_t flat_start_offset;
+     int64_t l1_table_offset;
-+qom_ss = qom_ss.apply(config_host, strict: false)
+     int64_t l1_backup_table_offset;
-+libqom = static_library('qom', qom_ss.sources() + genh,
+-    uint32_t *l1_table;
-+                        dependencies: [qom_ss.dependencies()],
++    void *l1_table;
-+                        name_suffix: 'fa')
+     uint32_t *l1_backup_table;
-+qom = declare_dependency(link_whole: libqom)
+     unsigned int l1_size;
-+
+     uint32_t l1_entry_sectors;
-+event_loop_base = files('event-loop-base.c')
-+event_loop_base = static_library('event-loop-base', sources: event_loop_base + genh,
+     unsigned int l2_size;
-+                                 build_by_default: true)
+-    uint32_t *l2_cache;
-+event_loop_base = declare_dependency(link_whole: event_loop_base,
++    void *l2_cache;
-+                                     dependencies: [qom])
+     uint32_t l2_cache_offsets[L2_CACHE_SIZE];
-+
+     uint32_t l2_cache_counts[L2_CACHE_SIZE];
- stub_ss = stub_ss.apply(config_all, strict: false)
+@@ -XXX,XX +XXX,XX @@ static int vmdk_add_extent(BlockDriverState *bs,
- util_ss.add_all(trace_ss)
+          *            minimal L2 table size: 512 entries
-@@ -XXX,XX +XXX,XX @@ subdir('monitor')
+          *            8 TB is still more than the maximal value supported for
- subdir('net')
+          *            VMDK3 & VMDK4 which is 2TB.
- subdir('replay')
++         *     64TB - for "ESXi seSparse Extent"
- subdir('semihosting')
++         *            minimal cluster size: 512B (default is 4KB)
--subdir('hw')
++         *            L2 table size: 4096 entries (const).
- subdir('tcg')
++         *            64TB is more than the maximal value supported for
- subdir('fpu')
++         *            seSparse VMDKs (which is slightly less than 64TB)
- subdir('accel')
+          */
-@@ -XXX,XX +XXX,XX @@ qemu_syms = custom_target('qemu.syms', output: 'qemu.syms',
+         error_setg(errp, "L1 size too big");
-                              capture: true,
+         return -EFBIG;
-                              command: [undefsym, nm, '@INPUT@'])
+@@ -XXX,XX +XXX,XX @@ static int vmdk_add_extent(BlockDriverState *bs,
+     extent->l2_size = l2_size;
--qom_ss = qom_ss.apply(config_host, strict: false)
+     extent->cluster_sectors = flat ? sectors : cluster_sectors;
--libqom = static_library('qom', qom_ss.sources() + genh,
+     extent->next_cluster_sector = ROUND_UP(nb_sectors, cluster_sectors);
--                        dependencies: [qom_ss.dependencies()],
++    extent->entry_size = sizeof(uint32_t);
--                        name_suffix: 'fa')
--
+     if (s->num_extents > 1) {
--qom = declare_dependency(link_whole: libqom)
+         extent->end_sector = (*(extent - 1)).end_sector + extent->sectors;
--
+@@ -XXX,XX +XXX,XX @@ static int vmdk_init_tables(BlockDriverState *bs, VmdkExtent *extent,
- authz_ss = authz_ss.apply(config_host, strict: false)
+     int i;
- libauthz = static_library('authz', authz_ss.sources() + genh,
-                           dependencies: [authz_ss.dependencies()],
+     /* read the L1 table */
-@@ -XXX,XX +XXX,XX @@ libblockdev = static_library('blockdev', blockdev_ss.sources() + genh,
+-    l1_size = extent->l1_size * sizeof(uint32_t);
-                              build_by_default: false)
++    l1_size = extent->l1_size * extent->entry_size;
+     extent->l1_table = g_try_malloc(l1_size);
- blockdev = declare_dependency(link_whole: [libblockdev],
+     if (l1_size && extent->l1_table == NULL) {
--                              dependencies: [block])
+         return -ENOMEM;
-+                              dependencies: [block, event_loop_base])
+@@ -XXX,XX +XXX,XX @@ static int vmdk_init_tables(BlockDriverState *bs, VmdkExtent *extent,
+         goto fail_l1;
- qmp_ss = qmp_ss.apply(config_host, strict: false)
+     }
- libqmp = static_library('qmp', qmp_ss.sources() + genh,
+     for (i = 0; i < extent->l1_size; i++) {
-diff --git a/include/sysemu/event-loop-base.h b/include/sysemu/event-loop-base.h
+-        le32_to_cpus(&extent->l1_table[i]);
-new file mode 100644
++        if (extent->entry_size == sizeof(uint64_t)) {
-index XXXXXXX..XXXXXXX
++            le64_to_cpus((uint64_t *)extent->l1_table + i);
---- /dev/null
++        } else {
-+++ b/include/sysemu/event-loop-base.h
++            assert(extent->entry_size == sizeof(uint32_t));
-@@ -XXX,XX +XXX,XX @@
++            le32_to_cpus((uint32_t *)extent->l1_table + i);
-+/*
++        }
-+ * QEMU event-loop backend
+     }
-+ *
-+ * Copyright (C) 2022 Red Hat Inc
+     if (extent->l1_backup_table_offset) {
-+ *
++        assert(!extent->sesparse);
-+ * Authors:
+         extent->l1_backup_table = g_try_malloc(l1_size);
-+ *  Nicolas Saenz Julienne <nsaenzju@redhat.com>
+         if (l1_size && extent->l1_backup_table == NULL) {
-+ *
+             ret = -ENOMEM;
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+@@ -XXX,XX +XXX,XX @@ static int vmdk_init_tables(BlockDriverState *bs, VmdkExtent *extent,
-+ * See the COPYING file in the top-level directory.
+     }
-+ */
-+#ifndef QEMU_EVENT_LOOP_BASE_H
+     extent->l2_cache =
-+#define QEMU_EVENT_LOOP_BASE_H
+-        g_new(uint32_t, extent->l2_size * L2_CACHE_SIZE);
-+
++        g_malloc(extent->entry_size * extent->l2_size * L2_CACHE_SIZE);
-+#include "qom/object.h"
+     return 0;
-+#include "block/aio.h"
+  fail_l1b:
-+#include "qemu/typedefs.h"
+     g_free(extent->l1_backup_table);
-+
+@@ -XXX,XX +XXX,XX @@ static int vmdk_open_vmfs_sparse(BlockDriverState *bs,
-+#define TYPE_EVENT_LOOP_BASE         "event-loop-base"
+     return ret;
-+OBJECT_DECLARE_TYPE(EventLoopBase, EventLoopBaseClass,
+ }
-+                    EVENT_LOOP_BASE)
-+
++#define SESPARSE_CONST_HEADER_MAGIC UINT64_C(0x00000000cafebabe)
-+struct EventLoopBaseClass {
++#define SESPARSE_VOLATILE_HEADER_MAGIC UINT64_C(0x00000000cafecafe)
-+    ObjectClass parent_class;
++
-+
++/* Strict checks - format not officially documented */
-+    void (*init)(EventLoopBase *base, Error **errp);
++static int check_se_sparse_const_header(VMDKSESparseConstHeader *header,
-+    void (*update_params)(EventLoopBase *base, Error **errp);
++                                        Error **errp)
 +};
 +
 +struct EventLoopBase {
 +    Object parent;
 +
 +    /* AioContext AIO engine parameters */
 +    int64_t aio_max_batch;
 +};
 +#endif
 diff --git a/include/sysemu/iothread.h b/include/sysemu/iothread.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/sysemu/iothread.h
 +++ b/include/sysemu/iothread.h
@@ -XXX,XX +XXX,XX @@
  #include "block/aio.h"
  #include "qemu/thread.h"
  #include "qom/object.h"
 +#include "sysemu/event-loop-base.h"
  #define TYPE_IOTHREAD "iothread"
  struct IOThread {
 -    Object parent_obj;
 +    EventLoopBase parent_obj;
      QemuThread thread;
      AioContext *ctx;
@@ -XXX,XX +XXX,XX @@ struct IOThread {
      int64_t poll_max_ns;
      int64_t poll_grow;
      int64_t poll_shrink;
 -
 -    /* AioContext AIO engine parameters */
 -    int64_t aio_max_batch;
  };
  typedef struct IOThread IOThread;
 diff --git a/event-loop-base.c b/event-loop-base.c
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/event-loop-base.c
@@ -XXX,XX +XXX,XX @@
 +/*
 + * QEMU event-loop base
 + *
 + * Copyright (C) 2022 Red Hat Inc
 + *
 + * Authors:
 + *  Stefan Hajnoczi <stefanha@redhat.com>
 + *  Nicolas Saenz Julienne <nsaenzju@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 "qom/object_interfaces.h"
 +#include "qapi/error.h"
 +#include "sysemu/event-loop-base.h"
 +
 +typedef struct {
 +    const char *name;
 +    ptrdiff_t offset; /* field's byte offset in EventLoopBase struct */
 +} EventLoopBaseParamInfo;
 +
 +static EventLoopBaseParamInfo aio_max_batch_info = {
 +    "aio-max-batch", offsetof(EventLoopBase, aio_max_batch),
 +};
 +
 +static void event_loop_base_get_param(Object *obj, Visitor *v,
 +        const char *name, void *opaque, Error **errp)
 +{
-+    EventLoopBase *event_loop_base = EVENT_LOOP_BASE(obj);
++    header->magic = le64_to_cpu(header->magic);
-+    EventLoopBaseParamInfo *info = opaque;
++    header->version = le64_to_cpu(header->version);
-+    int64_t *field = (void *)event_loop_base + info->offset;
++    header->grain_size = le64_to_cpu(header->grain_size);
-+
++    header->grain_table_size = le64_to_cpu(header->grain_table_size);
-+    visit_type_int64(v, name, field, errp);
++    header->flags = le64_to_cpu(header->flags);
 +    header->reserved1 = le64_to_cpu(header->reserved1);
 +    header->reserved2 = le64_to_cpu(header->reserved2);
 +    header->reserved3 = le64_to_cpu(header->reserved3);
 +    header->reserved4 = le64_to_cpu(header->reserved4);
 +
 +    header->volatile_header_offset =
 +        le64_to_cpu(header->volatile_header_offset);
 +    header->volatile_header_size = le64_to_cpu(header->volatile_header_size);
 +
 +    header->journal_header_offset = le64_to_cpu(header->journal_header_offset);
 +    header->journal_header_size = le64_to_cpu(header->journal_header_size);
 +
 +    header->journal_offset = le64_to_cpu(header->journal_offset);
 +    header->journal_size = le64_to_cpu(header->journal_size);
 +
 +    header->grain_dir_offset = le64_to_cpu(header->grain_dir_offset);
 +    header->grain_dir_size = le64_to_cpu(header->grain_dir_size);
 +
 +    header->grain_tables_offset = le64_to_cpu(header->grain_tables_offset);
 +    header->grain_tables_size = le64_to_cpu(header->grain_tables_size);
 +
 +    header->free_bitmap_offset = le64_to_cpu(header->free_bitmap_offset);
 +    header->free_bitmap_size = le64_to_cpu(header->free_bitmap_size);
 +
 +    header->backmap_offset = le64_to_cpu(header->backmap_offset);
 +    header->backmap_size = le64_to_cpu(header->backmap_size);
 +
 +    header->grains_offset = le64_to_cpu(header->grains_offset);
 +    header->grains_size = le64_to_cpu(header->grains_size);
 +
 +    if (header->magic != SESPARSE_CONST_HEADER_MAGIC) {
 +        error_setg(errp, "Bad const header magic: 0x%016" PRIx64,
 +                   header->magic);
 +        return -EINVAL;
 +    }
 +
 +    if (header->version != 0x0000000200000001) {
 +        error_setg(errp, "Unsupported version: 0x%016" PRIx64,
 +                   header->version);
 +        return -ENOTSUP;
 +    }
 +
 +    if (header->grain_size != 8) {
 +        error_setg(errp, "Unsupported grain size: %" PRIu64,
 +                   header->grain_size);
 +        return -ENOTSUP;
 +    }
 +
 +    if (header->grain_table_size != 64) {
 +        error_setg(errp, "Unsupported grain table size: %" PRIu64,
 +                   header->grain_table_size);
 +        return -ENOTSUP;
 +    }
 +
 +    if (header->flags != 0) {
 +        error_setg(errp, "Unsupported flags: 0x%016" PRIx64,
 +                   header->flags);
 +        return -ENOTSUP;
 +    }
 +
 +    if (header->reserved1 != 0 || header->reserved2 != 0 ||
 +        header->reserved3 != 0 || header->reserved4 != 0) {
 +        error_setg(errp, "Unsupported reserved bits:"
 +                   " 0x%016" PRIx64 " 0x%016" PRIx64
 +                   " 0x%016" PRIx64 " 0x%016" PRIx64,
 +                   header->reserved1, header->reserved2,
 +                   header->reserved3, header->reserved4);
 +        return -ENOTSUP;
 +    }
 +
 +    /* check that padding is 0 */
 +    if (!buffer_is_zero(header->pad, sizeof(header->pad))) {
 +        error_setg(errp, "Unsupported non-zero const header padding");
 +        return -ENOTSUP;
 +    }
 +
 +    return 0;
 +}
 +
-+static void event_loop_base_set_param(Object *obj, Visitor *v,
++static int check_se_sparse_volatile_header(VMDKSESparseVolatileHeader *header,
-+        const char *name, void *opaque, Error **errp)
++                                           Error **errp)
 +{
-+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_GET_CLASS(obj);
++    header->magic = le64_to_cpu(header->magic);
-+    EventLoopBase *base = EVENT_LOOP_BASE(obj);
++    header->free_gt_number = le64_to_cpu(header->free_gt_number);
-+    EventLoopBaseParamInfo *info = opaque;
++    header->next_txn_seq_number = le64_to_cpu(header->next_txn_seq_number);
-+    int64_t *field = (void *)base + info->offset;
++    header->replay_journal = le64_to_cpu(header->replay_journal);
-+    int64_t value;
++
-+
++    if (header->magic != SESPARSE_VOLATILE_HEADER_MAGIC) {
-+    if (!visit_type_int64(v, name, &value, errp)) {
++        error_setg(errp, "Bad volatile header magic: 0x%016" PRIx64,
-+        return;
++                   header->magic);
-+    }
++        return -EINVAL;
-+
++    }
-+    if (value < 0) {
++
-+        error_setg(errp, "%s value must be in range [0, %" PRId64 "]",
++    if (header->replay_journal) {
-+                   info->name, INT64_MAX);
++        error_setg(errp, "Image is dirty, Replaying journal not supported");
-+        return;
++        return -ENOTSUP;
 +    }
 +
-+    *field = value;
++    /* check that padding is 0 */
-+
++    if (!buffer_is_zero(header->pad, sizeof(header->pad))) {
-+    if (bc->update_params) {
++        error_setg(errp, "Unsupported non-zero volatile header padding");
-+        bc->update_params(base, errp);
++        return -ENOTSUP;
 +    }
 +
-+    return;
++    return 0;
 +}
 +
-+static void event_loop_base_complete(UserCreatable *uc, Error **errp)
++static int vmdk_open_se_sparse(BlockDriverState *bs,
 +                               BdrvChild *file,
 +                               int flags, Error **errp)
 +{
-+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_GET_CLASS(uc);
++    int ret;
-+    EventLoopBase *base = EVENT_LOOP_BASE(uc);
++    VMDKSESparseConstHeader const_header;
-+
++    VMDKSESparseVolatileHeader volatile_header;
-+    if (bc->init) {
++    VmdkExtent *extent;
-+        bc->init(base, errp);
++
-+    }
++    ret = bdrv_apply_auto_read_only(bs,
 +            "No write support for seSparse images available", errp);
 +    if (ret < 0) {
 +        return ret;
 +    }
 +
 +    assert(sizeof(const_header) == SECTOR_SIZE);
 +
 +    ret = bdrv_pread(file, 0, &const_header, sizeof(const_header));
 +    if (ret < 0) {
 +        bdrv_refresh_filename(file->bs);
 +        error_setg_errno(errp, -ret,
 +                         "Could not read const header from file '%s'",
 +                         file->bs->filename);
 +        return ret;
 +    }
 +
 +    /* check const header */
 +    ret = check_se_sparse_const_header(&const_header, errp);
 +    if (ret < 0) {
 +        return ret;
 +    }
 +
 +    assert(sizeof(volatile_header) == SECTOR_SIZE);
 +
 +    ret = bdrv_pread(file,
 +                     const_header.volatile_header_offset * SECTOR_SIZE,
 +                     &volatile_header, sizeof(volatile_header));
 +    if (ret < 0) {
 +        bdrv_refresh_filename(file->bs);
 +        error_setg_errno(errp, -ret,
 +                         "Could not read volatile header from file '%s'",
 +                         file->bs->filename);
 +        return ret;
 +    }
 +
 +    /* check volatile header */
 +    ret = check_se_sparse_volatile_header(&volatile_header, errp);
 +    if (ret < 0) {
 +        return ret;
 +    }
 +
 +    ret = vmdk_add_extent(bs, file, false,
 +                          const_header.capacity,
 +                          const_header.grain_dir_offset * SECTOR_SIZE,
 +                          0,
 +                          const_header.grain_dir_size *
 +                          SECTOR_SIZE / sizeof(uint64_t),
 +                          const_header.grain_table_size *
 +                          SECTOR_SIZE / sizeof(uint64_t),
 +                          const_header.grain_size,
 +                          &extent,
 +                          errp);
 +    if (ret < 0) {
 +        return ret;
 +    }
 +
 +    extent->sesparse = true;
 +    extent->sesparse_l2_tables_offset = const_header.grain_tables_offset;
 +    extent->sesparse_clusters_offset = const_header.grains_offset;
 +    extent->entry_size = sizeof(uint64_t);
 +
 +    ret = vmdk_init_tables(bs, extent, errp);
 +    if (ret) {
 +        /* free extent allocated by vmdk_add_extent */
 +        vmdk_free_last_extent(bs);
 +    }
 +
 +    return ret;
 +}
 +
-+static void event_loop_base_class_init(ObjectClass *klass, void *class_data)
+ static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf,
-+{
+                                QDict *options, Error **errp);
-+    UserCreatableClass *ucc = USER_CREATABLE_CLASS(klass);
-+    ucc->complete = event_loop_base_complete;
+@@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
-+
+          * RW [size in sectors] SPARSE "file-name.vmdk"
-+    object_class_property_add(klass, "aio-max-batch", "int",
+          * RW [size in sectors] VMFS "file-name.vmdk"
-+                              event_loop_base_get_param,
+          * RW [size in sectors] VMFSSPARSE "file-name.vmdk"
-+                              event_loop_base_set_param,
++         * RW [size in sectors] SESPARSE "file-name.vmdk"
-+                              NULL, &aio_max_batch_info);
+          */
-+}
+         flat_offset = -1;
-+
+         matches = sscanf(p, "%10s %" SCNd64 " %10s \"%511[^\n\r\"]\" %" SCNd64,
-+static const TypeInfo event_loop_base_info = {
+@@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
-+    .name = TYPE_EVENT_LOOP_BASE,
-+    .parent = TYPE_OBJECT,
+         if (sectors <= 0 ||
-+    .instance_size = sizeof(EventLoopBase),
+             (strcmp(type, "FLAT") && strcmp(type, "SPARSE") &&
-+    .class_size = sizeof(EventLoopBaseClass),
+-             strcmp(type, "VMFS") && strcmp(type, "VMFSSPARSE")) ||
-+    .class_init = event_loop_base_class_init,
++             strcmp(type, "VMFS") && strcmp(type, "VMFSSPARSE") &&
-+    .abstract = true,
++             strcmp(type, "SESPARSE")) ||
-+    .interfaces = (InterfaceInfo[]) {
+             (strcmp(access, "RW"))) {
-+        { TYPE_USER_CREATABLE },
+             continue;
-+        { }
+         }
-+    }
+@@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
-+};
+                 return ret;
-+
+             }
-+static void register_types(void)
+             extent = &s->extents[s->num_extents - 1];
-+{
++        } else if (!strcmp(type, "SESPARSE")) {
-+    type_register_static(&event_loop_base_info);
++            ret = vmdk_open_se_sparse(bs, extent_file, bs->open_flags, errp);
-+}
++            if (ret) {
-+type_init(register_types);
++                bdrv_unref_child(bs, extent_file);
-diff --git a/iothread.c b/iothread.c
++                return ret;
-index XXXXXXX..XXXXXXX 100644
++            }
---- a/iothread.c
++            extent = &s->extents[s->num_extents - 1];
-+++ b/iothread.c
+         } else {
-@@ -XXX,XX +XXX,XX @@
+             error_setg(errp, "Unsupported extent type '%s'", type);
- #include "qemu/module.h"
+             bdrv_unref_child(bs, extent_file);
- #include "block/aio.h"
+@@ -XXX,XX +XXX,XX @@ static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf,
- #include "block/block.h"
+     if (strcmp(ct, "monolithicFlat") &&
-+#include "sysemu/event-loop-base.h"
+         strcmp(ct, "vmfs") &&
- #include "sysemu/iothread.h"
+         strcmp(ct, "vmfsSparse") &&
- #include "qapi/error.h"
++        strcmp(ct, "seSparse") &&
- #include "qapi/qapi-commands-misc.h"
+         strcmp(ct, "twoGbMaxExtentSparse") &&
-@@ -XXX,XX +XXX,XX @@ static void iothread_init_gcontext(IOThread *iothread)
+         strcmp(ct, "twoGbMaxExtentFlat")) {
-     iothread->main_loop = g_main_loop_new(iothread->worker_context, TRUE);
+         error_setg(errp, "Unsupported image type '%s'", ct);
- }
+@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
 -static void iothread_set_aio_context_params(IOThread *iothread, Error **errp)
 +static void iothread_set_aio_context_params(EventLoopBase *base, Error **errp)
  {
-+    IOThread *iothread = IOTHREAD(base);
+     unsigned int l1_index, l2_offset, l2_index;
-     ERRP_GUARD();
+     int min_index, i, j;
+-    uint32_t min_count, *l2_table;
-+    if (!iothread->ctx) {
++    uint32_t min_count;
-+        return;
++    void *l2_table;
-+    }
+     bool zeroed = false;
-+
+     int64_t ret;
-     aio_context_set_poll_params(iothread->ctx,
+     int64_t cluster_sector;
-                                 iothread->poll_max_ns,
++    unsigned int l2_size_bytes = extent->l2_size * extent->entry_size;
-                                 iothread->poll_grow,
-@@ -XXX,XX +XXX,XX @@ static void iothread_set_aio_context_params(IOThread *iothread, Error **errp)
+     if (m_data) {
-     }
+         m_data->valid = 0;
+@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
-     aio_context_set_aio_params(iothread->ctx,
+     if (l1_index >= extent->l1_size) {
--                               iothread->aio_max_batch,
+         return VMDK_ERROR;
-+                               iothread->parent_obj.aio_max_batch,
+     }
-                                errp);
+-    l2_offset = extent->l1_table[l1_index];
- }
++    if (extent->sesparse) {
++        uint64_t l2_offset_u64;
--static void iothread_complete(UserCreatable *obj, Error **errp)
++
-+
++        assert(extent->entry_size == sizeof(uint64_t));
-+static void iothread_init(EventLoopBase *base, Error **errp)
++
- {
++        l2_offset_u64 = ((uint64_t *)extent->l1_table)[l1_index];
-     Error *local_error = NULL;
++        if (l2_offset_u64 == 0) {
--    IOThread *iothread = IOTHREAD(obj);
++            l2_offset = 0;
-+    IOThread *iothread = IOTHREAD(base);
++        } else if ((l2_offset_u64 & 0xffffffff00000000) != 0x1000000000000000) {
-     char *thread_name;
++            /*
++             * Top most nibble is 0x1 if grain table is allocated.
-     iothread->stopping = false;
++             * strict check - top most 4 bytes must be 0x10000000 since max
-@@ -XXX,XX +XXX,XX @@ static void iothread_complete(UserCreatable *obj, Error **errp)
++             * supported size is 64TB for disk - so no more than 64TB / 16MB
-      */
++             * grain directories which is smaller than uint32,
-     iothread_init_gcontext(iothread);
++             * where 16MB is the only supported default grain table coverage.
++             */
--    iothread_set_aio_context_params(iothread, &local_error);
++            return VMDK_ERROR;
-+    iothread_set_aio_context_params(base, &local_error);
++        } else {
-     if (local_error) {
++            l2_offset_u64 = l2_offset_u64 & 0x00000000ffffffff;
-         error_propagate(errp, local_error);
++            l2_offset_u64 = extent->sesparse_l2_tables_offset +
-         aio_context_unref(iothread->ctx);
++                l2_offset_u64 * l2_size_bytes / SECTOR_SIZE;
-@@ -XXX,XX +XXX,XX @@ static void iothread_complete(UserCreatable *obj, Error **errp)
++            if (l2_offset_u64 > 0x00000000ffffffff) {
-      * to inherit.
++                return VMDK_ERROR;
-      */
++            }
-     thread_name = g_strdup_printf("IO %s",
++            l2_offset = (unsigned int)(l2_offset_u64);
--                        object_get_canonical_path_component(OBJECT(obj)));
++        }
-+                        object_get_canonical_path_component(OBJECT(base)));
++    } else {
-     qemu_thread_create(&iothread->thread, thread_name, iothread_run,
++        assert(extent->entry_size == sizeof(uint32_t));
-                        iothread, QEMU_THREAD_JOINABLE);
++        l2_offset = ((uint32_t *)extent->l1_table)[l1_index];
-     g_free(thread_name);
++    }
-@@ -XXX,XX +XXX,XX @@ static IOThreadParamInfo poll_grow_info = {
+     if (!l2_offset) {
- static IOThreadParamInfo poll_shrink_info = {
+         return VMDK_UNALLOC;
-     "poll-shrink", offsetof(IOThread, poll_shrink),
+     }
- };
+@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
--static IOThreadParamInfo aio_max_batch_info = {
+                     extent->l2_cache_counts[j] >>= 1;
--    "aio-max-batch", offsetof(IOThread, aio_max_batch),
+                 }
--};
+             }
+-            l2_table = extent->l2_cache + (i * extent->l2_size);
- static void iothread_get_param(Object *obj, Visitor *v,
++            l2_table = (char *)extent->l2_cache + (i * l2_size_bytes);
-         const char *name, IOThreadParamInfo *info, Error **errp)
+             goto found;
-@@ -XXX,XX +XXX,XX @@ static void iothread_set_poll_param(Object *obj, Visitor *v,
+         }
      }
- }
+@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
+             min_index = i;
--static void iothread_get_aio_param(Object *obj, Visitor *v,
+         }
--        const char *name, void *opaque, Error **errp)
+     }
--{
+-    l2_table = extent->l2_cache + (min_index * extent->l2_size);
--    IOThreadParamInfo *info = opaque;
++    l2_table = (char *)extent->l2_cache + (min_index * l2_size_bytes);
--
+     BLKDBG_EVENT(extent->file, BLKDBG_L2_LOAD);
--    iothread_get_param(obj, v, name, info, errp);
+     if (bdrv_pread(extent->file,
--}
+                 (int64_t)l2_offset * 512,
--
+                 l2_table,
--static void iothread_set_aio_param(Object *obj, Visitor *v,
+-                extent->l2_size * sizeof(uint32_t)
--        const char *name, void *opaque, Error **errp)
+-            ) != extent->l2_size * sizeof(uint32_t)) {
--{
++                l2_size_bytes
--    IOThread *iothread = IOTHREAD(obj);
++            ) != l2_size_bytes) {
--    IOThreadParamInfo *info = opaque;
+         return VMDK_ERROR;
--
+     }
--    if (!iothread_set_param(obj, v, name, info, errp)) {
--        return;
+@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
--    }
+     extent->l2_cache_counts[min_index] = 1;
--
+  found:
--    if (iothread->ctx) {
+     l2_index = ((offset >> 9) / extent->cluster_sectors) % extent->l2_size;
--        aio_context_set_aio_params(iothread->ctx,
+-    cluster_sector = le32_to_cpu(l2_table[l2_index]);
--                                   iothread->aio_max_batch,
--                                   errp);
+-    if (extent->has_zero_grain && cluster_sector == VMDK_GTE_ZEROED) {
--    }
+-        zeroed = true;
--}
++    if (extent->sesparse) {
--
++        cluster_sector = le64_to_cpu(((uint64_t *)l2_table)[l2_index]);
- static void iothread_class_init(ObjectClass *klass, void *class_data)
++        switch (cluster_sector & 0xf000000000000000) {
- {
++        case 0x0000000000000000:
--    UserCreatableClass *ucc = USER_CREATABLE_CLASS(klass);
++            /* unallocated grain */
--    ucc->complete = iothread_complete;
++            if (cluster_sector != 0) {
-+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_CLASS(klass);
++                return VMDK_ERROR;
-+
++            }
-+    bc->init = iothread_init;
++            break;
-+    bc->update_params = iothread_set_aio_context_params;
++        case 0x1000000000000000:
++            /* scsi-unmapped grain - fallthrough */
-     object_class_property_add(klass, "poll-max-ns", "int",
++        case 0x2000000000000000:
-                               iothread_get_poll_param,
++            /* zero grain */
-@@ -XXX,XX +XXX,XX @@ static void iothread_class_init(ObjectClass *klass, void *class_data)
++            zeroed = true;
-                               iothread_get_poll_param,
++            break;
-                               iothread_set_poll_param,
++        case 0x3000000000000000:
-                               NULL, &poll_shrink_info);
++            /* allocated grain */
--    object_class_property_add(klass, "aio-max-batch", "int",
++            cluster_sector = (((cluster_sector & 0x0fff000000000000) >> 48) |
--                              iothread_get_aio_param,
++                              ((cluster_sector & 0x0000ffffffffffff) << 12));
--                              iothread_set_aio_param,
++            cluster_sector = extent->sesparse_clusters_offset +
--                              NULL, &aio_max_batch_info);
++                cluster_sector * extent->cluster_sectors;
- }
++            break;
++        default:
- static const TypeInfo iothread_info = {
++            return VMDK_ERROR;
-     .name = TYPE_IOTHREAD,
++        }
--    .parent = TYPE_OBJECT,
++    } else {
-+    .parent = TYPE_EVENT_LOOP_BASE,
++        cluster_sector = le32_to_cpu(((uint32_t *)l2_table)[l2_index]);
-     .class_init = iothread_class_init,
++
-     .instance_size = sizeof(IOThread),
++        if (extent->has_zero_grain && cluster_sector == VMDK_GTE_ZEROED) {
-     .instance_init = iothread_instance_init,
++            zeroed = true;
-     .instance_finalize = iothread_instance_finalize,
++        }
--    .interfaces = (InterfaceInfo[]) {
+     }
--        {TYPE_USER_CREATABLE},
--        {}
+     if (!cluster_sector || zeroed) {
--    },
+         if (!allocate) {
- };
+             return zeroed ? VMDK_ZEROED : VMDK_UNALLOC;
+         }
- static void iothread_register_types(void)
++        assert(!extent->sesparse);
-@@ -XXX,XX +XXX,XX @@ static int query_one_iothread(Object *object, void *opaque)
-     info->poll_max_ns = iothread->poll_max_ns;
+         if (extent->next_cluster_sector >= VMDK_EXTENT_MAX_SECTORS) {
-     info->poll_grow = iothread->poll_grow;
+             return VMDK_ERROR;
-     info->poll_shrink = iothread->poll_shrink;
+@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
--    info->aio_max_batch = iothread->aio_max_batch;
+             m_data->l1_index = l1_index;
-+    info->aio_max_batch = iothread->parent_obj.aio_max_batch;
+             m_data->l2_index = l2_index;
+             m_data->l2_offset = l2_offset;
-     QAPI_LIST_APPEND(*tail, info);
+-            m_data->l2_cache_entry = &l2_table[l2_index];
-     return 0;
++            m_data->l2_cache_entry = ((uint32_t *)l2_table) + l2_index;
          }
      }
      *cluster_offset = cluster_sector << BDRV_SECTOR_BITS;
@@ -XXX,XX +XXX,XX @@ static int vmdk_pwritev(BlockDriverState *bs, uint64_t offset,
          if (!extent) {
              return -EIO;
          }
 +        if (extent->sesparse) {
 +            return -ENOTSUP;
 +        }
          offset_in_cluster = vmdk_find_offset_in_cluster(extent, offset);
          n_bytes = MIN(bytes, extent->cluster_sectors * BDRV_SECTOR_SIZE
                               - offset_in_cluster);
 --
-.35.1
+.21.0

