The following changes since commit a26a98dfb9d448d7234d931ae3720feddf6f0651: Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20171006' into staging (2017-10-06 13:19:03 +0100) are available in the git repository at: git://repo.or.cz/qemu/kevin.git tags/for-upstream for you to fetch changes up to fc3fd63fc0573ffd2ee569591a2e7f6c7310fd18: Merge remote-tracking branch 'mreitz/tags/pull-block-2017-10-06' into queue-block (2017-10-06 16:32:08 +0200) ---------------------------------------------------------------- Block layer patches ---------------------------------------------------------------- Daniel P. Berrange (6): block: use 1 MB bounce buffers for crypto instead of 16KB crypto: expose encryption sector size in APIs block: fix data type casting for crypto payload offset block: convert crypto driver to bdrv_co_preadv|pwritev block: convert qcrypto_block_encrypt|decrypt to take bytes offset block: support passthrough of BDRV_REQ_FUA in crypto driver Eric Blake (27): block: Typo fix in copy_on_readv() block: Make bdrv_img_create() size selection easier to read hbitmap: Rename serialization_granularity to serialization_align qcow2: Ensure bitmap serialization is aligned dirty-bitmap: Drop unused functions dirty-bitmap: Avoid size query failure during truncate dirty-bitmap: Change bdrv_dirty_bitmap_size() to report bytes dirty-bitmap: Track bitmap size by bytes dirty-bitmap: Change bdrv_dirty_bitmap_*serialize*() to take bytes qcow2: Switch sectors_covered_by_bitmap_cluster() to byte-based dirty-bitmap: Set iterator start by offset, not sector dirty-bitmap: Change bdrv_dirty_iter_next() to report byte offset dirty-bitmap: Change bdrv_get_dirty_count() to report bytes dirty-bitmap: Change bdrv_get_dirty_locked() to take bytes dirty-bitmap: Change bdrv_[re]set_dirty_bitmap() to use bytes mirror: Switch mirror_dirty_init() to byte-based iteration qcow2: Switch qcow2_measure() to byte-based iteration qcow2: Switch load_bitmap_data() to byte-based iteration qcow2: Switch store_bitmap_data() to byte-based iteration dirty-bitmap: Switch bdrv_set_dirty() to bytes dirty-bitmap: Convert internal hbitmap size/granularity qemu-io: Add -C for opening with copy-on-read block: Uniform handling of 0-length bdrv_get_block_status() iotests: Restore stty settings on completion block: Add blkdebug hook for copy-on-read block: Perform copy-on-read in loop iotests: Add test 197 for covering copy-on-read Kevin Wolf (6): block: Introduce BdrvChildRole.update_filename commit: Support multiple roots above top node qemu-iotests: Allow QMP pretty printing in common.qemu qemu-iotests: Test commit block job where top has two parents commit: Remove overlay_bs Merge remote-tracking branch 'mreitz/tags/pull-block-2017-10-06' into queue-block Max Reitz (1): iotests: Fix 195 if IMGFMT is part of TEST_DIR Paolo Bonzini (10): qemu-iotests: remove dead code qemu-iotests: get rid of AWK_PROG qemu-iotests: move "check" code out of common.rc qemu-iotests: cleanup and fix search for programs qemu-iotests: limit non-_PROG-suffixed variables to common.rc qemu-iotests: do not include common.rc in "check" qemu-iotests: disintegrate more parts of common.config qemu-iotests: fix uninitialized variable qemu-iotests: get rid of $iam qemu-iotests: merge "check" and "common" Pavel Butsykin (2): qcow2: fix return error code in qcow2_truncate() qcow2: truncate the tail of the image file after shrinking the image Thomas Huth (1): hw/block/onenand: Remove dead code block Vladimir Sementsov-Ogievskiy (2): block/mirror: check backing in bdrv_mirror_top_refresh_filename block/mirror: check backing in bdrv_mirror_top_flush qapi/block-core.json | 5 +- block/qcow2.h | 1 + crypto/blockpriv.h | 5 +- include/block/block.h | 3 +- include/block/block_int.h | 8 +- include/block/dirty-bitmap.h | 43 +- include/crypto/block.h | 29 +- include/qemu/hbitmap.h | 8 +- block.c | 109 ++++-- block/backup.c | 7 +- block/commit.c | 64 +-- block/crypto.c | 130 +++--- block/dirty-bitmap.c | 134 ++----- block/io.c | 131 +++++-- block/mirror.c | 88 ++--- block/qcow.c | 11 +- block/qcow2-bitmap.c | 62 +-- block/qcow2-cluster.c | 8 +- block/qcow2-refcount.c | 22 ++ block/qcow2.c | 53 ++- crypto/block-luks.c | 18 +- crypto/block-qcow.c | 13 +- crypto/block.c | 26 +- hw/block/onenand.c | 4 - migration/block.c | 12 +- qemu-io.c | 15 +- tests/test-hbitmap.c | 10 +- util/hbitmap.c | 8 +- tests/qemu-iotests/030 | 4 - tests/qemu-iotests/039.out | 10 +- tests/qemu-iotests/061.out | 4 +- tests/qemu-iotests/137.out | 2 +- tests/qemu-iotests/165 | 2 +- tests/qemu-iotests/191 | 153 ++++++++ tests/qemu-iotests/191.out | 827 +++++++++++++++++++++++++++++++++++++++ tests/qemu-iotests/195 | 7 +- tests/qemu-iotests/197 | 109 ++++++ tests/qemu-iotests/197.out | 26 ++ tests/qemu-iotests/check | 585 ++++++++++++++++++++++++--- tests/qemu-iotests/common | 459 ---------------------- tests/qemu-iotests/common.config | 206 +--------- tests/qemu-iotests/common.filter | 1 + tests/qemu-iotests/common.qemu | 15 +- tests/qemu-iotests/common.rc | 205 +++++----- tests/qemu-iotests/group | 2 + 45 files changed, 2333 insertions(+), 1311 deletions(-) create mode 100755 tests/qemu-iotests/191 create mode 100644 tests/qemu-iotests/191.out create mode 100755 tests/qemu-iotests/197 create mode 100644 tests/qemu-iotests/197.out delete mode 100644 tests/qemu-iotests/common From: Eric Blake <eblake@redhat.com> Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_co_do_copy_on_readv(BdrvChild *child, /* FIXME We cannot require callers to have write permissions when all they * are doing is a read request. If we did things right, write permissions * would be obtained anyway, but internally by the copy-on-read code. As - * long as it is implemented here rather than in a separat filter driver, + * long as it is implemented here rather than in a separate filter driver, * the copy-on-read code doesn't have its own BdrvChild, however, for which * it could request permissions. Therefore we have to bypass the permission * system for the moment. */ -- 2.13.6 From: Eric Blake <eblake@redhat.com> All callers of bdrv_img_create() pass in a size, or -1 to read the size from the backing file. We then set that size as the QemuOpt default, which means we will reuse that default rather than the final parameter to qemu_opt_get_size() several lines later. But it is rather confusing to read subsequent checks of 'size == -1' when it looks (without seeing the full context) like size defaults to 0; it also doesn't help that a size of 0 is valid (for some formats). Rework the logic to make things more legible. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block.c b/block.c index XXXXXXX..XXXXXXX 100644 --- a/block.c +++ b/block.c @@ -XXX,XX +XXX,XX @@ void bdrv_img_create(const char *filename, const char *fmt, /* The size for the image must always be specified, unless we have a backing * file and we have not been forbidden from opening it. */ - size = qemu_opt_get_size(opts, BLOCK_OPT_SIZE, 0); + size = qemu_opt_get_size(opts, BLOCK_OPT_SIZE, img_size); if (backing_file && !(flags & BDRV_O_NO_BACKING)) { BlockDriverState *bs; char *full_backing = g_new0(char, PATH_MAX); -- 2.13.6 From: Eric Blake <eblake@redhat.com> The only client of hbitmap_serialization_granularity() is dirty-bitmap's bdrv_dirty_bitmap_serialization_align(). Keeping the two names consistent is worthwhile, and the shorter name is more representative of what the function returns (the required alignment to be used for start/count of other serialization functions, where violating the alignment causes assertion failures). Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/qemu/hbitmap.h | 8 ++++---- block/dirty-bitmap.c | 2 +- tests/test-hbitmap.c | 10 +++++----- util/hbitmap.c | 8 ++++---- 4 files changed, 14 insertions(+), 14 deletions(-) diff --git a/include/qemu/hbitmap.h b/include/qemu/hbitmap.h index XXXXXXX..XXXXXXX 100644 --- a/include/qemu/hbitmap.h +++ b/include/qemu/hbitmap.h @@ -XXX,XX +XXX,XX @@ bool hbitmap_get(const HBitmap *hb, uint64_t item); bool hbitmap_is_serializable(const HBitmap *hb); /** - * hbitmap_serialization_granularity: + * hbitmap_serialization_align: * @hb: HBitmap to operate on. * - * Granularity of serialization chunks, used by other serialization functions. - * For every chunk: + * Required alignment of serialization chunks, used by other serialization + * functions. For every chunk: * 1. Chunk start should be aligned to this granularity. * 2. Chunk size should be aligned too, except for last chunk (for which * start + count == hb->size) */ -uint64_t hbitmap_serialization_granularity(const HBitmap *hb); +uint64_t hbitmap_serialization_align(const HBitmap *hb); /** * hbitmap_serialization_size: diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ uint64_t bdrv_dirty_bitmap_serialization_size(const BdrvDirtyBitmap *bitmap, uint64_t bdrv_dirty_bitmap_serialization_align(const BdrvDirtyBitmap *bitmap) { - return hbitmap_serialization_granularity(bitmap->bitmap); + return hbitmap_serialization_align(bitmap->bitmap); } void bdrv_dirty_bitmap_serialize_part(const BdrvDirtyBitmap *bitmap, diff --git a/tests/test-hbitmap.c b/tests/test-hbitmap.c index XXXXXXX..XXXXXXX 100644 --- a/tests/test-hbitmap.c +++ b/tests/test-hbitmap.c @@ -XXX,XX +XXX,XX @@ static void test_hbitmap_meta_one(TestHBitmapData *data, const void *unused) } } -static void test_hbitmap_serialize_granularity(TestHBitmapData *data, - const void *unused) +static void test_hbitmap_serialize_align(TestHBitmapData *data, + const void *unused) { int r; hbitmap_test_init(data, L3 * 2, 3); g_assert(hbitmap_is_serializable(data->hb)); - r = hbitmap_serialization_granularity(data->hb); + r = hbitmap_serialization_align(data->hb); g_assert_cmpint(r, ==, 64 << 3); } @@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv) hbitmap_test_add("/hbitmap/meta/word", test_hbitmap_meta_word); hbitmap_test_add("/hbitmap/meta/sector", test_hbitmap_meta_sector); - hbitmap_test_add("/hbitmap/serialize/granularity", - test_hbitmap_serialize_granularity); + hbitmap_test_add("/hbitmap/serialize/align", + test_hbitmap_serialize_align); hbitmap_test_add("/hbitmap/serialize/basic", test_hbitmap_serialize_basic); hbitmap_test_add("/hbitmap/serialize/part", diff --git a/util/hbitmap.c b/util/hbitmap.c index XXXXXXX..XXXXXXX 100644 --- a/util/hbitmap.c +++ b/util/hbitmap.c @@ -XXX,XX +XXX,XX @@ bool hbitmap_is_serializable(const HBitmap *hb) { /* Every serialized chunk must be aligned to 64 bits so that endianness * requirements can be fulfilled on both 64 bit and 32 bit hosts. - * We have hbitmap_serialization_granularity() which converts this + * We have hbitmap_serialization_align() which converts this * alignment requirement from bitmap bits to items covered (e.g. sectors). * That value is: * 64 << hb->granularity * Since this value must not exceed UINT64_MAX, hb->granularity must be * less than 58 (== 64 - 6, where 6 is ld(64), i.e. 1 << 6 == 64). * - * In order for hbitmap_serialization_granularity() to always return a + * In order for hbitmap_serialization_align() to always return a * meaningful value, bitmaps that are to be serialized must have a * granularity of less than 58. */ @@ -XXX,XX +XXX,XX @@ bool hbitmap_get(const HBitmap *hb, uint64_t item) return (hb->levels[HBITMAP_LEVELS - 1][pos >> BITS_PER_LEVEL] & bit) != 0; } -uint64_t hbitmap_serialization_granularity(const HBitmap *hb) +uint64_t hbitmap_serialization_align(const HBitmap *hb) { assert(hbitmap_is_serializable(hb)); @@ -XXX,XX +XXX,XX @@ static void serialization_chunk(const HBitmap *hb, unsigned long **first_el, uint64_t *el_count) { uint64_t last = start + count - 1; - uint64_t gran = hbitmap_serialization_granularity(hb); + uint64_t gran = hbitmap_serialization_align(hb); assert((start & (gran - 1)) == 0); assert((last >> hb->granularity) < hb->size); -- 2.13.6 From: Eric Blake <eblake@redhat.com> When subdividing a bitmap serialization, the code in hbitmap.c enforces that start/count parameters are aligned (except that count can end early at end-of-bitmap). We exposed this required alignment through bdrv_dirty_bitmap_serialization_align(), but forgot to actually check that we comply with it. Fortunately, qcow2 is never dividing bitmap serialization smaller than one cluster (which is a minimum of 512 bytes); so we are always compliant with the serialization alignment (which insists that we partition at least 64 bits per chunk) because we are doing at least 4k bits per chunk. Still, it's safer to add an assertion (for the unlikely case that we'd ever support a cluster smaller than 512 bytes, or if the hbitmap implementation changes what it considers to be aligned), rather than leaving bdrv_dirty_bitmap_serialization_align() without a caller. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/qcow2-bitmap.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-bitmap.c +++ b/block/qcow2-bitmap.c @@ -XXX,XX +XXX,XX @@ static int free_bitmap_clusters(BlockDriverState *bs, Qcow2BitmapTable *tb) static uint64_t sectors_covered_by_bitmap_cluster(const BDRVQcow2State *s, const BdrvDirtyBitmap *bitmap) { - uint32_t sector_granularity = + uint64_t sector_granularity = bdrv_dirty_bitmap_granularity(bitmap) >> BDRV_SECTOR_BITS; + uint64_t sbc = sector_granularity * (s->cluster_size << 3); - return (uint64_t)sector_granularity * (s->cluster_size << 3); + assert(QEMU_IS_ALIGNED(sbc, + bdrv_dirty_bitmap_serialization_align(bitmap))); + return sbc; } /* load_bitmap_data -- 2.13.6 From: Eric Blake <eblake@redhat.com> We had several functions that no one is currently using, and which use sector-based interfaces. I'm trying to convert towards byte-based interfaces, so it's easier to just drop the unused functions: bdrv_dirty_bitmap_get_meta bdrv_dirty_bitmap_get_meta_locked bdrv_dirty_bitmap_reset_meta bdrv_dirty_bitmap_meta_granularity Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/dirty-bitmap.h | 10 ---------- block/dirty-bitmap.c | 44 -------------------------------------------- 2 files changed, 54 deletions(-) diff --git a/include/block/dirty-bitmap.h b/include/block/dirty-bitmap.