-[PULL 2/3] util/main-loop: Introduce the main loop into QOM
+[Qemu-devel] [PULL v2 7/8] ssh: switch from libssh2 to libssh
-From: Nicolas Saenz Julienne <nsaenzju@redhat.com>
+From: Pino Toscano <ptoscano@redhat.com>
-'event-loop-base' provides basic property handling for all 'AioContext'
+Rewrite the implementation of the ssh block driver to use libssh instead
-based event loops. So let's define a new 'MainLoopClass' that inherits
+of libssh2.  The libssh library has various advantages over libssh2:
-from it. This will permit tweaking the main loop's properties through
+- easier API for authentication (for example for using ssh-agent)
-qapi as well as through the command line using the '-object' keyword[1].
+- easier API for known_hosts handling
-Only one instance of 'MainLoopClass' might be created at any time.
+- supports newer types of keys in known_hosts
-'EventLoopBaseClass' learns a new callback, 'can_be_deleted()' so as to
+Use APIs/features available in libssh 0.8 conditionally, to support
-mark 'MainLoop' as non-deletable.
+older versions (which are not recommended though).
-[1] For example:
+Adjust the iotest 207 according to the different error message, and to
-      -object main-loop,id=main-loop,aio-max-batch=<value>
+find the default key type for localhost (to properly compare the
 fingerprint with).
 Contributed-by: Max Reitz <mreitz@redhat.com>
-Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
+Adjust the various Docker/Travis scripts to use libssh when available
-Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
+instead of libssh2. The mingw/mxe testing is dropped for now, as there
-Acked-by: Markus Armbruster <armbru@redhat.com>
+are no packages for it.
-Message-id: 20220425075723.20019-3-nsaenzju@redhat.com
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
+Signed-off-by: Pino Toscano <ptoscano@redhat.com>
 Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
 Acked-by: Alex Bennée <alex.bennee@linaro.org>
 Message-id: 20190620200840.17655-1-ptoscano@redhat.com
 Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
 Message-id: 5873173.t2JhDm7DL7@lindworm.usersys.redhat.com
 Signed-off-by: Max Reitz <mreitz@redhat.com>
 ---
- qapi/qom.json                    | 13 ++++++++
+ configure                                     |  65 +-
- meson.build                      |  3 +-
+ block/Makefile.objs                           |   6 +-
- include/qemu/main-loop.h         | 10 ++++++
+ block/ssh.c                                   | 652 ++++++++++--------
- include/sysemu/event-loop-base.h |  1 +
+ .travis.yml                                   |   4 +-
- event-loop-base.c                | 13 ++++++++
+ block/trace-events                            |  14 +-
- util/main-loop.c                 | 56 ++++++++++++++++++++++++++++++++
+ docs/qemu-block-drivers.texi                  |   2 +-
-files changed, 95 insertions(+), 1 deletion(-)
+ .../dockerfiles/debian-win32-cross.docker     |   1 -
  .../dockerfiles/debian-win64-cross.docker     |   1 -
  tests/docker/dockerfiles/fedora.docker        |   4 +-
  tests/docker/dockerfiles/ubuntu.docker        |   2 +-
  tests/docker/dockerfiles/ubuntu1804.docker    |   2 +-
  tests/qemu-iotests/207                        |  54 +-
  tests/qemu-iotests/207.out                    |   2 +-
 files changed, 449 insertions(+), 360 deletions(-)
-diff --git a/qapi/qom.json b/qapi/qom.json
+diff --git a/configure b/configure
 index XXXXXXX..XXXXXXX 100755
 --- a/configure
 +++ b/configure
@@ -XXX,XX +XXX,XX @@ auth_pam=""
  vte=""
  virglrenderer=""
  tpm=""
 -libssh2=""
 +libssh=""
  live_block_migration="yes"
  numa=""
  tcmalloc="no"
@@ -XXX,XX +XXX,XX @@ for opt do
    ;;
    --enable-tpm) tpm="yes"
    ;;
 -  --disable-libssh2) libssh2="no"
 +  --disable-libssh) libssh="no"
    ;;
 -  --enable-libssh2) libssh2="yes"
 +  --enable-libssh) libssh="yes"
    ;;
    --disable-live-block-migration) live_block_migration="no"
    ;;
@@ -XXX,XX +XXX,XX @@ disabled with --disable-FEATURE, default is enabled if available:
    coroutine-pool  coroutine freelist (better performance)
    glusterfs       GlusterFS backend
    tpm             TPM support
 -  libssh2         ssh block device support
 +  libssh          ssh block device support
    numa            libnuma support
    libxml2         for Parallels image format
    tcmalloc        tcmalloc support
@@ -XXX,XX +XXX,XX @@ EOF
  fi
  ##########################################
 -# libssh2 probe
 -min_libssh2_version=1.2.8
 -if test "$libssh2" != "no" ; then
 -  if $pkg_config --atleast-version=$min_libssh2_version libssh2; then
 -    libssh2_cflags=$($pkg_config libssh2 --cflags)
 -    libssh2_libs=$($pkg_config libssh2 --libs)
 -    libssh2=yes
 +# libssh probe
 +if test "$libssh" != "no" ; then
 +  if $pkg_config --exists libssh; then
 +    libssh_cflags=$($pkg_config libssh --cflags)
 +    libssh_libs=$($pkg_config libssh --libs)
 +    libssh=yes
    else
 -    if test "$libssh2" = "yes" ; then
 -      error_exit "libssh2 >= $min_libssh2_version required for --enable-libssh2"
 +    if test "$libssh" = "yes" ; then
 +      error_exit "libssh required for --enable-libssh"
      fi
 -    libssh2=no
 +    libssh=no
    fi
  fi
  ##########################################
 -# libssh2_sftp_fsync probe
 +# Check for libssh 0.8
 +# This is done like this instead of using the LIBSSH_VERSION_* and
 +# SSH_VERSION_* macros because some distributions in the past shipped
 +# snapshots of the future 0.8 from Git, and those snapshots did not
 +# have updated version numbers (still referring to 0.7.0).
 -if test "$libssh2" = "yes"; then
 +if test "$libssh" = "yes"; then
    cat > $TMPC <<EOF
 -#include <stdio.h>
 -#include <libssh2.h>
 -#include <libssh2_sftp.h>
 -int main(void) {
 -    LIBSSH2_SESSION *session;
 -    LIBSSH2_SFTP *sftp;
 -    LIBSSH2_SFTP_HANDLE *sftp_handle;
 -    session = libssh2_session_init ();
 -    sftp = libssh2_sftp_init (session);
 -    sftp_handle = libssh2_sftp_open (sftp, "/", 0, 0);
 -    libssh2_sftp_fsync (sftp_handle);
 -    return 0;
 -}
 +#include <libssh/libssh.h>
 +int main(void) { return ssh_get_server_publickey(NULL, NULL); }
  EOF
 -  # libssh2_cflags/libssh2_libs defined in previous test.
 -  if compile_prog "$libssh2_cflags" "$libssh2_libs" ; then
 -    QEMU_CFLAGS="-DHAS_LIBSSH2_SFTP_FSYNC $QEMU_CFLAGS"
 +  if compile_prog "$libssh_cflags" "$libssh_libs"; then
 +    libssh_cflags="-DHAVE_LIBSSH_0_8 $libssh_cflags"
    fi
  fi
@@ -XXX,XX +XXX,XX @@ echo "GlusterFS support $glusterfs"
  echo "gcov              $gcov_tool"
  echo "gcov enabled      $gcov"
  echo "TPM support       $tpm"
 -echo "libssh2 support   $libssh2"
 +echo "libssh support    $libssh"
  echo "QOM debugging     $qom_cast_debug"
  echo "Live block migration $live_block_migration"
  echo "lzo support       $lzo"
@@ -XXX,XX +XXX,XX @@ if test "$glusterfs_iocb_has_stat" = "yes" ; then
    echo "CONFIG_GLUSTERFS_IOCB_HAS_STAT=y" >> $config_host_mak
  fi
 -if test "$libssh2" = "yes" ; then
 -  echo "CONFIG_LIBSSH2=m" >> $config_host_mak
 -  echo "LIBSSH2_CFLAGS=$libssh2_cflags" >> $config_host_mak
 -  echo "LIBSSH2_LIBS=$libssh2_libs" >> $config_host_mak
 +if test "$libssh" = "yes" ; then
 +  echo "CONFIG_LIBSSH=m" >> $config_host_mak
 +  echo "LIBSSH_CFLAGS=$libssh_cflags" >> $config_host_mak
 +  echo "LIBSSH_LIBS=$libssh_libs" >> $config_host_mak
  fi
  if test "$live_block_migration" = "yes" ; then
 diff --git a/block/Makefile.objs b/block/Makefile.objs
 index XXXXXXX..XXXXXXX 100644
---- a/qapi/qom.json
+--- a/block/Makefile.objs
-+++ b/qapi/qom.json
++++ b/block/Makefile.objs
@@ -XXX,XX +XXX,XX @@ block-obj-$(CONFIG_CURL) += curl.o
  block-obj-$(CONFIG_RBD) += rbd.o
  block-obj-$(CONFIG_GLUSTERFS) += gluster.o
  block-obj-$(CONFIG_VXHS) += vxhs.o
 -block-obj-$(CONFIG_LIBSSH2) += ssh.o
 +block-obj-$(CONFIG_LIBSSH) += ssh.o
  block-obj-y += accounting.o dirty-bitmap.o
  block-obj-y += write-threshold.o
  block-obj-y += backup.o
@@ -XXX,XX +XXX,XX @@ rbd.o-libs         := $(RBD_LIBS)
  gluster.o-cflags   := $(GLUSTERFS_CFLAGS)
  gluster.o-libs     := $(GLUSTERFS_LIBS)
  vxhs.o-libs        := $(VXHS_LIBS)
 -ssh.o-cflags       := $(LIBSSH2_CFLAGS)
 -ssh.o-libs         := $(LIBSSH2_LIBS)
 +ssh.o-cflags       := $(LIBSSH_CFLAGS)
 +ssh.o-libs         := $(LIBSSH_LIBS)
  block-obj-dmg-bz2-$(CONFIG_BZIP2) += dmg-bz2.o
  block-obj-$(if $(CONFIG_DMG),m,n) += $(block-obj-dmg-bz2-y)
  dmg-bz2.o-libs     := $(BZIP2_LIBS)
 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 @@
-             '*poll-grow': 'int',
-             '*poll-shrink': 'int' } }
+ #include "qemu/osdep.h"
-+##
+-#include <libssh2.h>
-+# @MainLoopProperties:
+-#include <libssh2_sftp.h>
-+#
++#include <libssh/libssh.h>
-+# Properties for the main-loop object.
++#include <libssh/sftp.h>
-+#
-+# Since: 7.1
+ #include "block/block_int.h"
-+##
+ #include "block/qdict.h"
 +{ 'struct': 'MainLoopProperties',
 +  'base': 'EventLoopBaseProperties',
 +  'data': {} }
 +
  ##
  # @MemoryBackendProperties:
  #
 @@ -XXX,XX +XXX,XX @@
-     { 'name': 'input-linux',
+ #include "trace.h"
-       'if': 'CONFIG_LINUX' },
-     'iothread',
+ /*
-+    'main-loop',
+- * TRACE_LIBSSH2=<bitmask> enables tracing in libssh2 itself.  Note
-     { 'name': 'memory-backend-epc',
+- * that this requires that libssh2 was specially compiled with the
-       'if': 'CONFIG_LINUX' },
+- * `./configure --enable-debug' option, so most likely you will have
-     'memory-backend-file',
+- * to compile it yourself.  The meaning of <bitmask> is described
-@@ -XXX,XX +XXX,XX @@
+- * here: http://www.libssh2.org/libssh2_trace.html
-       'input-linux':                { 'type': 'InputLinuxProperties',
++ * TRACE_LIBSSH=<level> enables tracing in libssh itself.
-                                       'if': 'CONFIG_LINUX' },
++ * The meaning of <level> is described here:
-       'iothread':                   'IothreadProperties',
++ * http://api.libssh.org/master/group__libssh__log.html
-+      'main-loop':                  'MainLoopProperties',
+  */
-       'memory-backend-epc':         { 'type': 'MemoryBackendEpcProperties',
+-#define TRACE_LIBSSH2 0 /* or try: LIBSSH2_TRACE_SFTP */
-                                       'if': 'CONFIG_LINUX' },
++#define TRACE_LIBSSH  0 /* see: SSH_LOG_* */
-       'memory-backend-file':        'MemoryBackendFileProperties',
-diff --git a/meson.build b/meson.build
+ typedef struct BDRVSSHState {
-index XXXXXXX..XXXXXXX 100644
+     /* Coroutine. */
---- a/meson.build
+@@ -XXX,XX +XXX,XX @@ typedef struct BDRVSSHState {
-+++ b/meson.build
-@@ -XXX,XX +XXX,XX @@ libqemuutil = static_library('qemuutil',
+     /* SSH connection. */
-                              sources: util_ss.sources() + stub_ss.sources() + genh,
+     int sock;                         /* socket */
-                              dependencies: [util_ss.dependencies(), libm, threads, glib, socket, malloc, pixman])
+-    LIBSSH2_SESSION *session;         /* ssh session */
- qemuutil = declare_dependency(link_with: libqemuutil,
+-    LIBSSH2_SFTP *sftp;               /* sftp session */
--                              sources: genh + version_res)
+-    LIBSSH2_SFTP_HANDLE *sftp_handle; /* sftp remote file handle */
-+                              sources: genh + version_res,
++    ssh_session session;              /* ssh session */
-+                              dependencies: [event_loop_base])
++    sftp_session sftp;                /* sftp session */
++    sftp_file sftp_handle;            /* sftp remote file handle */
- if have_system or have_user
-   decodetree = generator(find_program('scripts/decodetree.py'),
+-    /* See ssh_seek() function below. */
-diff --git a/include/qemu/main-loop.h b/include/qemu/main-loop.h
+-    int64_t offset;
-index XXXXXXX..XXXXXXX 100644
+-    bool offset_op_read;
---- a/include/qemu/main-loop.h
+-
-+++ b/include/qemu/main-loop.h
+-    /* File attributes at open.  We try to keep the .filesize field
-@@ -XXX,XX +XXX,XX @@
++    /*
- #define QEMU_MAIN_LOOP_H
++     * File attributes at open.  We try to keep the .size field
+      * updated if it changes (eg by writing at the end of the file).
- #include "block/aio.h"
+      */
-+#include "qom/object.h"
+-    LIBSSH2_SFTP_ATTRIBUTES attrs;
-+#include "sysemu/event-loop-base.h"
++    sftp_attributes attrs;
- #define SIG_IPI SIGUSR1
+     InetSocketAddress *inet;
-+#define TYPE_MAIN_LOOP  "main-loop"
+@@ -XXX,XX +XXX,XX @@ static void ssh_state_init(BDRVSSHState *s)
-+OBJECT_DECLARE_TYPE(MainLoop, MainLoopClass, MAIN_LOOP)
+ {
-+
+     memset(s, 0, sizeof *s);
-+struct MainLoop {
+     s->sock = -1;
-+    EventLoopBase parent_obj;
+-    s->offset = -1;
-+};
+     qemu_co_mutex_init(&s->lock);
 +typedef struct MainLoop MainLoop;
 +
  /**
   * qemu_init_main_loop: Set up the process so that it can run the main loop.
   *
 diff --git a/include/sysemu/event-loop-base.h b/include/sysemu/event-loop-base.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/sysemu/event-loop-base.h
 +++ b/include/sysemu/event-loop-base.h
@@ -XXX,XX +XXX,XX @@ struct EventLoopBaseClass {
      void (*init)(EventLoopBase *base, Error **errp);
      void (*update_params)(EventLoopBase *base, Error **errp);
 +    bool (*can_be_deleted)(EventLoopBase *base);
  };
  struct EventLoopBase {
 diff --git a/event-loop-base.c b/event-loop-base.c
 index XXXXXXX..XXXXXXX 100644
 --- a/event-loop-base.c
 +++ b/event-loop-base.c
@@ -XXX,XX +XXX,XX @@ static void event_loop_base_complete(UserCreatable *uc, Error **errp)
      }
  }
-+static bool event_loop_base_can_be_deleted(UserCreatable *uc)
+@@ -XXX,XX +XXX,XX @@ static void ssh_state_free(BDRVSSHState *s)
-+{
+ {
-+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_GET_CLASS(uc);
+     g_free(s->user);
-+    EventLoopBase *backend = EVENT_LOOP_BASE(uc);
-+
++    if (s->attrs) {
-+    if (bc->can_be_deleted) {
++        sftp_attributes_free(s->attrs);
 +        return bc->can_be_deleted(backend);
 +    }
-+
+     if (s->sftp_handle) {
-+    return true;
+-        libssh2_sftp_close(s->sftp_handle);
-+}
++        sftp_close(s->sftp_handle);
-+
+     }
- static void event_loop_base_class_init(ObjectClass *klass, void *class_data)
+     if (s->sftp) {
 -        libssh2_sftp_shutdown(s->sftp);
 +        sftp_free(s->sftp);
      }
      if (s->session) {
 -        libssh2_session_disconnect(s->session,
 -                                   "from qemu ssh client: "
 -                                   "user closed the connection");
 -        libssh2_session_free(s->session);
 -    }
 -    if (s->sock >= 0) {
 -        close(s->sock);
 +        ssh_disconnect(s->session);
 +        ssh_free(s->session); /* This frees s->sock */
      }
  }
@@ -XXX,XX +XXX,XX @@ session_error_setg(Error **errp, BDRVSSHState *s, const char *fs, ...)
      va_end(args);
      if (s->session) {
 -        char *ssh_err;
 +        const char *ssh_err;
          int ssh_err_code;
 -        /* This is not an errno.  See <libssh2.h>. */
 -        ssh_err_code = libssh2_session_last_error(s->session,
 -                                                  &ssh_err, NULL, 0);
 -        error_setg(errp, "%s: %s (libssh2 error code: %d)",
 +        /* This is not an errno.  See <libssh/libssh.h>. */
 +        ssh_err = ssh_get_error(s->session);
 +        ssh_err_code = ssh_get_error_code(s->session);
 +        error_setg(errp, "%s: %s (libssh error code: %d)",
                     msg, ssh_err, ssh_err_code);
      } else {
          error_setg(errp, "%s", msg);
@@ -XXX,XX +XXX,XX @@ sftp_error_setg(Error **errp, BDRVSSHState *s, const char *fs, ...)
      va_end(args);
      if (s->sftp) {
 -        char *ssh_err;
 +        const char *ssh_err;
          int ssh_err_code;
 -        unsigned long sftp_err_code;
 +        int sftp_err_code;
 -        /* This is not an errno.  See <libssh2.h>. */
 -        ssh_err_code = libssh2_session_last_error(s->session,
 -                                                  &ssh_err, NULL, 0);
 -        /* See <libssh2_sftp.h>. */
 -        sftp_err_code = libssh2_sftp_last_error((s)->sftp);
 +        /* This is not an errno.  See <libssh/libssh.h>. */
 +        ssh_err = ssh_get_error(s->session);
 +        ssh_err_code = ssh_get_error_code(s->session);
 +        /* See <libssh/sftp.h>. */
 +        sftp_err_code = sftp_get_error(s->sftp);
          error_setg(errp,
 -                   "%s: %s (libssh2 error code: %d, sftp error code: %lu)",
 +                   "%s: %s (libssh error code: %d, sftp error code: %d)",
                     msg, ssh_err, ssh_err_code, sftp_err_code);
      } else {
          error_setg(errp, "%s", msg);
@@ -XXX,XX +XXX,XX @@ sftp_error_setg(Error **errp, BDRVSSHState *s, const char *fs, ...)
  static void sftp_error_trace(BDRVSSHState *s, const char *op)
  {
-     UserCreatableClass *ucc = USER_CREATABLE_CLASS(klass);
+-    char *ssh_err;
-     ucc->complete = event_loop_base_complete;
++    const char *ssh_err;
-+    ucc->can_be_deleted = event_loop_base_can_be_deleted;
+     int ssh_err_code;
+-    unsigned long sftp_err_code;
-     object_class_property_add(klass, "aio-max-batch", "int",
++    int sftp_err_code;
-                               event_loop_base_get_param,
-diff --git a/util/main-loop.c b/util/main-loop.c
+-    /* This is not an errno.  See <libssh2.h>. */
-index XXXXXXX..XXXXXXX 100644
+-    ssh_err_code = libssh2_session_last_error(s->session,
---- a/util/main-loop.c
+-                                              &ssh_err, NULL, 0);
-+++ b/util/main-loop.c
+-    /* See <libssh2_sftp.h>. */
-@@ -XXX,XX +XXX,XX @@
+-    sftp_err_code = libssh2_sftp_last_error((s)->sftp);
- #include "qemu/error-report.h"
++    /* This is not an errno.  See <libssh/libssh.h>. */
- #include "qemu/queue.h"
++    ssh_err = ssh_get_error(s->session);
- #include "qemu/compiler.h"
++    ssh_err_code = ssh_get_error_code(s->session);
-+#include "qom/object.h"
++    /* See <libssh/sftp.h>. */
++    sftp_err_code = sftp_get_error(s->sftp);
- #ifndef _WIN32
- #include <sys/wait.h>
+     trace_sftp_error(op, ssh_err, ssh_err_code, sftp_err_code);
-@@ -XXX,XX +XXX,XX @@ int qemu_init_main_loop(Error **errp)
+ }
@@ -XXX,XX +XXX,XX @@ static void ssh_parse_filename(const char *filename, QDict *options,
      parse_uri(filename, options, errp);
  }
 -static int check_host_key_knownhosts(BDRVSSHState *s,
 -                                     const char *host, int port, Error **errp)
 +static int check_host_key_knownhosts(BDRVSSHState *s, Error **errp)
  {
 -    const char *home;
 -    char *knh_file = NULL;
 -    LIBSSH2_KNOWNHOSTS *knh = NULL;
 -    struct libssh2_knownhost *found;
 -    int ret, r;
 -    const char *hostkey;
 -    size_t len;
 -    int type;
 -
 -    hostkey = libssh2_session_hostkey(s->session, &len, &type);
 -    if (!hostkey) {
 +    int ret;
 +#ifdef HAVE_LIBSSH_0_8
 +    enum ssh_known_hosts_e state;
 +    int r;
 +    ssh_key pubkey;
 +    enum ssh_keytypes_e pubkey_type;
 +    unsigned char *server_hash = NULL;
 +    size_t server_hash_len;
 +    char *fingerprint = NULL;
 +
 +    state = ssh_session_is_known_server(s->session);
 +    trace_ssh_server_status(state);
 +
 +    switch (state) {
 +    case SSH_KNOWN_HOSTS_OK:
 +        /* OK */
 +        trace_ssh_check_host_key_knownhosts();
 +        break;
 +    case SSH_KNOWN_HOSTS_CHANGED:
          ret = -EINVAL;
 -        session_error_setg(errp, s, "failed to read remote host key");
 +        r = ssh_get_server_publickey(s->session, &pubkey);
 +        if (r == 0) {
 +            r = ssh_get_publickey_hash(pubkey, SSH_PUBLICKEY_HASH_SHA256,
 +                                       &server_hash, &server_hash_len);
 +            pubkey_type = ssh_key_type(pubkey);
 +            ssh_key_free(pubkey);
 +        }
 +        if (r == 0) {
 +            fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA256,
 +                                                   server_hash,
 +                                                   server_hash_len);
 +            ssh_clean_pubkey_hash(&server_hash);
 +        }
 +        if (fingerprint) {
 +            error_setg(errp,
 +                       "host key (%s key with fingerprint %s) does not match "
 +                       "the one in known_hosts; this may be a possible attack",
 +                       ssh_key_type_to_char(pubkey_type), fingerprint);
 +            ssh_string_free_char(fingerprint);
 +        } else  {
 +            error_setg(errp,
 +                       "host key does not match the one in known_hosts; this "
 +                       "may be a possible attack");
 +        }
          goto out;
 -    }
 -
 -    knh = libssh2_knownhost_init(s->session);
 -    if (!knh) {
 +    case SSH_KNOWN_HOSTS_OTHER:
          ret = -EINVAL;
 -        session_error_setg(errp, s,
 -                           "failed to initialize known hosts support");
 +        error_setg(errp,
 +                   "host key for this server not found, another type exists");
 +        goto out;
 +    case SSH_KNOWN_HOSTS_UNKNOWN:
 +        ret = -EINVAL;
 +        error_setg(errp, "no host key was found in known_hosts");
 +        goto out;
 +    case SSH_KNOWN_HOSTS_NOT_FOUND:
 +        ret = -ENOENT;
 +        error_setg(errp, "known_hosts file not found");
 +        goto out;
 +    case SSH_KNOWN_HOSTS_ERROR:
 +        ret = -EINVAL;
 +        error_setg(errp, "error while checking the host");
 +        goto out;
 +    default:
 +        ret = -EINVAL;
 +        error_setg(errp, "error while checking for known server (%d)", state);
          goto out;
      }
 +#else /* !HAVE_LIBSSH_0_8 */
 +    int state;
 -    home = getenv("HOME");
 -    if (home) {
 -        knh_file = g_strdup_printf("%s/.ssh/known_hosts", home);
 -    } else {
 -        knh_file = g_strdup_printf("/root/.ssh/known_hosts");
 -    }
 -
 -    /* Read all known hosts from OpenSSH-style known_hosts file. */
 -    libssh2_knownhost_readfile(knh, knh_file, LIBSSH2_KNOWNHOST_FILE_OPENSSH);
 +    state = ssh_is_server_known(s->session);
 +    trace_ssh_server_status(state);
 -    r = libssh2_knownhost_checkp(knh, host, port, hostkey, len,
 -                                 LIBSSH2_KNOWNHOST_TYPE_PLAIN|
 -                                 LIBSSH2_KNOWNHOST_KEYENC_RAW,
 -                                 &found);
 -    switch (r) {
 -    case LIBSSH2_KNOWNHOST_CHECK_MATCH:
 +    switch (state) {
 +    case SSH_SERVER_KNOWN_OK:
          /* OK */
 -        trace_ssh_check_host_key_knownhosts(found->key);
 +        trace_ssh_check_host_key_knownhosts();
          break;
 -    case LIBSSH2_KNOWNHOST_CHECK_MISMATCH:
 +    case SSH_SERVER_KNOWN_CHANGED:
          ret = -EINVAL;
 -        session_error_setg(errp, s,
 -                      "host key does not match the one in known_hosts"
 -                      " (found key %s)", found->key);
 +        error_setg(errp,
 +                   "host key does not match the one in known_hosts; this "
 +                   "may be a possible attack");
          goto out;
 -    case LIBSSH2_KNOWNHOST_CHECK_NOTFOUND:
 +    case SSH_SERVER_FOUND_OTHER:
          ret = -EINVAL;
 -        session_error_setg(errp, s, "no host key was found in known_hosts");
 +        error_setg(errp,
 +                   "host key for this server not found, another type exists");
 +        goto out;
 +    case SSH_SERVER_FILE_NOT_FOUND:
 +        ret = -ENOENT;
 +        error_setg(errp, "known_hosts file not found");
          goto out;
 -    case LIBSSH2_KNOWNHOST_CHECK_FAILURE:
 +    case SSH_SERVER_NOT_KNOWN:
          ret = -EINVAL;
 -        session_error_setg(errp, s,
 -                      "failure matching the host key with known_hosts");
 +        error_setg(errp, "no host key was found in known_hosts");
 +        goto out;
 +    case SSH_SERVER_ERROR:
 +        ret = -EINVAL;
 +        error_setg(errp, "server error");
          goto out;
      default:
          ret = -EINVAL;
 -        session_error_setg(errp, s, "unknown error matching the host key"
 -                      " with known_hosts (%d)", r);
 +        error_setg(errp, "error while checking for known server (%d)", state);
          goto out;
      }
 +#endif /* !HAVE_LIBSSH_0_8 */
      /* known_hosts checking successful. */
      ret = 0;
   out:
 -    if (knh != NULL) {
 -        libssh2_knownhost_free(knh);
 -    }
 -    g_free(knh_file);
      return ret;
  }
@@ -XXX,XX +XXX,XX @@ static int compare_fingerprint(const unsigned char *fingerprint, size_t len,
  static int
  check_host_key_hash(BDRVSSHState *s, const char *hash,
 -                    int hash_type, size_t fingerprint_len, Error **errp)
 +                    enum ssh_publickey_hash_type type, Error **errp)
  {
 -    const char *fingerprint;
 -
 -    fingerprint = libssh2_hostkey_hash(s->session, hash_type);
 -    if (!fingerprint) {
 +    int r;
 +    ssh_key pubkey;
 +    unsigned char *server_hash;
 +    size_t server_hash_len;
 +
 +#ifdef HAVE_LIBSSH_0_8
 +    r = ssh_get_server_publickey(s->session, &pubkey);
 +#else
 +    r = ssh_get_publickey(s->session, &pubkey);
 +#endif
 +    if (r != SSH_OK) {
          session_error_setg(errp, s, "failed to read remote host key");
          return -EINVAL;
      }
 -    if(compare_fingerprint((unsigned char *) fingerprint, fingerprint_len,
 -                           hash) != 0) {
 +    r = ssh_get_publickey_hash(pubkey, type, &server_hash, &server_hash_len);
 +    ssh_key_free(pubkey);
 +    if (r != 0) {
 +        session_error_setg(errp, s,
 +                           "failed reading the hash of the server SSH key");
 +        return -EINVAL;
 +    }
 +
 +    r = compare_fingerprint(server_hash, server_hash_len, hash);
 +    ssh_clean_pubkey_hash(&server_hash);
 +    if (r != 0) {
          error_setg(errp, "remote host key does not match host_key_check '%s'",
                     hash);
          return -EPERM;
@@ -XXX,XX +XXX,XX @@ check_host_key_hash(BDRVSSHState *s, const char *hash,
      return 0;
  }
-+static void main_loop_update_params(EventLoopBase *base, Error **errp)
+-static int check_host_key(BDRVSSHState *s, const char *host, int port,
-+{
+-                          SshHostKeyCheck *hkc, Error **errp)
-+    if (!qemu_aio_context) {
++static int check_host_key(BDRVSSHState *s, SshHostKeyCheck *hkc, Error **errp)
-+        error_setg(errp, "qemu aio context not ready");
+ {
-+        return;
+     SshHostKeyCheckMode mode;
@@ -XXX,XX +XXX,XX @@ static int check_host_key(BDRVSSHState *s, const char *host, int port,
      case SSH_HOST_KEY_CHECK_MODE_HASH:
          if (hkc->u.hash.type == SSH_HOST_KEY_CHECK_HASH_TYPE_MD5) {
              return check_host_key_hash(s, hkc->u.hash.hash,
 -                                       LIBSSH2_HOSTKEY_HASH_MD5, 16, errp);
 +                                       SSH_PUBLICKEY_HASH_MD5, errp);
          } else if (hkc->u.hash.type == SSH_HOST_KEY_CHECK_HASH_TYPE_SHA1) {
              return check_host_key_hash(s, hkc->u.hash.hash,
 -                                       LIBSSH2_HOSTKEY_HASH_SHA1, 20, errp);
 +                                       SSH_PUBLICKEY_HASH_SHA1, errp);
          }
          g_assert_not_reached();
          break;
      case SSH_HOST_KEY_CHECK_MODE_KNOWN_HOSTS:
 -        return check_host_key_knownhosts(s, host, port, errp);
 +        return check_host_key_knownhosts(s, errp);
      default:
          g_assert_not_reached();
      }
@@ -XXX,XX +XXX,XX @@ static int check_host_key(BDRVSSHState *s, const char *host, int port,
      return -EINVAL;
  }
 -static int authenticate(BDRVSSHState *s, const char *user, Error **errp)
 +static int authenticate(BDRVSSHState *s, Error **errp)
  {
      int r, ret;
 -    const char *userauthlist;
 -    LIBSSH2_AGENT *agent = NULL;
 -    struct libssh2_agent_publickey *identity;
 -    struct libssh2_agent_publickey *prev_identity = NULL;
 +    int method;
 -    userauthlist = libssh2_userauth_list(s->session, user, strlen(user));
 -    if (strstr(userauthlist, "publickey") == NULL) {
 +    /* Try to authenticate with the "none" method. */
 +    r = ssh_userauth_none(s->session, NULL);
 +    if (r == SSH_AUTH_ERROR) {
          ret = -EPERM;
 -        error_setg(errp,
 -                "remote server does not support \"publickey\" authentication");
 +        session_error_setg(errp, s, "failed to authenticate using none "
 +                                    "authentication");
          goto out;
 -    }
 -
 -    /* Connect to ssh-agent and try each identity in turn. */
 -    agent = libssh2_agent_init(s->session);
 -    if (!agent) {
 -        ret = -EINVAL;
 -        session_error_setg(errp, s, "failed to initialize ssh-agent support");
 -        goto out;
 -    }
 -    if (libssh2_agent_connect(agent)) {
 -        ret = -ECONNREFUSED;
 -        session_error_setg(errp, s, "failed to connect to ssh-agent");
 -        goto out;
 -    }
 -    if (libssh2_agent_list_identities(agent)) {
 -        ret = -EINVAL;
 -        session_error_setg(errp, s,
 -                           "failed requesting identities from ssh-agent");
 +    } else if (r == SSH_AUTH_SUCCESS) {
 +        /* Authenticated! */
 +        ret = 0;
          goto out;
      }
 -    for(;;) {
 -        r = libssh2_agent_get_identity(agent, &identity, prev_identity);
 -        if (r == 1) {           /* end of list */
 -            break;
 -        }
 -        if (r < 0) {
 +    method = ssh_userauth_list(s->session, NULL);
 +    trace_ssh_auth_methods(method);
 +
 +    /*
 +     * Try to authenticate with publickey, using the ssh-agent
 +     * if available.
 +     */
 +    if (method & SSH_AUTH_METHOD_PUBLICKEY) {
 +        r = ssh_userauth_publickey_auto(s->session, NULL, NULL);
 +        if (r == SSH_AUTH_ERROR) {
              ret = -EINVAL;
 -            session_error_setg(errp, s,
 -                               "failed to obtain identity from ssh-agent");
 +            session_error_setg(errp, s, "failed to authenticate using "
 +                                        "publickey authentication");
              goto out;
 -        }
 -        r = libssh2_agent_userauth(agent, user, identity);
 -        if (r == 0) {
 +        } else if (r == SSH_AUTH_SUCCESS) {
              /* Authenticated! */
              ret = 0;
              goto out;
          }
 -        /* Failed to authenticate with this identity, try the next one. */
 -        prev_identity = identity;
      }
      ret = -EPERM;
@@ -XXX,XX +XXX,XX @@ static int authenticate(BDRVSSHState *s, const char *user, Error **errp)
                 "and the identities held by your ssh-agent");
   out:
 -    if (agent != NULL) {
 -        /* Note: libssh2 implementation implicitly calls
 -         * libssh2_agent_disconnect if necessary.
 -         */
 -        libssh2_agent_free(agent);
 -    }
 -
      return ret;
  }
@@ -XXX,XX +XXX,XX @@ static int connect_to_ssh(BDRVSSHState *s, BlockdevOptionsSsh *opts,
                            int ssh_flags, int creat_mode, Error **errp)
  {
      int r, ret;
 -    long port = 0;
 +    unsigned int port = 0;
 +    int new_sock = -1;
      if (opts->has_user) {
          s->user = g_strdup(opts->user);
@@ -XXX,XX +XXX,XX @@ static int connect_to_ssh(BDRVSSHState *s, BlockdevOptionsSsh *opts,
      s->inet = opts->server;
      opts->server = NULL;
 -    if (qemu_strtol(s->inet->port, NULL, 10, &port) < 0) {
 +    if (qemu_strtoui(s->inet->port, NULL, 10, &port) < 0) {
          error_setg(errp, "Use only numeric port value");
          ret = -EINVAL;
          goto err;
      }
      /* Open the socket and connect. */
 -    s->sock = inet_connect_saddr(s->inet, errp);
 -    if (s->sock < 0) {
 +    new_sock = inet_connect_saddr(s->inet, errp);
 +    if (new_sock < 0) {
          ret = -EIO;
          goto err;
      }
 +    /*
 +     * Try to disable the Nagle algorithm on TCP sockets to reduce latency,
 +     * but do not fail if it cannot be disabled.
 +     */
 +    r = socket_set_nodelay(new_sock);
 +    if (r < 0) {
 +        warn_report("can't set TCP_NODELAY for the ssh server %s: %s",
 +                    s->inet->host, strerror(errno));
 +    }
 +
-+    aio_context_set_aio_params(qemu_aio_context, base->aio_max_batch, errp);
+     /* Create SSH session. */
-+}
+-    s->session = libssh2_session_init();
-+
++    s->session = ssh_new();
-+MainLoop *mloop;
+     if (!s->session) {
-+
+         ret = -EINVAL;
-+static void main_loop_init(EventLoopBase *base, Error **errp)
+-        session_error_setg(errp, s, "failed to initialize libssh2 session");
-+{
++        session_error_setg(errp, s, "failed to initialize libssh session");
-+    MainLoop *m = MAIN_LOOP(base);
+         goto err;
-+
+     }
-+    if (mloop) {
-+        error_setg(errp, "only one main-loop instance allowed");
+-#if TRACE_LIBSSH2 != 0
-+        return;
+-    libssh2_trace(s->session, TRACE_LIBSSH2);
 -#endif
 +    /*
 +     * Make sure we are in blocking mode during the connection and
 +     * authentication phases.
 +     */
 +    ssh_set_blocking(s->session, 1);
 -    r = libssh2_session_handshake(s->session, s->sock);
 -    if (r != 0) {
 +    r = ssh_options_set(s->session, SSH_OPTIONS_USER, s->user);
 +    if (r < 0) {
 +        ret = -EINVAL;
 +        session_error_setg(errp, s,
 +                           "failed to set the user in the libssh session");
 +        goto err;
 +    }
 +
-+    main_loop_update_params(base, errp);
++    r = ssh_options_set(s->session, SSH_OPTIONS_HOST, s->inet->host);
-+
++    if (r < 0) {
-+    mloop = m;
++        ret = -EINVAL;
-+    return;
++        session_error_setg(errp, s,
-+}
++                           "failed to set the host in the libssh session");
-+
++        goto err;
-+static bool main_loop_can_be_deleted(EventLoopBase *base)
++    }
-+{
++
-+    return false;
++    if (port > 0) {
-+}
++        r = ssh_options_set(s->session, SSH_OPTIONS_PORT, &port);
-+
++        if (r < 0) {
-+static void main_loop_class_init(ObjectClass *oc, void *class_data)
++            ret = -EINVAL;
-+{
++            session_error_setg(errp, s,
-+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_CLASS(oc);
++                               "failed to set the port in the libssh session");
-+
++            goto err;
-+    bc->init = main_loop_init;
++        }
-+    bc->update_params = main_loop_update_params;
++    }
-+    bc->can_be_deleted = main_loop_can_be_deleted;
++
-+}
++    r = ssh_options_set(s->session, SSH_OPTIONS_COMPRESSION, "none");
-+
++    if (r < 0) {
-+static const TypeInfo main_loop_info = {
++        ret = -EINVAL;
-+    .name = TYPE_MAIN_LOOP,
++        session_error_setg(errp, s,
-+    .parent = TYPE_EVENT_LOOP_BASE,
++                           "failed to disable the compression in the libssh "
-+    .class_init = main_loop_class_init,
++                           "session");
-+    .instance_size = sizeof(MainLoop),
++        goto err;
-+};
++    }
 +
-+static void main_loop_register_types(void)
++    /* Read ~/.ssh/config. */
-+{
++    r = ssh_options_parse_config(s->session, NULL);
-+    type_register_static(&main_loop_info);
++    if (r < 0) {
-+}
++        ret = -EINVAL;
-+
++        session_error_setg(errp, s, "failed to parse ~/.ssh/config");
-+type_init(main_loop_register_types)
++        goto err;
-+
++    }
- static int max_priority;
++
++    r = ssh_options_set(s->session, SSH_OPTIONS_FD, &new_sock);
- #ifndef _WIN32
++    if (r < 0) {
 +        ret = -EINVAL;
 +        session_error_setg(errp, s,
 +                           "failed to set the socket in the libssh session");
 +        goto err;
 +    }
 +    /* libssh took ownership of the socket. */
 +    s->sock = new_sock;
 +    new_sock = -1;
 +
 +    /* Connect. */
 +    r = ssh_connect(s->session);
 +    if (r != SSH_OK) {
          ret = -EINVAL;
          session_error_setg(errp, s, "failed to establish SSH session");
          goto err;
      }
      /* Check the remote host's key against known_hosts. */
 -    ret = check_host_key(s, s->inet->host, port, opts->host_key_check, errp);
 +    ret = check_host_key(s, opts->host_key_check, errp);
      if (ret < 0) {
          goto err;
      }
      /* Authenticate. */
 -    ret = authenticate(s, s->user, errp);
 +    ret = authenticate(s, errp);
      if (ret < 0) {
          goto err;
      }
      /* Start SFTP. */
 -    s->sftp = libssh2_sftp_init(s->session);
 +    s->sftp = sftp_new(s->session);
      if (!s->sftp) {
 -        session_error_setg(errp, s, "failed to initialize sftp handle");
 +        session_error_setg(errp, s, "failed to create sftp handle");
 +        ret = -EINVAL;
 +        goto err;
 +    }
 +
 +    r = sftp_init(s->sftp);
 +    if (r < 0) {
 +        sftp_error_setg(errp, s, "failed to initialize sftp handle");
          ret = -EINVAL;
          goto err;
      }
      /* Open the remote file. */
      trace_ssh_connect_to_ssh(opts->path, ssh_flags, creat_mode);
 -    s->sftp_handle = libssh2_sftp_open(s->sftp, opts->path, ssh_flags,
 -                                       creat_mode);
 +    s->sftp_handle = sftp_open(s->sftp, opts->path, ssh_flags, creat_mode);
      if (!s->sftp_handle) {
 -        session_error_setg(errp, s, "failed to open remote file '%s'",
 -                           opts->path);
 +        sftp_error_setg(errp, s, "failed to open remote file '%s'",
 +                        opts->path);
          ret = -EINVAL;
          goto err;
      }
 -    r = libssh2_sftp_fstat(s->sftp_handle, &s->attrs);
 -    if (r < 0) {
 +    /* Make sure the SFTP file is handled in blocking mode. */
 +    sftp_file_set_blocking(s->sftp_handle);
 +
 +    s->attrs = sftp_fstat(s->sftp_handle);
 +    if (!s->attrs) {
          sftp_error_setg(errp, s, "failed to read file attributes");
          return -EINVAL;
      }
@@ -XXX,XX +XXX,XX @@ static int connect_to_ssh(BDRVSSHState *s, BlockdevOptionsSsh *opts,
      return 0;
   err:
 +    if (s->attrs) {
 +        sftp_attributes_free(s->attrs);
 +    }
 +    s->attrs = NULL;
      if (s->sftp_handle) {
 -        libssh2_sftp_close(s->sftp_handle);
 +        sftp_close(s->sftp_handle);
      }
      s->sftp_handle = NULL;
      if (s->sftp) {
 -        libssh2_sftp_shutdown(s->sftp);
 +        sftp_free(s->sftp);
      }
      s->sftp = NULL;
      if (s->session) {
 -        libssh2_session_disconnect(s->session,
 -                                   "from qemu ssh client: "
 -                                   "error opening connection");
 -        libssh2_session_free(s->session);
 +        ssh_disconnect(s->session);
 +        ssh_free(s->session);
      }
      s->session = NULL;
 +    s->sock = -1;
 +    if (new_sock >= 0) {
 +        close(new_sock);
 +    }
      return ret;
  }
@@ -XXX,XX +XXX,XX @@ static int ssh_file_open(BlockDriverState *bs, QDict *options, int bdrv_flags,
      ssh_state_init(s);
 -    ssh_flags = LIBSSH2_FXF_READ;
 +    ssh_flags = 0;
      if (bdrv_flags & BDRV_O_RDWR) {
 -        ssh_flags |= LIBSSH2_FXF_WRITE;
 +        ssh_flags |= O_RDWR;
 +    } else {
 +        ssh_flags |= O_RDONLY;
      }
      opts = ssh_parse_options(options, errp);
@@ -XXX,XX +XXX,XX @@ static int ssh_file_open(BlockDriverState *bs, QDict *options, int bdrv_flags,
      }
      /* Go non-blocking. */
 -    libssh2_session_set_blocking(s->session, 0);
 +    ssh_set_blocking(s->session, 0);
      qapi_free_BlockdevOptionsSsh(opts);
      return 0;
   err:
 -    if (s->sock >= 0) {
 -        close(s->sock);
 -    }
 -    s->sock = -1;
 -
      qapi_free_BlockdevOptionsSsh(opts);
      return ret;
@@ -XXX,XX +XXX,XX @@ static int ssh_grow_file(BDRVSSHState *s, int64_t offset, Error **errp)
  {
      ssize_t ret;
      char c[1] = { '\0' };
 -    int was_blocking = libssh2_session_get_blocking(s->session);
 +    int was_blocking = ssh_is_blocking(s->session);
      /* offset must be strictly greater than the current size so we do
       * not overwrite anything */
 -    assert(offset > 0 && offset > s->attrs.filesize);
 +    assert(offset > 0 && offset > s->attrs->size);
 -    libssh2_session_set_blocking(s->session, 1);
 +    ssh_set_blocking(s->session, 1);
 -    libssh2_sftp_seek64(s->sftp_handle, offset - 1);
 -    ret = libssh2_sftp_write(s->sftp_handle, c, 1);
 +    sftp_seek64(s->sftp_handle, offset - 1);
 +    ret = sftp_write(s->sftp_handle, c, 1);
 -    libssh2_session_set_blocking(s->session, was_blocking);
 +    ssh_set_blocking(s->session, was_blocking);
      if (ret < 0) {
          sftp_error_setg(errp, s, "Failed to grow file");
          return -EIO;
      }
 -    s->attrs.filesize = offset;
 +    s->attrs->size = offset;
      return 0;
  }
@@ -XXX,XX +XXX,XX @@ static int ssh_co_create(BlockdevCreateOptions *options, Error **errp)
      ssh_state_init(&s);
      ret = connect_to_ssh(&s, opts->location,
 -                         LIBSSH2_FXF_READ|LIBSSH2_FXF_WRITE|
 -                         LIBSSH2_FXF_CREAT|LIBSSH2_FXF_TRUNC,
 +                         O_RDWR | O_CREAT | O_TRUNC,
 , errp);
      if (ret < 0) {
          goto fail;
@@ -XXX,XX +XXX,XX @@ static int ssh_has_zero_init(BlockDriverState *bs)
      /* Assume false, unless we can positively prove it's true. */
      int has_zero_init = 0;
 -    if (s->attrs.flags & LIBSSH2_SFTP_ATTR_PERMISSIONS) {
 -        if (s->attrs.permissions & LIBSSH2_SFTP_S_IFREG) {
 -            has_zero_init = 1;
 -        }
 +    if (s->attrs->type == SSH_FILEXFER_TYPE_REGULAR) {
 +        has_zero_init = 1;
      }
      return has_zero_init;
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void co_yield(BDRVSSHState *s, BlockDriverState *bs)
          .co = qemu_coroutine_self()
      };
 -    r = libssh2_session_block_directions(s->session);
 +    r = ssh_get_poll_flags(s->session);
 -    if (r & LIBSSH2_SESSION_BLOCK_INBOUND) {
 +    if (r & SSH_READ_PENDING) {
          rd_handler = restart_coroutine;
      }
 -    if (r & LIBSSH2_SESSION_BLOCK_OUTBOUND) {
 +    if (r & SSH_WRITE_PENDING) {
          wr_handler = restart_coroutine;
      }
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void co_yield(BDRVSSHState *s, BlockDriverState *bs)
      trace_ssh_co_yield_back(s->sock);
  }
 -/* SFTP has a function `libssh2_sftp_seek64' which seeks to a position
 - * in the remote file.  Notice that it just updates a field in the
 - * sftp_handle structure, so there is no network traffic and it cannot
 - * fail.
 - *
 - * However, `libssh2_sftp_seek64' does have a catastrophic effect on
 - * performance since it causes the handle to throw away all in-flight
 - * reads and buffered readahead data.  Therefore this function tries
 - * to be intelligent about when to call the underlying libssh2 function.
 - */
 -#define SSH_SEEK_WRITE 0
 -#define SSH_SEEK_READ  1
 -#define SSH_SEEK_FORCE 2
 -
 -static void ssh_seek(BDRVSSHState *s, int64_t offset, int flags)
 -{
 -    bool op_read = (flags & SSH_SEEK_READ) != 0;
 -    bool force = (flags & SSH_SEEK_FORCE) != 0;
 -
 -    if (force || op_read != s->offset_op_read || offset != s->offset) {
 -        trace_ssh_seek(offset);
 -        libssh2_sftp_seek64(s->sftp_handle, offset);
 -        s->offset = offset;
 -        s->offset_op_read = op_read;
 -    }
 -}
 -
  static coroutine_fn int ssh_read(BDRVSSHState *s, BlockDriverState *bs,
                                   int64_t offset, size_t size,
                                   QEMUIOVector *qiov)
@@ -XXX,XX +XXX,XX @@ static coroutine_fn int ssh_read(BDRVSSHState *s, BlockDriverState *bs,
      trace_ssh_read(offset, size);
 -    ssh_seek(s, offset, SSH_SEEK_READ);
 +    trace_ssh_seek(offset);
 +    sftp_seek64(s->sftp_handle, offset);
      /* This keeps track of the current iovec element ('i'), where we
       * will write to next ('buf'), and the end of the current iovec
@@ -XXX,XX +XXX,XX @@ static coroutine_fn int ssh_read(BDRVSSHState *s, BlockDriverState *bs,
      buf = i->iov_base;
      end_of_vec = i->iov_base + i->iov_len;
 -    /* libssh2 has a hard-coded limit of 2000 bytes per request,
 -     * although it will also do readahead behind our backs.  Therefore
 -     * we may have to do repeated reads here until we have read 'size'
 -     * bytes.
 -     */
      for (got = 0; got < size; ) {
 +        size_t request_read_size;
      again:
 -        trace_ssh_read_buf(buf, end_of_vec - buf);
 -        r = libssh2_sftp_read(s->sftp_handle, buf, end_of_vec - buf);
 -        trace_ssh_read_return(r);
 +        /*
 +         * The size of SFTP packets is limited to 32K bytes, so limit
 +         * the amount of data requested to 16K, as libssh currently
 +         * does not handle multiple requests on its own.
 +         */
 +        request_read_size = MIN(end_of_vec - buf, 16384);
 +        trace_ssh_read_buf(buf, end_of_vec - buf, request_read_size);
 +        r = sftp_read(s->sftp_handle, buf, request_read_size);
 +        trace_ssh_read_return(r, sftp_get_error(s->sftp));
 -        if (r == LIBSSH2_ERROR_EAGAIN || r == LIBSSH2_ERROR_TIMEOUT) {
 +        if (r == SSH_AGAIN) {
              co_yield(s, bs);
              goto again;
          }
 -        if (r < 0) {
 -            sftp_error_trace(s, "read");
 -            s->offset = -1;
 -            return -EIO;
 -        }
 -        if (r == 0) {
 +        if (r == SSH_EOF || (r == 0 && sftp_get_error(s->sftp) == SSH_FX_EOF)) {
              /* EOF: Short read so pad the buffer with zeroes and return it. */
              qemu_iovec_memset(qiov, got, 0, size - got);
              return 0;
          }
 +        if (r <= 0) {
 +            sftp_error_trace(s, "read");
 +            return -EIO;
 +        }
          got += r;
          buf += r;
 -        s->offset += r;
          if (buf >= end_of_vec && got < size) {
              i++;
              buf = i->iov_base;
@@ -XXX,XX +XXX,XX @@ static int ssh_write(BDRVSSHState *s, BlockDriverState *bs,
      trace_ssh_write(offset, size);
 -    ssh_seek(s, offset, SSH_SEEK_WRITE);
 +    trace_ssh_seek(offset);
 +    sftp_seek64(s->sftp_handle, offset);
      /* This keeps track of the current iovec element ('i'), where we
       * will read from next ('buf'), and the end of the current iovec
@@ -XXX,XX +XXX,XX @@ static int ssh_write(BDRVSSHState *s, BlockDriverState *bs,
      end_of_vec = i->iov_base + i->iov_len;
      for (written = 0; written < size; ) {
 +        size_t request_write_size;
      again:
 -        trace_ssh_write_buf(buf, end_of_vec - buf);
 -        r = libssh2_sftp_write(s->sftp_handle, buf, end_of_vec - buf);
 -        trace_ssh_write_return(r);
 +        /*
 +         * Avoid too large data packets, as libssh currently does not
 +         * handle multiple requests on its own.
 +         */
 +        request_write_size = MIN(end_of_vec - buf, 131072);
 +        trace_ssh_write_buf(buf, end_of_vec - buf, request_write_size);
 +        r = sftp_write(s->sftp_handle, buf, request_write_size);
 +        trace_ssh_write_return(r, sftp_get_error(s->sftp));
 -        if (r == LIBSSH2_ERROR_EAGAIN || r == LIBSSH2_ERROR_TIMEOUT) {
 +        if (r == SSH_AGAIN) {
              co_yield(s, bs);
              goto again;
          }
          if (r < 0) {
              sftp_error_trace(s, "write");
 -            s->offset = -1;
              return -EIO;
          }
 -        /* The libssh2 API is very unclear about this.  A comment in
 -         * the code says "nothing was acked, and no EAGAIN was
 -         * received!" which apparently means that no data got sent
 -         * out, and the underlying channel didn't return any EAGAIN
 -         * indication.  I think this is a bug in either libssh2 or
 -         * OpenSSH (server-side).  In any case, forcing a seek (to
 -         * discard libssh2 internal buffers), and then trying again
 -         * works for me.
 -         */
 -        if (r == 0) {
 -            ssh_seek(s, offset + written, SSH_SEEK_WRITE|SSH_SEEK_FORCE);
 -            co_yield(s, bs);
 -            goto again;
 -        }
          written += r;
          buf += r;
 -        s->offset += r;
          if (buf >= end_of_vec && written < size) {
              i++;
              buf = i->iov_base;
              end_of_vec = i->iov_base + i->iov_len;
          }
 -        if (offset + written > s->attrs.filesize)
 -            s->attrs.filesize = offset + written;
 +        if (offset + written > s->attrs->size) {
 +            s->attrs->size = offset + written;
 +        }
      }
      return 0;
@@ -XXX,XX +XXX,XX @@ static void unsafe_flush_warning(BDRVSSHState *s, const char *what)
      }
  }
 -#ifdef HAS_LIBSSH2_SFTP_FSYNC
 +#ifdef HAVE_LIBSSH_0_8
  static coroutine_fn int ssh_flush(BDRVSSHState *s, BlockDriverState *bs)
  {
      int r;
      trace_ssh_flush();
 +
 +    if (!sftp_extension_supported(s->sftp, "fsync@openssh.com", "1")) {
 +        unsafe_flush_warning(s, "OpenSSH >= 6.3");
 +        return 0;
 +    }
   again:
 -    r = libssh2_sftp_fsync(s->sftp_handle);
 -    if (r == LIBSSH2_ERROR_EAGAIN || r == LIBSSH2_ERROR_TIMEOUT) {
 +    r = sftp_fsync(s->sftp_handle);
 +    if (r == SSH_AGAIN) {
          co_yield(s, bs);
          goto again;
      }
 -    if (r == LIBSSH2_ERROR_SFTP_PROTOCOL &&
 -        libssh2_sftp_last_error(s->sftp) == LIBSSH2_FX_OP_UNSUPPORTED) {
 -        unsafe_flush_warning(s, "OpenSSH >= 6.3");
 -        return 0;
 -    }
      if (r < 0) {
          sftp_error_trace(s, "fsync");
          return -EIO;
@@ -XXX,XX +XXX,XX @@ static coroutine_fn int ssh_co_flush(BlockDriverState *bs)
      return ret;
  }
 -#else /* !HAS_LIBSSH2_SFTP_FSYNC */
 +#else /* !HAVE_LIBSSH_0_8 */
  static coroutine_fn int ssh_co_flush(BlockDriverState *bs)
  {
      BDRVSSHState *s = bs->opaque;
 -    unsafe_flush_warning(s, "libssh2 >= 1.4.4");
 +    unsafe_flush_warning(s, "libssh >= 0.8.0");
      return 0;
  }
 -#endif /* !HAS_LIBSSH2_SFTP_FSYNC */
 +#endif /* !HAVE_LIBSSH_0_8 */
  static int64_t ssh_getlength(BlockDriverState *bs)
  {
      BDRVSSHState *s = bs->opaque;
      int64_t length;
 -    /* Note we cannot make a libssh2 call here. */
 -    length = (int64_t) s->attrs.filesize;
 +    /* Note we cannot make a libssh call here. */
 +    length = (int64_t) s->attrs->size;
      trace_ssh_getlength(length);
      return length;
@@ -XXX,XX +XXX,XX @@ static int coroutine_fn ssh_co_truncate(BlockDriverState *bs, int64_t offset,
          return -ENOTSUP;
      }
 -    if (offset < s->attrs.filesize) {
 +    if (offset < s->attrs->size) {
          error_setg(errp, "ssh driver does not support shrinking files");
          return -ENOTSUP;
      }
 -    if (offset == s->attrs.filesize) {
 +    if (offset == s->attrs->size) {
          return 0;
      }
@@ -XXX,XX +XXX,XX @@ static void bdrv_ssh_init(void)
  {
      int r;
 -    r = libssh2_init(0);
 +    r = ssh_init();
      if (r != 0) {
 -        fprintf(stderr, "libssh2 initialization failed, %d\n", r);
 +        fprintf(stderr, "libssh initialization failed, %d\n", r);
          exit(EXIT_FAILURE);
      }
 +#if TRACE_LIBSSH != 0
 +    ssh_set_log_level(TRACE_LIBSSH);
 +#endif
 +
      bdrv_register(&bdrv_ssh);
  }
 diff --git a/.travis.yml b/.travis.yml
 index XXXXXXX..XXXXXXX 100644
 --- a/.travis.yml
 +++ b/.travis.yml
@@ -XXX,XX +XXX,XX @@ addons:
        - libseccomp-dev
        - libspice-protocol-dev
        - libspice-server-dev
 -      - libssh2-1-dev
 +      - libssh-dev
        - liburcu-dev
        - libusb-1.0-0-dev
        - libvte-2.91-dev
@@ -XXX,XX +XXX,XX @@ matrix:
              - libseccomp-dev
              - libspice-protocol-dev
              - libspice-server-dev
 -            - libssh2-1-dev
 +            - libssh-dev
              - liburcu-dev
              - libusb-1.0-0-dev
              - libvte-2.91-dev
 diff --git a/block/trace-events b/block/trace-events
 index XXXXXXX..XXXXXXX 100644
 --- a/block/trace-events
 +++ b/block/trace-events
@@ -XXX,XX +XXX,XX @@ nbd_client_connect_success(const char *export_name) "export '%s'"
  # ssh.c
  ssh_restart_coroutine(void *co) "co=%p"
  ssh_flush(void) "fsync"
 -ssh_check_host_key_knownhosts(const char *key) "host key OK: %s"
 +ssh_check_host_key_knownhosts(void) "host key OK"
  ssh_connect_to_ssh(char *path, int flags, int mode) "opening file %s flags=0x%x creat_mode=0%o"
  ssh_co_yield(int sock, void *rd_handler, void *wr_handler) "s->sock=%d rd_handler=%p wr_handler=%p"
  ssh_co_yield_back(int sock) "s->sock=%d - back"
  ssh_getlength(int64_t length) "length=%" PRIi64
  ssh_co_create_opts(uint64_t size) "total_size=%" PRIu64
  ssh_read(int64_t offset, size_t size) "offset=%" PRIi64 " size=%zu"
 -ssh_read_buf(void *buf, size_t size) "sftp_read buf=%p size=%zu"
 -ssh_read_return(ssize_t ret) "sftp_read returned %zd"
 +ssh_read_buf(void *buf, size_t size, size_t actual_size) "sftp_read buf=%p size=%zu (actual size=%zu)"
 +ssh_read_return(ssize_t ret, int sftp_err) "sftp_read returned %zd (sftp error=%d)"
  ssh_write(int64_t offset, size_t size) "offset=%" PRIi64 " size=%zu"
 -ssh_write_buf(void *buf, size_t size) "sftp_write buf=%p size=%zu"
 -ssh_write_return(ssize_t ret) "sftp_write returned %zd"
 +ssh_write_buf(void *buf, size_t size, size_t actual_size) "sftp_write buf=%p size=%zu (actual size=%zu)"
 +ssh_write_return(ssize_t ret, int sftp_err) "sftp_write returned %zd (sftp error=%d)"
  ssh_seek(int64_t offset) "seeking to offset=%" PRIi64
 +ssh_auth_methods(int methods) "auth methods=0x%x"
 +ssh_server_status(int status) "server status=%d"
  # curl.c
  curl_timer_cb(long timeout_ms) "timer callback timeout_ms %ld"
@@ -XXX,XX +XXX,XX @@ sheepdog_snapshot_create(const char *sn_name, const char *id) "%s %s"
  sheepdog_snapshot_create_inode(const char *name, uint32_t snap, uint32_t vdi) "s->inode: name %s snap_id 0x%" PRIx32 " vdi 0x%" PRIx32
  # ssh.c
 -sftp_error(const char *op, const char *ssh_err, int ssh_err_code, unsigned long sftp_err_code) "%s failed: %s (libssh2 error code: %d, sftp error code: %lu)"
 +sftp_error(const char *op, const char *ssh_err, int ssh_err_code, int sftp_err_code) "%s failed: %s (libssh error code: %d, sftp error code: %d)"
 diff --git a/docs/qemu-block-drivers.texi b/docs/qemu-block-drivers.texi
 index XXXXXXX..XXXXXXX 100644
 --- a/docs/qemu-block-drivers.texi
 +++ b/docs/qemu-block-drivers.texi
@@ -XXX,XX +XXX,XX @@ print a warning when @code{fsync} is not supported:
  warning: ssh server @code{ssh.example.com:22} does not support fsync
 -With sufficiently new versions of libssh2 and OpenSSH, @code{fsync} is
 +With sufficiently new versions of libssh and OpenSSH, @code{fsync} is
  supported.
  @node disk_images_nvme
 diff --git a/tests/docker/dockerfiles/debian-win32-cross.docker b/tests/docker/dockerfiles/debian-win32-cross.docker
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/docker/dockerfiles/debian-win32-cross.docker
 +++ b/tests/docker/dockerfiles/debian-win32-cross.docker
@@ -XXX,XX +XXX,XX @@ RUN DEBIAN_FRONTEND=noninteractive eatmydata \
          mxe-$TARGET-w64-mingw32.shared-curl \
          mxe-$TARGET-w64-mingw32.shared-glib \
          mxe-$TARGET-w64-mingw32.shared-libgcrypt \
 -        mxe-$TARGET-w64-mingw32.shared-libssh2 \
          mxe-$TARGET-w64-mingw32.shared-libusb1 \
          mxe-$TARGET-w64-mingw32.shared-lzo \
          mxe-$TARGET-w64-mingw32.shared-nettle \
 diff --git a/tests/docker/dockerfiles/debian-win64-cross.docker b/tests/docker/dockerfiles/debian-win64-cross.docker
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/docker/dockerfiles/debian-win64-cross.docker
 +++ b/tests/docker/dockerfiles/debian-win64-cross.docker
@@ -XXX,XX +XXX,XX @@ RUN DEBIAN_FRONTEND=noninteractive eatmydata \
          mxe-$TARGET-w64-mingw32.shared-curl \
          mxe-$TARGET-w64-mingw32.shared-glib \
          mxe-$TARGET-w64-mingw32.shared-libgcrypt \
 -        mxe-$TARGET-w64-mingw32.shared-libssh2 \
          mxe-$TARGET-w64-mingw32.shared-libusb1 \
          mxe-$TARGET-w64-mingw32.shared-lzo \
          mxe-$TARGET-w64-mingw32.shared-nettle \
 diff --git a/tests/docker/dockerfiles/fedora.docker b/tests/docker/dockerfiles/fedora.docker
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/docker/dockerfiles/fedora.docker
 +++ b/tests/docker/dockerfiles/fedora.docker
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES \
      libpng-devel \
      librbd-devel \
      libseccomp-devel \
 -    libssh2-devel \
 +    libssh-devel \
      libubsan \
      libusbx-devel \
      libxml2-devel \
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES \
      mingw32-gtk3 \
      mingw32-libjpeg-turbo \
      mingw32-libpng \
 -    mingw32-libssh2 \
      mingw32-libtasn1 \
      mingw32-nettle \
      mingw32-pixman \
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES \
      mingw64-gtk3 \
      mingw64-libjpeg-turbo \
      mingw64-libpng \
 -    mingw64-libssh2 \
      mingw64-libtasn1 \
      mingw64-nettle \
      mingw64-pixman \
 diff --git a/tests/docker/dockerfiles/ubuntu.docker b/tests/docker/dockerfiles/ubuntu.docker
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/docker/dockerfiles/ubuntu.docker
 +++ b/tests/docker/dockerfiles/ubuntu.docker
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES flex bison \
      libsnappy-dev \
      libspice-protocol-dev \
      libspice-server-dev \
 -    libssh2-1-dev \
 +    libssh-dev \
      libusb-1.0-0-dev \
      libusbredirhost-dev \
      libvdeplug-dev \
 diff --git a/tests/docker/dockerfiles/ubuntu1804.docker b/tests/docker/dockerfiles/ubuntu1804.docker
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/docker/dockerfiles/ubuntu1804.docker
 +++ b/tests/docker/dockerfiles/ubuntu1804.docker
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES flex bison \
      libsnappy-dev \
      libspice-protocol-dev \
      libspice-server-dev \
 -    libssh2-1-dev \
 +    libssh-dev \
      libusb-1.0-0-dev \
      libusbredirhost-dev \
      libvdeplug-dev \
 diff --git a/tests/qemu-iotests/207 b/tests/qemu-iotests/207
 index XXXXXXX..XXXXXXX 100755
 --- a/tests/qemu-iotests/207
 +++ b/tests/qemu-iotests/207
@@ -XXX,XX +XXX,XX @@ with iotests.FilePath('t.img') as disk_path, \
      iotests.img_info_log(remote_path)
 -    md5_key = subprocess.check_output(
 -        'ssh-keyscan -t rsa 127.0.0.1 2>/dev/null | grep -v "\\^#" | ' +
 -        'cut -d" " -f3 | base64 -d | md5sum -b | cut -d" " -f1',
 -        shell=True).rstrip().decode('ascii')
 +    keys = subprocess.check_output(
 +        'ssh-keyscan 127.0.0.1 2>/dev/null | grep -v "\\^#" | ' +
 +        'cut -d" " -f3',
 +        shell=True).rstrip().decode('ascii').split('\n')
 +
 +    # Mappings of base64 representations to digests
 +    md5_keys = {}
 +    sha1_keys = {}
 +
 +    for key in keys:
 +        md5_keys[key] = subprocess.check_output(
 +            'echo %s | base64 -d | md5sum -b | cut -d" " -f1' % key,
 +            shell=True).rstrip().decode('ascii')
 +
 +        sha1_keys[key] = subprocess.check_output(
 +            'echo %s | base64 -d | sha1sum -b | cut -d" " -f1' % key,
 +            shell=True).rstrip().decode('ascii')
      vm.launch()
 +
 +    # Find correct key first
 +    matching_key = None
 +    for key in keys:
 +        result = vm.qmp('blockdev-add',
 +                        driver='ssh', node_name='node0', path=disk_path,
 +                        server={
 +                             'host': '127.0.0.1',
 +                             'port': '22',
 +                        }, host_key_check={
 +                             'mode': 'hash',
 +                             'type': 'md5',
 +                             'hash': md5_keys[key],
 +                        })
 +
 +        if 'error' not in result:
 +            vm.qmp('blockdev-del', node_name='node0')
 +            matching_key = key
 +            break
 +
 +    if matching_key is None:
 +        vm.shutdown()
 +        iotests.notrun('Did not find a key that fits 127.0.0.1')
 +
      blockdev_create(vm, { 'driver': 'ssh',
                            'location': {
                                'path': disk_path,
@@ -XXX,XX +XXX,XX @@ with iotests.FilePath('t.img') as disk_path, \
                                'host-key-check': {
                                    'mode': 'hash',
                                    'type': 'md5',
 -                                  'hash': md5_key,
 +                                  'hash': md5_keys[matching_key],
                                }
                            },
                            'size': 8388608 })
@@ -XXX,XX +XXX,XX @@ with iotests.FilePath('t.img') as disk_path, \
      iotests.img_info_log(remote_path)
 -    sha1_key = subprocess.check_output(
 -        'ssh-keyscan -t rsa 127.0.0.1 2>/dev/null | grep -v "\\^#" | ' +
 -        'cut -d" " -f3 | base64 -d | sha1sum -b | cut -d" " -f1',
 -        shell=True).rstrip().decode('ascii')
 -
      vm.launch()
      blockdev_create(vm, { 'driver': 'ssh',
                            'location': {
@@ -XXX,XX +XXX,XX @@ with iotests.FilePath('t.img') as disk_path, \
                                'host-key-check': {
                                    'mode': 'hash',
                                    'type': 'sha1',
 -                                  'hash': sha1_key,
 +                                  'hash': sha1_keys[matching_key],
                                }
                            },
                            'size': 4194304 })
 diff --git a/tests/qemu-iotests/207.out b/tests/qemu-iotests/207.out
 index XXXXXXX..XXXXXXX 100644
 --- a/tests/qemu-iotests/207.out
 +++ b/tests/qemu-iotests/207.out
@@ -XXX,XX +XXX,XX @@ virtual size: 4 MiB (4194304 bytes)
  {"execute": "blockdev-create", "arguments": {"job-id": "job0", "options": {"driver": "ssh", "location": {"host-key-check": {"mode": "none"}, "path": "/this/is/not/an/existing/path", "server": {"host": "127.0.0.1", "port": "22"}}, "size": 4194304}}}
  {"return": {}}
 -Job failed: failed to open remote file '/this/is/not/an/existing/path': Failed opening remote file (libssh2 error code: -31)
 +Job failed: failed to open remote file '/this/is/not/an/existing/path': SFTP server: No such file (libssh error code: 1, sftp error code: 2)
  {"execute": "job-dismiss", "arguments": {"id": "job0"}}
  {"return": {}}
 --
-.35.1
+.21.0