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/dirty-bitmap.h +++ b/include/block/dirty-bitmap.h @@ -XXX,XX +XXX,XX @@ void bdrv_enable_dirty_bitmap(BdrvDirtyBitmap *bitmap); BlockDirtyInfoList *bdrv_query_dirty_bitmaps(BlockDriverState *bs); uint32_t bdrv_get_default_bitmap_granularity(BlockDriverState *bs); uint32_t bdrv_dirty_bitmap_granularity(const BdrvDirtyBitmap *bitmap); -uint32_t bdrv_dirty_bitmap_meta_granularity(BdrvDirtyBitmap *bitmap); bool bdrv_dirty_bitmap_enabled(BdrvDirtyBitmap *bitmap); bool bdrv_dirty_bitmap_frozen(BdrvDirtyBitmap *bitmap); const char *bdrv_dirty_bitmap_name(const BdrvDirtyBitmap *bitmap); @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty_bitmap(BdrvDirtyBitmap *bitmap, int64_t cur_sector, int64_t nr_sectors); void bdrv_reset_dirty_bitmap(BdrvDirtyBitmap *bitmap, int64_t cur_sector, int64_t nr_sectors); -int bdrv_dirty_bitmap_get_meta(BlockDriverState *bs, - BdrvDirtyBitmap *bitmap, int64_t sector, - int nb_sectors); -int bdrv_dirty_bitmap_get_meta_locked(BlockDriverState *bs, - BdrvDirtyBitmap *bitmap, int64_t sector, - int nb_sectors); -void bdrv_dirty_bitmap_reset_meta(BlockDriverState *bs, - BdrvDirtyBitmap *bitmap, int64_t sector, - int nb_sectors); BdrvDirtyBitmapIter *bdrv_dirty_meta_iter_new(BdrvDirtyBitmap *bitmap); BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap, uint64_t first_sector); diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ void bdrv_release_meta_dirty_bitmap(BdrvDirtyBitmap *bitmap) qemu_mutex_unlock(bitmap->mutex); } -int bdrv_dirty_bitmap_get_meta_locked(BlockDriverState *bs, - BdrvDirtyBitmap *bitmap, int64_t sector, - int nb_sectors) -{ - uint64_t i; - int sectors_per_bit = 1 << hbitmap_granularity(bitmap->meta); - - /* To optimize: we can make hbitmap to internally check the range in a - * coarse level, or at least do it word by word. */ - for (i = sector; i < sector + nb_sectors; i += sectors_per_bit) { - if (hbitmap_get(bitmap->meta, i)) { - return true; - } - } - return false; -} - -int bdrv_dirty_bitmap_get_meta(BlockDriverState *bs, - BdrvDirtyBitmap *bitmap, int64_t sector, - int nb_sectors) -{ - bool dirty; - - qemu_mutex_lock(bitmap->mutex); - dirty = bdrv_dirty_bitmap_get_meta_locked(bs, bitmap, sector, nb_sectors); - qemu_mutex_unlock(bitmap->mutex); - - return dirty; -} - -void bdrv_dirty_bitmap_reset_meta(BlockDriverState *bs, - BdrvDirtyBitmap *bitmap, int64_t sector, - int nb_sectors) -{ - qemu_mutex_lock(bitmap->mutex); - hbitmap_reset(bitmap->meta, sector, nb_sectors); - qemu_mutex_unlock(bitmap->mutex); -} - int64_t bdrv_dirty_bitmap_size(const BdrvDirtyBitmap *bitmap) { return bitmap->size; @@ -XXX,XX +XXX,XX @@ uint32_t bdrv_dirty_bitmap_granularity(const BdrvDirtyBitmap *bitmap) return BDRV_SECTOR_SIZE << hbitmap_granularity(bitmap->bitmap); } -uint32_t bdrv_dirty_bitmap_meta_granularity(BdrvDirtyBitmap *bitmap) -{ - return BDRV_SECTOR_SIZE << hbitmap_granularity(bitmap->meta); -} - BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap, uint64_t first_sector) { -- 2.13.6 From: Eric Blake <eblake@redhat.com> We've previously fixed several places where we failed to account for possible errors from bdrv_nb_sectors(). Fix another one by making bdrv_dirty_bitmap_truncate() take the new size from the caller instead of querying itself; then adjust the sole caller bdrv_truncate() to pass the size just determined by a successful resize, or to reuse the size given to the original truncate operation when refresh_total_sectors() was not able to confirm the actual size (the two sizes can potentially differ according to rounding constraints), thus avoiding sizing the bitmaps to -1. This also fixes a bug where not all failure paths in bdrv_truncate() would set errp. Note that bdrv_truncate() is still a bit awkward. We may want to revisit it later and clean up things to better guarantee that a resize attempt either fails cleanly up front, or cannot fail after guest-visible changes have been made (if temporary changes are made, then they need to be cleanly rolled back). But that is a task for another day; for now, the goal is the bare minimum fix to ensure that just bdrv_dirty_bitmap_truncate() cannot fail. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/dirty-bitmap.h | 2 +- block.c | 16 +++++++++++----- block/dirty-bitmap.c | 6 +++--- 3 files changed, 15 insertions(+), 9 deletions(-) diff --git a/include/block/dirty-bitmap.h b/include/block/dirty-bitmap.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/dirty-bitmap.h +++ b/include/block/dirty-bitmap.h @@ -XXX,XX +XXX,XX @@ int64_t bdrv_dirty_iter_next(BdrvDirtyBitmapIter *iter); void bdrv_set_dirty_iter(BdrvDirtyBitmapIter *hbi, int64_t sector_num); int64_t bdrv_get_dirty_count(BdrvDirtyBitmap *bitmap); int64_t bdrv_get_meta_dirty_count(BdrvDirtyBitmap *bitmap); -void bdrv_dirty_bitmap_truncate(BlockDriverState *bs); +void bdrv_dirty_bitmap_truncate(BlockDriverState *bs, int64_t bytes); bool bdrv_dirty_bitmap_readonly(const BdrvDirtyBitmap *bitmap); bool bdrv_has_readonly_bitmaps(BlockDriverState *bs); bool bdrv_dirty_bitmap_get_autoload(const BdrvDirtyBitmap *bitmap); diff --git a/block.c b/block.c index XXXXXXX..XXXXXXX 100644 --- a/block.c +++ b/block.c @@ -XXX,XX +XXX,XX @@ int bdrv_truncate(BdrvChild *child, int64_t offset, PreallocMode prealloc, assert(!(bs->open_flags & BDRV_O_INACTIVE)); ret = drv->bdrv_truncate(bs, offset, prealloc, errp); - if (ret == 0) { - ret = refresh_total_sectors(bs, offset >> BDRV_SECTOR_BITS); - bdrv_dirty_bitmap_truncate(bs); - bdrv_parent_cb_resize(bs); - atomic_inc(&bs->write_gen); + if (ret < 0) { + return ret; + } + ret = refresh_total_sectors(bs, offset >> BDRV_SECTOR_BITS); + if (ret < 0) { + error_setg_errno(errp, -ret, "Could not refresh total sector count"); + } else { + offset = bs->total_sectors * BDRV_SECTOR_SIZE; } + bdrv_dirty_bitmap_truncate(bs, offset); + bdrv_parent_cb_resize(bs); + atomic_inc(&bs->write_gen); return ret; } diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ /* * Block Dirty Bitmap * - * Copyright (c) 2016 Red Hat. Inc + * Copyright (c) 2016-2017 Red Hat. Inc * * Permission is hereby granted, free of charge, to any person obtaining a copy * of this software and associated documentation files (the "Software"), to deal @@ -XXX,XX +XXX,XX @@ BdrvDirtyBitmap *bdrv_reclaim_dirty_bitmap(BlockDriverState *bs, * Truncates _all_ bitmaps attached to a BDS. * Called with BQL taken. */ -void bdrv_dirty_bitmap_truncate(BlockDriverState *bs) +void bdrv_dirty_bitmap_truncate(BlockDriverState *bs, int64_t bytes) { BdrvDirtyBitmap *bitmap; - uint64_t size = bdrv_nb_sectors(bs); + int64_t size = DIV_ROUND_UP(bytes, BDRV_SECTOR_SIZE); bdrv_dirty_bitmaps_lock(bs); QLIST_FOREACH(bitmap, &bs->dirty_bitmaps, list) { -- 2.13.6 From: Eric Blake <eblake@redhat.com> We're already reporting bytes for bdrv_dirty_bitmap_granularity(); mixing bytes and sectors in our return values is a recipe for confusion. A later cleanup will convert dirty bitmap internals to be entirely byte-based, but in the meantime, we should report the bitmap size in bytes. The only external caller in qcow2-bitmap.c is temporarily more verbose (because it is still using sector-based math), but will later be switched to track progress by bytes instead of sectors. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/dirty-bitmap.c | 2 +- block/qcow2-bitmap.c | 14 ++++++++------ 2 files changed, 9 insertions(+), 7 deletions(-) diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ void bdrv_release_meta_dirty_bitmap(BdrvDirtyBitmap *bitmap) int64_t bdrv_dirty_bitmap_size(const BdrvDirtyBitmap *bitmap) { - return bitmap->size; + return bitmap->size * BDRV_SECTOR_SIZE; } const char *bdrv_dirty_bitmap_name(const BdrvDirtyBitmap *bitmap) diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-bitmap.c +++ b/block/qcow2-bitmap.c @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, BDRVQcow2State *s = bs->opaque; uint64_t sector, sbc; uint64_t bm_size = bdrv_dirty_bitmap_size(bitmap); + uint64_t bm_sectors = DIV_ROUND_UP(bm_size, BDRV_SECTOR_SIZE); uint8_t *buf = NULL; uint64_t i, tab_size = size_to_clusters(s, - bdrv_dirty_bitmap_serialization_size(bitmap, 0, bm_size)); + bdrv_dirty_bitmap_serialization_size(bitmap, 0, bm_sectors)); if (tab_size != bitmap_table_size || tab_size > BME_MAX_TABLE_SIZE) { return -EINVAL; @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, buf = g_malloc(s->cluster_size); sbc = sectors_covered_by_bitmap_cluster(s, bitmap); for (i = 0, sector = 0; i < tab_size; ++i, sector += sbc) { - uint64_t count = MIN(bm_size - sector, sbc); + uint64_t count = MIN(bm_sectors - sector, sbc); uint64_t entry = bitmap_table[i]; uint64_t offset = entry & BME_TABLE_ENTRY_OFFSET_MASK; @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, int64_t sector; uint64_t sbc; uint64_t bm_size = bdrv_dirty_bitmap_size(bitmap); + uint64_t bm_sectors = DIV_ROUND_UP(bm_size, BDRV_SECTOR_SIZE); const char *bm_name = bdrv_dirty_bitmap_name(bitmap); uint8_t *buf = NULL; BdrvDirtyBitmapIter *dbi; uint64_t *tb; uint64_t tb_size = size_to_clusters(s, - bdrv_dirty_bitmap_serialization_size(bitmap, 0, bm_size)); + bdrv_dirty_bitmap_serialization_size(bitmap, 0, bm_sectors)); if (tb_size > BME_MAX_TABLE_SIZE || tb_size * s->cluster_size > BME_MAX_PHYS_SIZE) @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, dbi = bdrv_dirty_iter_new(bitmap, 0); buf = g_malloc(s->cluster_size); sbc = sectors_covered_by_bitmap_cluster(s, bitmap); - assert(DIV_ROUND_UP(bm_size, sbc) == tb_size); + assert(DIV_ROUND_UP(bm_sectors, sbc) == tb_size); while ((sector = bdrv_dirty_iter_next(dbi)) != -1) { uint64_t cluster = sector / sbc; @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, int64_t off; sector = cluster * sbc; - end = MIN(bm_size, sector + sbc); + end = MIN(bm_sectors, sector + sbc); write_size = bdrv_dirty_bitmap_serialization_size(bitmap, sector, end - sector); assert(write_size <= s->cluster_size); @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, goto fail; } - if (end >= bm_size) { + if (end >= bm_sectors) { break; } -- 2.13.6 From: Eric Blake <eblake@redhat.com> We are still using an internal hbitmap that tracks a size in sectors, with the granularity scaled down accordingly, because it lets us use a shortcut for our iterators which are currently sector-based. But there's no reason we can't track the dirty bitmap size in bytes, since it is (mostly) an internal-only variable (remember, the size is how many bytes are covered by the bitmap, not how many bytes the bitmap occupies). A later cleanup will convert dirty bitmap internals to be entirely byte-based, eliminating the intermediate sector rounding added here; and technically, since bdrv_getlength() already rounds up to sectors, our use of DIV_ROUND_UP is more for theoretical completeness than for any actual rounding. Use is_power_of_2() while at it, instead of open-coding that. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/dirty-bitmap.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ struct BdrvDirtyBitmap { HBitmap *meta; /* Meta dirty bitmap */ BdrvDirtyBitmap *successor; /* Anonymous child; implies frozen status */ char *name; /* Optional non-empty unique ID */ - int64_t size; /* Size of the bitmap (Number of sectors) */ + int64_t size; /* Size of the bitmap, in bytes */ bool disabled; /* Bitmap is disabled. It ignores all writes to the device */ int active_iterators; /* How many iterators are active */ @@ -XXX,XX +XXX,XX @@ BdrvDirtyBitmap *bdrv_create_dirty_bitmap(BlockDriverState *bs, { int64_t bitmap_size; BdrvDirtyBitmap *bitmap; - uint32_t sector_granularity; - assert((granularity & (granularity - 1)) == 0); + assert(is_power_of_2(granularity) && granularity >= BDRV_SECTOR_SIZE); if (name && bdrv_find_dirty_bitmap(bs, name)) { error_setg(errp, "Bitmap already exists: %s", name); return NULL; } - sector_granularity = granularity >> BDRV_SECTOR_BITS; - assert(sector_granularity); - bitmap_size = bdrv_nb_sectors(bs); + bitmap_size = bdrv_getlength(bs); if (bitmap_size < 0) { error_setg_errno(errp, -bitmap_size, "could not get length of device"); errno = -bitmap_size; @@ -XXX,XX +XXX,XX @@ BdrvDirtyBitmap *bdrv_create_dirty_bitmap(BlockDriverState *bs, } bitmap = g_new0(BdrvDirtyBitmap, 1); bitmap->mutex = &bs->dirty_bitmap_mutex; - bitmap->bitmap = hbitmap_alloc(bitmap_size, ctz32(sector_granularity)); + /* + * TODO - let hbitmap track full granularity. For now, it is tracking + * only sector granularity, as a shortcut for our iterators. + */ + bitmap->bitmap = hbitmap_alloc(DIV_ROUND_UP(bitmap_size, BDRV_SECTOR_SIZE), + ctz32(granularity) - BDRV_SECTOR_BITS); bitmap->size = bitmap_size; bitmap->name = g_strdup(name); bitmap->disabled = false; @@ -XXX,XX +XXX,XX @@ void bdrv_release_meta_dirty_bitmap(BdrvDirtyBitmap *bitmap) int64_t bdrv_dirty_bitmap_size(const BdrvDirtyBitmap *bitmap) { - return bitmap->size * BDRV_SECTOR_SIZE; + return bitmap->size; } const char *bdrv_dirty_bitmap_name(const BdrvDirtyBitmap *bitmap) @@ -XXX,XX +XXX,XX @@ BdrvDirtyBitmap *bdrv_reclaim_dirty_bitmap(BlockDriverState *bs, void bdrv_dirty_bitmap_truncate(BlockDriverState *bs, int64_t bytes) { BdrvDirtyBitmap *bitmap; - int64_t size = DIV_ROUND_UP(bytes, BDRV_SECTOR_SIZE); bdrv_dirty_bitmaps_lock(bs); QLIST_FOREACH(bitmap, &bs->dirty_bitmaps, list) { assert(!bdrv_dirty_bitmap_frozen(bitmap)); assert(!bitmap->active_iterators); - hbitmap_truncate(bitmap->bitmap, size); - bitmap->size = size; + hbitmap_truncate(bitmap->bitmap, DIV_ROUND_UP(bytes, BDRV_SECTOR_SIZE)); + bitmap->size = bytes; } bdrv_dirty_bitmaps_unlock(bs); } @@ -XXX,XX +XXX,XX @@ void bdrv_clear_dirty_bitmap(BdrvDirtyBitmap *bitmap, HBitmap **out) hbitmap_reset_all(bitmap->bitmap); } else { HBitmap *backup = bitmap->bitmap; - bitmap->bitmap = hbitmap_alloc(bitmap->size, + bitmap->bitmap = hbitmap_alloc(DIV_ROUND_UP(bitmap->size, + BDRV_SECTOR_SIZE), hbitmap_granularity(backup)); *out = backup; } -- 2.13.6 From: Eric Blake <eblake@redhat.com> Right now, the dirty-bitmap code exposes the fact that we use a scale of sector granularity in the underlying hbitmap to anything that wants to serialize a dirty bitmap. It's nicer to uniformly expose bytes as our dirty-bitmap interface, matching the previous change to bitmap size. The only caller to serialization is currently qcow2-cluster.c, which becomes a bit more verbose because it is still tracking sectors for other reasons, but a later patch will fix that to more uniformly use byte offsets everywhere. Likewise, within dirty-bitmap, we have to add more assertions that we are not truncating incorrectly, which can go away once the internal hbitmap is byte-based rather than sector-based. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/dirty-bitmap.h | 14 +++++++------- block/dirty-bitmap.c | 37 ++++++++++++++++++++++++------------- block/qcow2-bitmap.c | 22 ++++++++++++++-------- 3 files changed, 45 insertions(+), 28 deletions(-) diff --git a/include/block/dirty-bitmap.h b/include/block/dirty-bitmap.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/dirty-bitmap.h +++ b/include/block/dirty-bitmap.h @@ -XXX,XX +XXX,XX @@ BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap, void bdrv_dirty_iter_free(BdrvDirtyBitmapIter *iter); uint64_t bdrv_dirty_bitmap_serialization_size(const BdrvDirtyBitmap *bitmap, - uint64_t start, uint64_t count); + uint64_t offset, uint64_t bytes); uint64_t bdrv_dirty_bitmap_serialization_align(const BdrvDirtyBitmap *bitmap); void bdrv_dirty_bitmap_serialize_part(const BdrvDirtyBitmap *bitmap, - uint8_t *buf, uint64_t start, - uint64_t count); + uint8_t *buf, uint64_t offset, + uint64_t bytes); void bdrv_dirty_bitmap_deserialize_part(BdrvDirtyBitmap *bitmap, - uint8_t *buf, uint64_t start, - uint64_t count, bool finish); + uint8_t *buf, uint64_t offset, + uint64_t bytes, bool finish); void bdrv_dirty_bitmap_deserialize_zeroes(BdrvDirtyBitmap *bitmap, - uint64_t start, uint64_t count, + uint64_t offset, uint64_t bytes, bool finish); void bdrv_dirty_bitmap_deserialize_ones(BdrvDirtyBitmap *bitmap, - uint64_t start, uint64_t count, + uint64_t offset, uint64_t bytes, bool finish); void bdrv_dirty_bitmap_deserialize_finish(BdrvDirtyBitmap *bitmap); diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ void bdrv_undo_clear_dirty_bitmap(BdrvDirtyBitmap *bitmap, HBitmap *in) } uint64_t bdrv_dirty_bitmap_serialization_size(const BdrvDirtyBitmap *bitmap, - uint64_t start, uint64_t count) + uint64_t offset, uint64_t bytes) { - return hbitmap_serialization_size(bitmap->bitmap, start, count); + assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); + return hbitmap_serialization_size(bitmap->bitmap, + offset >> BDRV_SECTOR_BITS, + bytes >> BDRV_SECTOR_BITS); } uint64_t bdrv_dirty_bitmap_serialization_align(const BdrvDirtyBitmap *bitmap) { - return hbitmap_serialization_align(bitmap->bitmap); + return hbitmap_serialization_align(bitmap->bitmap) * BDRV_SECTOR_SIZE; } void bdrv_dirty_bitmap_serialize_part(const BdrvDirtyBitmap *bitmap, - uint8_t *buf, uint64_t start, - uint64_t count) + uint8_t *buf, uint64_t offset, + uint64_t bytes) { - hbitmap_serialize_part(bitmap->bitmap, buf, start, count); + assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); + hbitmap_serialize_part(bitmap->bitmap, buf, offset >> BDRV_SECTOR_BITS, + bytes >> BDRV_SECTOR_BITS); } void bdrv_dirty_bitmap_deserialize_part(BdrvDirtyBitmap *bitmap, - uint8_t *buf, uint64_t start, - uint64_t count, bool finish) + uint8_t *buf, uint64_t offset, + uint64_t bytes, bool finish) { - hbitmap_deserialize_part(bitmap->bitmap, buf, start, count, finish); + assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); + hbitmap_deserialize_part(bitmap->bitmap, buf, offset >> BDRV_SECTOR_BITS, + bytes >> BDRV_SECTOR_BITS, finish); } void bdrv_dirty_bitmap_deserialize_zeroes(BdrvDirtyBitmap *bitmap, - uint64_t start, uint64_t count, + uint64_t offset, uint64_t bytes, bool finish) { - hbitmap_deserialize_zeroes(bitmap->bitmap, start, count, finish); + assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); + hbitmap_deserialize_zeroes(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, + bytes >> BDRV_SECTOR_BITS, finish); } void bdrv_dirty_bitmap_deserialize_ones(BdrvDirtyBitmap *bitmap, - uint64_t start, uint64_t count, + uint64_t offset, uint64_t bytes, bool finish) { - hbitmap_deserialize_ones(bitmap->bitmap, start, count, finish); + assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); + hbitmap_deserialize_ones(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, + bytes >> BDRV_SECTOR_BITS, finish); } void bdrv_dirty_bitmap_deserialize_finish(BdrvDirtyBitmap *bitmap) diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-bitmap.c +++ b/block/qcow2-bitmap.c @@ -XXX,XX +XXX,XX @@ static uint64_t sectors_covered_by_bitmap_cluster(const BDRVQcow2State *s, bdrv_dirty_bitmap_granularity(bitmap) >> BDRV_SECTOR_BITS; uint64_t sbc = sector_granularity * (s->cluster_size << 3); - assert(QEMU_IS_ALIGNED(sbc, + assert(QEMU_IS_ALIGNED(sbc << BDRV_SECTOR_BITS, bdrv_dirty_bitmap_serialization_align(bitmap))); return sbc; } @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, uint8_t *buf = NULL; uint64_t i, tab_size = size_to_clusters(s, - bdrv_dirty_bitmap_serialization_size(bitmap, 0, bm_sectors)); + bdrv_dirty_bitmap_serialization_size(bitmap, 0, bm_size)); if (tab_size != bitmap_table_size || tab_size > BME_MAX_TABLE_SIZE) { return -EINVAL; @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, if (offset == 0) { if (entry & BME_TABLE_ENTRY_FLAG_ALL_ONES) { - bdrv_dirty_bitmap_deserialize_ones(bitmap, sector, count, + bdrv_dirty_bitmap_deserialize_ones(bitmap, + sector * BDRV_SECTOR_SIZE, + count * BDRV_SECTOR_SIZE, false); } else { /* No need to deserialize zeros because the dirty bitmap is @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, if (ret < 0) { goto finish; } - bdrv_dirty_bitmap_deserialize_part(bitmap, buf, sector, count, + bdrv_dirty_bitmap_deserialize_part(bitmap, buf, + sector * BDRV_SECTOR_SIZE, + count * BDRV_SECTOR_SIZE, false); } } @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, uint64_t *tb; uint64_t tb_size = size_to_clusters(s, - bdrv_dirty_bitmap_serialization_size(bitmap, 0, bm_sectors)); + bdrv_dirty_bitmap_serialization_size(bitmap, 0, bm_size)); if (tb_size > BME_MAX_TABLE_SIZE || tb_size * s->cluster_size > BME_MAX_PHYS_SIZE) @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, sector = cluster * sbc; end = MIN(bm_sectors, sector + sbc); - write_size = - bdrv_dirty_bitmap_serialization_size(bitmap, sector, end - sector); + write_size = bdrv_dirty_bitmap_serialization_size(bitmap, + sector * BDRV_SECTOR_SIZE, (end - sector) * BDRV_SECTOR_SIZE); assert(write_size <= s->cluster_size); off = qcow2_alloc_clusters(bs, s->cluster_size); @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, } tb[cluster] = off; - bdrv_dirty_bitmap_serialize_part(bitmap, buf, sector, end - sector); + bdrv_dirty_bitmap_serialize_part(bitmap, buf, + sector * BDRV_SECTOR_SIZE, + (end - sector) * BDRV_SECTOR_SIZE); if (write_size < s->cluster_size) { memset(buf + write_size, 0, s->cluster_size - write_size); } -- 2.13.6 From: Eric Blake <eblake@redhat.com> We are gradually converting to byte-based interfaces, as they are easier to reason about than sector-based. Change the qcow2 bitmap helper function sectors_covered_by_bitmap_cluster(), renaming it to bytes_covered_by_bitmap_cluster() in the process. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/qcow2-bitmap.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-bitmap.c +++ b/block/qcow2-bitmap.c @@ -XXX,XX +XXX,XX @@ static int free_bitmap_clusters(BlockDriverState *bs, Qcow2BitmapTable *tb) return 0; } -/* This function returns the number of disk sectors covered by a single qcow2 - * cluster of bitmap data. */ -static uint64_t sectors_covered_by_bitmap_cluster(const BDRVQcow2State *s, - const BdrvDirtyBitmap *bitmap) +/* Return the disk size covered by a single qcow2 cluster of bitmap data. */ +static uint64_t bytes_covered_by_bitmap_cluster(const BDRVQcow2State *s, + const BdrvDirtyBitmap *bitmap) { - uint64_t sector_granularity = - bdrv_dirty_bitmap_granularity(bitmap) >> BDRV_SECTOR_BITS; - uint64_t sbc = sector_granularity * (s->cluster_size << 3); + uint64_t granularity = bdrv_dirty_bitmap_granularity(bitmap); + uint64_t limit = granularity * (s->cluster_size << 3); - assert(QEMU_IS_ALIGNED(sbc << BDRV_SECTOR_BITS, + assert(QEMU_IS_ALIGNED(limit, bdrv_dirty_bitmap_serialization_align(bitmap))); - return sbc; + return limit; } /* load_bitmap_data @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, { int ret = 0; BDRVQcow2State *s = bs->opaque; - uint64_t sector, sbc; + uint64_t sector, limit, sbc; uint64_t bm_size = bdrv_dirty_bitmap_size(bitmap); uint64_t bm_sectors = DIV_ROUND_UP(bm_size, BDRV_SECTOR_SIZE); uint8_t *buf = NULL; @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, } buf = g_malloc(s->cluster_size); - sbc = sectors_covered_by_bitmap_cluster(s, bitmap); + limit = bytes_covered_by_bitmap_cluster(s, bitmap); + sbc = limit >> BDRV_SECTOR_BITS; for (i = 0, sector = 0; i < tab_size; ++i, sector += sbc) { uint64_t count = MIN(bm_sectors - sector, sbc); uint64_t entry = bitmap_table[i]; @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, int ret; BDRVQcow2State *s = bs->opaque; int64_t sector; - uint64_t sbc; + uint64_t limit, sbc; uint64_t bm_size = bdrv_dirty_bitmap_size(bitmap); uint64_t bm_sectors = DIV_ROUND_UP(bm_size, BDRV_SECTOR_SIZE); const char *bm_name = bdrv_dirty_bitmap_name(bitmap); @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, dbi = bdrv_dirty_iter_new(bitmap, 0); buf = g_malloc(s->cluster_size); - sbc = sectors_covered_by_bitmap_cluster(s, bitmap); - assert(DIV_ROUND_UP(bm_sectors, sbc) == tb_size); + limit = bytes_covered_by_bitmap_cluster(s, bitmap); + sbc = limit >> BDRV_SECTOR_BITS; + assert(DIV_ROUND_UP(bm_size, limit) == tb_size); while ((sector = bdrv_dirty_iter_next(dbi)) != -1) { uint64_t cluster = sector / sbc; -- 2.13.6 From: Eric Blake <eblake@redhat.com> All callers to bdrv_dirty_iter_new() passed 0 for their initial starting point, drop that parameter. Most callers to bdrv_set_dirty_iter() were scaling a byte offset to a sector number; the exception qcow2-bitmap will be converted later to use byte rather than sector iteration. Move the scaling to occur internally to dirty bitmap code instead, so that callers now pass in bytes. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/dirty-bitmap.h | 5 ++--- block/backup.c | 5 ++--- block/dirty-bitmap.c | 9 ++++----- block/mirror.c | 4 ++-- block/qcow2-bitmap.c | 4 ++-- 5 files changed, 12 insertions(+), 15 deletions(-) diff --git a/include/block/dirty-bitmap.h b/include/block/dirty-bitmap.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/dirty-bitmap.h +++ b/include/block/dirty-bitmap.h @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty_bitmap(BdrvDirtyBitmap *bitmap, void bdrv_reset_dirty_bitmap(BdrvDirtyBitmap *bitmap, int64_t cur_sector, int64_t nr_sectors); BdrvDirtyBitmapIter *bdrv_dirty_meta_iter_new(BdrvDirtyBitmap *bitmap); -BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap, - uint64_t first_sector); +BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap); void bdrv_dirty_iter_free(BdrvDirtyBitmapIter *iter); uint64_t bdrv_dirty_bitmap_serialization_size(const BdrvDirtyBitmap *bitmap, @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, void bdrv_reset_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, int64_t cur_sector, int64_t nr_sectors); int64_t bdrv_dirty_iter_next(BdrvDirtyBitmapIter *iter); -void bdrv_set_dirty_iter(BdrvDirtyBitmapIter *hbi, int64_t sector_num); +void bdrv_set_dirty_iter(BdrvDirtyBitmapIter *hbi, int64_t offset); int64_t bdrv_get_dirty_count(BdrvDirtyBitmap *bitmap); int64_t bdrv_get_meta_dirty_count(BdrvDirtyBitmap *bitmap); void bdrv_dirty_bitmap_truncate(BlockDriverState *bs, int64_t bytes); diff --git a/block/backup.c b/block/backup.c index XXXXXXX..XXXXXXX 100644 --- a/block/backup.c +++ b/block/backup.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn backup_run_incremental(BackupBlockJob *job) granularity = bdrv_dirty_bitmap_granularity(job->sync_bitmap); clusters_per_iter = MAX((granularity / job->cluster_size), 1); - dbi = bdrv_dirty_iter_new(job->sync_bitmap, 0); + dbi = bdrv_dirty_iter_new(job->sync_bitmap); /* Find the next dirty sector(s) */ while ((offset = bdrv_dirty_iter_next(dbi) * BDRV_SECTOR_SIZE) >= 0) { @@ -XXX,XX +XXX,XX @@ static int coroutine_fn backup_run_incremental(BackupBlockJob *job) /* If the bitmap granularity is smaller than the backup granularity, * we need to advance the iterator pointer to the next cluster. */ if (granularity < job->cluster_size) { - bdrv_set_dirty_iter(dbi, - cluster * job->cluster_size / BDRV_SECTOR_SIZE); + bdrv_set_dirty_iter(dbi, cluster * job->cluster_size); } last_cluster = cluster - 1; diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ uint32_t bdrv_dirty_bitmap_granularity(const BdrvDirtyBitmap *bitmap) return BDRV_SECTOR_SIZE << hbitmap_granularity(bitmap->bitmap); } -BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap, - uint64_t first_sector) +BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap) { BdrvDirtyBitmapIter *iter = g_new(BdrvDirtyBitmapIter, 1); - hbitmap_iter_init(&iter->hbi, bitmap->bitmap, first_sector); + hbitmap_iter_init(&iter->hbi, bitmap->bitmap, 0); iter->bitmap = bitmap; bitmap->active_iterators++; return iter; @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector, /** * Advance a BdrvDirtyBitmapIter to an arbitrary offset. */ -void bdrv_set_dirty_iter(BdrvDirtyBitmapIter *iter, int64_t sector_num) +void bdrv_set_dirty_iter(BdrvDirtyBitmapIter *iter, int64_t offset) { - hbitmap_iter_init(&iter->hbi, iter->hbi.hb, sector_num); + hbitmap_iter_init(&iter->hbi, iter->hbi.hb, offset >> BDRV_SECTOR_BITS); } int64_t bdrv_get_dirty_count(BdrvDirtyBitmap *bitmap) diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s) next_dirty = bdrv_dirty_iter_next(s->dbi) * BDRV_SECTOR_SIZE; if (next_dirty > next_offset || next_dirty < 0) { /* The bitmap iterator's cache is stale, refresh it */ - bdrv_set_dirty_iter(s->dbi, next_offset >> BDRV_SECTOR_BITS); + bdrv_set_dirty_iter(s->dbi, next_offset); next_dirty = bdrv_dirty_iter_next(s->dbi) * BDRV_SECTOR_SIZE; } assert(next_dirty == next_offset); @@ -XXX,XX +XXX,XX @@ static void coroutine_fn mirror_run(void *opaque) } assert(!s->dbi); - s->dbi = bdrv_dirty_iter_new(s->dirty_bitmap, 0); + s->dbi = bdrv_dirty_iter_new(s->dirty_bitmap); for (;;) { uint64_t delay_ns = 0; int64_t cnt, delta; diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-bitmap.c +++ b/block/qcow2-bitmap.c @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, return NULL; } - dbi = bdrv_dirty_iter_new(bitmap, 0); + dbi = bdrv_dirty_iter_new(bitmap); buf = g_malloc(s->cluster_size); limit = bytes_covered_by_bitmap_cluster(s, bitmap); sbc = limit >> BDRV_SECTOR_BITS; @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, break; } - bdrv_set_dirty_iter(dbi, end); + bdrv_set_dirty_iter(dbi, end * BDRV_SECTOR_SIZE); } *bitmap_table_size = tb_size; -- 2.13.6 From: Eric Blake <eblake@redhat.com> Thanks to recent cleanups, most callers were scaling a return value of sectors into bytes (the exception, in qcow2-bitmap, will be converted to byte-based iteration later). Update the interface to do the scaling internally instead. In qcow2-bitmap, the code was specifically checking for an error return of -1. To avoid a regression, we either have to make sure we continue to return -1 (rather than a scaled -512) on error, or we have to fix the caller to treat all negative values as error rather than just one magic value. It's easy enough to make both changes at the same time, even though either one in isolation would work. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/backup.c | 2 +- block/dirty-bitmap.c | 3 ++- block/mirror.c | 8 ++++---- block/qcow2-bitmap.c | 2 +- 4 files changed, 8 insertions(+), 7 deletions(-) diff --git a/block/backup.c b/block/backup.c index XXXXXXX..XXXXXXX 100644 --- a/block/backup.c +++ b/block/backup.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn backup_run_incremental(BackupBlockJob *job) dbi = bdrv_dirty_iter_new(job->sync_bitmap); /* Find the next dirty sector(s) */ - while ((offset = bdrv_dirty_iter_next(dbi) * BDRV_SECTOR_SIZE) >= 0) { + while ((offset = bdrv_dirty_iter_next(dbi)) >= 0) { cluster = offset / job->cluster_size; /* Fake progress updates for any clusters we skipped */ diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ void bdrv_dirty_iter_free(BdrvDirtyBitmapIter *iter) int64_t bdrv_dirty_iter_next(BdrvDirtyBitmapIter *iter) { - return hbitmap_iter_next(&iter->hbi); + int64_t ret = hbitmap_iter_next(&iter->hbi); + return ret < 0 ? -1 : ret * BDRV_SECTOR_SIZE; } /* Called within bdrv_dirty_bitmap_lock..unlock */ diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s) int max_io_bytes = MAX(s->buf_size / MAX_IN_FLIGHT, MAX_IO_BYTES); bdrv_dirty_bitmap_lock(s->dirty_bitmap); - offset = bdrv_dirty_iter_next(s->dbi) * BDRV_SECTOR_SIZE; + offset = bdrv_dirty_iter_next(s->dbi); if (offset < 0) { bdrv_set_dirty_iter(s->dbi, 0); - offset = bdrv_dirty_iter_next(s->dbi) * BDRV_SECTOR_SIZE; + offset = bdrv_dirty_iter_next(s->dbi); trace_mirror_restart_iter(s, bdrv_get_dirty_count(s->dirty_bitmap) * BDRV_SECTOR_SIZE); assert(offset >= 0); @@ -XXX,XX +XXX,XX @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s) break; } - next_dirty = bdrv_dirty_iter_next(s->dbi) * BDRV_SECTOR_SIZE; + next_dirty = bdrv_dirty_iter_next(s->dbi); if (next_dirty > next_offset || next_dirty < 0) { /* The bitmap iterator's cache is stale, refresh it */ bdrv_set_dirty_iter(s->dbi, next_offset); - next_dirty = bdrv_dirty_iter_next(s->dbi) * BDRV_SECTOR_SIZE; + next_dirty = bdrv_dirty_iter_next(s->dbi); } assert(next_dirty == next_offset); nb_chunks++; diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-bitmap.c +++ b/block/qcow2-bitmap.c @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, sbc = limit >> BDRV_SECTOR_BITS; assert(DIV_ROUND_UP(bm_size, limit) == tb_size); - while ((sector = bdrv_dirty_iter_next(dbi)) != -1) { + while ((sector = bdrv_dirty_iter_next(dbi) >> BDRV_SECTOR_BITS) >= 0) { uint64_t cluster = sector / sbc; uint64_t end, write_size; int64_t off; -- 2.13.6 From: Eric Blake <eblake@redhat.com> Thanks to recent cleanups, all callers were scaling a return value of sectors into bytes; do the scaling internally instead. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/dirty-bitmap.c | 4 ++-- block/mirror.c | 16 ++++++---------- migration/block.c | 2 +- 3 files changed, 9 insertions(+), 13 deletions(-) diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ BlockDirtyInfoList *bdrv_query_dirty_bitmaps(BlockDriverState *bs) QLIST_FOREACH(bm, &bs->dirty_bitmaps, list) { BlockDirtyInfo *info = g_new0(BlockDirtyInfo, 1); BlockDirtyInfoList *entry = g_new0(BlockDirtyInfoList, 1); - info->count = bdrv_get_dirty_count(bm) << BDRV_SECTOR_BITS; + info->count = bdrv_get_dirty_count(bm); info->granularity = bdrv_dirty_bitmap_granularity(bm); info->has_name = !!bm->name; info->name = g_strdup(bm->name); @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty_iter(BdrvDirtyBitmapIter *iter, int64_t offset) int64_t bdrv_get_dirty_count(BdrvDirtyBitmap *bitmap) { - return hbitmap_count(bitmap->bitmap); + return hbitmap_count(bitmap->bitmap) << BDRV_SECTOR_BITS; } int64_t bdrv_get_meta_dirty_count(BdrvDirtyBitmap *bitmap) diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s) if (offset < 0) { bdrv_set_dirty_iter(s->dbi, 0); offset = bdrv_dirty_iter_next(s->dbi); - trace_mirror_restart_iter(s, bdrv_get_dirty_count(s->dirty_bitmap) * - BDRV_SECTOR_SIZE); + trace_mirror_restart_iter(s, bdrv_get_dirty_count(s->dirty_bitmap)); assert(offset >= 0); } bdrv_dirty_bitmap_unlock(s->dirty_bitmap); @@ -XXX,XX +XXX,XX @@ static void coroutine_fn mirror_run(void *opaque) cnt = bdrv_get_dirty_count(s->dirty_bitmap); /* s->common.offset contains the number of bytes already processed so - * far, cnt is the number of dirty sectors remaining and + * far, cnt is the number of dirty bytes remaining and * s->bytes_in_flight is the number of bytes currently being * processed; together those are the current total operation length */ - s->common.len = s->common.offset + s->bytes_in_flight + - cnt * BDRV_SECTOR_SIZE; + s->common.len = s->common.offset + s->bytes_in_flight + cnt; /* Note that even when no rate limit is applied we need to yield * periodically with no pending I/O so that bdrv_drain_all() returns. @@ -XXX,XX +XXX,XX @@ static void coroutine_fn mirror_run(void *opaque) s->common.iostatus == BLOCK_DEVICE_IO_STATUS_OK) { if (s->in_flight >= MAX_IN_FLIGHT || s->buf_free_count == 0 || (cnt == 0 && s->in_flight > 0)) { - trace_mirror_yield(s, cnt * BDRV_SECTOR_SIZE, - s->buf_free_count, s->in_flight); + trace_mirror_yield(s, cnt, s->buf_free_count, s->in_flight); mirror_wait_for_io(s); continue; } else if (cnt != 0) { @@ -XXX,XX +XXX,XX @@ static void coroutine_fn mirror_run(void *opaque) * whether to switch to target check one last time if I/O has * come in the meanwhile, and if not flush the data to disk. */ - trace_mirror_before_drain(s, cnt * BDRV_SECTOR_SIZE); + trace_mirror_before_drain(s, cnt); bdrv_drained_begin(bs); cnt = bdrv_get_dirty_count(s->dirty_bitmap); @@ -XXX,XX +XXX,XX @@ static void coroutine_fn mirror_run(void *opaque) } ret = 0; - trace_mirror_before_sleep(s, cnt * BDRV_SECTOR_SIZE, - s->synced, delay_ns); + trace_mirror_before_sleep(s, cnt, s->synced, delay_ns); if (!s->synced) { block_job_sleep_ns(&s->common, QEMU_CLOCK_REALTIME, delay_ns); if (block_job_is_cancelled(&s->common)) { diff --git a/migration/block.c b/migration/block.c index XXXXXXX..XXXXXXX 100644 --- a/migration/block.c +++ b/migration/block.c @@ -XXX,XX +XXX,XX @@ static int64_t get_remaining_dirty(void) aio_context_release(blk_get_aio_context(bmds->blk)); } - return dirty << BDRV_SECTOR_BITS; + return dirty; } -- 2.13.6 From: Eric Blake <eblake@redhat.com> Half the callers were already scaling bytes to sectors; the other half can eventually be simplified to use byte iteration. Both callers were already using the result as a bool, so make that explicit. Making the change also makes it easier for a future dirty-bitmap patch to offload scaling over to the internal hbitmap. Remember, asking whether a byte is dirty is effectively asking whether the entire granularity containing the byte is dirty, since we only track dirtiness by granularity. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/dirty-bitmap.h | 4 ++-- block/dirty-bitmap.c | 8 ++++---- block/mirror.c | 3 +-- migration/block.c | 3 ++- 4 files changed, 9 insertions(+), 9 deletions(-) diff --git a/include/block/dirty-bitmap.h b/include/block/dirty-bitmap.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/dirty-bitmap.h +++ b/include/block/dirty-bitmap.h @@ -XXX,XX +XXX,XX @@ void bdrv_dirty_bitmap_set_persistance(BdrvDirtyBitmap *bitmap, /* Functions that require manual locking. */ void bdrv_dirty_bitmap_lock(BdrvDirtyBitmap *bitmap); void bdrv_dirty_bitmap_unlock(BdrvDirtyBitmap *bitmap); -int bdrv_get_dirty_locked(BlockDriverState *bs, BdrvDirtyBitmap *bitmap, - int64_t sector); +bool bdrv_get_dirty_locked(BlockDriverState *bs, BdrvDirtyBitmap *bitmap, + int64_t offset); void bdrv_set_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, int64_t cur_sector, int64_t nr_sectors); void bdrv_reset_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ BlockDirtyInfoList *bdrv_query_dirty_bitmaps(BlockDriverState *bs) } /* Called within bdrv_dirty_bitmap_lock..unlock */ -int bdrv_get_dirty_locked(BlockDriverState *bs, BdrvDirtyBitmap *bitmap, - int64_t sector) +bool bdrv_get_dirty_locked(BlockDriverState *bs, BdrvDirtyBitmap *bitmap, + int64_t offset) { if (bitmap) { - return hbitmap_get(bitmap->bitmap, sector); + return hbitmap_get(bitmap->bitmap, offset >> BDRV_SECTOR_BITS); } else { - return 0; + return false; } } diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s) int64_t next_offset = offset + nb_chunks * s->granularity; int64_t next_chunk = next_offset / s->granularity; if (next_offset >= s->bdev_length || - !bdrv_get_dirty_locked(source, s->dirty_bitmap, - next_offset >> BDRV_SECTOR_BITS)) { + !bdrv_get_dirty_locked(source, s->dirty_bitmap, next_offset)) { break; } if (test_bit(next_chunk, s->in_flight_bitmap)) { diff --git a/migration/block.c b/migration/block.c index XXXXXXX..XXXXXXX 100644 --- a/migration/block.c +++ b/migration/block.c @@ -XXX,XX +XXX,XX @@ static int mig_save_device_dirty(QEMUFile *f, BlkMigDevState *bmds, blk_mig_unlock(); } bdrv_dirty_bitmap_lock(bmds->dirty_bitmap); - if (bdrv_get_dirty_locked(bs, bmds->dirty_bitmap, sector)) { + if (bdrv_get_dirty_locked(bs, bmds->dirty_bitmap, + sector * BDRV_SECTOR_SIZE)) { if (total_sectors - sector < BDRV_SECTORS_PER_DIRTY_CHUNK) { nr_sectors = total_sectors - sector; } else { -- 2.13.6 From: Eric Blake <eblake@redhat.com> Some of the callers were already scaling bytes to sectors; others can be easily converted to pass byte offsets, all in our shift towards a consistent byte interface everywhere. Making the change will also make it easier to write the hold-out callers to use byte rather than sectors for their iterations; it also makes it easier for a future dirty-bitmap patch to offload scaling over to the internal hbitmap. Although all callers happen to pass sector-aligned values, make the internal scaling robust to any sub-sector requests. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/dirty-bitmap.h | 8 ++++---- block/dirty-bitmap.c | 22 ++++++++++++++-------- block/mirror.c | 16 ++++++++-------- migration/block.c | 7 +++++-- 4 files changed, 31 insertions(+), 22 deletions(-) diff --git a/include/block/dirty-bitmap.h b/include/block/dirty-bitmap.