-New patch
+[Qemu-devel] [PULL v2 8/8] iotests: Fix 205 for concurrent runs
+Tests should place their files into the test directory.  This includes
+Unix sockets.  205 currently fails to do so, which prevents it from
+being run concurrently.
+Signed-off-by: Max Reitz <mreitz@redhat.com>
+Message-id: 20190618210238.9524-1-mreitz@redhat.com
+Reviewed-by: Eric Blake <eblake@redhat.com>
+Signed-off-by: Max Reitz <mreitz@redhat.com>
+---
+ tests/qemu-iotests/205 | 2 +-
+file changed, 1 insertion(+), 1 deletion(-)
+diff --git a/tests/qemu-iotests/205 b/tests/qemu-iotests/205
+index XXXXXXX..XXXXXXX 100755
+--- a/tests/qemu-iotests/205
++++ b/tests/qemu-iotests/205
+@@ -XXX,XX +XXX,XX @@ import iotests
+ import time
+ from iotests import qemu_img_create, qemu_io, filter_qemu_io, QemuIoInteractive
+-nbd_sock = 'nbd_sock'
++nbd_sock = os.path.join(iotests.test_dir, 'nbd_sock')
+ nbd_uri = 'nbd+unix:///exp?socket=' + nbd_sock
+ disk = os.path.join(iotests.test_dir, 'disk')
+--
+.21.0

The following changes since commit 9cf289af47bcfae5c75de37d8e5d6fd23705322c:

Merge tag 'qga-pull-request' of gitlab.com:marcandre.lureau/qemu into staging (2022-05-04 03:42:49 -0700)

are available in the Git repository at:

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

for you to fetch changes up to bef2e050d6a7feb865854c65570c496ac5a8cf53:

util/event-loop-base: Introduce options to set the thread pool size (2022-05-04 17:02:19 +0100)

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

Add new thread-pool-min/thread-pool-max parameters to control the thread pool
used for async I/O.

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

Nicolas Saenz Julienne (3):
  Introduce event-loop-base abstract class
  util/main-loop: Introduce the main loop into QOM
  util/event-loop-base: Introduce options to set the thread pool size

-- 
2.35.1

From: Nicolas Saenz Julienne <nsaenzju@redhat.com>

Introduce the 'event-loop-base' abstract class, it'll hold the
properties common to all event loops and provide the necessary hooks for
their creation and maintenance. Then have iothread inherit from it.

EventLoopBaseClass is defined as user creatable and provides a hook for
its children to attach themselves to the user creatable class 'complete'
function. It also provides an update_params() callback to propagate
property changes onto its children.

The new 'event-loop-base' class will live in the root directory. It is
built on its own using the 'link_whole' option (there are no direct
function dependencies between the class and its children, it all happens
trough 'constructor' magic). And also imposes new compilation
dependencies:

qom <- event-loop-base <- blockdev (iothread.c)

And in subsequent patches:

qom <- event-loop-base <- qemuutil (util/main-loop.c)

All this forced some amount of reordering in meson.build:

- Moved qom build definition before qemuutil. Doing it the other way
   around (i.e. moving qemuutil after qom) isn't possible as a lot of
   core libraries that live in between the two depend on it.

- Process the 'hw' subdir earlier, as it introduces files into the
   'qom' source set.

No functional changes intended.

Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Message-id: 20220425075723.20019-2-nsaenzju@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 qapi/qom.json                    |  22 +++++--
 meson.build                      |  23 ++++---
 include/sysemu/event-loop-base.h |  36 +++++++++++
 include/sysemu/iothread.h        |   6 +-
 event-loop-base.c                | 104 +++++++++++++++++++++++++++++++
 iothread.c                       |  65 ++++++-------------
 6 files changed, 192 insertions(+), 64 deletions(-)
 create mode 100644 include/sysemu/event-loop-base.h
 create mode 100644 event-loop-base.c

diff --git a/qapi/qom.json b/qapi/qom.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/qom.json
+++ b/qapi/qom.json
@@ -XXX,XX +XXX,XX @@
             '*repeat': 'bool',
             '*grab-toggle': 'GrabToggleKeys' } }
 
+##
+# @EventLoopBaseProperties:
+#
+# Common properties for event loops
+#
+# @aio-max-batch: maximum number of requests in a batch for the AIO engine,
+#                 0 means that the engine will use its default.
+#                 (default: 0)
+#
+# Since: 7.1
+##
+{ 'struct': 'EventLoopBaseProperties',
+  'data': { '*aio-max-batch': 'int' } }
+
 ##
 # @IothreadProperties:
 #
@@ -XXX,XX +XXX,XX @@
 #               algorithm detects it is spending too long polling without
 #               encountering events. 0 selects a default behaviour (default: 0)
 #
-# @aio-max-batch: maximum number of requests in a batch for the AIO engine,
-#                 0 means that the engine will use its default
-#                 (default:0, since 6.1)
+# The @aio-max-batch option is available since 6.1.
 #
 # Since: 2.0
 ##
 { 'struct': 'IothreadProperties',
+  'base': 'EventLoopBaseProperties',
   'data': { '*poll-max-ns': 'int',
             '*poll-grow': 'int',
-            '*poll-shrink': 'int',
-            '*aio-max-batch': 'int' } }
+            '*poll-shrink': 'int' } }
 
 ##
 # @MemoryBackendProperties:
diff --git a/meson.build b/meson.build
index XXXXXXX..XXXXXXX 100644
--- a/meson.build
+++ b/meson.build
@@ -XXX,XX +XXX,XX @@ subdir('qom')
 subdir('authz')
 subdir('crypto')
 subdir('ui')
+subdir('hw')
 
 
 if enable_modules
@@ -XXX,XX +XXX,XX @@ if enable_modules
   modulecommon = declare_dependency(link_whole: libmodulecommon, compile_args: '-DBUILD_DSO')
 endif
 
+qom_ss = qom_ss.apply(config_host, strict: false)
+libqom = static_library('qom', qom_ss.sources() + genh,
+                        dependencies: [qom_ss.dependencies()],
+                        name_suffix: 'fa')
+qom = declare_dependency(link_whole: libqom)
+
+event_loop_base = files('event-loop-base.c')
+event_loop_base = static_library('event-loop-base', sources: event_loop_base + genh,
+                                 build_by_default: true)
+event_loop_base = declare_dependency(link_whole: event_loop_base,
+                                     dependencies: [qom])
+
 stub_ss = stub_ss.apply(config_all, strict: false)
 
 util_ss.add_all(trace_ss)
@@ -XXX,XX +XXX,XX @@ subdir('monitor')
 subdir('net')
 subdir('replay')
 subdir('semihosting')
-subdir('hw')
 subdir('tcg')
 subdir('fpu')
 subdir('accel')
@@ -XXX,XX +XXX,XX @@ qemu_syms = custom_target('qemu.syms', output: 'qemu.syms',
                              capture: true,
                              command: [undefsym, nm, '@INPUT@'])
 
-qom_ss = qom_ss.apply(config_host, strict: false)
-libqom = static_library('qom', qom_ss.sources() + genh,
-                        dependencies: [qom_ss.dependencies()],
-                        name_suffix: 'fa')
-
-qom = declare_dependency(link_whole: libqom)
-
 authz_ss = authz_ss.apply(config_host, strict: false)
 libauthz = static_library('authz', authz_ss.sources() + genh,
                           dependencies: [authz_ss.dependencies()],
@@ -XXX,XX +XXX,XX @@ libblockdev = static_library('blockdev', blockdev_ss.sources() + genh,
                              build_by_default: false)
 
 blockdev = declare_dependency(link_whole: [libblockdev],
-                              dependencies: [block])
+                              dependencies: [block, event_loop_base])
 
 qmp_ss = qmp_ss.apply(config_host, strict: false)
 libqmp = static_library('qmp', qmp_ss.sources() + genh,
diff --git a/include/sysemu/event-loop-base.h b/include/sysemu/event-loop-base.h
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/include/sysemu/event-loop-base.h
@@ -XXX,XX +XXX,XX @@
+/*
+ * QEMU event-loop backend
+ *
+ * Copyright (C) 2022 Red Hat Inc
+ *
+ * Authors:
+ *  Nicolas Saenz Julienne <nsaenzju@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 QEMU_EVENT_LOOP_BASE_H
+#define QEMU_EVENT_LOOP_BASE_H
+
+#include "qom/object.h"
+#include "block/aio.h"
+#include "qemu/typedefs.h"
+
+#define TYPE_EVENT_LOOP_BASE         "event-loop-base"
+OBJECT_DECLARE_TYPE(EventLoopBase, EventLoopBaseClass,
+                    EVENT_LOOP_BASE)
+
+struct EventLoopBaseClass {
+    ObjectClass parent_class;
+
+    void (*init)(EventLoopBase *base, Error **errp);
+    void (*update_params)(EventLoopBase *base, Error **errp);
+};
+
+struct EventLoopBase {
+    Object parent;
+
+    /* AioContext AIO engine parameters */
+    int64_t aio_max_batch;
+};
+#endif
diff --git a/include/sysemu/iothread.h b/include/sysemu/iothread.h
index XXXXXXX..XXXXXXX 100644
--- a/include/sysemu/iothread.h
+++ b/include/sysemu/iothread.h
@@ -XXX,XX +XXX,XX @@
 #include "block/aio.h"
 #include "qemu/thread.h"
 #include "qom/object.h"
+#include "sysemu/event-loop-base.h"
 
 #define TYPE_IOTHREAD "iothread"
 
 struct IOThread {
-    Object parent_obj;
+    EventLoopBase parent_obj;
 
     QemuThread thread;
     AioContext *ctx;
@@ -XXX,XX +XXX,XX @@ struct IOThread {
     int64_t poll_max_ns;
     int64_t poll_grow;
     int64_t poll_shrink;
-
-    /* AioContext AIO engine parameters */
-    int64_t aio_max_batch;
 };
 typedef struct IOThread IOThread;
 
diff --git a/event-loop-base.c b/event-loop-base.c
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/event-loop-base.c
@@ -XXX,XX +XXX,XX @@
+/*
+ * QEMU event-loop base
+ *
+ * Copyright (C) 2022 Red Hat Inc
+ *
+ * Authors:
+ *  Stefan Hajnoczi <stefanha@redhat.com>
+ *  Nicolas Saenz Julienne <nsaenzju@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 "qom/object_interfaces.h"
+#include "qapi/error.h"
+#include "sysemu/event-loop-base.h"
+
+typedef struct {
+    const char *name;
+    ptrdiff_t offset; /* field's byte offset in EventLoopBase struct */
+} EventLoopBaseParamInfo;
+
+static EventLoopBaseParamInfo aio_max_batch_info = {
+    "aio-max-batch", offsetof(EventLoopBase, aio_max_batch),
+};
+
+static void event_loop_base_get_param(Object *obj, Visitor *v,
+        const char *name, void *opaque, Error **errp)
+{
+    EventLoopBase *event_loop_base = EVENT_LOOP_BASE(obj);
+    EventLoopBaseParamInfo *info = opaque;
+    int64_t *field = (void *)event_loop_base + info->offset;
+
+    visit_type_int64(v, name, field, errp);
+}
+
+static void event_loop_base_set_param(Object *obj, Visitor *v,
+        const char *name, void *opaque, Error **errp)
+{
+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_GET_CLASS(obj);
+    EventLoopBase *base = EVENT_LOOP_BASE(obj);
+    EventLoopBaseParamInfo *info = opaque;
+    int64_t *field = (void *)base + info->offset;
+    int64_t value;
+
+    if (!visit_type_int64(v, name, &value, errp)) {
+        return;
+    }
+
+    if (value < 0) {
+        error_setg(errp, "%s value must be in range [0, %" PRId64 "]",
+                   info->name, INT64_MAX);
+        return;
+    }
+
+    *field = value;
+
+    if (bc->update_params) {
+        bc->update_params(base, errp);
+    }
+
+    return;
+}
+
+static void event_loop_base_complete(UserCreatable *uc, Error **errp)
+{
+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_GET_CLASS(uc);
+    EventLoopBase *base = EVENT_LOOP_BASE(uc);
+
+    if (bc->init) {
+        bc->init(base, errp);
+    }
+}
+
+static void event_loop_base_class_init(ObjectClass *klass, void *class_data)
+{
+    UserCreatableClass *ucc = USER_CREATABLE_CLASS(klass);
+    ucc->complete = event_loop_base_complete;
+
+    object_class_property_add(klass, "aio-max-batch", "int",
+                              event_loop_base_get_param,
+                              event_loop_base_set_param,
+                              NULL, &aio_max_batch_info);
+}
+
+static const TypeInfo event_loop_base_info = {
+    .name = TYPE_EVENT_LOOP_BASE,
+    .parent = TYPE_OBJECT,
+    .instance_size = sizeof(EventLoopBase),
+    .class_size = sizeof(EventLoopBaseClass),
+    .class_init = event_loop_base_class_init,
+    .abstract = true,
+    .interfaces = (InterfaceInfo[]) {
+        { TYPE_USER_CREATABLE },
+        { }
+    }
+};
+
+static void register_types(void)
+{
+    type_register_static(&event_loop_base_info);
+}
+type_init(register_types);
diff --git a/iothread.c b/iothread.c
index XXXXXXX..XXXXXXX 100644
--- a/iothread.c
+++ b/iothread.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/module.h"
 #include "block/aio.h"
 #include "block/block.h"
+#include "sysemu/event-loop-base.h"
 #include "sysemu/iothread.h"
 #include "qapi/error.h"
 #include "qapi/qapi-commands-misc.h"
@@ -XXX,XX +XXX,XX @@ static void iothread_init_gcontext(IOThread *iothread)
     iothread->main_loop = g_main_loop_new(iothread->worker_context, TRUE);
 }
 
-static void iothread_set_aio_context_params(IOThread *iothread, Error **errp)
+static void iothread_set_aio_context_params(EventLoopBase *base, Error **errp)
 {
+    IOThread *iothread = IOTHREAD(base);
     ERRP_GUARD();
 
+    if (!iothread->ctx) {
+        return;
+    }
+
     aio_context_set_poll_params(iothread->ctx,
                                 iothread->poll_max_ns,
                                 iothread->poll_grow,
@@ -XXX,XX +XXX,XX @@ static void iothread_set_aio_context_params(IOThread *iothread, Error **errp)
     }
 
     aio_context_set_aio_params(iothread->ctx,
-                               iothread->aio_max_batch,
+                               iothread->parent_obj.aio_max_batch,
                                errp);
 }
 
-static void iothread_complete(UserCreatable *obj, Error **errp)
+
+static void iothread_init(EventLoopBase *base, Error **errp)
 {
     Error *local_error = NULL;
-    IOThread *iothread = IOTHREAD(obj);
+    IOThread *iothread = IOTHREAD(base);
     char *thread_name;
 
     iothread->stopping = false;
@@ -XXX,XX +XXX,XX @@ static void iothread_complete(UserCreatable *obj, Error **errp)
      */
     iothread_init_gcontext(iothread);
 
-    iothread_set_aio_context_params(iothread, &local_error);
+    iothread_set_aio_context_params(base, &local_error);
     if (local_error) {
         error_propagate(errp, local_error);
         aio_context_unref(iothread->ctx);
@@ -XXX,XX +XXX,XX @@ static void iothread_complete(UserCreatable *obj, Error **errp)
      * to inherit.
      */
     thread_name = g_strdup_printf("IO %s",
-                        object_get_canonical_path_component(OBJECT(obj)));
+                        object_get_canonical_path_component(OBJECT(base)));
     qemu_thread_create(&iothread->thread, thread_name, iothread_run,
                        iothread, QEMU_THREAD_JOINABLE);
     g_free(thread_name);
@@ -XXX,XX +XXX,XX @@ static IOThreadParamInfo poll_grow_info = {
 static IOThreadParamInfo poll_shrink_info = {
     "poll-shrink", offsetof(IOThread, poll_shrink),
 };
-static IOThreadParamInfo aio_max_batch_info = {
-    "aio-max-batch", offsetof(IOThread, aio_max_batch),
-};
 
 static void iothread_get_param(Object *obj, Visitor *v,
         const char *name, IOThreadParamInfo *info, Error **errp)
@@ -XXX,XX +XXX,XX @@ static void iothread_set_poll_param(Object *obj, Visitor *v,
     }
 }
 
-static void iothread_get_aio_param(Object *obj, Visitor *v,
-        const char *name, void *opaque, Error **errp)
-{
-    IOThreadParamInfo *info = opaque;
-
-    iothread_get_param(obj, v, name, info, errp);
-}
-
-static void iothread_set_aio_param(Object *obj, Visitor *v,
-        const char *name, void *opaque, Error **errp)
-{
-    IOThread *iothread = IOTHREAD(obj);
-    IOThreadParamInfo *info = opaque;
-
-    if (!iothread_set_param(obj, v, name, info, errp)) {
-        return;
-    }
-
-    if (iothread->ctx) {
-        aio_context_set_aio_params(iothread->ctx,
-                                   iothread->aio_max_batch,
-                                   errp);
-    }
-}
-
 static void iothread_class_init(ObjectClass *klass, void *class_data)
 {
-    UserCreatableClass *ucc = USER_CREATABLE_CLASS(klass);
-    ucc->complete = iothread_complete;
+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_CLASS(klass);
+
+    bc->init = iothread_init;
+    bc->update_params = iothread_set_aio_context_params;
 
     object_class_property_add(klass, "poll-max-ns", "int",
                               iothread_get_poll_param,
@@ -XXX,XX +XXX,XX @@ static void iothread_class_init(ObjectClass *klass, void *class_data)
                               iothread_get_poll_param,
                               iothread_set_poll_param,
                               NULL, &poll_shrink_info);
-    object_class_property_add(klass, "aio-max-batch", "int",
-                              iothread_get_aio_param,
-                              iothread_set_aio_param,
-                              NULL, &aio_max_batch_info);
 }
 
 static const TypeInfo iothread_info = {
     .name = TYPE_IOTHREAD,
-    .parent = TYPE_OBJECT,
+    .parent = TYPE_EVENT_LOOP_BASE,
     .class_init = iothread_class_init,
     .instance_size = sizeof(IOThread),
     .instance_init = iothread_instance_init,
     .instance_finalize = iothread_instance_finalize,
-    .interfaces = (InterfaceInfo[]) {
-        {TYPE_USER_CREATABLE},
-        {}
-    },
 };
 
 static void iothread_register_types(void)
@@ -XXX,XX +XXX,XX @@ static int query_one_iothread(Object *object, void *opaque)
     info->poll_max_ns = iothread->poll_max_ns;
     info->poll_grow = iothread->poll_grow;
     info->poll_shrink = iothread->poll_shrink;
-    info->aio_max_batch = iothread->aio_max_batch;
+    info->aio_max_batch = iothread->parent_obj.aio_max_batch;
 
     QAPI_LIST_APPEND(*tail, info);
     return 0;
-- 
2.35.1

From: Nicolas Saenz Julienne <nsaenzju@redhat.com>

'event-loop-base' provides basic property handling for all 'AioContext'
based event loops. So let's define a new 'MainLoopClass' that inherits
from it. This will permit tweaking the main loop's properties through
qapi as well as through the command line using the '-object' keyword[1].
Only one instance of 'MainLoopClass' might be created at any time.

'EventLoopBaseClass' learns a new callback, 'can_be_deleted()' so as to
mark 'MainLoop' as non-deletable.

[1] For example:
      -object main-loop,id=main-loop,aio-max-batch=<value>

Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Message-id: 20220425075723.20019-3-nsaenzju@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 qapi/qom.json                    | 13 ++++++++
 meson.build                      |  3 +-
 include/qemu/main-loop.h         | 10 ++++++
 include/sysemu/event-loop-base.h |  1 +
 event-loop-base.c                | 13 ++++++++
 util/main-loop.c                 | 56 ++++++++++++++++++++++++++++++++
 6 files changed, 95 insertions(+), 1 deletion(-)

diff --git a/qapi/qom.json b/qapi/qom.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/qom.json
+++ b/qapi/qom.json
@@ -XXX,XX +XXX,XX @@
             '*poll-grow': 'int',
             '*poll-shrink': 'int' } }
 
+##
+# @MainLoopProperties:
+#
+# Properties for the main-loop object.
+#
+# Since: 7.1
+##
+{ 'struct': 'MainLoopProperties',
+  'base': 'EventLoopBaseProperties',
+  'data': {} }
+
 ##
 # @MemoryBackendProperties:
 #
@@ -XXX,XX +XXX,XX @@
     { 'name': 'input-linux',
       'if': 'CONFIG_LINUX' },
     'iothread',
+    'main-loop',
     { 'name': 'memory-backend-epc',
       'if': 'CONFIG_LINUX' },
     'memory-backend-file',
@@ -XXX,XX +XXX,XX @@
       'input-linux':                { 'type': 'InputLinuxProperties',
                                       'if': 'CONFIG_LINUX' },
       'iothread':                   'IothreadProperties',
+      'main-loop':                  'MainLoopProperties',
       'memory-backend-epc':         { 'type': 'MemoryBackendEpcProperties',
                                       'if': 'CONFIG_LINUX' },
       'memory-backend-file':        'MemoryBackendFileProperties',
diff --git a/meson.build b/meson.build
index XXXXXXX..XXXXXXX 100644
--- a/meson.build
+++ b/meson.build
@@ -XXX,XX +XXX,XX @@ libqemuutil = static_library('qemuutil',
                              sources: util_ss.sources() + stub_ss.sources() + genh,
                              dependencies: [util_ss.dependencies(), libm, threads, glib, socket, malloc, pixman])
 qemuutil = declare_dependency(link_with: libqemuutil,
-                              sources: genh + version_res)
+                              sources: genh + version_res,
+                              dependencies: [event_loop_base])
 
 if have_system or have_user
   decodetree = generator(find_program('scripts/decodetree.py'),
diff --git a/include/qemu/main-loop.h b/include/qemu/main-loop.h
index XXXXXXX..XXXXXXX 100644
--- a/include/qemu/main-loop.h
+++ b/include/qemu/main-loop.h
@@ -XXX,XX +XXX,XX @@
 #define QEMU_MAIN_LOOP_H
 
 #include "block/aio.h"
+#include "qom/object.h"
+#include "sysemu/event-loop-base.h"
 
 #define SIG_IPI SIGUSR1
 
+#define TYPE_MAIN_LOOP  "main-loop"
+OBJECT_DECLARE_TYPE(MainLoop, MainLoopClass, MAIN_LOOP)
+
+struct MainLoop {
+    EventLoopBase parent_obj;
+};
+typedef struct MainLoop MainLoop;
+
 /**
  * qemu_init_main_loop: Set up the process so that it can run the main loop.
  *
diff --git a/include/sysemu/event-loop-base.h b/include/sysemu/event-loop-base.h
index XXXXXXX..XXXXXXX 100644
--- a/include/sysemu/event-loop-base.h
+++ b/include/sysemu/event-loop-base.h
@@ -XXX,XX +XXX,XX @@ struct EventLoopBaseClass {
 
     void (*init)(EventLoopBase *base, Error **errp);
     void (*update_params)(EventLoopBase *base, Error **errp);
+    bool (*can_be_deleted)(EventLoopBase *base);
 };
 
 struct EventLoopBase {
diff --git a/event-loop-base.c b/event-loop-base.c
index XXXXXXX..XXXXXXX 100644
--- a/event-loop-base.c
+++ b/event-loop-base.c
@@ -XXX,XX +XXX,XX @@ static void event_loop_base_complete(UserCreatable *uc, Error **errp)
     }
 }
 
+static bool event_loop_base_can_be_deleted(UserCreatable *uc)
+{
+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_GET_CLASS(uc);
+    EventLoopBase *backend = EVENT_LOOP_BASE(uc);
+
+    if (bc->can_be_deleted) {
+        return bc->can_be_deleted(backend);
+    }
+
+    return true;
+}
+
 static void event_loop_base_class_init(ObjectClass *klass, void *class_data)
 {
     UserCreatableClass *ucc = USER_CREATABLE_CLASS(klass);
     ucc->complete = event_loop_base_complete;
+    ucc->can_be_deleted = event_loop_base_can_be_deleted;
 
     object_class_property_add(klass, "aio-max-batch", "int",
                               event_loop_base_get_param,
diff --git a/util/main-loop.c b/util/main-loop.c
index XXXXXXX..XXXXXXX 100644
--- a/util/main-loop.c
+++ b/util/main-loop.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/error-report.h"
 #include "qemu/queue.h"
 #include "qemu/compiler.h"
+#include "qom/object.h"
 
 #ifndef _WIN32
 #include <sys/wait.h>
@@ -XXX,XX +XXX,XX @@ int qemu_init_main_loop(Error **errp)
     return 0;
 }
 
+static void main_loop_update_params(EventLoopBase *base, Error **errp)
+{
+    if (!qemu_aio_context) {
+        error_setg(errp, "qemu aio context not ready");
+        return;
+    }
+
+    aio_context_set_aio_params(qemu_aio_context, base->aio_max_batch, errp);
+}
+
+MainLoop *mloop;
+
+static void main_loop_init(EventLoopBase *base, Error **errp)
+{
+    MainLoop *m = MAIN_LOOP(base);
+
+    if (mloop) {
+        error_setg(errp, "only one main-loop instance allowed");
+        return;
+    }
+
+    main_loop_update_params(base, errp);
+
+    mloop = m;
+    return;
+}
+
+static bool main_loop_can_be_deleted(EventLoopBase *base)
+{
+    return false;
+}
+
+static void main_loop_class_init(ObjectClass *oc, void *class_data)
+{
+    EventLoopBaseClass *bc = EVENT_LOOP_BASE_CLASS(oc);
+
+    bc->init = main_loop_init;
+    bc->update_params = main_loop_update_params;
+    bc->can_be_deleted = main_loop_can_be_deleted;
+}
+
+static const TypeInfo main_loop_info = {
+    .name = TYPE_MAIN_LOOP,
+    .parent = TYPE_EVENT_LOOP_BASE,
+    .class_init = main_loop_class_init,
+    .instance_size = sizeof(MainLoop),
+};
+
+static void main_loop_register_types(void)
+{
+    type_register_static(&main_loop_info);
+}
+
+type_init(main_loop_register_types)
+
 static int max_priority;
 
 #ifndef _WIN32
-- 
2.35.1

From: Nicolas Saenz Julienne <nsaenzju@redhat.com>

The thread pool regulates itself: when idle, it kills threads until
empty, when in demand, it creates new threads until full. This behaviour
doesn't play well with latency sensitive workloads where the price of
creating a new thread is too high. For example, when paired with qemu's
'-mlock', or using safety features like SafeStack, creating a new thread
has been measured take multiple milliseconds.

In order to mitigate this let's introduce a new 'EventLoopBase'
property to set the thread pool size. The threads will be created during
the pool's initialization or upon updating the property's value, remain
available during its lifetime regardless of demand, and destroyed upon
freeing it. A properly characterized workload will then be able to
configure the pool to avoid any latency spikes.

Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Message-id: 20220425075723.20019-4-nsaenzju@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 qapi/qom.json                    | 10 +++++-
 include/block/aio.h              | 10 ++++++
 include/block/thread-pool.h      |  3 ++
 include/sysemu/event-loop-base.h |  4 +++
 event-loop-base.c                | 23 +++++++++++++
 iothread.c                       |  3 ++
 util/aio-posix.c                 |  1 +
 util/async.c                     | 20 ++++++++++++
 util/main-loop.c                 |  9 ++++++
 util/thread-pool.c               | 55 +++++++++++++++++++++++++++++---
 10 files changed, 133 insertions(+), 5 deletions(-)

diff --git a/qapi/qom.json b/qapi/qom.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/qom.json
+++ b/qapi/qom.json
@@ -XXX,XX +XXX,XX @@
 #                 0 means that the engine will use its default.
 #                 (default: 0)
 #
+# @thread-pool-min: minimum number of threads reserved in the thread pool
+#                   (default:0)
+#
+# @thread-pool-max: maximum number of threads the thread pool can contain
+#                   (default:64)
+#
 # Since: 7.1
 ##
 { 'struct': 'EventLoopBaseProperties',
-  'data': { '*aio-max-batch': 'int' } }
+  'data': { '*aio-max-batch': 'int',
+            '*thread-pool-min': 'int',
+            '*thread-pool-max': 'int' } }
 
 ##
 # @IothreadProperties:
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 @@ struct AioContext {
     QSLIST_HEAD(, Coroutine) scheduled_coroutines;
     QEMUBH *co_schedule_bh;
 
+    int thread_pool_min;
+    int thread_pool_max;
     /* Thread pool for performing work and receiving completion callbacks.
      * Has its own locking.
      */
@@ -XXX,XX +XXX,XX @@ void aio_context_set_poll_params(AioContext *ctx, int64_t max_ns,
 void aio_context_set_aio_params(AioContext *ctx, int64_t max_batch,
                                 Error **errp);
 
+/**
+ * aio_context_set_thread_pool_params:
+ * @ctx: the aio context
+ * @min: min number of threads to have readily available in the thread pool
+ * @min: max number of threads the thread pool can contain
+ */
+void aio_context_set_thread_pool_params(AioContext *ctx, int64_t min,
+                                        int64_t max, Error **errp);
 #endif
diff --git a/include/block/thread-pool.h b/include/block/thread-pool.h
index XXXXXXX..XXXXXXX 100644
--- a/include/block/thread-pool.h
+++ b/include/block/thread-pool.h
@@ -XXX,XX +XXX,XX @@
 
 #include "block/block.h"
 
+#define THREAD_POOL_MAX_THREADS_DEFAULT         64
+
 typedef int ThreadPoolFunc(void *opaque);
 
 typedef struct ThreadPool ThreadPool;
@@ -XXX,XX +XXX,XX @@ BlockAIOCB *thread_pool_submit_aio(ThreadPool *pool,
 int coroutine_fn thread_pool_submit_co(ThreadPool *pool,
         ThreadPoolFunc *func, void *arg);
 void thread_pool_submit(ThreadPool *pool, ThreadPoolFunc *func, void *arg);
+void thread_pool_update_params(ThreadPool *pool, struct AioContext *ctx);
 
 #endif
diff --git a/include/sysemu/event-loop-base.h b/include/sysemu/event-loop-base.h
index XXXXXXX..XXXXXXX 100644
--- a/include/sysemu/event-loop-base.h
+++ b/include/sysemu/event-loop-base.h
@@ -XXX,XX +XXX,XX @@ struct EventLoopBase {
 
     /* AioContext AIO engine parameters */
     int64_t aio_max_batch;
+
+    /* AioContext thread pool parameters */
+    int64_t thread_pool_min;
+    int64_t thread_pool_max;
 };
 #endif
diff --git a/event-loop-base.c b/event-loop-base.c
index XXXXXXX..XXXXXXX 100644
--- a/event-loop-base.c
+++ b/event-loop-base.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/osdep.h"
 #include "qom/object_interfaces.h"
 #include "qapi/error.h"
+#include "block/thread-pool.h"
 #include "sysemu/event-loop-base.h"
 
 typedef struct {
@@ -XXX,XX +XXX,XX @@ typedef struct {
     ptrdiff_t offset; /* field's byte offset in EventLoopBase struct */
 } EventLoopBaseParamInfo;
 
+static void event_loop_base_instance_init(Object *obj)
+{
+    EventLoopBase *base = EVENT_LOOP_BASE(obj);
+
+    base->thread_pool_max = THREAD_POOL_MAX_THREADS_DEFAULT;
+}
+
 static EventLoopBaseParamInfo aio_max_batch_info = {
     "aio-max-batch", offsetof(EventLoopBase, aio_max_batch),
 };
+static EventLoopBaseParamInfo thread_pool_min_info = {
+    "thread-pool-min", offsetof(EventLoopBase, thread_pool_min),
+};
+static EventLoopBaseParamInfo thread_pool_max_info = {
+    "thread-pool-max", offsetof(EventLoopBase, thread_pool_max),
+};
 
 static void event_loop_base_get_param(Object *obj, Visitor *v,
         const char *name, void *opaque, Error **errp)
@@ -XXX,XX +XXX,XX @@ static void event_loop_base_class_init(ObjectClass *klass, void *class_data)
                               event_loop_base_get_param,
                               event_loop_base_set_param,
                               NULL, &aio_max_batch_info);
+    object_class_property_add(klass, "thread-pool-min", "int",
+                              event_loop_base_get_param,
+                              event_loop_base_set_param,
+                              NULL, &thread_pool_min_info);
+    object_class_property_add(klass, "thread-pool-max", "int",
+                              event_loop_base_get_param,
+                              event_loop_base_set_param,
+                              NULL, &thread_pool_max_info);
 }
 
 static const TypeInfo event_loop_base_info = {
     .name = TYPE_EVENT_LOOP_BASE,
     .parent = TYPE_OBJECT,
     .instance_size = sizeof(EventLoopBase),
+    .instance_init = event_loop_base_instance_init,
     .class_size = sizeof(EventLoopBaseClass),
     .class_init = event_loop_base_class_init,
     .abstract = true,
diff --git a/iothread.c b/iothread.c
index XXXXXXX..XXXXXXX 100644
--- a/iothread.c
+++ b/iothread.c
@@ -XXX,XX +XXX,XX @@ static void iothread_set_aio_context_params(EventLoopBase *base, Error **errp)
     aio_context_set_aio_params(iothread->ctx,
                                iothread->parent_obj.aio_max_batch,
                                errp);
+
+    aio_context_set_thread_pool_params(iothread->ctx, base->thread_pool_min,
+                                       base->thread_pool_max, errp);
 }
 
 
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 @@
 
 #include "qemu/osdep.h"
 #include "block/block.h"
+#include "block/thread-pool.h"
 #include "qemu/main-loop.h"
 #include "qemu/rcu.h"
 #include "qemu/rcu_queue.h"
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 @@ AioContext *aio_context_new(Error **errp)
 
     ctx->aio_max_batch = 0;
 
+    ctx->thread_pool_min = 0;
+    ctx->thread_pool_max = THREAD_POOL_MAX_THREADS_DEFAULT;
+
     return ctx;
 fail:
     g_source_destroy(&ctx->source);
@@ -XXX,XX +XXX,XX @@ void qemu_set_current_aio_context(AioContext *ctx)
     assert(!get_my_aiocontext());
     set_my_aiocontext(ctx);
 }
+
+void aio_context_set_thread_pool_params(AioContext *ctx, int64_t min,
+                                        int64_t max, Error **errp)
+{
+
+    if (min > max || !max || min > INT_MAX || max > INT_MAX) {
+        error_setg(errp, "bad thread-pool-min/thread-pool-max values");
+        return;
+    }
+
+    ctx->thread_pool_min = min;
+    ctx->thread_pool_max = max;
+
+    if (ctx->thread_pool) {
+        thread_pool_update_params(ctx->thread_pool, ctx);
+    }
+}
diff --git a/util/main-loop.c b/util/main-loop.c
index XXXXXXX..XXXXXXX 100644
--- a/util/main-loop.c
+++ b/util/main-loop.c
@@ -XXX,XX +XXX,XX @@
 #include "sysemu/replay.h"
 #include "qemu/main-loop.h"
 #include "block/aio.h"
+#include "block/thread-pool.h"
 #include "qemu/error-report.h"
 #include "qemu/queue.h"
 #include "qemu/compiler.h"
@@ -XXX,XX +XXX,XX @@ int qemu_init_main_loop(Error **errp)
 
 static void main_loop_update_params(EventLoopBase *base, Error **errp)
 {
+    ERRP_GUARD();
+
     if (!qemu_aio_context) {
         error_setg(errp, "qemu aio context not ready");
         return;
     }
 
     aio_context_set_aio_params(qemu_aio_context, base->aio_max_batch, errp);
+    if (*errp) {
+        return;
+    }
+
+    aio_context_set_thread_pool_params(qemu_aio_context, base->thread_pool_min,
+                                       base->thread_pool_max, errp);
 }
 
 MainLoop *mloop;
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 @@ struct ThreadPool {
     QemuMutex lock;
     QemuCond worker_stopped;
     QemuSemaphore sem;
-    int max_threads;
     QEMUBH *new_thread_bh;
 
     /* The following variables are only accessed from one AioContext. */
@@ -XXX,XX +XXX,XX @@ struct ThreadPool {
     int new_threads;     /* backlog of threads we need to create */
     int pending_threads; /* threads created but not running yet */
     bool stopping;
+    int min_threads;
+    int max_threads;
 };
 
+static inline bool back_to_sleep(ThreadPool *pool, int ret)
+{
+    /*
+     * The semaphore timed out, we should exit the loop except when:
+     *  - There is work to do, we raced with the signal.
+     *  - The max threads threshold just changed, we raced with the signal.
+     *  - The thread pool forces a minimum number of readily available threads.
+     */
+    if (ret == -1 && (!QTAILQ_EMPTY(&pool->request_list) ||
+            pool->cur_threads > pool->max_threads ||
+            pool->cur_threads <= pool->min_threads)) {
+            return true;
+    }
+
+    return false;
+}
+
 static void *worker_thread(void *opaque)
 {
     ThreadPool *pool = opaque;
@@ -XXX,XX +XXX,XX @@ static void *worker_thread(void *opaque)
             ret = qemu_sem_timedwait(&pool->sem, 10000);
             qemu_mutex_lock(&pool->lock);
             pool->idle_threads--;
-        } while (ret == -1 && !QTAILQ_EMPTY(&pool->request_list));
-        if (ret == -1 || pool->stopping) {
+        } while (back_to_sleep(pool, ret));
+        if (ret == -1 || pool->stopping ||
+            pool->cur_threads > pool->max_threads) {
             break;
         }
 
@@ -XXX,XX +XXX,XX @@ void thread_pool_submit(ThreadPool *pool, ThreadPoolFunc *func, void *arg)
     thread_pool_submit_aio(pool, func, arg, NULL, NULL);
 }
 
+void thread_pool_update_params(ThreadPool *pool, AioContext *ctx)
+{
+    qemu_mutex_lock(&pool->lock);
+
+    pool->min_threads = ctx->thread_pool_min;
+    pool->max_threads = ctx->thread_pool_max;
+
+    /*
+     * We either have to:
+     *  - Increase the number available of threads until over the min_threads
+     *    threshold.
+     *  - Decrease the number of available threads until under the max_threads
+     *    threshold.
+     *  - Do nothing. The current number of threads fall in between the min and
+     *    max thresholds. We'll let the pool manage itself.
+     */
+    for (int i = pool->cur_threads; i < pool->min_threads; i++) {
+        spawn_thread(pool);
+    }
+
+    for (int i = pool->cur_threads; i > pool->max_threads; i--) {
+        qemu_sem_post(&pool->sem);
+    }
+
+    qemu_mutex_unlock(&pool->lock);
+}
+
 static void thread_pool_init_one(ThreadPool *pool, AioContext *ctx)
 {
     if (!ctx) {
@@ -XXX,XX +XXX,XX @@ static void thread_pool_init_one(ThreadPool *pool, AioContext *ctx)
     qemu_mutex_init(&pool->lock);
     qemu_cond_init(&pool->worker_stopped);
     qemu_sem_init(&pool->sem, 0);
-    pool->max_threads = 64;
     pool->new_thread_bh = aio_bh_new(ctx, spawn_thread_bh_fn, pool);
 
     QLIST_INIT(&pool->head);
     QTAILQ_INIT(&pool->request_list);
+
+    thread_pool_update_params(pool, ctx);
 }
 
 ThreadPool *thread_pool_new(AioContext *ctx)
-- 
2.35.1

The following changes since commit 474f3938d79ab36b9231c9ad3b5a9314c2aeacde:

Merge remote-tracking branch 'remotes/amarkovic/tags/mips-queue-jun-21-2019' into staging (2019-06-21 15:40:50 +0100)

are available in the Git repository at:

https://github.com/XanClic/qemu.git tags/pull-block-2019-06-24

for you to fetch changes up to ab5d4a30f7f3803ca5106b370969c1b7b54136f8:

iotests: Fix 205 for concurrent runs (2019-06-24 16:01:40 +0200)

----------------------------------------------------------------
Block patches:
- The SSH block driver now uses libssh instead of libssh2
- The VMDK block driver gets read-only support for the seSparse
  subformat
- Various fixes

---

v2:
- Squashed Pino's fix for pre-0.8 libssh into the libssh patch

----------------------------------------------------------------
Anton Nefedov (1):
  iotest 134: test cluster-misaligned encrypted write

Klaus Birkelund Jensen (1):
  nvme: do not advertise support for unsupported arbitration mechanism

Max Reitz (1):
  iotests: Fix 205 for concurrent runs

Pino Toscano (1):
  ssh: switch from libssh2 to libssh

Sam Eiderman (3):
  vmdk: Fix comment regarding max l1_size coverage
  vmdk: Reduce the max bound for L1 table size
  vmdk: Add read-only support for seSparse snapshots

Vladimir Sementsov-Ogievskiy (1):
  blockdev: enable non-root nodes for transaction drive-backup source

-- 
2.21.0

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

We forget to enable it for transaction .prepare, while it is already
enabled in do_drive_backup since commit a2d665c1bc362
    "blockdev: loosen restrictions on drive-backup source node"

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20190618140804.59214-1-vsementsov@virtuozzo.com
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 blockdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/blockdev.c b/blockdev.c
index XXXXXXX..XXXXXXX 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -XXX,XX +XXX,XX @@ static void drive_backup_prepare(BlkActionState *common, Error **errp)
     assert(common->action->type == TRANSACTION_ACTION_KIND_DRIVE_BACKUP);
     backup = common->action->u.drive_backup.data;
 
-    bs = qmp_get_root_bs(backup->device, errp);
+    bs = bdrv_lookup_bs(backup->device, backup->device, errp);
     if (!bs) {
         return;
     }
-- 
2.21.0

From: Anton Nefedov <anton.nefedov@virtuozzo.com>

COW (even empty/zero) areas require encryption too

Signed-off-by: Anton Nefedov <anton.nefedov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Message-id: 20190516143028.81155-1-anton.nefedov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 tests/qemu-iotests/134     |  9 +++++++++
 tests/qemu-iotests/134.out | 10 ++++++++++
 2 files changed, 19 insertions(+)

diff --git a/tests/qemu-iotests/134 b/tests/qemu-iotests/134
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/134
+++ b/tests/qemu-iotests/134
@@ -XXX,XX +XXX,XX @@ echo
 echo "== reading whole image =="
 $QEMU_IO --object $SECRET -c "read 0 $size" --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
 
+echo
+echo "== rewriting cluster part =="
+$QEMU_IO --object $SECRET -c "write -P 0xb 512 512" --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
+
+echo
+echo "== verify pattern =="
+$QEMU_IO --object $SECRET -c "read -P 0 0 512"  --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
+$QEMU_IO --object $SECRET -c "read -P 0xb 512 512"  --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
+
 echo
 echo "== rewriting whole image =="
 $QEMU_IO --object $SECRET -c "write -P 0xa 0 $size" --image-opts $IMGSPEC | _filter_qemu_io | _filter_testdir
diff --git a/tests/qemu-iotests/134.out b/tests/qemu-iotests/134.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/134.out
+++ b/tests/qemu-iotests/134.out
@@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 encryption=on encrypt.
 read 134217728/134217728 bytes at offset 0
 128 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
+== rewriting cluster part ==
+wrote 512/512 bytes at offset 512
+512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+== verify pattern ==
+read 512/512 bytes at offset 0
+512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 512/512 bytes at offset 512
+512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
 == rewriting whole image ==
 wrote 134217728/134217728 bytes at offset 0
 128 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-- 
2.21.0

From: Sam Eiderman <shmuel.eiderman@oracle.com>

Commit b0651b8c246d ("vmdk: Move l1_size check into vmdk_add_extent")
extended the l1_size check from VMDK4 to VMDK3 but did not update the
default coverage in the moved comment.

The previous vmdk4 calculation:

(512 * 1024 * 1024) * 512(l2 entries) * 65536(grain) = 16PB

The added vmdk3 calculation:

(512 * 1024 * 1024) * 4096(l2 entries) * 512(grain) = 1PB

Adding the calculation of vmdk3 to the comment.

In any case, VMware does not offer virtual disks more than 2TB for
vmdk4/vmdk3 or 64TB for the new undocumented seSparse format which is
not implemented yet in qemu.

Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Reviewed-by: Eyal Moscovici <eyal.moscovici@oracle.com>
Reviewed-by: Liran Alon <liran.alon@oracle.com>
Reviewed-by: Arbel Moshe <arbel.moshe@oracle.com>
Signed-off-by: Sam Eiderman <shmuel.eiderman@oracle.com>
Message-id: 20190620091057.47441-2-shmuel.eiderman@oracle.com
Reviewed-by: yuchenlin <yuchenlin@synology.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/vmdk.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/block/vmdk.c b/block/vmdk.c
index XXXXXXX..XXXXXXX 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -XXX,XX +XXX,XX @@ static int vmdk_add_extent(BlockDriverState *bs,
         return -EFBIG;
     }
     if (l1_size > 512 * 1024 * 1024) {
-        /* Although with big capacity and small l1_entry_sectors, we can get a
+        /*
+         * Although with big capacity and small l1_entry_sectors, we can get a
          * big l1_size, we don't want unbounded value to allocate the table.
-         * Limit it to 512M, which is 16PB for default cluster and L2 table
-         * size */
+         * Limit it to 512M, which is:
+         *     16PB - for default "Hosted Sparse Extent" (VMDK4)
+         *            cluster size: 64KB, L2 table size: 512 entries
+         *     1PB  - for default "ESXi Host Sparse Extent" (VMDK3/vmfsSparse)
+         *            cluster size: 512B, L2 table size: 4096 entries
+         */
         error_setg(errp, "L1 size too big");
         return -EFBIG;
     }
-- 
2.21.0

From: Sam Eiderman <shmuel.eiderman@oracle.com>

512M of L1 entries is a very loose bound, only 32M are required to store
the maximal supported VMDK file size of 2TB.

Fixed qemu-iotest 59# - now failure occures before on impossible L1
table size.

Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Reviewed-by: Eyal Moscovici <eyal.moscovici@oracle.com>
Reviewed-by: Liran Alon <liran.alon@oracle.com>
Reviewed-by: Arbel Moshe <arbel.moshe@oracle.com>
Signed-off-by: Sam Eiderman <shmuel.eiderman@oracle.com>
Message-id: 20190620091057.47441-3-shmuel.eiderman@oracle.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/vmdk.c               | 13 +++++++------
 tests/qemu-iotests/059.out |  2 +-
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/block/vmdk.c b/block/vmdk.c
index XXXXXXX..XXXXXXX 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -XXX,XX +XXX,XX @@ static int vmdk_add_extent(BlockDriverState *bs,
         error_setg(errp, "Invalid granularity, image may be corrupt");
         return -EFBIG;
     }
-    if (l1_size > 512 * 1024 * 1024) {
+    if (l1_size > 32 * 1024 * 1024) {
         /*
          * Although with big capacity and small l1_entry_sectors, we can get a
          * big l1_size, we don't want unbounded value to allocate the table.
-         * Limit it to 512M, which is:
-         *     16PB - for default "Hosted Sparse Extent" (VMDK4)
-         *            cluster size: 64KB, L2 table size: 512 entries
-         *     1PB  - for default "ESXi Host Sparse Extent" (VMDK3/vmfsSparse)
-         *            cluster size: 512B, L2 table size: 4096 entries
+         * Limit it to 32M, which is enough to store:
+         *     8TB  - for both VMDK3 & VMDK4 with
+         *            minimal cluster size: 512B
+         *            minimal L2 table size: 512 entries
+         *            8 TB is still more than the maximal value supported for
+         *            VMDK3 & VMDK4 which is 2TB.
          */
         error_setg(errp, "L1 size too big");
         return -EFBIG;
diff --git a/tests/qemu-iotests/059.out b/tests/qemu-iotests/059.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/059.out
+++ b/tests/qemu-iotests/059.out
@@ -XXX,XX +XXX,XX @@ Offset          Length          Mapped to       File
 0x140000000     0x10000         0x50000         TEST_DIR/t-s003.vmdk
 
 === Testing afl image with a very large capacity ===
-qemu-img: Can't get image size 'TEST_DIR/afl9.IMGFMT': File too large
+qemu-img: Could not open 'TEST_DIR/afl9.IMGFMT': L1 size too big
 *** done
-- 
2.21.0

From: Sam Eiderman <shmuel.eiderman@oracle.com>

Until ESXi 6.5 VMware used the vmfsSparse format for snapshots (VMDK3 in
QEMU).

This format was lacking in the following:

* Grain directory (L1) and grain table (L2) entries were 32-bit,
      allowing access to only 2TB (slightly less) of data.
    * The grain size (default) was 512 bytes - leading to data
      fragmentation and many grain tables.
    * For space reclamation purposes, it was necessary to find all the
      grains which are not pointed to by any grain table - so a reverse
      mapping of "offset of grain in vmdk" to "grain table" must be
      constructed - which takes large amounts of CPU/RAM.

The format specification can be found in VMware's documentation:
https://www.vmware.com/support/developer/vddk/vmdk_50_technote.pdf

In ESXi 6.5, to support snapshot files larger than 2TB, a new format was
introduced: SESparse (Space Efficient).

This format fixes the above issues:

* All entries are now 64-bit.
    * The grain size (default) is 4KB.
    * Grain directory and grain tables are now located at the beginning
      of the file.
      + seSparse format reserves space for all grain tables.
      + Grain tables can be addressed using an index.
      + Grains are located in the end of the file and can also be
        addressed with an index.
      - seSparse vmdks of large disks (64TB) have huge preallocated
        headers - mainly due to L2 tables, even for empty snapshots.
    * The header contains a reverse mapping ("backmap") of "offset of
      grain in vmdk" to "grain table" and a bitmap ("free bitmap") which
      specifies for each grain - whether it is allocated or not.
      Using these data structures we can implement space reclamation
      efficiently.
    * Due to the fact that the header now maintains two mappings:
        * The regular one (grain directory & grain tables)
        * A reverse one (backmap and free bitmap)
      These data structures can lose consistency upon crash and result
      in a corrupted VMDK.
      Therefore, a journal is also added to the VMDK and is replayed
      when the VMware reopens the file after a crash.

Since ESXi 6.7 - SESparse is the only snapshot format available.

Unfortunately, VMware does not provide documentation regarding the new
seSparse format.

This commit is based on black-box research of the seSparse format.
Various in-guest block operations and their effect on the snapshot file
were tested.

The only VMware provided source of information (regarding the underlying
implementation) was a log file on the ESXi:

/var/log/hostd.log

Whenever an seSparse snapshot is created - the log is being populated
with seSparse records.

Relevant log records are of the form:

[...] Const Header:
[...]  constMagic     = 0xcafebabe
[...]  version        = 2.1
[...]  capacity       = 204800
[...]  grainSize      = 8
[...]  grainTableSize = 64
[...]  flags          = 0
[...] Extents:
[...]  Header         : <1 : 1>
[...]  JournalHdr     : <2 : 2>
[...]  Journal        : <2048 : 2048>
[...]  GrainDirectory : <4096 : 2048>
[...]  GrainTables    : <6144 : 2048>
[...]  FreeBitmap     : <8192 : 2048>
[...]  BackMap        : <10240 : 2048>
[...]  Grain          : <12288 : 204800>
[...] Volatile Header:
[...] volatileMagic     = 0xcafecafe
[...] FreeGTNumber      = 0
[...] nextTxnSeqNumber  = 0
[...] replayJournal     = 0

The sizes that are seen in the log file are in sectors.
Extents are of the following format: <offset : size>

This commit is a strict implementation which enforces:
    * magics
    * version number 2.1
    * grain size of 8 sectors  (4KB)
    * grain table size of 64 sectors
    * zero flags
    * extent locations

Additionally, this commit proivdes only a subset of the functionality
offered by seSparse's format:
    * Read-only
    * No journal replay
    * No space reclamation
    * No unmap support

Hence, journal header, journal, free bitmap and backmap extents are
unused, only the "classic" (L1 -> L2 -> data) grain access is
implemented.

However there are several differences in the grain access itself.
Grain directory (L1):
    * Grain directory entries are indexes (not offsets) to grain
      tables.
    * Valid grain directory entries have their highest nibble set to
      0x1.
    * Since grain tables are always located in the beginning of the
      file - the index can fit into 32 bits - so we can use its low
      part if it's valid.
Grain table (L2):
    * Grain table entries are indexes (not offsets) to grains.
    * If the highest nibble of the entry is:
        0x0:
            The grain in not allocated.
            The rest of the bytes are 0.
        0x1:
            The grain is unmapped - guest sees a zero grain.
            The rest of the bits point to the previously mapped grain,
            see 0x3 case.
        0x2:
            The grain is zero.
        0x3:
            The grain is allocated - to get the index calculate:
            ((entry & 0x0fff000000000000) >> 48) |
            ((entry & 0x0000ffffffffffff) << 12)
    * The difference between 0x1 and 0x2 is that 0x1 is an unallocated
      grain which results from the guest using sg_unmap to unmap the
      grain - but the grain itself still exists in the grain extent - a
      space reclamation procedure should delete it.
      Unmapping a zero grain has no effect (0x2 will not change to 0x1)
      but unmapping an unallocated grain will (0x0 to 0x1) - naturally.

In order to implement seSparse some fields had to be changed to support
both 32-bit and 64-bit entry sizes.

Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Reviewed-by: Eyal Moscovici <eyal.moscovici@oracle.com>
Reviewed-by: Arbel Moshe <arbel.moshe@oracle.com>
Signed-off-by: Sam Eiderman <shmuel.eiderman@oracle.com>
Message-id: 20190620091057.47441-4-shmuel.eiderman@oracle.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block/vmdk.c | 358 ++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 342 insertions(+), 16 deletions(-)

diff --git a/block/vmdk.c b/block/vmdk.c
index XXXXXXX..XXXXXXX 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -XXX,XX +XXX,XX @@ typedef struct {
     uint16_t compressAlgorithm;
 } QEMU_PACKED VMDK4Header;
 
+typedef struct VMDKSESparseConstHeader {
+    uint64_t magic;
+    uint64_t version;
+    uint64_t capacity;
+    uint64_t grain_size;
+    uint64_t grain_table_size;
+    uint64_t flags;
+    uint64_t reserved1;
+    uint64_t reserved2;
+    uint64_t reserved3;
+    uint64_t reserved4;
+    uint64_t volatile_header_offset;
+    uint64_t volatile_header_size;
+    uint64_t journal_header_offset;
+    uint64_t journal_header_size;
+    uint64_t journal_offset;
+    uint64_t journal_size;
+    uint64_t grain_dir_offset;
+    uint64_t grain_dir_size;
+    uint64_t grain_tables_offset;
+    uint64_t grain_tables_size;
+    uint64_t free_bitmap_offset;
+    uint64_t free_bitmap_size;
+    uint64_t backmap_offset;
+    uint64_t backmap_size;
+    uint64_t grains_offset;
+    uint64_t grains_size;
+    uint8_t pad[304];
+} QEMU_PACKED VMDKSESparseConstHeader;
+
+typedef struct VMDKSESparseVolatileHeader {
+    uint64_t magic;
+    uint64_t free_gt_number;
+    uint64_t next_txn_seq_number;
+    uint64_t replay_journal;
+    uint8_t pad[480];
+} QEMU_PACKED VMDKSESparseVolatileHeader;
+
 #define L2_CACHE_SIZE 16
 
 typedef struct VmdkExtent {
@@ -XXX,XX +XXX,XX @@ typedef struct VmdkExtent {
     bool compressed;
     bool has_marker;
     bool has_zero_grain;
+    bool sesparse;
+    uint64_t sesparse_l2_tables_offset;
+    uint64_t sesparse_clusters_offset;
+    int32_t entry_size;
     int version;
     int64_t sectors;
     int64_t end_sector;
     int64_t flat_start_offset;
     int64_t l1_table_offset;
     int64_t l1_backup_table_offset;
-    uint32_t *l1_table;
+    void *l1_table;
     uint32_t *l1_backup_table;
     unsigned int l1_size;
     uint32_t l1_entry_sectors;
 
     unsigned int l2_size;
-    uint32_t *l2_cache;
+    void *l2_cache;
     uint32_t l2_cache_offsets[L2_CACHE_SIZE];
     uint32_t l2_cache_counts[L2_CACHE_SIZE];
 
@@ -XXX,XX +XXX,XX @@ static int vmdk_add_extent(BlockDriverState *bs,
          *            minimal L2 table size: 512 entries
          *            8 TB is still more than the maximal value supported for
          *            VMDK3 & VMDK4 which is 2TB.
+         *     64TB - for "ESXi seSparse Extent"
+         *            minimal cluster size: 512B (default is 4KB)
+         *            L2 table size: 4096 entries (const).
+         *            64TB is more than the maximal value supported for
+         *            seSparse VMDKs (which is slightly less than 64TB)
          */
         error_setg(errp, "L1 size too big");
         return -EFBIG;
@@ -XXX,XX +XXX,XX @@ static int vmdk_add_extent(BlockDriverState *bs,
     extent->l2_size = l2_size;
     extent->cluster_sectors = flat ? sectors : cluster_sectors;
     extent->next_cluster_sector = ROUND_UP(nb_sectors, cluster_sectors);
+    extent->entry_size = sizeof(uint32_t);
 
     if (s->num_extents > 1) {
         extent->end_sector = (*(extent - 1)).end_sector + extent->sectors;
@@ -XXX,XX +XXX,XX @@ static int vmdk_init_tables(BlockDriverState *bs, VmdkExtent *extent,
     int i;
 
     /* read the L1 table */
-    l1_size = extent->l1_size * sizeof(uint32_t);
+    l1_size = extent->l1_size * extent->entry_size;
     extent->l1_table = g_try_malloc(l1_size);
     if (l1_size && extent->l1_table == NULL) {
         return -ENOMEM;
@@ -XXX,XX +XXX,XX @@ static int vmdk_init_tables(BlockDriverState *bs, VmdkExtent *extent,
         goto fail_l1;
     }
     for (i = 0; i < extent->l1_size; i++) {
-        le32_to_cpus(&extent->l1_table[i]);
+        if (extent->entry_size == sizeof(uint64_t)) {
+            le64_to_cpus((uint64_t *)extent->l1_table + i);
+        } else {
+            assert(extent->entry_size == sizeof(uint32_t));
+            le32_to_cpus((uint32_t *)extent->l1_table + i);
+        }
     }
 
     if (extent->l1_backup_table_offset) {
+        assert(!extent->sesparse);
         extent->l1_backup_table = g_try_malloc(l1_size);
         if (l1_size && extent->l1_backup_table == NULL) {
             ret = -ENOMEM;
@@ -XXX,XX +XXX,XX @@ static int vmdk_init_tables(BlockDriverState *bs, VmdkExtent *extent,
     }
 
     extent->l2_cache =
-        g_new(uint32_t, extent->l2_size * L2_CACHE_SIZE);
+        g_malloc(extent->entry_size * extent->l2_size * L2_CACHE_SIZE);
     return 0;
  fail_l1b:
     g_free(extent->l1_backup_table);
@@ -XXX,XX +XXX,XX @@ static int vmdk_open_vmfs_sparse(BlockDriverState *bs,
     return ret;
 }
 
+#define SESPARSE_CONST_HEADER_MAGIC UINT64_C(0x00000000cafebabe)
+#define SESPARSE_VOLATILE_HEADER_MAGIC UINT64_C(0x00000000cafecafe)
+
+/* Strict checks - format not officially documented */
+static int check_se_sparse_const_header(VMDKSESparseConstHeader *header,
+                                        Error **errp)
+{
+    header->magic = le64_to_cpu(header->magic);
+    header->version = le64_to_cpu(header->version);
+    header->grain_size = le64_to_cpu(header->grain_size);
+    header->grain_table_size = le64_to_cpu(header->grain_table_size);
+    header->flags = le64_to_cpu(header->flags);
+    header->reserved1 = le64_to_cpu(header->reserved1);
+    header->reserved2 = le64_to_cpu(header->reserved2);
+    header->reserved3 = le64_to_cpu(header->reserved3);
+    header->reserved4 = le64_to_cpu(header->reserved4);
+
+    header->volatile_header_offset =
+        le64_to_cpu(header->volatile_header_offset);
+    header->volatile_header_size = le64_to_cpu(header->volatile_header_size);
+
+    header->journal_header_offset = le64_to_cpu(header->journal_header_offset);
+    header->journal_header_size = le64_to_cpu(header->journal_header_size);
+
+    header->journal_offset = le64_to_cpu(header->journal_offset);
+    header->journal_size = le64_to_cpu(header->journal_size);
+
+    header->grain_dir_offset = le64_to_cpu(header->grain_dir_offset);
+    header->grain_dir_size = le64_to_cpu(header->grain_dir_size);
+
+    header->grain_tables_offset = le64_to_cpu(header->grain_tables_offset);
+    header->grain_tables_size = le64_to_cpu(header->grain_tables_size);
+
+    header->free_bitmap_offset = le64_to_cpu(header->free_bitmap_offset);
+    header->free_bitmap_size = le64_to_cpu(header->free_bitmap_size);
+
+    header->backmap_offset = le64_to_cpu(header->backmap_offset);
+    header->backmap_size = le64_to_cpu(header->backmap_size);
+
+    header->grains_offset = le64_to_cpu(header->grains_offset);
+    header->grains_size = le64_to_cpu(header->grains_size);
+
+    if (header->magic != SESPARSE_CONST_HEADER_MAGIC) {
+        error_setg(errp, "Bad const header magic: 0x%016" PRIx64,
+                   header->magic);
+        return -EINVAL;
+    }
+
+    if (header->version != 0x0000000200000001) {
+        error_setg(errp, "Unsupported version: 0x%016" PRIx64,
+                   header->version);
+        return -ENOTSUP;
+    }
+
+    if (header->grain_size != 8) {
+        error_setg(errp, "Unsupported grain size: %" PRIu64,
+                   header->grain_size);
+        return -ENOTSUP;
+    }
+
+    if (header->grain_table_size != 64) {
+        error_setg(errp, "Unsupported grain table size: %" PRIu64,
+                   header->grain_table_size);
+        return -ENOTSUP;
+    }
+
+    if (header->flags != 0) {
+        error_setg(errp, "Unsupported flags: 0x%016" PRIx64,
+                   header->flags);
+        return -ENOTSUP;
+    }
+
+    if (header->reserved1 != 0 || header->reserved2 != 0 ||
+        header->reserved3 != 0 || header->reserved4 != 0) {
+        error_setg(errp, "Unsupported reserved bits:"
+                   " 0x%016" PRIx64 " 0x%016" PRIx64
+                   " 0x%016" PRIx64 " 0x%016" PRIx64,
+                   header->reserved1, header->reserved2,
+                   header->reserved3, header->reserved4);
+        return -ENOTSUP;
+    }
+
+    /* check that padding is 0 */
+    if (!buffer_is_zero(header->pad, sizeof(header->pad))) {
+        error_setg(errp, "Unsupported non-zero const header padding");
+        return -ENOTSUP;
+    }
+
+    return 0;
+}
+
+static int check_se_sparse_volatile_header(VMDKSESparseVolatileHeader *header,
+                                           Error **errp)
+{
+    header->magic = le64_to_cpu(header->magic);
+    header->free_gt_number = le64_to_cpu(header->free_gt_number);
+    header->next_txn_seq_number = le64_to_cpu(header->next_txn_seq_number);
+    header->replay_journal = le64_to_cpu(header->replay_journal);
+
+    if (header->magic != SESPARSE_VOLATILE_HEADER_MAGIC) {
+        error_setg(errp, "Bad volatile header magic: 0x%016" PRIx64,
+                   header->magic);
+        return -EINVAL;
+    }
+
+    if (header->replay_journal) {
+        error_setg(errp, "Image is dirty, Replaying journal not supported");
+        return -ENOTSUP;
+    }
+
+    /* check that padding is 0 */
+    if (!buffer_is_zero(header->pad, sizeof(header->pad))) {
+        error_setg(errp, "Unsupported non-zero volatile header padding");
+        return -ENOTSUP;
+    }
+
+    return 0;
+}
+
+static int vmdk_open_se_sparse(BlockDriverState *bs,
+                               BdrvChild *file,
+                               int flags, Error **errp)
+{
+    int ret;
+    VMDKSESparseConstHeader const_header;
+    VMDKSESparseVolatileHeader volatile_header;
+    VmdkExtent *extent;
+
+    ret = bdrv_apply_auto_read_only(bs,
+            "No write support for seSparse images available", errp);
+    if (ret < 0) {
+        return ret;
+    }
+
+    assert(sizeof(const_header) == SECTOR_SIZE);
+
+    ret = bdrv_pread(file, 0, &const_header, sizeof(const_header));
+    if (ret < 0) {
+        bdrv_refresh_filename(file->bs);
+        error_setg_errno(errp, -ret,
+                         "Could not read const header from file '%s'",
+                         file->bs->filename);
+        return ret;
+    }
+
+    /* check const header */
+    ret = check_se_sparse_const_header(&const_header, errp);
+    if (ret < 0) {
+        return ret;
+    }
+
+    assert(sizeof(volatile_header) == SECTOR_SIZE);
+
+    ret = bdrv_pread(file,
+                     const_header.volatile_header_offset * SECTOR_SIZE,
+                     &volatile_header, sizeof(volatile_header));
+    if (ret < 0) {
+        bdrv_refresh_filename(file->bs);
+        error_setg_errno(errp, -ret,
+                         "Could not read volatile header from file '%s'",
+                         file->bs->filename);
+        return ret;
+    }
+
+    /* check volatile header */
+    ret = check_se_sparse_volatile_header(&volatile_header, errp);
+    if (ret < 0) {
+        return ret;
+    }
+
+    ret = vmdk_add_extent(bs, file, false,
+                          const_header.capacity,
+                          const_header.grain_dir_offset * SECTOR_SIZE,
+                          0,
+                          const_header.grain_dir_size *
+                          SECTOR_SIZE / sizeof(uint64_t),
+                          const_header.grain_table_size *
+                          SECTOR_SIZE / sizeof(uint64_t),
+                          const_header.grain_size,
+                          &extent,
+                          errp);
+    if (ret < 0) {
+        return ret;
+    }
+
+    extent->sesparse = true;
+    extent->sesparse_l2_tables_offset = const_header.grain_tables_offset;
+    extent->sesparse_clusters_offset = const_header.grains_offset;
+    extent->entry_size = sizeof(uint64_t);
+
+    ret = vmdk_init_tables(bs, extent, errp);
+    if (ret) {
+        /* free extent allocated by vmdk_add_extent */
+        vmdk_free_last_extent(bs);
+    }
+
+    return ret;
+}
+
 static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf,
                                QDict *options, Error **errp);
 
@@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
          * RW [size in sectors] SPARSE "file-name.vmdk"
          * RW [size in sectors] VMFS "file-name.vmdk"
          * RW [size in sectors] VMFSSPARSE "file-name.vmdk"
+         * RW [size in sectors] SESPARSE "file-name.vmdk"
          */
         flat_offset = -1;
         matches = sscanf(p, "%10s %" SCNd64 " %10s \"%511[^\n\r\"]\" %" SCNd64,
@@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
 
         if (sectors <= 0 ||
             (strcmp(type, "FLAT") && strcmp(type, "SPARSE") &&
-             strcmp(type, "VMFS") && strcmp(type, "VMFSSPARSE")) ||
+             strcmp(type, "VMFS") && strcmp(type, "VMFSSPARSE") &&
+             strcmp(type, "SESPARSE")) ||
             (strcmp(access, "RW"))) {
             continue;
         }
@@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
                 return ret;
             }
             extent = &s->extents[s->num_extents - 1];
+        } else if (!strcmp(type, "SESPARSE")) {
+            ret = vmdk_open_se_sparse(bs, extent_file, bs->open_flags, errp);
+            if (ret) {
+                bdrv_unref_child(bs, extent_file);
+                return ret;
+            }
+            extent = &s->extents[s->num_extents - 1];
         } else {
             error_setg(errp, "Unsupported extent type '%s'", type);
             bdrv_unref_child(bs, extent_file);
@@ -XXX,XX +XXX,XX @@ static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf,
     if (strcmp(ct, "monolithicFlat") &&
         strcmp(ct, "vmfs") &&
         strcmp(ct, "vmfsSparse") &&
+        strcmp(ct, "seSparse") &&
         strcmp(ct, "twoGbMaxExtentSparse") &&
         strcmp(ct, "twoGbMaxExtentFlat")) {
         error_setg(errp, "Unsupported image type '%s'", ct);
@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
 {
     unsigned int l1_index, l2_offset, l2_index;
     int min_index, i, j;
-    uint32_t min_count, *l2_table;
+    uint32_t min_count;
+    void *l2_table;
     bool zeroed = false;
     int64_t ret;
     int64_t cluster_sector;
+    unsigned int l2_size_bytes = extent->l2_size * extent->entry_size;
 
     if (m_data) {
         m_data->valid = 0;
@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
     if (l1_index >= extent->l1_size) {
         return VMDK_ERROR;
     }
-    l2_offset = extent->l1_table[l1_index];
+    if (extent->sesparse) {
+        uint64_t l2_offset_u64;
+
+        assert(extent->entry_size == sizeof(uint64_t));
+
+        l2_offset_u64 = ((uint64_t *)extent->l1_table)[l1_index];
+        if (l2_offset_u64 == 0) {
+            l2_offset = 0;
+        } else if ((l2_offset_u64 & 0xffffffff00000000) != 0x1000000000000000) {
+            /*
+             * Top most nibble is 0x1 if grain table is allocated.
+             * strict check - top most 4 bytes must be 0x10000000 since max
+             * supported size is 64TB for disk - so no more than 64TB / 16MB
+             * grain directories which is smaller than uint32,
+             * where 16MB is the only supported default grain table coverage.
+             */
+            return VMDK_ERROR;
+        } else {
+            l2_offset_u64 = l2_offset_u64 & 0x00000000ffffffff;
+            l2_offset_u64 = extent->sesparse_l2_tables_offset +
+                l2_offset_u64 * l2_size_bytes / SECTOR_SIZE;
+            if (l2_offset_u64 > 0x00000000ffffffff) {
+                return VMDK_ERROR;
+            }
+            l2_offset = (unsigned int)(l2_offset_u64);
+        }
+    } else {
+        assert(extent->entry_size == sizeof(uint32_t));
+        l2_offset = ((uint32_t *)extent->l1_table)[l1_index];
+    }
     if (!l2_offset) {
         return VMDK_UNALLOC;
     }
@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
                     extent->l2_cache_counts[j] >>= 1;
                 }
             }
-            l2_table = extent->l2_cache + (i * extent->l2_size);
+            l2_table = (char *)extent->l2_cache + (i * l2_size_bytes);
             goto found;
         }
     }
@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
             min_index = i;
         }
     }
-    l2_table = extent->l2_cache + (min_index * extent->l2_size);
+    l2_table = (char *)extent->l2_cache + (min_index * l2_size_bytes);
     BLKDBG_EVENT(extent->file, BLKDBG_L2_LOAD);
     if (bdrv_pread(extent->file,
                 (int64_t)l2_offset * 512,
                 l2_table,
-                extent->l2_size * sizeof(uint32_t)
-            ) != extent->l2_size * sizeof(uint32_t)) {
+                l2_size_bytes
+            ) != l2_size_bytes) {
         return VMDK_ERROR;
     }
 
@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
     extent->l2_cache_counts[min_index] = 1;
  found:
     l2_index = ((offset >> 9) / extent->cluster_sectors) % extent->l2_size;
-    cluster_sector = le32_to_cpu(l2_table[l2_index]);
 
-    if (extent->has_zero_grain && cluster_sector == VMDK_GTE_ZEROED) {
-        zeroed = true;
+    if (extent->sesparse) {
+        cluster_sector = le64_to_cpu(((uint64_t *)l2_table)[l2_index]);
+        switch (cluster_sector & 0xf000000000000000) {
+        case 0x0000000000000000:
+            /* unallocated grain */
+            if (cluster_sector != 0) {
+                return VMDK_ERROR;
+            }
+            break;
+        case 0x1000000000000000:
+            /* scsi-unmapped grain - fallthrough */
+        case 0x2000000000000000:
+            /* zero grain */
+            zeroed = true;
+            break;
+        case 0x3000000000000000:
+            /* allocated grain */
+            cluster_sector = (((cluster_sector & 0x0fff000000000000) >> 48) |
+                              ((cluster_sector & 0x0000ffffffffffff) << 12));
+            cluster_sector = extent->sesparse_clusters_offset +
+                cluster_sector * extent->cluster_sectors;
+            break;
+        default:
+            return VMDK_ERROR;
+        }
+    } else {
+        cluster_sector = le32_to_cpu(((uint32_t *)l2_table)[l2_index]);
+
+        if (extent->has_zero_grain && cluster_sector == VMDK_GTE_ZEROED) {
+            zeroed = true;
+        }
     }
 
     if (!cluster_sector || zeroed) {
         if (!allocate) {
             return zeroed ? VMDK_ZEROED : VMDK_UNALLOC;
         }
+        assert(!extent->sesparse);
 
         if (extent->next_cluster_sector >= VMDK_EXTENT_MAX_SECTORS) {
             return VMDK_ERROR;
@@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs,
             m_data->l1_index = l1_index;
             m_data->l2_index = l2_index;
             m_data->l2_offset = l2_offset;
-            m_data->l2_cache_entry = &l2_table[l2_index];
+            m_data->l2_cache_entry = ((uint32_t *)l2_table) + l2_index;
         }
     }
     *cluster_offset = cluster_sector << BDRV_SECTOR_BITS;
@@ -XXX,XX +XXX,XX @@ static int vmdk_pwritev(BlockDriverState *bs, uint64_t offset,
         if (!extent) {
             return -EIO;
         }
+        if (extent->sesparse) {
+            return -ENOTSUP;
+        }
         offset_in_cluster = vmdk_find_offset_in_cluster(extent, offset);
         n_bytes = MIN(bytes, extent->cluster_sectors * BDRV_SECTOR_SIZE
                              - offset_in_cluster);
-- 
2.21.0

From: Pino Toscano <ptoscano@redhat.com>

Rewrite the implementation of the ssh block driver to use libssh instead
of libssh2.  The libssh library has various advantages over libssh2:
- easier API for authentication (for example for using ssh-agent)
- easier API for known_hosts handling
- supports newer types of keys in known_hosts

Use APIs/features available in libssh 0.8 conditionally, to support
older versions (which are not recommended though).

Adjust the iotest 207 according to the different error message, and to
find the default key type for localhost (to properly compare the
fingerprint with).
Contributed-by: Max Reitz <mreitz@redhat.com>

Adjust the various Docker/Travis scripts to use libssh when available
instead of libssh2. The mingw/mxe testing is dropped for now, as there
are no packages for it.

Signed-off-by: Pino Toscano <ptoscano@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Alex Bennée <alex.bennee@linaro.org>
Message-id: 20190620200840.17655-1-ptoscano@redhat.com
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 5873173.t2JhDm7DL7@lindworm.usersys.redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 configure                                     |  65 +-
 block/Makefile.objs                           |   6 +-
 block/ssh.c                                   | 652 ++++++++++--------
 .travis.yml                                   |   4 +-
 block/trace-events                            |  14 +-
 docs/qemu-block-drivers.texi                  |   2 +-
 .../dockerfiles/debian-win32-cross.docker     |   1 -
 .../dockerfiles/debian-win64-cross.docker     |   1 -
 tests/docker/dockerfiles/fedora.docker        |   4 +-
 tests/docker/dockerfiles/ubuntu.docker        |   2 +-
 tests/docker/dockerfiles/ubuntu1804.docker    |   2 +-
 tests/qemu-iotests/207                        |  54 +-
 tests/qemu-iotests/207.out                    |   2 +-
 13 files changed, 449 insertions(+), 360 deletions(-)

diff --git a/configure b/configure
index XXXXXXX..XXXXXXX 100755
--- a/configure
+++ b/configure
@@ -XXX,XX +XXX,XX @@ auth_pam=""
 vte=""
 virglrenderer=""
 tpm=""
-libssh2=""
+libssh=""
 live_block_migration="yes"
 numa=""
 tcmalloc="no"
@@ -XXX,XX +XXX,XX @@ for opt do
   ;;
   --enable-tpm) tpm="yes"
   ;;
-  --disable-libssh2) libssh2="no"
+  --disable-libssh) libssh="no"
   ;;
-  --enable-libssh2) libssh2="yes"
+  --enable-libssh) libssh="yes"
   ;;
   --disable-live-block-migration) live_block_migration="no"
   ;;
@@ -XXX,XX +XXX,XX @@ disabled with --disable-FEATURE, default is enabled if available:
   coroutine-pool  coroutine freelist (better performance)
   glusterfs       GlusterFS backend
   tpm             TPM support
-  libssh2         ssh block device support
+  libssh          ssh block device support
   numa            libnuma support
   libxml2         for Parallels image format
   tcmalloc        tcmalloc support
@@ -XXX,XX +XXX,XX @@ EOF
 fi
 
 ##########################################
-# libssh2 probe
-min_libssh2_version=1.2.8
-if test "$libssh2" != "no" ; then
-  if $pkg_config --atleast-version=$min_libssh2_version libssh2; then
-    libssh2_cflags=$($pkg_config libssh2 --cflags)
-    libssh2_libs=$($pkg_config libssh2 --libs)
-    libssh2=yes
+# libssh probe
+if test "$libssh" != "no" ; then
+  if $pkg_config --exists libssh; then
+    libssh_cflags=$($pkg_config libssh --cflags)
+    libssh_libs=$($pkg_config libssh --libs)
+    libssh=yes
   else
-    if test "$libssh2" = "yes" ; then
-      error_exit "libssh2 >= $min_libssh2_version required for --enable-libssh2"
+    if test "$libssh" = "yes" ; then
+      error_exit "libssh required for --enable-libssh"
     fi
-    libssh2=no
+    libssh=no
   fi
 fi
 
 ##########################################
-# libssh2_sftp_fsync probe
+# Check for libssh 0.8
+# This is done like this instead of using the LIBSSH_VERSION_* and
+# SSH_VERSION_* macros because some distributions in the past shipped
+# snapshots of the future 0.8 from Git, and those snapshots did not
+# have updated version numbers (still referring to 0.7.0).
 
-if test "$libssh2" = "yes"; then
+if test "$libssh" = "yes"; then
   cat > $TMPC <<EOF
-#include <stdio.h>
-#include <libssh2.h>
-#include <libssh2_sftp.h>
-int main(void) {
-    LIBSSH2_SESSION *session;
-    LIBSSH2_SFTP *sftp;
-    LIBSSH2_SFTP_HANDLE *sftp_handle;
-    session = libssh2_session_init ();
-    sftp = libssh2_sftp_init (session);
-    sftp_handle = libssh2_sftp_open (sftp, "/", 0, 0);
-    libssh2_sftp_fsync (sftp_handle);
-    return 0;
-}
+#include <libssh/libssh.h>
+int main(void) { return ssh_get_server_publickey(NULL, NULL); }
 EOF
-  # libssh2_cflags/libssh2_libs defined in previous test.
-  if compile_prog "$libssh2_cflags" "$libssh2_libs" ; then
-    QEMU_CFLAGS="-DHAS_LIBSSH2_SFTP_FSYNC $QEMU_CFLAGS"
+  if compile_prog "$libssh_cflags" "$libssh_libs"; then
+    libssh_cflags="-DHAVE_LIBSSH_0_8 $libssh_cflags"
   fi
 fi
 
@@ -XXX,XX +XXX,XX @@ echo "GlusterFS support $glusterfs"
 echo "gcov              $gcov_tool"
 echo "gcov enabled      $gcov"
 echo "TPM support       $tpm"
-echo "libssh2 support   $libssh2"
+echo "libssh support    $libssh"
 echo "QOM debugging     $qom_cast_debug"
 echo "Live block migration $live_block_migration"
 echo "lzo support       $lzo"
@@ -XXX,XX +XXX,XX @@ if test "$glusterfs_iocb_has_stat" = "yes" ; then
   echo "CONFIG_GLUSTERFS_IOCB_HAS_STAT=y" >> $config_host_mak
 fi
 
-if test "$libssh2" = "yes" ; then
-  echo "CONFIG_LIBSSH2=m" >> $config_host_mak
-  echo "LIBSSH2_CFLAGS=$libssh2_cflags" >> $config_host_mak
-  echo "LIBSSH2_LIBS=$libssh2_libs" >> $config_host_mak
+if test "$libssh" = "yes" ; then
+  echo "CONFIG_LIBSSH=m" >> $config_host_mak
+  echo "LIBSSH_CFLAGS=$libssh_cflags" >> $config_host_mak
+  echo "LIBSSH_LIBS=$libssh_libs" >> $config_host_mak
 fi
 
 if test "$live_block_migration" = "yes" ; then
diff --git a/block/Makefile.objs b/block/Makefile.objs
index XXXXXXX..XXXXXXX 100644
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -XXX,XX +XXX,XX @@ block-obj-$(CONFIG_CURL) += curl.o
 block-obj-$(CONFIG_RBD) += rbd.o
 block-obj-$(CONFIG_GLUSTERFS) += gluster.o
 block-obj-$(CONFIG_VXHS) += vxhs.o
-block-obj-$(CONFIG_LIBSSH2) += ssh.o
+block-obj-$(CONFIG_LIBSSH) += ssh.o
 block-obj-y += accounting.o dirty-bitmap.o
 block-obj-y += write-threshold.o
 block-obj-y += backup.o
@@ -XXX,XX +XXX,XX @@ rbd.o-libs         := $(RBD_LIBS)
 gluster.o-cflags   := $(GLUSTERFS_CFLAGS)
 gluster.o-libs     := $(GLUSTERFS_LIBS)
 vxhs.o-libs        := $(VXHS_LIBS)
-ssh.o-cflags       := $(LIBSSH2_CFLAGS)
-ssh.o-libs         := $(LIBSSH2_LIBS)
+ssh.o-cflags       := $(LIBSSH_CFLAGS)
+ssh.o-libs         := $(LIBSSH_LIBS)
 block-obj-dmg-bz2-$(CONFIG_BZIP2) += dmg-bz2.o
 block-obj-$(if $(CONFIG_DMG),m,n) += $(block-obj-dmg-bz2-y)
 dmg-bz2.o-libs     := $(BZIP2_LIBS)
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 @@
 
 #include "qemu/osdep.h"
 
-#include <libssh2.h>
-#include <libssh2_sftp.h>
+#include <libssh/libssh.h>
+#include <libssh/sftp.h>
 
 #include "block/block_int.h"
 #include "block/qdict.h"
@@ -XXX,XX +XXX,XX @@
 #include "trace.h"
 
 /*
- * TRACE_LIBSSH2=<bitmask> enables tracing in libssh2 itself.  Note
- * that this requires that libssh2 was specially compiled with the
- * `./configure --enable-debug' option, so most likely you will have
- * to compile it yourself.  The meaning of <bitmask> is described
- * here: http://www.libssh2.org/libssh2_trace.html
+ * TRACE_LIBSSH=<level> enables tracing in libssh itself.
+ * The meaning of <level> is described here:
+ * http://api.libssh.org/master/group__libssh__log.html
  */
-#define TRACE_LIBSSH2 0 /* or try: LIBSSH2_TRACE_SFTP */
+#define TRACE_LIBSSH  0 /* see: SSH_LOG_* */
 
 typedef struct BDRVSSHState {
     /* Coroutine. */
@@ -XXX,XX +XXX,XX @@ typedef struct BDRVSSHState {
 
     /* SSH connection. */
     int sock;                         /* socket */
-    LIBSSH2_SESSION *session;         /* ssh session */
-    LIBSSH2_SFTP *sftp;               /* sftp session */
-    LIBSSH2_SFTP_HANDLE *sftp_handle; /* sftp remote file handle */
+    ssh_session session;              /* ssh session */
+    sftp_session sftp;                /* sftp session */
+    sftp_file sftp_handle;            /* sftp remote file handle */
 
-    /* See ssh_seek() function below. */
-    int64_t offset;
-    bool offset_op_read;
-
-    /* File attributes at open.  We try to keep the .filesize field
+    /*
+     * File attributes at open.  We try to keep the .size field
      * updated if it changes (eg by writing at the end of the file).
      */
-    LIBSSH2_SFTP_ATTRIBUTES attrs;
+    sftp_attributes attrs;
 
     InetSocketAddress *inet;
 
@@ -XXX,XX +XXX,XX @@ static void ssh_state_init(BDRVSSHState *s)
 {
     memset(s, 0, sizeof *s);
     s->sock = -1;
-    s->offset = -1;
     qemu_co_mutex_init(&s->lock);
 }
 
@@ -XXX,XX +XXX,XX @@ static void ssh_state_free(BDRVSSHState *s)
 {
     g_free(s->user);
 
+    if (s->attrs) {
+        sftp_attributes_free(s->attrs);
+    }
     if (s->sftp_handle) {
-        libssh2_sftp_close(s->sftp_handle);
+        sftp_close(s->sftp_handle);
     }
     if (s->sftp) {
-        libssh2_sftp_shutdown(s->sftp);
+        sftp_free(s->sftp);
     }
     if (s->session) {
-        libssh2_session_disconnect(s->session,
-                                   "from qemu ssh client: "
-                                   "user closed the connection");
-        libssh2_session_free(s->session);
-    }
-    if (s->sock >= 0) {
-        close(s->sock);
+        ssh_disconnect(s->session);
+        ssh_free(s->session); /* This frees s->sock */
     }
 }
 
@@ -XXX,XX +XXX,XX @@ session_error_setg(Error **errp, BDRVSSHState *s, const char *fs, ...)
     va_end(args);
 
     if (s->session) {
-        char *ssh_err;
+        const char *ssh_err;
         int ssh_err_code;
 
-        /* This is not an errno.  See <libssh2.h>. */
-        ssh_err_code = libssh2_session_last_error(s->session,
-                                                  &ssh_err, NULL, 0);
-        error_setg(errp, "%s: %s (libssh2 error code: %d)",
+        /* This is not an errno.  See <libssh/libssh.h>. */
+        ssh_err = ssh_get_error(s->session);
+        ssh_err_code = ssh_get_error_code(s->session);
+        error_setg(errp, "%s: %s (libssh error code: %d)",
                    msg, ssh_err, ssh_err_code);
     } else {
         error_setg(errp, "%s", msg);
@@ -XXX,XX +XXX,XX @@ sftp_error_setg(Error **errp, BDRVSSHState *s, const char *fs, ...)
     va_end(args);
 
     if (s->sftp) {
-        char *ssh_err;
+        const char *ssh_err;
         int ssh_err_code;
-        unsigned long sftp_err_code;
+        int sftp_err_code;
 
-        /* This is not an errno.  See <libssh2.h>. */
-        ssh_err_code = libssh2_session_last_error(s->session,
-                                                  &ssh_err, NULL, 0);
-        /* See <libssh2_sftp.h>. */
-        sftp_err_code = libssh2_sftp_last_error((s)->sftp);
+        /* This is not an errno.  See <libssh/libssh.h>. */
+        ssh_err = ssh_get_error(s->session);
+        ssh_err_code = ssh_get_error_code(s->session);
+        /* See <libssh/sftp.h>. */
+        sftp_err_code = sftp_get_error(s->sftp);
 
         error_setg(errp,
-                   "%s: %s (libssh2 error code: %d, sftp error code: %lu)",
+                   "%s: %s (libssh error code: %d, sftp error code: %d)",
                    msg, ssh_err, ssh_err_code, sftp_err_code);
     } else {
         error_setg(errp, "%s", msg);
@@ -XXX,XX +XXX,XX @@ sftp_error_setg(Error **errp, BDRVSSHState *s, const char *fs, ...)
 
 static void sftp_error_trace(BDRVSSHState *s, const char *op)
 {
-    char *ssh_err;
+    const char *ssh_err;
     int ssh_err_code;
-    unsigned long sftp_err_code;
+    int sftp_err_code;
 
-    /* This is not an errno.  See <libssh2.h>. */
-    ssh_err_code = libssh2_session_last_error(s->session,
-                                              &ssh_err, NULL, 0);
-    /* See <libssh2_sftp.h>. */
-    sftp_err_code = libssh2_sftp_last_error((s)->sftp);
+    /* This is not an errno.  See <libssh/libssh.h>. */
+    ssh_err = ssh_get_error(s->session);
+    ssh_err_code = ssh_get_error_code(s->session);
+    /* See <libssh/sftp.h>. */
+    sftp_err_code = sftp_get_error(s->sftp);
 
     trace_sftp_error(op, ssh_err, ssh_err_code, sftp_err_code);
 }
@@ -XXX,XX +XXX,XX @@ static void ssh_parse_filename(const char *filename, QDict *options,
     parse_uri(filename, options, errp);
 }
 
-static int check_host_key_knownhosts(BDRVSSHState *s,
-                                     const char *host, int port, Error **errp)
+static int check_host_key_knownhosts(BDRVSSHState *s, Error **errp)
 {
-    const char *home;
-    char *knh_file = NULL;
-    LIBSSH2_KNOWNHOSTS *knh = NULL;
-    struct libssh2_knownhost *found;
-    int ret, r;
-    const char *hostkey;
-    size_t len;
-    int type;
-
-    hostkey = libssh2_session_hostkey(s->session, &len, &type);
-    if (!hostkey) {
+    int ret;
+#ifdef HAVE_LIBSSH_0_8
+    enum ssh_known_hosts_e state;
+    int r;
+    ssh_key pubkey;
+    enum ssh_keytypes_e pubkey_type;
+    unsigned char *server_hash = NULL;
+    size_t server_hash_len;
+    char *fingerprint = NULL;
+
+    state = ssh_session_is_known_server(s->session);
+    trace_ssh_server_status(state);
+
+    switch (state) {
+    case SSH_KNOWN_HOSTS_OK:
+        /* OK */
+        trace_ssh_check_host_key_knownhosts();
+        break;
+    case SSH_KNOWN_HOSTS_CHANGED:
         ret = -EINVAL;
-        session_error_setg(errp, s, "failed to read remote host key");
+        r = ssh_get_server_publickey(s->session, &pubkey);
+        if (r == 0) {
+            r = ssh_get_publickey_hash(pubkey, SSH_PUBLICKEY_HASH_SHA256,
+                                       &server_hash, &server_hash_len);
+            pubkey_type = ssh_key_type(pubkey);
+            ssh_key_free(pubkey);
+        }
+        if (r == 0) {
+            fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA256,
+                                                   server_hash,
+                                                   server_hash_len);
+            ssh_clean_pubkey_hash(&server_hash);
+        }
+        if (fingerprint) {
+            error_setg(errp,
+                       "host key (%s key with fingerprint %s) does not match "
+                       "the one in known_hosts; this may be a possible attack",
+                       ssh_key_type_to_char(pubkey_type), fingerprint);
+            ssh_string_free_char(fingerprint);
+        } else  {
+            error_setg(errp,
+                       "host key does not match the one in known_hosts; this "
+                       "may be a possible attack");
+        }
         goto out;
-    }
-
-    knh = libssh2_knownhost_init(s->session);
-    if (!knh) {
+    case SSH_KNOWN_HOSTS_OTHER:
         ret = -EINVAL;
-        session_error_setg(errp, s,
-                           "failed to initialize known hosts support");
+        error_setg(errp,
+                   "host key for this server not found, another type exists");
+        goto out;
+    case SSH_KNOWN_HOSTS_UNKNOWN:
+        ret = -EINVAL;
+        error_setg(errp, "no host key was found in known_hosts");
+        goto out;
+    case SSH_KNOWN_HOSTS_NOT_FOUND:
+        ret = -ENOENT;
+        error_setg(errp, "known_hosts file not found");
+        goto out;
+    case SSH_KNOWN_HOSTS_ERROR:
+        ret = -EINVAL;
+        error_setg(errp, "error while checking the host");
+        goto out;
+    default:
+        ret = -EINVAL;
+        error_setg(errp, "error while checking for known server (%d)", state);
         goto out;
     }
+#else /* !HAVE_LIBSSH_0_8 */
+    int state;
 
-    home = getenv("HOME");
-    if (home) {
-        knh_file = g_strdup_printf("%s/.ssh/known_hosts", home);
-    } else {
-        knh_file = g_strdup_printf("/root/.ssh/known_hosts");
-    }
-
-    /* Read all known hosts from OpenSSH-style known_hosts file. */
-    libssh2_knownhost_readfile(knh, knh_file, LIBSSH2_KNOWNHOST_FILE_OPENSSH);
+    state = ssh_is_server_known(s->session);
+    trace_ssh_server_status(state);
 
-    r = libssh2_knownhost_checkp(knh, host, port, hostkey, len,
-                                 LIBSSH2_KNOWNHOST_TYPE_PLAIN|
-                                 LIBSSH2_KNOWNHOST_KEYENC_RAW,
-                                 &found);
-    switch (r) {
-    case LIBSSH2_KNOWNHOST_CHECK_MATCH:
+    switch (state) {
+    case SSH_SERVER_KNOWN_OK:
         /* OK */
-        trace_ssh_check_host_key_knownhosts(found->key);
+        trace_ssh_check_host_key_knownhosts();
         break;
-    case LIBSSH2_KNOWNHOST_CHECK_MISMATCH:
+    case SSH_SERVER_KNOWN_CHANGED:
         ret = -EINVAL;
-        session_error_setg(errp, s,
-                      "host key does not match the one in known_hosts"
-                      " (found key %s)", found->key);
+        error_setg(errp,
+                   "host key does not match the one in known_hosts; this "
+                   "may be a possible attack");
         goto out;
-    case LIBSSH2_KNOWNHOST_CHECK_NOTFOUND:
+    case SSH_SERVER_FOUND_OTHER:
         ret = -EINVAL;
-        session_error_setg(errp, s, "no host key was found in known_hosts");
+        error_setg(errp,
+                   "host key for this server not found, another type exists");
+        goto out;
+    case SSH_SERVER_FILE_NOT_FOUND:
+        ret = -ENOENT;
+        error_setg(errp, "known_hosts file not found");
         goto out;
-    case LIBSSH2_KNOWNHOST_CHECK_FAILURE:
+    case SSH_SERVER_NOT_KNOWN:
         ret = -EINVAL;
-        session_error_setg(errp, s,
-                      "failure matching the host key with known_hosts");
+        error_setg(errp, "no host key was found in known_hosts");
+        goto out;
+    case SSH_SERVER_ERROR:
+        ret = -EINVAL;
+        error_setg(errp, "server error");
         goto out;
     default:
         ret = -EINVAL;
-        session_error_setg(errp, s, "unknown error matching the host key"
-                      " with known_hosts (%d)", r);
+        error_setg(errp, "error while checking for known server (%d)", state);
         goto out;
     }
+#endif /* !HAVE_LIBSSH_0_8 */
 
     /* known_hosts checking successful. */
     ret = 0;
 
  out:
-    if (knh != NULL) {
-        libssh2_knownhost_free(knh);
-    }
-    g_free(knh_file);
     return ret;
 }
 
@@ -XXX,XX +XXX,XX @@ static int compare_fingerprint(const unsigned char *fingerprint, size_t len,
 
 static int
 check_host_key_hash(BDRVSSHState *s, const char *hash,
-                    int hash_type, size_t fingerprint_len, Error **errp)
+                    enum ssh_publickey_hash_type type, Error **errp)
 {
-    const char *fingerprint;
-
-    fingerprint = libssh2_hostkey_hash(s->session, hash_type);
-    if (!fingerprint) {
+    int r;
+    ssh_key pubkey;
+    unsigned char *server_hash;
+    size_t server_hash_len;
+
+#ifdef HAVE_LIBSSH_0_8
+    r = ssh_get_server_publickey(s->session, &pubkey);
+#else
+    r = ssh_get_publickey(s->session, &pubkey);
+#endif
+    if (r != SSH_OK) {
         session_error_setg(errp, s, "failed to read remote host key");
         return -EINVAL;
     }
 
-    if(compare_fingerprint((unsigned char *) fingerprint, fingerprint_len,
-                           hash) != 0) {
+    r = ssh_get_publickey_hash(pubkey, type, &server_hash, &server_hash_len);
+    ssh_key_free(pubkey);
+    if (r != 0) {
+        session_error_setg(errp, s,
+                           "failed reading the hash of the server SSH key");
+        return -EINVAL;
+    }
+
+    r = compare_fingerprint(server_hash, server_hash_len, hash);
+    ssh_clean_pubkey_hash(&server_hash);
+    if (r != 0) {
         error_setg(errp, "remote host key does not match host_key_check '%s'",
                    hash);
         return -EPERM;
@@ -XXX,XX +XXX,XX @@ check_host_key_hash(BDRVSSHState *s, const char *hash,
     return 0;
 }
 
-static int check_host_key(BDRVSSHState *s, const char *host, int port,
-                          SshHostKeyCheck *hkc, Error **errp)
+static int check_host_key(BDRVSSHState *s, SshHostKeyCheck *hkc, Error **errp)
 {
     SshHostKeyCheckMode mode;
 
@@ -XXX,XX +XXX,XX @@ static int check_host_key(BDRVSSHState *s, const char *host, int port,
     case SSH_HOST_KEY_CHECK_MODE_HASH:
         if (hkc->u.hash.type == SSH_HOST_KEY_CHECK_HASH_TYPE_MD5) {
             return check_host_key_hash(s, hkc->u.hash.hash,
-                                       LIBSSH2_HOSTKEY_HASH_MD5, 16, errp);
+                                       SSH_PUBLICKEY_HASH_MD5, errp);
         } else if (hkc->u.hash.type == SSH_HOST_KEY_CHECK_HASH_TYPE_SHA1) {
             return check_host_key_hash(s, hkc->u.hash.hash,
-                                       LIBSSH2_HOSTKEY_HASH_SHA1, 20, errp);
+                                       SSH_PUBLICKEY_HASH_SHA1, errp);
         }
         g_assert_not_reached();
         break;
     case SSH_HOST_KEY_CHECK_MODE_KNOWN_HOSTS:
-        return check_host_key_knownhosts(s, host, port, errp);
+        return check_host_key_knownhosts(s, errp);
     default:
         g_assert_not_reached();
     }
@@ -XXX,XX +XXX,XX @@ static int check_host_key(BDRVSSHState *s, const char *host, int port,
     return -EINVAL;
 }
 
-static int authenticate(BDRVSSHState *s, const char *user, Error **errp)
+static int authenticate(BDRVSSHState *s, Error **errp)
 {
     int r, ret;
-    const char *userauthlist;
-    LIBSSH2_AGENT *agent = NULL;
-    struct libssh2_agent_publickey *identity;
-    struct libssh2_agent_publickey *prev_identity = NULL;
+    int method;
 
-    userauthlist = libssh2_userauth_list(s->session, user, strlen(user));
-    if (strstr(userauthlist, "publickey") == NULL) {
+    /* Try to authenticate with the "none" method. */
+    r = ssh_userauth_none(s->session, NULL);
+    if (r == SSH_AUTH_ERROR) {
         ret = -EPERM;
-        error_setg(errp,
-                "remote server does not support \"publickey\" authentication");
+        session_error_setg(errp, s, "failed to authenticate using none "
+                                    "authentication");
         goto out;
-    }
-
-    /* Connect to ssh-agent and try each identity in turn. */
-    agent = libssh2_agent_init(s->session);
-    if (!agent) {
-        ret = -EINVAL;
-        session_error_setg(errp, s, "failed to initialize ssh-agent support");
-        goto out;
-    }
-    if (libssh2_agent_connect(agent)) {
-        ret = -ECONNREFUSED;
-        session_error_setg(errp, s, "failed to connect to ssh-agent");
-        goto out;
-    }
-    if (libssh2_agent_list_identities(agent)) {
-        ret = -EINVAL;
-        session_error_setg(errp, s,
-                           "failed requesting identities from ssh-agent");
+    } else if (r == SSH_AUTH_SUCCESS) {
+        /* Authenticated! */
+        ret = 0;
         goto out;
     }
 
-    for(;;) {
-        r = libssh2_agent_get_identity(agent, &identity, prev_identity);
-        if (r == 1) {           /* end of list */
-            break;
-        }
-        if (r < 0) {
+    method = ssh_userauth_list(s->session, NULL);
+    trace_ssh_auth_methods(method);
+
+    /*
+     * Try to authenticate with publickey, using the ssh-agent
+     * if available.
+     */
+    if (method & SSH_AUTH_METHOD_PUBLICKEY) {
+        r = ssh_userauth_publickey_auto(s->session, NULL, NULL);
+        if (r == SSH_AUTH_ERROR) {
             ret = -EINVAL;
-            session_error_setg(errp, s,
-                               "failed to obtain identity from ssh-agent");
+            session_error_setg(errp, s, "failed to authenticate using "
+                                        "publickey authentication");
             goto out;
-        }
-        r = libssh2_agent_userauth(agent, user, identity);
-        if (r == 0) {
+        } else if (r == SSH_AUTH_SUCCESS) {
             /* Authenticated! */
             ret = 0;
             goto out;
         }
-        /* Failed to authenticate with this identity, try the next one. */
-        prev_identity = identity;
     }
 
     ret = -EPERM;
@@ -XXX,XX +XXX,XX @@ static int authenticate(BDRVSSHState *s, const char *user, Error **errp)
                "and the identities held by your ssh-agent");
 
  out:
-    if (agent != NULL) {
-        /* Note: libssh2 implementation implicitly calls
-         * libssh2_agent_disconnect if necessary.
-         */
-        libssh2_agent_free(agent);
-    }
-
     return ret;
 }
 
@@ -XXX,XX +XXX,XX @@ static int connect_to_ssh(BDRVSSHState *s, BlockdevOptionsSsh *opts,
                           int ssh_flags, int creat_mode, Error **errp)
 {
     int r, ret;
-    long port = 0;
+    unsigned int port = 0;
+    int new_sock = -1;
 
     if (opts->has_user) {
         s->user = g_strdup(opts->user);
@@ -XXX,XX +XXX,XX @@ static int connect_to_ssh(BDRVSSHState *s, BlockdevOptionsSsh *opts,
     s->inet = opts->server;
     opts->server = NULL;
 
-    if (qemu_strtol(s->inet->port, NULL, 10, &port) < 0) {
+    if (qemu_strtoui(s->inet->port, NULL, 10, &port) < 0) {
         error_setg(errp, "Use only numeric port value");
         ret = -EINVAL;
         goto err;
     }
 
     /* Open the socket and connect. */
-    s->sock = inet_connect_saddr(s->inet, errp);
-    if (s->sock < 0) {
+    new_sock = inet_connect_saddr(s->inet, errp);
+    if (new_sock < 0) {
         ret = -EIO;
         goto err;
     }
 
+    /*
+     * Try to disable the Nagle algorithm on TCP sockets to reduce latency,
+     * but do not fail if it cannot be disabled.
+     */
+    r = socket_set_nodelay(new_sock);
+    if (r < 0) {
+        warn_report("can't set TCP_NODELAY for the ssh server %s: %s",
+                    s->inet->host, strerror(errno));
+    }
+
     /* Create SSH session. */
-    s->session = libssh2_session_init();
+    s->session = ssh_new();
     if (!s->session) {
         ret = -EINVAL;
-        session_error_setg(errp, s, "failed to initialize libssh2 session");
+        session_error_setg(errp, s, "failed to initialize libssh session");
         goto err;
     }
 
-#if TRACE_LIBSSH2 != 0
-    libssh2_trace(s->session, TRACE_LIBSSH2);
-#endif
+    /*
+     * Make sure we are in blocking mode during the connection and
+     * authentication phases.
+     */
+    ssh_set_blocking(s->session, 1);
 
-    r = libssh2_session_handshake(s->session, s->sock);
-    if (r != 0) {
+    r = ssh_options_set(s->session, SSH_OPTIONS_USER, s->user);
+    if (r < 0) {
+        ret = -EINVAL;
+        session_error_setg(errp, s,
+                           "failed to set the user in the libssh session");
+        goto err;
+    }
+
+    r = ssh_options_set(s->session, SSH_OPTIONS_HOST, s->inet->host);
+    if (r < 0) {
+        ret = -EINVAL;
+        session_error_setg(errp, s,
+                           "failed to set the host in the libssh session");
+        goto err;
+    }
+
+    if (port > 0) {
+        r = ssh_options_set(s->session, SSH_OPTIONS_PORT, &port);
+        if (r < 0) {
+            ret = -EINVAL;
+            session_error_setg(errp, s,
+                               "failed to set the port in the libssh session");
+            goto err;
+        }
+    }
+
+    r = ssh_options_set(s->session, SSH_OPTIONS_COMPRESSION, "none");
+    if (r < 0) {
+        ret = -EINVAL;
+        session_error_setg(errp, s,
+                           "failed to disable the compression in the libssh "
+                           "session");
+        goto err;
+    }
+
+    /* Read ~/.ssh/config. */
+    r = ssh_options_parse_config(s->session, NULL);
+    if (r < 0) {
+        ret = -EINVAL;
+        session_error_setg(errp, s, "failed to parse ~/.ssh/config");
+        goto err;
+    }
+
+    r = ssh_options_set(s->session, SSH_OPTIONS_FD, &new_sock);
+    if (r < 0) {
+        ret = -EINVAL;
+        session_error_setg(errp, s,
+                           "failed to set the socket in the libssh session");
+        goto err;
+    }
+    /* libssh took ownership of the socket. */
+    s->sock = new_sock;
+    new_sock = -1;
+
+    /* Connect. */
+    r = ssh_connect(s->session);
+    if (r != SSH_OK) {
         ret = -EINVAL;
         session_error_setg(errp, s, "failed to establish SSH session");
         goto err;
     }
 
     /* Check the remote host's key against known_hosts. */
-    ret = check_host_key(s, s->inet->host, port, opts->host_key_check, errp);
+    ret = check_host_key(s, opts->host_key_check, errp);
     if (ret < 0) {
         goto err;
     }
 
     /* Authenticate. */
-    ret = authenticate(s, s->user, errp);
+    ret = authenticate(s, errp);
     if (ret < 0) {
         goto err;
     }
 
     /* Start SFTP. */
-    s->sftp = libssh2_sftp_init(s->session);
+    s->sftp = sftp_new(s->session);
     if (!s->sftp) {
-        session_error_setg(errp, s, "failed to initialize sftp handle");
+        session_error_setg(errp, s, "failed to create sftp handle");
+        ret = -EINVAL;
+        goto err;
+    }
+
+    r = sftp_init(s->sftp);
+    if (r < 0) {
+        sftp_error_setg(errp, s, "failed to initialize sftp handle");
         ret = -EINVAL;
         goto err;
     }
 
     /* Open the remote file. */
     trace_ssh_connect_to_ssh(opts->path, ssh_flags, creat_mode);
-    s->sftp_handle = libssh2_sftp_open(s->sftp, opts->path, ssh_flags,
-                                       creat_mode);
+    s->sftp_handle = sftp_open(s->sftp, opts->path, ssh_flags, creat_mode);
     if (!s->sftp_handle) {
-        session_error_setg(errp, s, "failed to open remote file '%s'",
-                           opts->path);
+        sftp_error_setg(errp, s, "failed to open remote file '%s'",
+                        opts->path);
         ret = -EINVAL;
         goto err;
     }
 
-    r = libssh2_sftp_fstat(s->sftp_handle, &s->attrs);
-    if (r < 0) {
+    /* Make sure the SFTP file is handled in blocking mode. */
+    sftp_file_set_blocking(s->sftp_handle);
+
+    s->attrs = sftp_fstat(s->sftp_handle);
+    if (!s->attrs) {
         sftp_error_setg(errp, s, "failed to read file attributes");
         return -EINVAL;
     }
@@ -XXX,XX +XXX,XX @@ static int connect_to_ssh(BDRVSSHState *s, BlockdevOptionsSsh *opts,
     return 0;
 
  err:
+    if (s->attrs) {
+        sftp_attributes_free(s->attrs);
+    }
+    s->attrs = NULL;
     if (s->sftp_handle) {
-        libssh2_sftp_close(s->sftp_handle);
+        sftp_close(s->sftp_handle);
     }
     s->sftp_handle = NULL;
     if (s->sftp) {
-        libssh2_sftp_shutdown(s->sftp);
+        sftp_free(s->sftp);
     }
     s->sftp = NULL;
     if (s->session) {
-        libssh2_session_disconnect(s->session,
-                                   "from qemu ssh client: "
-                                   "error opening connection");
-        libssh2_session_free(s->session);
+        ssh_disconnect(s->session);
+        ssh_free(s->session);
     }
     s->session = NULL;
+    s->sock = -1;
+    if (new_sock >= 0) {
+        close(new_sock);
+    }
 
     return ret;
 }
@@ -XXX,XX +XXX,XX @@ static int ssh_file_open(BlockDriverState *bs, QDict *options, int bdrv_flags,
 
     ssh_state_init(s);
 
-    ssh_flags = LIBSSH2_FXF_READ;
+    ssh_flags = 0;
     if (bdrv_flags & BDRV_O_RDWR) {
-        ssh_flags |= LIBSSH2_FXF_WRITE;
+        ssh_flags |= O_RDWR;
+    } else {
+        ssh_flags |= O_RDONLY;
     }
 
     opts = ssh_parse_options(options, errp);
@@ -XXX,XX +XXX,XX @@ static int ssh_file_open(BlockDriverState *bs, QDict *options, int bdrv_flags,
     }
 
     /* Go non-blocking. */
-    libssh2_session_set_blocking(s->session, 0);
+    ssh_set_blocking(s->session, 0);
 
     qapi_free_BlockdevOptionsSsh(opts);
 
     return 0;
 
  err:
-    if (s->sock >= 0) {
-        close(s->sock);
-    }
-    s->sock = -1;
-
     qapi_free_BlockdevOptionsSsh(opts);
 
     return ret;
@@ -XXX,XX +XXX,XX @@ static int ssh_grow_file(BDRVSSHState *s, int64_t offset, Error **errp)
 {
     ssize_t ret;
     char c[1] = { '\0' };
-    int was_blocking = libssh2_session_get_blocking(s->session);
+    int was_blocking = ssh_is_blocking(s->session);
 
     /* offset must be strictly greater than the current size so we do
      * not overwrite anything */
-    assert(offset > 0 && offset > s->attrs.filesize);
+    assert(offset > 0 && offset > s->attrs->size);
 
-    libssh2_session_set_blocking(s->session, 1);
+    ssh_set_blocking(s->session, 1);
 
-    libssh2_sftp_seek64(s->sftp_handle, offset - 1);
-    ret = libssh2_sftp_write(s->sftp_handle, c, 1);
+    sftp_seek64(s->sftp_handle, offset - 1);
+    ret = sftp_write(s->sftp_handle, c, 1);
 
-    libssh2_session_set_blocking(s->session, was_blocking);
+    ssh_set_blocking(s->session, was_blocking);
 
     if (ret < 0) {
         sftp_error_setg(errp, s, "Failed to grow file");
         return -EIO;
     }
 
-    s->attrs.filesize = offset;
+    s->attrs->size = offset;
     return 0;
 }
 
@@ -XXX,XX +XXX,XX @@ static int ssh_co_create(BlockdevCreateOptions *options, Error **errp)
     ssh_state_init(&s);
 
     ret = connect_to_ssh(&s, opts->location,
-                         LIBSSH2_FXF_READ|LIBSSH2_FXF_WRITE|
-                         LIBSSH2_FXF_CREAT|LIBSSH2_FXF_TRUNC,
+                         O_RDWR | O_CREAT | O_TRUNC,
                          0644, errp);
     if (ret < 0) {
         goto fail;
@@ -XXX,XX +XXX,XX @@ static int ssh_has_zero_init(BlockDriverState *bs)
     /* Assume false, unless we can positively prove it's true. */
     int has_zero_init = 0;
 
-    if (s->attrs.flags & LIBSSH2_SFTP_ATTR_PERMISSIONS) {
-        if (s->attrs.permissions & LIBSSH2_SFTP_S_IFREG) {
-            has_zero_init = 1;
-        }
+    if (s->attrs->type == SSH_FILEXFER_TYPE_REGULAR) {
+        has_zero_init = 1;
     }
 
     return has_zero_init;
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void co_yield(BDRVSSHState *s, BlockDriverState *bs)
         .co = qemu_coroutine_self()
     };
 
-    r = libssh2_session_block_directions(s->session);
+    r = ssh_get_poll_flags(s->session);
 
-    if (r & LIBSSH2_SESSION_BLOCK_INBOUND) {
+    if (r & SSH_READ_PENDING) {
         rd_handler = restart_coroutine;
     }
-    if (r & LIBSSH2_SESSION_BLOCK_OUTBOUND) {
+    if (r & SSH_WRITE_PENDING) {
         wr_handler = restart_coroutine;
     }
 
@@ -XXX,XX +XXX,XX @@ static coroutine_fn void co_yield(BDRVSSHState *s, BlockDriverState *bs)
     trace_ssh_co_yield_back(s->sock);
 }
 
-/* SFTP has a function `libssh2_sftp_seek64' which seeks to a position
- * in the remote file.  Notice that it just updates a field in the
- * sftp_handle structure, so there is no network traffic and it cannot
- * fail.
- *
- * However, `libssh2_sftp_seek64' does have a catastrophic effect on
- * performance since it causes the handle to throw away all in-flight
- * reads and buffered readahead data.  Therefore this function tries
- * to be intelligent about when to call the underlying libssh2 function.
- */
-#define SSH_SEEK_WRITE 0
-#define SSH_SEEK_READ  1
-#define SSH_SEEK_FORCE 2
-
-static void ssh_seek(BDRVSSHState *s, int64_t offset, int flags)
-{
-    bool op_read = (flags & SSH_SEEK_READ) != 0;
-    bool force = (flags & SSH_SEEK_FORCE) != 0;
-
-    if (force || op_read != s->offset_op_read || offset != s->offset) {
-        trace_ssh_seek(offset);
-        libssh2_sftp_seek64(s->sftp_handle, offset);
-        s->offset = offset;
-        s->offset_op_read = op_read;
-    }
-}
-
 static coroutine_fn int ssh_read(BDRVSSHState *s, BlockDriverState *bs,
                                  int64_t offset, size_t size,
                                  QEMUIOVector *qiov)
@@ -XXX,XX +XXX,XX @@ static coroutine_fn int ssh_read(BDRVSSHState *s, BlockDriverState *bs,
 
     trace_ssh_read(offset, size);
 
-    ssh_seek(s, offset, SSH_SEEK_READ);
+    trace_ssh_seek(offset);
+    sftp_seek64(s->sftp_handle, offset);
 
     /* This keeps track of the current iovec element ('i'), where we
      * will write to next ('buf'), and the end of the current iovec
@@ -XXX,XX +XXX,XX @@ static coroutine_fn int ssh_read(BDRVSSHState *s, BlockDriverState *bs,
     buf = i->iov_base;
     end_of_vec = i->iov_base + i->iov_len;
 
-    /* libssh2 has a hard-coded limit of 2000 bytes per request,
-     * although it will also do readahead behind our backs.  Therefore
-     * we may have to do repeated reads here until we have read 'size'
-     * bytes.
-     */
     for (got = 0; got < size; ) {
+        size_t request_read_size;
     again:
-        trace_ssh_read_buf(buf, end_of_vec - buf);
-        r = libssh2_sftp_read(s->sftp_handle, buf, end_of_vec - buf);
-        trace_ssh_read_return(r);
+        /*
+         * The size of SFTP packets is limited to 32K bytes, so limit
+         * the amount of data requested to 16K, as libssh currently
+         * does not handle multiple requests on its own.
+         */
+        request_read_size = MIN(end_of_vec - buf, 16384);
+        trace_ssh_read_buf(buf, end_of_vec - buf, request_read_size);
+        r = sftp_read(s->sftp_handle, buf, request_read_size);
+        trace_ssh_read_return(r, sftp_get_error(s->sftp));
 
-        if (r == LIBSSH2_ERROR_EAGAIN || r == LIBSSH2_ERROR_TIMEOUT) {
+        if (r == SSH_AGAIN) {
             co_yield(s, bs);
             goto again;
         }
-        if (r < 0) {
-            sftp_error_trace(s, "read");
-            s->offset = -1;
-            return -EIO;
-        }
-        if (r == 0) {
+        if (r == SSH_EOF || (r == 0 && sftp_get_error(s->sftp) == SSH_FX_EOF)) {
             /* EOF: Short read so pad the buffer with zeroes and return it. */
             qemu_iovec_memset(qiov, got, 0, size - got);
             return 0;
         }
+        if (r <= 0) {
+            sftp_error_trace(s, "read");
+            return -EIO;
+        }
 
         got += r;
         buf += r;
-        s->offset += r;
         if (buf >= end_of_vec && got < size) {
             i++;
             buf = i->iov_base;
@@ -XXX,XX +XXX,XX @@ static int ssh_write(BDRVSSHState *s, BlockDriverState *bs,
 
     trace_ssh_write(offset, size);
 
-    ssh_seek(s, offset, SSH_SEEK_WRITE);
+    trace_ssh_seek(offset);
+    sftp_seek64(s->sftp_handle, offset);
 
     /* This keeps track of the current iovec element ('i'), where we
      * will read from next ('buf'), and the end of the current iovec
@@ -XXX,XX +XXX,XX @@ static int ssh_write(BDRVSSHState *s, BlockDriverState *bs,
     end_of_vec = i->iov_base + i->iov_len;
 
     for (written = 0; written < size; ) {
+        size_t request_write_size;
     again:
-        trace_ssh_write_buf(buf, end_of_vec - buf);
-        r = libssh2_sftp_write(s->sftp_handle, buf, end_of_vec - buf);
-        trace_ssh_write_return(r);
+        /*
+         * Avoid too large data packets, as libssh currently does not
+         * handle multiple requests on its own.
+         */
+        request_write_size = MIN(end_of_vec - buf, 131072);
+        trace_ssh_write_buf(buf, end_of_vec - buf, request_write_size);
+        r = sftp_write(s->sftp_handle, buf, request_write_size);
+        trace_ssh_write_return(r, sftp_get_error(s->sftp));
 
-        if (r == LIBSSH2_ERROR_EAGAIN || r == LIBSSH2_ERROR_TIMEOUT) {
+        if (r == SSH_AGAIN) {
             co_yield(s, bs);
             goto again;
         }
         if (r < 0) {
             sftp_error_trace(s, "write");
-            s->offset = -1;
             return -EIO;
         }
-        /* The libssh2 API is very unclear about this.  A comment in
-         * the code says "nothing was acked, and no EAGAIN was
-         * received!" which apparently means that no data got sent
-         * out, and the underlying channel didn't return any EAGAIN
-         * indication.  I think this is a bug in either libssh2 or
-         * OpenSSH (server-side).  In any case, forcing a seek (to
-         * discard libssh2 internal buffers), and then trying again
-         * works for me.
-         */
-        if (r == 0) {
-            ssh_seek(s, offset + written, SSH_SEEK_WRITE|SSH_SEEK_FORCE);
-            co_yield(s, bs);
-            goto again;
-        }
 
         written += r;
         buf += r;
-        s->offset += r;
         if (buf >= end_of_vec && written < size) {
             i++;
             buf = i->iov_base;
             end_of_vec = i->iov_base + i->iov_len;
         }
 
-        if (offset + written > s->attrs.filesize)
-            s->attrs.filesize = offset + written;
+        if (offset + written > s->attrs->size) {
+            s->attrs->size = offset + written;
+        }
     }
 
     return 0;
@@ -XXX,XX +XXX,XX @@ static void unsafe_flush_warning(BDRVSSHState *s, const char *what)
     }
 }
 
-#ifdef HAS_LIBSSH2_SFTP_FSYNC
+#ifdef HAVE_LIBSSH_0_8
 
 static coroutine_fn int ssh_flush(BDRVSSHState *s, BlockDriverState *bs)
 {
     int r;
 
     trace_ssh_flush();
+
+    if (!sftp_extension_supported(s->sftp, "fsync@openssh.com", "1")) {
+        unsafe_flush_warning(s, "OpenSSH >= 6.3");
+        return 0;
+    }
  again:
-    r = libssh2_sftp_fsync(s->sftp_handle);
-    if (r == LIBSSH2_ERROR_EAGAIN || r == LIBSSH2_ERROR_TIMEOUT) {
+    r = sftp_fsync(s->sftp_handle);
+    if (r == SSH_AGAIN) {
         co_yield(s, bs);
         goto again;
     }
-    if (r == LIBSSH2_ERROR_SFTP_PROTOCOL &&
-        libssh2_sftp_last_error(s->sftp) == LIBSSH2_FX_OP_UNSUPPORTED) {
-        unsafe_flush_warning(s, "OpenSSH >= 6.3");
-        return 0;
-    }
     if (r < 0) {
         sftp_error_trace(s, "fsync");
         return -EIO;
@@ -XXX,XX +XXX,XX @@ static coroutine_fn int ssh_co_flush(BlockDriverState *bs)
     return ret;
 }
 
-#else /* !HAS_LIBSSH2_SFTP_FSYNC */
+#else /* !HAVE_LIBSSH_0_8 */
 
 static coroutine_fn int ssh_co_flush(BlockDriverState *bs)
 {
     BDRVSSHState *s = bs->opaque;
 
-    unsafe_flush_warning(s, "libssh2 >= 1.4.4");
+    unsafe_flush_warning(s, "libssh >= 0.8.0");
     return 0;
 }
 
-#endif /* !HAS_LIBSSH2_SFTP_FSYNC */
+#endif /* !HAVE_LIBSSH_0_8 */
 
 static int64_t ssh_getlength(BlockDriverState *bs)
 {
     BDRVSSHState *s = bs->opaque;
     int64_t length;
 
-    /* Note we cannot make a libssh2 call here. */
-    length = (int64_t) s->attrs.filesize;
+    /* Note we cannot make a libssh call here. */
+    length = (int64_t) s->attrs->size;
     trace_ssh_getlength(length);
 
     return length;
@@ -XXX,XX +XXX,XX @@ static int coroutine_fn ssh_co_truncate(BlockDriverState *bs, int64_t offset,
         return -ENOTSUP;
     }
 
-    if (offset < s->attrs.filesize) {
+    if (offset < s->attrs->size) {
         error_setg(errp, "ssh driver does not support shrinking files");
         return -ENOTSUP;
     }
 
-    if (offset == s->attrs.filesize) {
+    if (offset == s->attrs->size) {
         return 0;
     }
 
@@ -XXX,XX +XXX,XX @@ static void bdrv_ssh_init(void)
 {
     int r;
 
-    r = libssh2_init(0);
+    r = ssh_init();
     if (r != 0) {
-        fprintf(stderr, "libssh2 initialization failed, %d\n", r);
+        fprintf(stderr, "libssh initialization failed, %d\n", r);
         exit(EXIT_FAILURE);
     }
 
+#if TRACE_LIBSSH != 0
+    ssh_set_log_level(TRACE_LIBSSH);
+#endif
+
     bdrv_register(&bdrv_ssh);
 }
 
diff --git a/.travis.yml b/.travis.yml
index XXXXXXX..XXXXXXX 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -XXX,XX +XXX,XX @@ addons:
       - libseccomp-dev
       - libspice-protocol-dev
       - libspice-server-dev
-      - libssh2-1-dev
+      - libssh-dev
       - liburcu-dev
       - libusb-1.0-0-dev
       - libvte-2.91-dev
@@ -XXX,XX +XXX,XX @@ matrix:
             - libseccomp-dev
             - libspice-protocol-dev
             - libspice-server-dev
-            - libssh2-1-dev
+            - libssh-dev
             - liburcu-dev
             - libusb-1.0-0-dev
             - libvte-2.91-dev
diff --git a/block/trace-events b/block/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/block/trace-events
+++ b/block/trace-events
@@ -XXX,XX +XXX,XX @@ nbd_client_connect_success(const char *export_name) "export '%s'"
 # ssh.c
 ssh_restart_coroutine(void *co) "co=%p"
 ssh_flush(void) "fsync"
-ssh_check_host_key_knownhosts(const char *key) "host key OK: %s"
+ssh_check_host_key_knownhosts(void) "host key OK"
 ssh_connect_to_ssh(char *path, int flags, int mode) "opening file %s flags=0x%x creat_mode=0%o"
 ssh_co_yield(int sock, void *rd_handler, void *wr_handler) "s->sock=%d rd_handler=%p wr_handler=%p"
 ssh_co_yield_back(int sock) "s->sock=%d - back"
 ssh_getlength(int64_t length) "length=%" PRIi64
 ssh_co_create_opts(uint64_t size) "total_size=%" PRIu64
 ssh_read(int64_t offset, size_t size) "offset=%" PRIi64 " size=%zu"
-ssh_read_buf(void *buf, size_t size) "sftp_read buf=%p size=%zu"
-ssh_read_return(ssize_t ret) "sftp_read returned %zd"
+ssh_read_buf(void *buf, size_t size, size_t actual_size) "sftp_read buf=%p size=%zu (actual size=%zu)"
+ssh_read_return(ssize_t ret, int sftp_err) "sftp_read returned %zd (sftp error=%d)"
 ssh_write(int64_t offset, size_t size) "offset=%" PRIi64 " size=%zu"
-ssh_write_buf(void *buf, size_t size) "sftp_write buf=%p size=%zu"
-ssh_write_return(ssize_t ret) "sftp_write returned %zd"
+ssh_write_buf(void *buf, size_t size, size_t actual_size) "sftp_write buf=%p size=%zu (actual size=%zu)"
+ssh_write_return(ssize_t ret, int sftp_err) "sftp_write returned %zd (sftp error=%d)"
 ssh_seek(int64_t offset) "seeking to offset=%" PRIi64
+ssh_auth_methods(int methods) "auth methods=0x%x"
+ssh_server_status(int status) "server status=%d"
 
 # curl.c
 curl_timer_cb(long timeout_ms) "timer callback timeout_ms %ld"
@@ -XXX,XX +XXX,XX @@ sheepdog_snapshot_create(const char *sn_name, const char *id) "%s %s"
 sheepdog_snapshot_create_inode(const char *name, uint32_t snap, uint32_t vdi) "s->inode: name %s snap_id 0x%" PRIx32 " vdi 0x%" PRIx32
 
 # ssh.c
-sftp_error(const char *op, const char *ssh_err, int ssh_err_code, unsigned long sftp_err_code) "%s failed: %s (libssh2 error code: %d, sftp error code: %lu)"
+sftp_error(const char *op, const char *ssh_err, int ssh_err_code, int sftp_err_code) "%s failed: %s (libssh error code: %d, sftp error code: %d)"
diff --git a/docs/qemu-block-drivers.texi b/docs/qemu-block-drivers.texi
index XXXXXXX..XXXXXXX 100644
--- a/docs/qemu-block-drivers.texi
+++ b/docs/qemu-block-drivers.texi
@@ -XXX,XX +XXX,XX @@ print a warning when @code{fsync} is not supported:
 
 warning: ssh server @code{ssh.example.com:22} does not support fsync
 
-With sufficiently new versions of libssh2 and OpenSSH, @code{fsync} is
+With sufficiently new versions of libssh and OpenSSH, @code{fsync} is
 supported.
 
 @node disk_images_nvme
diff --git a/tests/docker/dockerfiles/debian-win32-cross.docker b/tests/docker/dockerfiles/debian-win32-cross.docker
index XXXXXXX..XXXXXXX 100644
--- a/tests/docker/dockerfiles/debian-win32-cross.docker
+++ b/tests/docker/dockerfiles/debian-win32-cross.docker
@@ -XXX,XX +XXX,XX @@ RUN DEBIAN_FRONTEND=noninteractive eatmydata \
         mxe-$TARGET-w64-mingw32.shared-curl \
         mxe-$TARGET-w64-mingw32.shared-glib \
         mxe-$TARGET-w64-mingw32.shared-libgcrypt \
-        mxe-$TARGET-w64-mingw32.shared-libssh2 \
         mxe-$TARGET-w64-mingw32.shared-libusb1 \
         mxe-$TARGET-w64-mingw32.shared-lzo \
         mxe-$TARGET-w64-mingw32.shared-nettle \
diff --git a/tests/docker/dockerfiles/debian-win64-cross.docker b/tests/docker/dockerfiles/debian-win64-cross.docker
index XXXXXXX..XXXXXXX 100644
--- a/tests/docker/dockerfiles/debian-win64-cross.docker
+++ b/tests/docker/dockerfiles/debian-win64-cross.docker
@@ -XXX,XX +XXX,XX @@ RUN DEBIAN_FRONTEND=noninteractive eatmydata \
         mxe-$TARGET-w64-mingw32.shared-curl \
         mxe-$TARGET-w64-mingw32.shared-glib \
         mxe-$TARGET-w64-mingw32.shared-libgcrypt \
-        mxe-$TARGET-w64-mingw32.shared-libssh2 \
         mxe-$TARGET-w64-mingw32.shared-libusb1 \
         mxe-$TARGET-w64-mingw32.shared-lzo \
         mxe-$TARGET-w64-mingw32.shared-nettle \
diff --git a/tests/docker/dockerfiles/fedora.docker b/tests/docker/dockerfiles/fedora.docker
index XXXXXXX..XXXXXXX 100644
--- a/tests/docker/dockerfiles/fedora.docker
+++ b/tests/docker/dockerfiles/fedora.docker
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES \
     libpng-devel \
     librbd-devel \
     libseccomp-devel \
-    libssh2-devel \
+    libssh-devel \
     libubsan \
     libusbx-devel \
     libxml2-devel \
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES \
     mingw32-gtk3 \
     mingw32-libjpeg-turbo \
     mingw32-libpng \
-    mingw32-libssh2 \
     mingw32-libtasn1 \
     mingw32-nettle \
     mingw32-pixman \
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES \
     mingw64-gtk3 \
     mingw64-libjpeg-turbo \
     mingw64-libpng \
-    mingw64-libssh2 \
     mingw64-libtasn1 \
     mingw64-nettle \
     mingw64-pixman \
diff --git a/tests/docker/dockerfiles/ubuntu.docker b/tests/docker/dockerfiles/ubuntu.docker
index XXXXXXX..XXXXXXX 100644
--- a/tests/docker/dockerfiles/ubuntu.docker
+++ b/tests/docker/dockerfiles/ubuntu.docker
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES flex bison \
     libsnappy-dev \
     libspice-protocol-dev \
     libspice-server-dev \
-    libssh2-1-dev \
+    libssh-dev \
     libusb-1.0-0-dev \
     libusbredirhost-dev \
     libvdeplug-dev \
diff --git a/tests/docker/dockerfiles/ubuntu1804.docker b/tests/docker/dockerfiles/ubuntu1804.docker
index XXXXXXX..XXXXXXX 100644
--- a/tests/docker/dockerfiles/ubuntu1804.docker
+++ b/tests/docker/dockerfiles/ubuntu1804.docker
@@ -XXX,XX +XXX,XX @@ ENV PACKAGES flex bison \
     libsnappy-dev \
     libspice-protocol-dev \
     libspice-server-dev \
-    libssh2-1-dev \
+    libssh-dev \
     libusb-1.0-0-dev \
     libusbredirhost-dev \
     libvdeplug-dev \
diff --git a/tests/qemu-iotests/207 b/tests/qemu-iotests/207
index XXXXXXX..XXXXXXX 100755
--- a/tests/qemu-iotests/207
+++ b/tests/qemu-iotests/207
@@ -XXX,XX +XXX,XX @@ with iotests.FilePath('t.img') as disk_path, \
 
     iotests.img_info_log(remote_path)
 
-    md5_key = subprocess.check_output(
-        'ssh-keyscan -t rsa 127.0.0.1 2>/dev/null | grep -v "\\^#" | ' +
-        'cut -d" " -f3 | base64 -d | md5sum -b | cut -d" " -f1',
-        shell=True).rstrip().decode('ascii')
+    keys = subprocess.check_output(
+        'ssh-keyscan 127.0.0.1 2>/dev/null | grep -v "\\^#" | ' +
+        'cut -d" " -f3',
+        shell=True).rstrip().decode('ascii').split('\n')
+
+    # Mappings of base64 representations to digests
+    md5_keys = {}
+    sha1_keys = {}
+
+    for key in keys:
+        md5_keys[key] = subprocess.check_output(
+            'echo %s | base64 -d | md5sum -b | cut -d" " -f1' % key,
+            shell=True).rstrip().decode('ascii')
+
+        sha1_keys[key] = subprocess.check_output(
+            'echo %s | base64 -d | sha1sum -b | cut -d" " -f1' % key,
+            shell=True).rstrip().decode('ascii')
 
     vm.launch()
+
+    # Find correct key first
+    matching_key = None
+    for key in keys:
+        result = vm.qmp('blockdev-add',
+                        driver='ssh', node_name='node0', path=disk_path,
+                        server={
+                             'host': '127.0.0.1',
+                             'port': '22',
+                        }, host_key_check={
+                             'mode': 'hash',
+                             'type': 'md5',
+                             'hash': md5_keys[key],
+                        })
+
+        if 'error' not in result:
+            vm.qmp('blockdev-del', node_name='node0')
+            matching_key = key
+            break
+
+    if matching_key is None:
+        vm.shutdown()
+        iotests.notrun('Did not find a key that fits 127.0.0.1')
+
     blockdev_create(vm, { 'driver': 'ssh',
                           'location': {
                               'path': disk_path,
@@ -XXX,XX +XXX,XX @@ with iotests.FilePath('t.img') as disk_path, \
                               'host-key-check': {
                                   'mode': 'hash',
                                   'type': 'md5',
-                                  'hash': md5_key,
+                                  'hash': md5_keys[matching_key],
                               }
                           },
                           'size': 8388608 })
@@ -XXX,XX +XXX,XX @@ with iotests.FilePath('t.img') as disk_path, \
 
     iotests.img_info_log(remote_path)
 
-    sha1_key = subprocess.check_output(
-        'ssh-keyscan -t rsa 127.0.0.1 2>/dev/null | grep -v "\\^#" | ' +
-        'cut -d" " -f3 | base64 -d | sha1sum -b | cut -d" " -f1',
-        shell=True).rstrip().decode('ascii')
-
     vm.launch()
     blockdev_create(vm, { 'driver': 'ssh',
                           'location': {
@@ -XXX,XX +XXX,XX @@ with iotests.FilePath('t.img') as disk_path, \
                               'host-key-check': {
                                   'mode': 'hash',
                                   'type': 'sha1',
-                                  'hash': sha1_key,
+                                  'hash': sha1_keys[matching_key],
                               }
                           },
                           'size': 4194304 })
diff --git a/tests/qemu-iotests/207.out b/tests/qemu-iotests/207.out
index XXXXXXX..XXXXXXX 100644
--- a/tests/qemu-iotests/207.out
+++ b/tests/qemu-iotests/207.out
@@ -XXX,XX +XXX,XX @@ virtual size: 4 MiB (4194304 bytes)
 
 {"execute": "blockdev-create", "arguments": {"job-id": "job0", "options": {"driver": "ssh", "location": {"host-key-check": {"mode": "none"}, "path": "/this/is/not/an/existing/path", "server": {"host": "127.0.0.1", "port": "22"}}, "size": 4194304}}}
 {"return": {}}
-Job failed: failed to open remote file '/this/is/not/an/existing/path': Failed opening remote file (libssh2 error code: -31)
+Job failed: failed to open remote file '/this/is/not/an/existing/path': SFTP server: No such file (libssh error code: 1, sftp error code: 2)
 {"execute": "job-dismiss", "arguments": {"id": "job0"}}
 {"return": {}}
 
-- 
2.21.0