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/dirty-bitmap.h +++ b/include/block/dirty-bitmap.h @@ -XXX,XX +XXX,XX @@ const char *bdrv_dirty_bitmap_name(const BdrvDirtyBitmap *bitmap); int64_t bdrv_dirty_bitmap_size(const BdrvDirtyBitmap *bitmap); DirtyBitmapStatus bdrv_dirty_bitmap_status(BdrvDirtyBitmap *bitmap); void bdrv_set_dirty_bitmap(BdrvDirtyBitmap *bitmap, - int64_t cur_sector, int64_t nr_sectors); + int64_t offset, int64_t bytes); void bdrv_reset_dirty_bitmap(BdrvDirtyBitmap *bitmap, - int64_t cur_sector, int64_t nr_sectors); + int64_t offset, int64_t bytes); BdrvDirtyBitmapIter *bdrv_dirty_meta_iter_new(BdrvDirtyBitmap *bitmap); BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap); void bdrv_dirty_iter_free(BdrvDirtyBitmapIter *iter); @@ -XXX,XX +XXX,XX @@ void bdrv_dirty_bitmap_unlock(BdrvDirtyBitmap *bitmap); bool bdrv_get_dirty_locked(BlockDriverState *bs, BdrvDirtyBitmap *bitmap, int64_t offset); void bdrv_set_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, - int64_t cur_sector, int64_t nr_sectors); + int64_t offset, int64_t bytes); void bdrv_reset_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, - int64_t cur_sector, int64_t nr_sectors); + int64_t offset, int64_t bytes); int64_t bdrv_dirty_iter_next(BdrvDirtyBitmapIter *iter); void bdrv_set_dirty_iter(BdrvDirtyBitmapIter *hbi, int64_t offset); int64_t bdrv_get_dirty_count(BdrvDirtyBitmap *bitmap); diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ int64_t bdrv_dirty_iter_next(BdrvDirtyBitmapIter *iter) /* Called within bdrv_dirty_bitmap_lock..unlock */ void bdrv_set_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, - int64_t cur_sector, int64_t nr_sectors) + int64_t offset, int64_t bytes) { + int64_t end_sector = DIV_ROUND_UP(offset + bytes, BDRV_SECTOR_SIZE); + assert(bdrv_dirty_bitmap_enabled(bitmap)); assert(!bdrv_dirty_bitmap_readonly(bitmap)); - hbitmap_set(bitmap->bitmap, cur_sector, nr_sectors); + hbitmap_set(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, + end_sector - (offset >> BDRV_SECTOR_BITS)); } void bdrv_set_dirty_bitmap(BdrvDirtyBitmap *bitmap, - int64_t cur_sector, int64_t nr_sectors) + int64_t offset, int64_t bytes) { bdrv_dirty_bitmap_lock(bitmap); - bdrv_set_dirty_bitmap_locked(bitmap, cur_sector, nr_sectors); + bdrv_set_dirty_bitmap_locked(bitmap, offset, bytes); bdrv_dirty_bitmap_unlock(bitmap); } /* Called within bdrv_dirty_bitmap_lock..unlock */ void bdrv_reset_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, - int64_t cur_sector, int64_t nr_sectors) + int64_t offset, int64_t bytes) { + int64_t end_sector = DIV_ROUND_UP(offset + bytes, BDRV_SECTOR_SIZE); + assert(bdrv_dirty_bitmap_enabled(bitmap)); assert(!bdrv_dirty_bitmap_readonly(bitmap)); - hbitmap_reset(bitmap->bitmap, cur_sector, nr_sectors); + hbitmap_reset(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, + end_sector - (offset >> BDRV_SECTOR_BITS)); } void bdrv_reset_dirty_bitmap(BdrvDirtyBitmap *bitmap, - int64_t cur_sector, int64_t nr_sectors) + int64_t offset, int64_t bytes) { bdrv_dirty_bitmap_lock(bitmap); - bdrv_reset_dirty_bitmap_locked(bitmap, cur_sector, nr_sectors); + bdrv_reset_dirty_bitmap_locked(bitmap, offset, bytes); bdrv_dirty_bitmap_unlock(bitmap); } diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static void mirror_write_complete(void *opaque, int ret) if (ret < 0) { BlockErrorAction action; - bdrv_set_dirty_bitmap(s->dirty_bitmap, op->offset >> BDRV_SECTOR_BITS, - op->bytes >> BDRV_SECTOR_BITS); + bdrv_set_dirty_bitmap(s->dirty_bitmap, op->offset, op->bytes); action = mirror_error_action(s, false, -ret); if (action == BLOCK_ERROR_ACTION_REPORT && s->ret >= 0) { s->ret = ret; @@ -XXX,XX +XXX,XX @@ static void mirror_read_complete(void *opaque, int ret) if (ret < 0) { BlockErrorAction action; - bdrv_set_dirty_bitmap(s->dirty_bitmap, op->offset >> BDRV_SECTOR_BITS, - op->bytes >> BDRV_SECTOR_BITS); + bdrv_set_dirty_bitmap(s->dirty_bitmap, op->offset, op->bytes); action = mirror_error_action(s, true, -ret); if (action == BLOCK_ERROR_ACTION_REPORT && s->ret >= 0) { s->ret = ret; @@ -XXX,XX +XXX,XX @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s) * calling bdrv_get_block_status_above could yield - if some blocks are * marked dirty in this window, we need to know. */ - bdrv_reset_dirty_bitmap_locked(s->dirty_bitmap, offset >> BDRV_SECTOR_BITS, - nb_chunks * sectors_per_chunk); + bdrv_reset_dirty_bitmap_locked(s->dirty_bitmap, offset, + nb_chunks * s->granularity); bdrv_dirty_bitmap_unlock(s->dirty_bitmap); bitmap_set(s->in_flight_bitmap, offset / s->granularity, nb_chunks); @@ -XXX,XX +XXX,XX @@ static int coroutine_fn mirror_dirty_init(MirrorBlockJob *s) if (base == NULL && !bdrv_has_zero_init(target_bs)) { if (!bdrv_can_write_zeroes_with_unmap(target_bs)) { - bdrv_set_dirty_bitmap(s->dirty_bitmap, 0, end); + bdrv_set_dirty_bitmap(s->dirty_bitmap, 0, s->bdev_length); return 0; } @@ -XXX,XX +XXX,XX @@ static int coroutine_fn mirror_dirty_init(MirrorBlockJob *s) n = count >> BDRV_SECTOR_BITS; assert(n > 0); if (ret == 1) { - bdrv_set_dirty_bitmap(s->dirty_bitmap, sector_num, n); + bdrv_set_dirty_bitmap(s->dirty_bitmap, + sector_num * BDRV_SECTOR_SIZE, + n * BDRV_SECTOR_SIZE); } sector_num += n; } diff --git a/migration/block.c b/migration/block.c index XXXXXXX..XXXXXXX 100644 --- a/migration/block.c +++ b/migration/block.c @@ -XXX,XX +XXX,XX @@ static int mig_save_device_bulk(QEMUFile *f, BlkMigDevState *bmds) blk->aiocb = blk_aio_preadv(bb, cur_sector * BDRV_SECTOR_SIZE, &blk->qiov, 0, blk_mig_read_cb, blk); - bdrv_reset_dirty_bitmap(bmds->dirty_bitmap, cur_sector, nr_sectors); + bdrv_reset_dirty_bitmap(bmds->dirty_bitmap, cur_sector * BDRV_SECTOR_SIZE, + nr_sectors * BDRV_SECTOR_SIZE); aio_context_release(blk_get_aio_context(bmds->blk)); qemu_mutex_unlock_iothread(); @@ -XXX,XX +XXX,XX @@ static int mig_save_device_dirty(QEMUFile *f, BlkMigDevState *bmds, } else { nr_sectors = BDRV_SECTORS_PER_DIRTY_CHUNK; } - bdrv_reset_dirty_bitmap_locked(bmds->dirty_bitmap, sector, nr_sectors); + bdrv_reset_dirty_bitmap_locked(bmds->dirty_bitmap, + sector * BDRV_SECTOR_SIZE, + nr_sectors * BDRV_SECTOR_SIZE); bdrv_dirty_bitmap_unlock(bmds->dirty_bitmap); blk = g_new(BlkMigBlock, 1); -- 2.13.6 From: Eric Blake <eblake@redhat.com> Now that we have adjusted the majority of the calls this function makes to be byte-based, it is easier to read the code if it makes passes over the image using bytes rather than sectors. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/mirror.c | 38 ++++++++++++++------------------------ 1 file changed, 14 insertions(+), 24 deletions(-) diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static void mirror_throttle(MirrorBlockJob *s) static int coroutine_fn mirror_dirty_init(MirrorBlockJob *s) { - int64_t sector_num, end; + int64_t offset; BlockDriverState *base = s->base; BlockDriverState *bs = s->source; BlockDriverState *target_bs = blk_bs(s->target); - int ret, n; + int ret; int64_t count; - end = s->bdev_length / BDRV_SECTOR_SIZE; - if (base == NULL && !bdrv_has_zero_init(target_bs)) { if (!bdrv_can_write_zeroes_with_unmap(target_bs)) { bdrv_set_dirty_bitmap(s->dirty_bitmap, 0, s->bdev_length); @@ -XXX,XX +XXX,XX @@ static int coroutine_fn mirror_dirty_init(MirrorBlockJob *s) } s->initial_zeroing_ongoing = true; - for (sector_num = 0; sector_num < end; ) { - int nb_sectors = MIN(end - sector_num, - QEMU_ALIGN_DOWN(INT_MAX, s->granularity) >> BDRV_SECTOR_BITS); + for (offset = 0; offset < s->bdev_length; ) { + int bytes = MIN(s->bdev_length - offset, + QEMU_ALIGN_DOWN(INT_MAX, s->granularity)); mirror_throttle(s); @@ -XXX,XX +XXX,XX @@ static int coroutine_fn mirror_dirty_init(MirrorBlockJob *s) continue; } - mirror_do_zero_or_discard(s, sector_num * BDRV_SECTOR_SIZE, - nb_sectors * BDRV_SECTOR_SIZE, false); - sector_num += nb_sectors; + mirror_do_zero_or_discard(s, offset, bytes, false); + offset += bytes; } mirror_wait_for_all_io(s); @@ -XXX,XX +XXX,XX @@ static int coroutine_fn mirror_dirty_init(MirrorBlockJob *s) } /* First part, loop on the sectors and initialize the dirty bitmap. */ - for (sector_num = 0; sector_num < end; ) { + for (offset = 0; offset < s->bdev_length; ) { /* Just to make sure we are not exceeding int limit. */ - int nb_sectors = MIN(INT_MAX >> BDRV_SECTOR_BITS, - end - sector_num); + int bytes = MIN(s->bdev_length - offset, + QEMU_ALIGN_DOWN(INT_MAX, s->granularity)); mirror_throttle(s); @@ -XXX,XX +XXX,XX @@ static int coroutine_fn mirror_dirty_init(MirrorBlockJob *s) return 0; } - ret = bdrv_is_allocated_above(bs, base, sector_num * BDRV_SECTOR_SIZE, - nb_sectors * BDRV_SECTOR_SIZE, &count); + ret = bdrv_is_allocated_above(bs, base, offset, bytes, &count); if (ret < 0) { return ret; } - /* TODO: Relax this once bdrv_is_allocated_above and dirty - * bitmaps no longer require sector alignment. */ - assert(QEMU_IS_ALIGNED(count, BDRV_SECTOR_SIZE)); - n = count >> BDRV_SECTOR_BITS; - assert(n > 0); + assert(count); if (ret == 1) { - bdrv_set_dirty_bitmap(s->dirty_bitmap, - sector_num * BDRV_SECTOR_SIZE, - n * BDRV_SECTOR_SIZE); + bdrv_set_dirty_bitmap(s->dirty_bitmap, offset, count); } - sector_num += n; + offset += count; } return 0; } -- 2.13.6 From: Eric Blake <eblake@redhat.com> This is new code, but it is easier to read if it makes passes over the image using bytes rather than sectors (and will get easier in the future when bdrv_get_block_status is converted to byte-based). Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/qcow2.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ static BlockMeasureInfo *qcow2_measure(QemuOpts *opts, BlockDriverState *in_bs, */ required = virtual_size; } else { - int cluster_sectors = cluster_size / BDRV_SECTOR_SIZE; - int64_t sector_num; + int64_t offset; int pnum = 0; - for (sector_num = 0; - sector_num < ssize / BDRV_SECTOR_SIZE; - sector_num += pnum) { - int nb_sectors = MIN(ssize / BDRV_SECTOR_SIZE - sector_num, - BDRV_REQUEST_MAX_SECTORS); + for (offset = 0; offset < ssize; + offset += pnum * BDRV_SECTOR_SIZE) { + int nb_sectors = MIN(ssize - offset, + BDRV_REQUEST_MAX_BYTES) / BDRV_SECTOR_SIZE; BlockDriverState *file; int64_t ret; ret = bdrv_get_block_status_above(in_bs, NULL, - sector_num, nb_sectors, + offset >> BDRV_SECTOR_BITS, + nb_sectors, &pnum, &file); if (ret < 0) { error_setg_errno(&local_err, -ret, @@ -XXX,XX +XXX,XX @@ static BlockMeasureInfo *qcow2_measure(QemuOpts *opts, BlockDriverState *in_bs, } else if ((ret & (BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED)) == (BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED)) { /* Extend pnum to end of cluster for next iteration */ - pnum = ROUND_UP(sector_num + pnum, cluster_sectors) - - sector_num; + pnum = (ROUND_UP(offset + pnum * BDRV_SECTOR_SIZE, + cluster_size) - offset) >> BDRV_SECTOR_BITS; /* Count clusters we've seen */ - required += (sector_num % cluster_sectors + pnum) * - BDRV_SECTOR_SIZE; + required += offset % cluster_size + pnum * BDRV_SECTOR_SIZE; } } } -- 2.13.6 From: Eric Blake <eblake@redhat.com> Now that we have adjusted the majority of the calls this function makes to be byte-based, it is easier to read the code if it makes passes over the image using bytes rather than sectors. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/qcow2-bitmap.c | 22 ++++++++-------------- 1 file changed, 8 insertions(+), 14 deletions(-) diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-bitmap.c +++ b/block/qcow2-bitmap.c @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, { int ret = 0; BDRVQcow2State *s = bs->opaque; - uint64_t sector, limit, sbc; + uint64_t offset, limit; uint64_t bm_size = bdrv_dirty_bitmap_size(bitmap); - uint64_t bm_sectors = DIV_ROUND_UP(bm_size, BDRV_SECTOR_SIZE); uint8_t *buf = NULL; uint64_t i, tab_size = size_to_clusters(s, @@ -XXX,XX +XXX,XX @@ static int load_bitmap_data(BlockDriverState *bs, buf = g_malloc(s->cluster_size); limit = bytes_covered_by_bitmap_cluster(s, bitmap); - sbc = limit >> BDRV_SECTOR_BITS; - for (i = 0, sector = 0; i < tab_size; ++i, sector += sbc) { - uint64_t count = MIN(bm_sectors - sector, sbc); + for (i = 0, offset = 0; i < tab_size; ++i, offset += limit) { + uint64_t count = MIN(bm_size - offset, limit); uint64_t entry = bitmap_table[i]; - uint64_t offset = entry & BME_TABLE_ENTRY_OFFSET_MASK; + uint64_t data_offset = entry & BME_TABLE_ENTRY_OFFSET_MASK; assert(check_table_entry(entry, s->cluster_size) == 0); - if (offset == 0) { + if (data_offset == 0) { if (entry & BME_TABLE_ENTRY_FLAG_ALL_ONES) { - bdrv_dirty_bitmap_deserialize_ones(bitmap, - sector * BDRV_SECTOR_SIZE, - count * BDRV_SECTOR_SIZE, + bdrv_dirty_bitmap_deserialize_ones(bitmap, offset, count, false); } else { /* No need to deserialize zeros because the dirty bitmap is * already cleared */ } } else { - ret = bdrv_pread(bs->file, offset, buf, s->cluster_size); + ret = bdrv_pread(bs->file, data_offset, buf, s->cluster_size); if (ret < 0) { goto finish; } - bdrv_dirty_bitmap_deserialize_part(bitmap, buf, - sector * BDRV_SECTOR_SIZE, - count * BDRV_SECTOR_SIZE, + bdrv_dirty_bitmap_deserialize_part(bitmap, buf, offset, count, false); } } -- 2.13.6 From: Eric Blake <eblake@redhat.com> Now that we have adjusted the majority of the calls this function makes to be byte-based, it is easier to read the code if it makes passes over the image using bytes rather than sectors. iotests 165 was rather weak - on a default 64k-cluster image, where bitmap granularity also defaults to 64k bytes, a single cluster of the bitmap table thus covers (64*1024*8) bits which each cover 64k bytes, or 32G of image space. But the test only uses a 1G image, so it cannot trigger any more than one loop of the code in store_bitmap_data(); and it was writing to the first cluster. In order to test that we are properly aligning which portions of the bitmap are being written to the file, we really want to test a case where the first dirty bit returned by bdrv_dirty_iter_next() is not aligned to the start of a cluster, which we can do by modifying the test to write data that doesn't happen to fall in the first cluster of the image. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/qcow2-bitmap.c | 31 ++++++++++++++++--------------- tests/qemu-iotests/165 | 2 +- 2 files changed, 17 insertions(+), 16 deletions(-) diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-bitmap.c +++ b/block/qcow2-bitmap.c @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, { int ret; BDRVQcow2State *s = bs->opaque; - int64_t sector; - uint64_t limit, sbc; + int64_t offset; + uint64_t limit; uint64_t bm_size = bdrv_dirty_bitmap_size(bitmap); - uint64_t bm_sectors = DIV_ROUND_UP(bm_size, BDRV_SECTOR_SIZE); const char *bm_name = bdrv_dirty_bitmap_name(bitmap); uint8_t *buf = NULL; BdrvDirtyBitmapIter *dbi; @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, dbi = bdrv_dirty_iter_new(bitmap); buf = g_malloc(s->cluster_size); limit = bytes_covered_by_bitmap_cluster(s, bitmap); - sbc = limit >> BDRV_SECTOR_BITS; assert(DIV_ROUND_UP(bm_size, limit) == tb_size); - while ((sector = bdrv_dirty_iter_next(dbi) >> BDRV_SECTOR_BITS) >= 0) { - uint64_t cluster = sector / sbc; + while ((offset = bdrv_dirty_iter_next(dbi)) >= 0) { + uint64_t cluster = offset / limit; uint64_t end, write_size; int64_t off; - sector = cluster * sbc; - end = MIN(bm_sectors, sector + sbc); - write_size = bdrv_dirty_bitmap_serialization_size(bitmap, - sector * BDRV_SECTOR_SIZE, (end - sector) * BDRV_SECTOR_SIZE); + /* + * We found the first dirty offset, but want to write out the + * entire cluster of the bitmap that includes that offset, + * including any leading zero bits. + */ + offset = QEMU_ALIGN_DOWN(offset, limit); + end = MIN(bm_size, offset + limit); + write_size = bdrv_dirty_bitmap_serialization_size(bitmap, offset, + end - offset); assert(write_size <= s->cluster_size); off = qcow2_alloc_clusters(bs, s->cluster_size); @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, } tb[cluster] = off; - bdrv_dirty_bitmap_serialize_part(bitmap, buf, - sector * BDRV_SECTOR_SIZE, - (end - sector) * BDRV_SECTOR_SIZE); + bdrv_dirty_bitmap_serialize_part(bitmap, buf, offset, end - offset); if (write_size < s->cluster_size) { memset(buf + write_size, 0, s->cluster_size - write_size); } @@ -XXX,XX +XXX,XX @@ static uint64_t *store_bitmap_data(BlockDriverState *bs, goto fail; } - if (end >= bm_sectors) { + if (end >= bm_size) { break; } - bdrv_set_dirty_iter(dbi, end * BDRV_SECTOR_SIZE); + bdrv_set_dirty_iter(dbi, end); } *bitmap_table_size = tb_size; diff --git a/tests/qemu-iotests/165 b/tests/qemu-iotests/165 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/165 +++ b/tests/qemu-iotests/165 @@ -XXX,XX +XXX,XX @@ disk = os.path.join(iotests.test_dir, 'disk') disk_size = 0x40000000 # 1G # regions for qemu_io: (start, count) in bytes -regions1 = ((0, 0x100000), +regions1 = ((0x0fff00, 0x10000), (0x200000, 0x100000)) regions2 = ((0x10000000, 0x20000), -- 2.13.6 From: Eric Blake <eblake@redhat.com> Both callers already had bytes available, but were scaling to sectors. Move the scaling to internal code. In the case of bdrv_aligned_pwritev(), we are now passing the exact offset rather than a rounded sector-aligned value, but that's okay as long as dirty bitmap widens start/bytes to granularity boundaries. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/block_int.h | 2 +- block/dirty-bitmap.c | 7 ++++--- block/io.c | 6 ++---- 3 files changed, 7 insertions(+), 8 deletions(-) diff --git a/include/block/block_int.h b/include/block/block_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block_int.h +++ b/include/block/block_int.h @@ -XXX,XX +XXX,XX @@ void blk_dev_eject_request(BlockBackend *blk, bool force); bool blk_dev_is_tray_open(BlockBackend *blk); bool blk_dev_is_medium_locked(BlockBackend *blk); -void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector, int64_t nr_sect); +void bdrv_set_dirty(BlockDriverState *bs, int64_t offset, int64_t bytes); bool bdrv_requests_pending(BlockDriverState *bs); void bdrv_clear_dirty_bitmap(BdrvDirtyBitmap *bitmap, HBitmap **out); diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ void bdrv_dirty_bitmap_deserialize_finish(BdrvDirtyBitmap *bitmap) hbitmap_deserialize_finish(bitmap->bitmap); } -void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector, - int64_t nr_sectors) +void bdrv_set_dirty(BlockDriverState *bs, int64_t offset, int64_t bytes) { BdrvDirtyBitmap *bitmap; + int64_t end_sector = DIV_ROUND_UP(offset + bytes, BDRV_SECTOR_SIZE); if (QLIST_EMPTY(&bs->dirty_bitmaps)) { return; @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector, continue; } assert(!bdrv_dirty_bitmap_readonly(bitmap)); - hbitmap_set(bitmap->bitmap, cur_sector, nr_sectors); + hbitmap_set(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, + end_sector - (offset >> BDRV_SECTOR_BITS)); } bdrv_dirty_bitmaps_unlock(bs); } diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_aligned_pwritev(BdrvChild *child, bool waited; int ret; - int64_t start_sector = offset >> BDRV_SECTOR_BITS; int64_t end_sector = DIV_ROUND_UP(offset + bytes, BDRV_SECTOR_SIZE); uint64_t bytes_remaining = bytes; int max_transfer; @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_aligned_pwritev(BdrvChild *child, bdrv_debug_event(bs, BLKDBG_PWRITEV_DONE); atomic_inc(&bs->write_gen); - bdrv_set_dirty(bs, start_sector, end_sector - start_sector); + bdrv_set_dirty(bs, offset, bytes); stat64_max(&bs->wr_highest_offset, offset + bytes); @@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_pdiscard(BlockDriverState *bs, int64_t offset, ret = 0; out: atomic_inc(&bs->write_gen); - bdrv_set_dirty(bs, req.offset >> BDRV_SECTOR_BITS, - req.bytes >> BDRV_SECTOR_BITS); + bdrv_set_dirty(bs, req.offset, req.bytes); tracked_request_end(&req); bdrv_dec_in_flight(bs); return ret; -- 2.13.6 From: Eric Blake <eblake@redhat.com> Now that all callers are using byte-based interfaces, there's no reason for our internal hbitmap to remain with sector-based granularity. It also simplifies our internal scaling, since we already know that hbitmap widens requests out to granularity boundaries. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/dirty-bitmap.c | 62 +++++++++++++++------------------------------------- 1 file changed, 18 insertions(+), 44 deletions(-) diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index XXXXXXX..XXXXXXX 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -XXX,XX +XXX,XX @@ */ struct BdrvDirtyBitmap { QemuMutex *mutex; - HBitmap *bitmap; /* Dirty sector bitmap implementation */ + HBitmap *bitmap; /* Dirty bitmap implementation */ HBitmap *meta; /* Meta dirty bitmap */ BdrvDirtyBitmap *successor; /* Anonymous child; implies frozen status */ char *name; /* Optional non-empty unique ID */ @@ -XXX,XX +XXX,XX @@ BdrvDirtyBitmap *bdrv_create_dirty_bitmap(BlockDriverState *bs, } bitmap = g_new0(BdrvDirtyBitmap, 1); bitmap->mutex = &bs->dirty_bitmap_mutex; - /* - * TODO - let hbitmap track full granularity. For now, it is tracking - * only sector granularity, as a shortcut for our iterators. - */ - bitmap->bitmap = hbitmap_alloc(DIV_ROUND_UP(bitmap_size, BDRV_SECTOR_SIZE), - ctz32(granularity) - BDRV_SECTOR_BITS); + bitmap->bitmap = hbitmap_alloc(bitmap_size, ctz32(granularity)); bitmap->size = bitmap_size; bitmap->name = g_strdup(name); bitmap->disabled = false; @@ -XXX,XX +XXX,XX @@ void bdrv_dirty_bitmap_truncate(BlockDriverState *bs, int64_t bytes) QLIST_FOREACH(bitmap, &bs->dirty_bitmaps, list) { assert(!bdrv_dirty_bitmap_frozen(bitmap)); assert(!bitmap->active_iterators); - hbitmap_truncate(bitmap->bitmap, DIV_ROUND_UP(bytes, BDRV_SECTOR_SIZE)); + hbitmap_truncate(bitmap->bitmap, bytes); bitmap->size = bytes; } bdrv_dirty_bitmaps_unlock(bs); @@ -XXX,XX +XXX,XX @@ bool bdrv_get_dirty_locked(BlockDriverState *bs, BdrvDirtyBitmap *bitmap, int64_t offset) { if (bitmap) { - return hbitmap_get(bitmap->bitmap, offset >> BDRV_SECTOR_BITS); + return hbitmap_get(bitmap->bitmap, offset); } else { return false; } @@ -XXX,XX +XXX,XX @@ uint32_t bdrv_get_default_bitmap_granularity(BlockDriverState *bs) uint32_t bdrv_dirty_bitmap_granularity(const BdrvDirtyBitmap *bitmap) { - return BDRV_SECTOR_SIZE << hbitmap_granularity(bitmap->bitmap); + return 1U << hbitmap_granularity(bitmap->bitmap); } BdrvDirtyBitmapIter *bdrv_dirty_iter_new(BdrvDirtyBitmap *bitmap) @@ -XXX,XX +XXX,XX @@ void bdrv_dirty_iter_free(BdrvDirtyBitmapIter *iter) int64_t bdrv_dirty_iter_next(BdrvDirtyBitmapIter *iter) { - int64_t ret = hbitmap_iter_next(&iter->hbi); - return ret < 0 ? -1 : ret * BDRV_SECTOR_SIZE; + return hbitmap_iter_next(&iter->hbi); } /* Called within bdrv_dirty_bitmap_lock..unlock */ void bdrv_set_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, int64_t offset, int64_t bytes) { - int64_t end_sector = DIV_ROUND_UP(offset + bytes, BDRV_SECTOR_SIZE); - assert(bdrv_dirty_bitmap_enabled(bitmap)); assert(!bdrv_dirty_bitmap_readonly(bitmap)); - hbitmap_set(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, - end_sector - (offset >> BDRV_SECTOR_BITS)); + hbitmap_set(bitmap->bitmap, offset, bytes); } void bdrv_set_dirty_bitmap(BdrvDirtyBitmap *bitmap, @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty_bitmap(BdrvDirtyBitmap *bitmap, void bdrv_reset_dirty_bitmap_locked(BdrvDirtyBitmap *bitmap, int64_t offset, int64_t bytes) { - int64_t end_sector = DIV_ROUND_UP(offset + bytes, BDRV_SECTOR_SIZE); - assert(bdrv_dirty_bitmap_enabled(bitmap)); assert(!bdrv_dirty_bitmap_readonly(bitmap)); - hbitmap_reset(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, - end_sector - (offset >> BDRV_SECTOR_BITS)); + hbitmap_reset(bitmap->bitmap, offset, bytes); } void bdrv_reset_dirty_bitmap(BdrvDirtyBitmap *bitmap, @@ -XXX,XX +XXX,XX @@ void bdrv_clear_dirty_bitmap(BdrvDirtyBitmap *bitmap, HBitmap **out) hbitmap_reset_all(bitmap->bitmap); } else { HBitmap *backup = bitmap->bitmap; - bitmap->bitmap = hbitmap_alloc(DIV_ROUND_UP(bitmap->size, - BDRV_SECTOR_SIZE), + bitmap->bitmap = hbitmap_alloc(bitmap->size, hbitmap_granularity(backup)); *out = backup; } @@ -XXX,XX +XXX,XX @@ void bdrv_undo_clear_dirty_bitmap(BdrvDirtyBitmap *bitmap, HBitmap *in) uint64_t bdrv_dirty_bitmap_serialization_size(const BdrvDirtyBitmap *bitmap, uint64_t offset, uint64_t bytes) { - assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); - return hbitmap_serialization_size(bitmap->bitmap, - offset >> BDRV_SECTOR_BITS, - bytes >> BDRV_SECTOR_BITS); + return hbitmap_serialization_size(bitmap->bitmap, offset, bytes); } uint64_t bdrv_dirty_bitmap_serialization_align(const BdrvDirtyBitmap *bitmap) { - return hbitmap_serialization_align(bitmap->bitmap) * BDRV_SECTOR_SIZE; + return hbitmap_serialization_align(bitmap->bitmap); } void bdrv_dirty_bitmap_serialize_part(const BdrvDirtyBitmap *bitmap, uint8_t *buf, uint64_t offset, uint64_t bytes) { - assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); - hbitmap_serialize_part(bitmap->bitmap, buf, offset >> BDRV_SECTOR_BITS, - bytes >> BDRV_SECTOR_BITS); + hbitmap_serialize_part(bitmap->bitmap, buf, offset, bytes); } void bdrv_dirty_bitmap_deserialize_part(BdrvDirtyBitmap *bitmap, uint8_t *buf, uint64_t offset, uint64_t bytes, bool finish) { - assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); - hbitmap_deserialize_part(bitmap->bitmap, buf, offset >> BDRV_SECTOR_BITS, - bytes >> BDRV_SECTOR_BITS, finish); + hbitmap_deserialize_part(bitmap->bitmap, buf, offset, bytes, finish); } void bdrv_dirty_bitmap_deserialize_zeroes(BdrvDirtyBitmap *bitmap, uint64_t offset, uint64_t bytes, bool finish) { - assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); - hbitmap_deserialize_zeroes(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, - bytes >> BDRV_SECTOR_BITS, finish); + hbitmap_deserialize_zeroes(bitmap->bitmap, offset, bytes, finish); } void bdrv_dirty_bitmap_deserialize_ones(BdrvDirtyBitmap *bitmap, uint64_t offset, uint64_t bytes, bool finish) { - assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)); - hbitmap_deserialize_ones(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, - bytes >> BDRV_SECTOR_BITS, finish); + hbitmap_deserialize_ones(bitmap->bitmap, offset, bytes, finish); } void bdrv_dirty_bitmap_deserialize_finish(BdrvDirtyBitmap *bitmap) @@ -XXX,XX +XXX,XX @@ void bdrv_dirty_bitmap_deserialize_finish(BdrvDirtyBitmap *bitmap) void bdrv_set_dirty(BlockDriverState *bs, int64_t offset, int64_t bytes) { BdrvDirtyBitmap *bitmap; - int64_t end_sector = DIV_ROUND_UP(offset + bytes, BDRV_SECTOR_SIZE); if (QLIST_EMPTY(&bs->dirty_bitmaps)) { return; @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty(BlockDriverState *bs, int64_t offset, int64_t bytes) continue; } assert(!bdrv_dirty_bitmap_readonly(bitmap)); - hbitmap_set(bitmap->bitmap, offset >> BDRV_SECTOR_BITS, - end_sector - (offset >> BDRV_SECTOR_BITS)); + hbitmap_set(bitmap->bitmap, offset, bytes); } bdrv_dirty_bitmaps_unlock(bs); } @@ -XXX,XX +XXX,XX @@ void bdrv_set_dirty(BlockDriverState *bs, int64_t offset, int64_t bytes) */ void bdrv_set_dirty_iter(BdrvDirtyBitmapIter *iter, int64_t offset) { - hbitmap_iter_init(&iter->hbi, iter->hbi.hb, offset >> BDRV_SECTOR_BITS); + hbitmap_iter_init(&iter->hbi, iter->hbi.hb, offset); } int64_t bdrv_get_dirty_count(BdrvDirtyBitmap *bitmap) { - return hbitmap_count(bitmap->bitmap) << BDRV_SECTOR_BITS; + return hbitmap_count(bitmap->bitmap); } int64_t bdrv_get_meta_dirty_count(BdrvDirtyBitmap *bitmap) -- 2.13.6 From: Thomas Huth <thuth@redhat.com> The condition of the for-loop makes sure that b is always smaller than s->blocks, so the "if (b >= s->blocks)" statement is completely superfluous here. Buglink: https://bugs.launchpad.net/qemu/+bug/1715007 Signed-off-by: Thomas Huth <thuth@redhat.com> Reviewed-by: Laurent Vivier <lvivier@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- hw/block/onenand.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/hw/block/onenand.c b/hw/block/onenand.c index XXXXXXX..XXXXXXX 100644 --- a/hw/block/onenand.c +++ b/hw/block/onenand.c @@ -XXX,XX +XXX,XX @@ static void onenand_command(OneNANDState *s) s->intstatus |= ONEN_INT; for (b = 0; b < s->blocks; b ++) { - if (b >= s->blocks) { - s->status |= ONEN_ERR_CMD; - break; - } if (s->blockwp[b] == ONEN_LOCK_LOCKTIGHTEN) break; -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> This includes shell function, shell variables and command line options (randomize.awk does not exist). Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/check | 28 ----------------- tests/qemu-iotests/common | 23 -------------- tests/qemu-iotests/common.config | 26 --------------- tests/qemu-iotests/common.rc | 68 ---------------------------------------- 4 files changed, 145 deletions(-) diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -XXX,XX +XXX,XX @@ then export SOCKET_SCM_HELPER="$build_iotests/socket_scm_helper" fi -# if ./qemu exists, it should be prioritized and will be chosen by common.config if [[ -z "$QEMU_PROG" && ! -x './qemu' ]] then arch=$(uname -m 2> /dev/null) @@ -XXX,XX +XXX,XX @@ _timestamp() _wrapup() { - # for hangcheck ... - # remove files that were used by hangcheck - # - [ -f "${TEST_DIR}"/check.pid ] && rm -rf "${TEST_DIR}"/check.pid - [ -f "${TEST_DIR}"/check.sts ] && rm -rf "${TEST_DIR}"/check.sts - if $showme then : @@ -XXX,XX +XXX,XX @@ END { if (NR > 0) { trap "_wrapup; exit \$status" 0 1 2 3 15 -# for hangcheck ... -# Save pid of check in a well known place, so that hangcheck can be sure it -# has the right pid (getting the pid from ps output is not reliable enough). -# -rm -rf "${TEST_DIR}"/check.pid -echo $$ > "${TEST_DIR}"/check.pid - -# for hangcheck ... -# Save the status of check in a well known place, so that hangcheck can be -# sure to know where check is up to (getting test number from ps output is -# not reliable enough since the trace stuff has been introduced). -# -rm -rf "${TEST_DIR}"/check.sts -echo "preamble" > "${TEST_DIR}"/check.sts - -# don't leave old full output behind on a clean run -rm -f check.full - [ -f $TIMESTAMP_FILE ] || touch $TIMESTAMP_FILE FULL_IMGFMT_DETAILS=`_full_imgfmt_details` @@ -XXX,XX +XXX,XX @@ do fi rm -f core $seq.notrun - # for hangcheck ... - echo "$seq" > "${TEST_DIR}"/check.sts - start=`_wallclock` $timestamp && printf %s " [$(date "+%T")]" diff --git a/tests/qemu-iotests/common b/tests/qemu-iotests/common index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common +++ b/tests/qemu-iotests/common @@ -XXX,XX +XXX,XX @@ # common procedures for QA scripts # -_setenvironment() -{ - MSGVERB="text:action" - export MSGVERB -} - -rm -f "$OUTPUT_DIR/$iam.out" -_setenvironment - -check=${check-true} - diff="diff -u" verbose=false debug=false @@ -XXX,XX +XXX,XX @@ showme=false sortme=false expunge=true have_test_arg=false -randomize=false cachemode=false rm -f $tmp.list $tmp.tmp $tmp.sed @@ -XXX,XX +XXX,XX @@ other options -n show me, do not run tests -o options -o options to pass to qemu-img create/convert -T output timestamps - -r randomize test order -c mode cache mode testlist options @@ -XXX,XX +XXX,XX @@ testlist options cachemode=true xpand=false ;; - -r) # randomize test order - randomize=true - xpand=false - ;; - -T) # turn on timestamp output timestamp=true xpand=false @@ -XXX,XX +XXX,XX @@ fi list=`sort $tmp.list` rm -f $tmp.list $tmp.tmp $tmp.sed -if $randomize -then - list=`echo $list | awk -f randomize.awk` -fi - [ "$QEMU" = "" ] && _fatal "qemu not found" [ "$QEMU_IMG" = "" ] && _fatal "qemu-img not found" [ "$QEMU_IO" = "" ] && _fatal "qemu-io not found" diff --git a/tests/qemu-iotests/common.config b/tests/qemu-iotests/common.config index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.config +++ b/tests/qemu-iotests/common.config @@ -XXX,XX +XXX,XX @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. # -# -# setup and check for config parameters, and in particular -# -# EMAIL - email of the script runner. -# TEST_DIR - scratch test directory -# -# - These can be added to $HOST_CONFIG_DIR (witch default to ./config) -# below or a separate local configuration file can be used (using -# the HOST_OPTIONS variable). -# - This script is shared by the stress test system and the auto-qa -# system (includes both regression test and benchmark components). -# - this script shouldn't make any assertions about filesystem -# validity or mountedness. -# - # all tests should use a common language setting to prevent golden # output mismatches. export LANG=C PATH=".:$PATH" -HOST=`hostname -s 2> /dev/null` HOSTOS=`uname -s` -EMAIL=root@localhost # where auto-qa will send its status messages -export HOST_OPTIONS=${HOST_OPTIONS:=local.config} -export CHECK_OPTIONS=${CHECK_OPTIONS:="-g auto"} export PWD=`pwd` export _QEMU_HANDLE=0 @@ -XXX,XX +XXX,XX @@ _fatal() export AWK_PROG="`set_prog_path awk`" [ "$AWK_PROG" = "" ] && _fatal "awk not found" -export SED_PROG="`set_prog_path sed`" -[ "$SED_PROG" = "" ] && _fatal "sed not found" - -export PS_ALL_FLAGS="-ef" - if [ -z "$QEMU_PROG" ]; then export QEMU_PROG="`set_prog_path qemu`" fi @@ -XXX,XX +XXX,XX @@ fi export QEMU_DEFAULT_MACHINE="$default_machine" -[ -f /etc/qemu-iotest.config ] && . /etc/qemu-iotest.config - if [ -z "$TEST_DIR" ]; then TEST_DIR=`pwd`/scratch fi diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.rc +++ b/tests/qemu-iotests/common.rc @@ -XXX,XX +XXX,XX @@ _img_info() done } -_get_pids_by_name() -{ - if [ $# -ne 1 ] - then - echo "Usage: _get_pids_by_name process-name" 1>&2 - exit 1 - fi - - # Algorithm ... all ps(1) variants have a time of the form MM:SS or - # HH:MM:SS before the psargs field, use this as the search anchor. - # - # Matches with $1 (process-name) occur if the first psarg is $1 - # or ends in /$1 ... the matching uses sed's regular expressions, - # so passing a regex into $1 will work. - - ps $PS_ALL_FLAGS \ - | sed -n \ - -e 's/$/ /' \ - -e 's/[ ][ ]*/ /g' \ - -e 's/^ //' \ - -e 's/^[^ ]* //' \ - -e "/[0-9]:[0-9][0-9] *[^ ]*\/$1 /s/ .*//p" \ - -e "/[0-9]:[0-9][0-9] *$1 /s/ .*//p" -} - -# fqdn for localhost -# -_get_fqdn() -{ - host=`hostname` - $NSLOOKUP_PROG $host | $AWK_PROG '{ if ($1 == "Name:") print $2 }' -} - -# check if run as root -# -_need_to_be_root() -{ - id=`id | $SED_PROG -e 's/(.*//' -e 's/.*=//'` - if [ "$id" -ne 0 ] - then - echo "Arrgh ... you need to be root (not uid=$id) to run this test" - exit 1 - fi -} - # bail out, setting up .notrun file # _notrun() @@ -XXX,XX +XXX,XX @@ _full_platform_details() echo "$os/$platform $host $kernel" } -_link_out_file() -{ - if [ -z "$1" ]; then - echo Error must pass \$seq. - exit - fi - rm -f $1 - if [ "`uname`" == "IRIX64" ] || [ "`uname`" == "IRIX" ]; then - ln -s $1.irix $1 - elif [ "`uname`" == "Linux" ]; then - ln -s $1.linux $1 - else - echo Error test $seq does not run on the operating system: `uname` - exit - fi -} - -_die() -{ - echo $@ - exit 1 -} - # make sure this script returns success true -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/check | 4 ++-- tests/qemu-iotests/common | 2 +- tests/qemu-iotests/common.config | 3 --- 3 files changed, 3 insertions(+), 6 deletions(-) diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -XXX,XX +XXX,XX @@ tmp="${TEST_DIR}"/$$ _wallclock() { - date "+%H %M %S" | $AWK_PROG '{ print $1*3600 + $2*60 + $3 }' + date "+%H %M %S" | awk '{ print $1*3600 + $2*60 + $3 }' } _timestamp() @@ -XXX,XX +XXX,XX @@ _wrapup() if [ -f $TIMESTAMP_FILE -a -f $tmp.time ] then cat $TIMESTAMP_FILE $tmp.time \ - | $AWK_PROG ' + | awk ' { t[$1] = $2 } END { if (NR > 0) { for (i in t) print i " " t[i] diff --git a/tests/qemu-iotests/common b/tests/qemu-iotests/common index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common +++ b/tests/qemu-iotests/common @@ -XXX,XX +XXX,XX @@ testlist options if $xpand then have_test_arg=true - $AWK_PROG </dev/null ' + awk </dev/null ' BEGIN { for (t='$start'; t<='$end'; t++) printf "%03d\n",t }' \ | while read id do diff --git a/tests/qemu-iotests/common.config b/tests/qemu-iotests/common.config index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.config +++ b/tests/qemu-iotests/common.config @@ -XXX,XX +XXX,XX @@ _fatal() exit 1 } -export AWK_PROG="`set_prog_path awk`" -[ "$AWK_PROG" = "" ] && _fatal "awk not found" - if [ -z "$QEMU_PROG" ]; then export QEMU_PROG="`set_prog_path qemu`" fi -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> Some functions in common.rc are never used by the tests. Move them out of that file and into common, which is already included only by "check". Code that actually *is* common to "check" and tests can be placed in common.config. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/common | 25 ++++++++++++++++++++++++- tests/qemu-iotests/common.config | 12 ++++++++++++ tests/qemu-iotests/common.rc | 40 ---------------------------------------- 3 files changed, 36 insertions(+), 41 deletions(-) diff --git a/tests/qemu-iotests/common b/tests/qemu-iotests/common index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common +++ b/tests/qemu-iotests/common @@ -XXX,XX +XXX,XX @@ # common procedures for QA scripts # +_full_imgfmt_details() +{ + if [ -n "$IMGOPTS" ]; then + echo "$IMGFMT ($IMGOPTS)" + else + echo "$IMGFMT" + fi +} + +_full_platform_details() +{ + os=`uname -s` + host=`hostname -s` + kernel=`uname -r` + platform=`uname -m` + echo "$os/$platform $host $kernel" +} + diff="diff -u" verbose=false debug=false @@ -XXX,XX +XXX,XX @@ if [ "$IMGOPTSSYNTAX" != "true" ]; then fi # Set default options for qemu-img create -o if they were not specified -_set_default_imgopts +if [ "$IMGFMT" == "qcow2" ] && ! (echo "$IMGOPTS" | grep "compat=" > /dev/null); then + IMGOPTS=$(_optstr_add "$IMGOPTS" "compat=1.1") +fi +if [ "$IMGFMT" == "luks" ] && ! (echo "$IMGOPTS" | grep "iter-time=" > /dev/null); then + IMGOPTS=$(_optstr_add "$IMGOPTS" "iter-time=10") +fi if [ -s $tmp.list ] then diff --git a/tests/qemu-iotests/common.config b/tests/qemu-iotests/common.config index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.config +++ b/tests/qemu-iotests/common.config @@ -XXX,XX +XXX,XX @@ export PWD=`pwd` export _QEMU_HANDLE=0 +# make sure we have a standard umask +umask 022 + # $1 = prog to look for, $2* = default pathnames if not found in $PATH set_prog_path() { @@ -XXX,XX +XXX,XX @@ set_prog_path() return 1 } +_optstr_add() +{ + if [ -n "$1" ]; then + echo "$1,$2" + else + echo "$2" + fi +} + _fatal() { echo "$*" diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.rc +++ b/tests/qemu-iotests/common.rc @@ -XXX,XX +XXX,XX @@ then fi fi -# make sure we have a standard umask -umask 022 - if [ "$IMGOPTSSYNTAX" = "true" ]; then DRIVER="driver=$IMGFMT" if [ "$IMGFMT" = "luks" ]; then @@ -XXX,XX +XXX,XX @@ else fi ORIG_TEST_IMG="$TEST_IMG" -_optstr_add() -{ - if [ -n "$1" ]; then - echo "$1,$2" - else - echo "$2" - fi -} - -_set_default_imgopts() -{ - if [ "$IMGFMT" == "qcow2" ] && ! (echo "$IMGOPTS" | grep "compat=" > /dev/null); then - IMGOPTS=$(_optstr_add "$IMGOPTS" "compat=1.1") - fi - if [ "$IMGFMT" == "luks" ] && ! (echo "$IMGOPTS" | grep "iter-time=" > /dev/null); then - IMGOPTS=$(_optstr_add "$IMGOPTS" "iter-time=10") - fi -} - _use_sample_img() { SAMPLE_IMG_FILE="${1%\.bz2}" @@ -XXX,XX +XXX,XX @@ _require_command() [ -x "$c" ] || _notrun "$1 utility required, skipped this test" } -_full_imgfmt_details() -{ - if [ -n "$IMGOPTS" ]; then - echo "$IMGFMT ($IMGOPTS)" - else - echo "$IMGFMT" - fi -} - -_full_platform_details() -{ - os=`uname -s` - host=`hostname -s` - kernel=`uname -r` - platform=`uname -m` - echo "$os/$platform $host $kernel" -} - # make sure this script returns success true -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> Instead of ./check failing when a binary is missing, we try each test case now and each one fails with tons of test case diffs. Also, all the variables were initialized by "check" prior to "common" being sourced, and then (uselessly) checked for emptiness again in "check". Centralize the search for programs in "common" (which will soon be one with "check"), including the "realpath" invocation which can be done just once in "check" rather than in the tests. For qnio_server, move the detection to "common", simplifying set_prog_path to stop handling the unused second argument, and embedding the "realpath" pass. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/check | 41 --------------------- tests/qemu-iotests/common | 77 +++++++++++++++++++++++++++++++++++++--- tests/qemu-iotests/common.config | 61 +------------------------------ 3 files changed, 73 insertions(+), 106 deletions(-) diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -XXX,XX +XXX,XX @@ fi build_root="$build_iotests/../.." -if [ -x "$build_iotests/socket_scm_helper" ] -then - export SOCKET_SCM_HELPER="$build_iotests/socket_scm_helper" -fi - -if [[ -z "$QEMU_PROG" && ! -x './qemu' ]] -then - arch=$(uname -m 2> /dev/null) - - if [[ -n $arch && -x "$build_root/$arch-softmmu/qemu-system-$arch" ]] - then - export QEMU_PROG="$build_root/$arch-softmmu/qemu-system-$arch" - else - pushd "$build_root" > /dev/null - for binary in *-softmmu/qemu-system-* - do - if [ -x "$binary" ] - then - export QEMU_PROG="$build_root/$binary" - break - fi - done - popd > /dev/null - fi -fi - -if [[ -z $QEMU_IMG_PROG && -x "$build_root/qemu-img" && ! -x './qemu-img' ]] -then - export QEMU_IMG_PROG="$build_root/qemu-img" -fi - -if [[ -z $QEMU_IO_PROG && -x "$build_root/qemu-io" && ! -x './qemu-io' ]] -then - export QEMU_IO_PROG="$build_root/qemu-io" -fi - -if [[ -z $QEMU_NBD_PROG && -x "$build_root/qemu-nbd" && ! -x './qemu-nbd' ]] -then - export QEMU_NBD_PROG="$build_root/qemu-nbd" -fi - # we need common.env if ! . "$build_iotests/common.env" then diff --git a/tests/qemu-iotests/common b/tests/qemu-iotests/common index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common +++ b/tests/qemu-iotests/common @@ -XXX,XX +XXX,XX @@ _full_platform_details() echo "$os/$platform $host $kernel" } +# $1 = prog to look for +set_prog_path() +{ + p=`command -v $1 2> /dev/null` + if [ -n "$p" -a -x "$p" ]; then + realpath -- "$(type -p "$p")" + else + return 1 + fi +} + diff="diff -u" verbose=false debug=false @@ -XXX,XX +XXX,XX @@ fi list=`sort $tmp.list` rm -f $tmp.list $tmp.tmp $tmp.sed -[ "$QEMU" = "" ] && _fatal "qemu not found" -[ "$QEMU_IMG" = "" ] && _fatal "qemu-img not found" -[ "$QEMU_IO" = "" ] && _fatal "qemu-io not found" +if [ -z "$QEMU_PROG" ] +then + if [ -x "$build_iotests/qemu" ]; then + export QEMU_PROG="$build_iotests/qemu" + elif [ -x "$build_root/$arch-softmmu/qemu-system-$arch" ]; then + export QEMU_PROG="$build_root/$arch-softmmu/qemu-system-$arch" + else + pushd "$build_root" > /dev/null + for binary in *-softmmu/qemu-system-* + do + if [ -x "$binary" ] + then + export QEMU_PROG="$build_root/$binary" + break + fi + done + popd > /dev/null + [ "$QEMU_PROG" = "" ] && _init_error "qemu not found" + fi +fi +export QEMU_PROG=$(realpath -- "$(type -p "$QEMU_PROG")") + +if [ -z "$QEMU_IMG_PROG" ]; then + if [ -x "$build_iotests/qemu-img" ]; then + export QEMU_IMG_PROG="$build_iotests/qemu-img" + elif [ -x "$build_root/qemu-img" ]; then + export QEMU_IMG_PROG="$build_root/qemu-img" + else + _init_error "qemu-img not found" + fi +fi +export QEMU_IMG_PROG=$(realpath -- "$(type -p "$QEMU_IMG_PROG")") + +if [ -z "$QEMU_IO_PROG" ]; then + if [ -x "$build_iotests/qemu-io" ]; then + export QEMU_IO_PROG="$build_iotests/qemu-io" + elif [ -x "$build_root/qemu-io" ]; then + export QEMU_IO_PROG="$build_root/qemu-io" + else + _init_error "qemu-io not found" + fi +fi +export QEMU_IO_PROG=$(realpath -- "$(type -p "$QEMU_IO_PROG")") -if [ "$IMGPROTO" = "nbd" ] ; then - [ "$QEMU_NBD" = "" ] && _fatal "qemu-nbd not found" +if [ -z $QEMU_NBD_PROG ]; then + if [ -x "$build_iotests/qemu-nbd" ]; then + export QEMU_NBD_PROG="$build_iotests/qemu-nbd" + elif [ -x "$build_root/qemu-nbd" ]; then + export QEMU_NBD_PROG="$build_root/qemu-nbd" + else + _init_error "qemu-nbd not found" + fi +fi +export QEMU_NBD_PROG=$(realpath -- "$(type -p "$QEMU_NBD_PROG")") + +if [ -z "$QEMU_VXHS_PROG" ]; then + export QEMU_VXHS_PROG="`set_prog_path qnio_server`" +fi + +if [ -x "$build_iotests/socket_scm_helper" ] +then + export SOCKET_SCM_HELPER="$build_iotests/socket_scm_helper" fi diff --git a/tests/qemu-iotests/common.config b/tests/qemu-iotests/common.config index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.config +++ b/tests/qemu-iotests/common.config @@ -XXX,XX +XXX,XX @@ export LANG=C PATH=".:$PATH" HOSTOS=`uname -s` +arch=`uname -m` export PWD=`pwd` @@ -XXX,XX +XXX,XX @@ export _QEMU_HANDLE=0 # make sure we have a standard umask umask 022 -# $1 = prog to look for, $2* = default pathnames if not found in $PATH -set_prog_path() -{ - p=`command -v $1 2> /dev/null` - if [ -n "$p" -a -x "$p" ]; then - echo $p - return 0 - fi - p=$1 - - shift - for f; do - if [ -x $f ]; then - echo $f - return 0 - fi - done - - echo "" - return 1 -} - _optstr_add() { if [ -n "$1" ]; then @@ -XXX,XX +XXX,XX @@ _optstr_add() fi } -_fatal() -{ - echo "$*" - status=1 - exit 1 -} - -if [ -z "$QEMU_PROG" ]; then - export QEMU_PROG="`set_prog_path qemu`" -fi - -if [ -z "$QEMU_IMG_PROG" ]; then - export QEMU_IMG_PROG="`set_prog_path qemu-img`" -fi - -if [ -z "$QEMU_IO_PROG" ]; then - export QEMU_IO_PROG="`set_prog_path qemu-io`" -fi - -if [ -z "$QEMU_NBD_PROG" ]; then - export QEMU_NBD_PROG="`set_prog_path qemu-nbd`" -fi - -if [ -z "$QEMU_VXHS_PROG" ]; then - export QEMU_VXHS_PROG="`set_prog_path qnio_server`" -fi - -export QEMU_PROG=$(realpath -- "$(type -p "$QEMU_PROG")") -export QEMU_IMG_PROG=$(realpath -- "$(type -p "$QEMU_IMG_PROG")") -export QEMU_IO_PROG=$(realpath -- "$(type -p "$QEMU_IO_PROG")") -export QEMU_NBD_PROG=$(realpath -- "$(type -p "$QEMU_NBD_PROG")") - -# This program is not built as part of qemu but (possibly) provided by the -# system, so it may not be present at all -if [ -n "$QEMU_VXHS_PROG" ]; then - export QEMU_VXHS_PROG=$(realpath -- "$(type -p "$QEMU_VXHS_PROG")") -fi - _qemu_wrapper() { ( -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> These are never used by "check", with one exception that does not need $QEMU_OPTIONS. Keep them in common.rc, which will be soon included only by the tests. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/039.out | 10 +++--- tests/qemu-iotests/061.out | 4 +-- tests/qemu-iotests/137.out | 2 +- tests/qemu-iotests/common.config | 69 ++-------------------------------------- tests/qemu-iotests/common.rc | 65 +++++++++++++++++++++++++++++++++++++ 5 files changed, 75 insertions(+), 75 deletions(-) diff --git a/tests/qemu-iotests/039.out b/tests/qemu-iotests/039.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/039.out +++ b/tests/qemu-iotests/039.out @@ -XXX,XX +XXX,XX @@ No errors were found on the image. Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 wrote 512/512 bytes at offset 0 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -./common.config: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then +./common.rc: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; else exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; @@ -XXX,XX +XXX,XX @@ read 512/512 bytes at offset 0 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 wrote 512/512 bytes at offset 0 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -./common.config: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then +./common.rc: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; else exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; @@ -XXX,XX +XXX,XX @@ incompatible_features 0x0 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 wrote 512/512 bytes at offset 0 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -./common.config: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then +./common.rc: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; else exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; @@ -XXX,XX +XXX,XX @@ No errors were found on the image. Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 wrote 512/512 bytes at offset 0 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -./common.config: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then +./common.rc: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; else exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; @@ -XXX,XX +XXX,XX @@ Data may be corrupted, or further writes to the image may corrupt it. Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728 wrote 512/512 bytes at offset 0 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -./common.config: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then +./common.rc: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; else exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; diff --git a/tests/qemu-iotests/061.out b/tests/qemu-iotests/061.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/061.out +++ b/tests/qemu-iotests/061.out @@ -XXX,XX +XXX,XX @@ No errors were found on the image. Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 wrote 131072/131072 bytes at offset 0 128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -./common.config: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then +./common.rc: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; else exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; @@ -XXX,XX +XXX,XX @@ No errors were found on the image. Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 wrote 131072/131072 bytes at offset 0 128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -./common.config: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then +./common.rc: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; else exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; diff --git a/tests/qemu-iotests/137.out b/tests/qemu-iotests/137.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/137.out +++ b/tests/qemu-iotests/137.out @@ -XXX,XX +XXX,XX @@ Cache clean interval too big Unsupported value 'blubb' for qcow2 option 'overlap-check'. Allowed are any of the following: none, constant, cached, all wrote 512/512 bytes at offset 0 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -./common.config: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then +./common.rc: Killed ( if [ "${VALGRIND_QEMU}" == "y" ]; then exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; else exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"; diff --git a/tests/qemu-iotests/common.config b/tests/qemu-iotests/common.config index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.config +++ b/tests/qemu-iotests/common.config @@ -XXX,XX +XXX,XX @@ _optstr_add() fi } -_qemu_wrapper() -{ - ( - if [ -n "${QEMU_NEED_PID}" ]; then - echo $BASHPID > "${QEMU_TEST_DIR}/qemu-${_QEMU_HANDLE}.pid" - fi - exec "$QEMU_PROG" $QEMU_OPTIONS "$@" - ) -} - -_qemu_img_wrapper() -{ - (exec "$QEMU_IMG_PROG" $QEMU_IMG_OPTIONS "$@") -} - -_qemu_io_wrapper() -{ - local VALGRIND_LOGFILE="${TEST_DIR}"/$$.valgrind - local QEMU_IO_ARGS="$QEMU_IO_OPTIONS" - if [ "$IMGOPTSSYNTAX" = "true" ]; then - QEMU_IO_ARGS="--image-opts $QEMU_IO_ARGS" - if [ -n "$IMGKEYSECRET" ]; then - QEMU_IO_ARGS="--object secret,id=keysec0,data=$IMGKEYSECRET $QEMU_IO_ARGS" - fi - fi - local RETVAL - ( - if [ "${VALGRIND_QEMU}" == "y" ]; then - exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" - else - exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" - fi - ) - RETVAL=$? - if [ "${VALGRIND_QEMU}" == "y" ]; then - if [ $RETVAL == 99 ]; then - cat "${VALGRIND_LOGFILE}" - fi - rm -f "${VALGRIND_LOGFILE}" - fi - (exit $RETVAL) -} - -_qemu_nbd_wrapper() -{ - ( - echo $BASHPID > "${QEMU_TEST_DIR}/qemu-nbd.pid" - exec "$QEMU_NBD_PROG" $QEMU_NBD_OPTIONS "$@" - ) -} - -_qemu_vxhs_wrapper() -{ - ( - echo $BASHPID > "${TEST_DIR}/qemu-vxhs.pid" - exec "$QEMU_VXHS_PROG" $QEMU_VXHS_OPTIONS "$@" - ) -} - -export QEMU=_qemu_wrapper -export QEMU_IMG=_qemu_img_wrapper -export QEMU_IO=_qemu_io_wrapper -export QEMU_NBD=_qemu_nbd_wrapper -export QEMU_VXHS=_qemu_vxhs_wrapper - QEMU_IMG_EXTRA_ARGS= if [ "$IMGOPTSSYNTAX" = "true" ]; then QEMU_IMG_EXTRA_ARGS="--image-opts $QEMU_IMG_EXTRA_ARGS" @@ -XXX,XX +XXX,XX @@ fi export QEMU_IMG_EXTRA_ARGS -default_machine=$($QEMU -machine help | sed -n '/(default)/ s/ .*//p') -default_alias_machine=$($QEMU -machine help | \ +default_machine=$($QEMU_PROG -machine help | sed -n '/(default)/ s/ .*//p') +default_alias_machine=$($QEMU_PROG -machine help | \ sed -n "/(alias of $default_machine)/ { s/ .*//p; q; }") if [[ "$default_alias_machine" ]]; then default_machine="$default_alias_machine" diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.rc +++ b/tests/qemu-iotests/common.rc @@ -XXX,XX +XXX,XX @@ then fi fi +_qemu_wrapper() +{ + ( + if [ -n "${QEMU_NEED_PID}" ]; then + echo $BASHPID > "${QEMU_TEST_DIR}/qemu-${_QEMU_HANDLE}.pid" + fi + exec "$QEMU_PROG" $QEMU_OPTIONS "$@" + ) +} + +_qemu_img_wrapper() +{ + (exec "$QEMU_IMG_PROG" $QEMU_IMG_OPTIONS "$@") +} + +_qemu_io_wrapper() +{ + local VALGRIND_LOGFILE="${TEST_DIR}"/$$.valgrind + local QEMU_IO_ARGS="$QEMU_IO_OPTIONS" + if [ "$IMGOPTSSYNTAX" = "true" ]; then + QEMU_IO_ARGS="--image-opts $QEMU_IO_ARGS" + if [ -n "$IMGKEYSECRET" ]; then + QEMU_IO_ARGS="--object secret,id=keysec0,data=$IMGKEYSECRET $QEMU_IO_ARGS" + fi + fi + local RETVAL + ( + if [ "${VALGRIND_QEMU}" == "y" ]; then + exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" + else + exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@" + fi + ) + RETVAL=$? + if [ "${VALGRIND_QEMU}" == "y" ]; then + if [ $RETVAL == 99 ]; then + cat "${VALGRIND_LOGFILE}" + fi + rm -f "${VALGRIND_LOGFILE}" + fi + (exit $RETVAL) +} + +_qemu_nbd_wrapper() +{ + ( + echo $BASHPID > "${QEMU_TEST_DIR}/qemu-nbd.pid" + exec "$QEMU_NBD_PROG" $QEMU_NBD_OPTIONS "$@" + ) +} + +_qemu_vxhs_wrapper() +{ + ( + echo $BASHPID > "${TEST_DIR}/qemu-vxhs.pid" + exec "$QEMU_VXHS_PROG" $QEMU_VXHS_OPTIONS "$@" + ) +} + +export QEMU=_qemu_wrapper +export QEMU_IMG=_qemu_img_wrapper +export QEMU_IO=_qemu_io_wrapper +export QEMU_NBD=_qemu_nbd_wrapper +export QEMU_VXHS=_qemu_vxhs_wrapper + if [ "$IMGOPTSSYNTAX" = "true" ]; then DRIVER="driver=$IMGFMT" if [ "$IMGFMT" = "luks" ]; then -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> It only provides functions used by the test programs. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/check | 6 ------ tests/qemu-iotests/common.rc | 13 +++++-------- 2 files changed, 5 insertions(+), 14 deletions(-) diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -XXX,XX +XXX,XX @@ then _init_error "failed to source common.config" fi -# we need common.rc -if ! . "$source_iotests/common.rc" -then - _init_error "failed to source common.rc" -fi - # we need common . "$source_iotests/common" diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.rc +++ b/tests/qemu-iotests/common.rc @@ -XXX,XX +XXX,XX @@ poke_file() printf "$3" | dd "of=$1" bs=1 "seek=$2" conv=notrunc &>/dev/null } -# we need common.config -if [ "$iam" != "check" ] -then - if ! . ./common.config - then - echo "$iam: failed to source common.config" - exit 1 - fi + +if ! . ./common.config + then + echo "$iam: failed to source common.config" + exit 1 fi _qemu_wrapper() -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> Split "check" parts from tests part. For the directory setup, the actual computation of directories goes in "check", while the sanity checks go in the tests. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/common | 24 ++++++++++++++++++++ tests/qemu-iotests/common.config | 49 ---------------------------------------- tests/qemu-iotests/common.qemu | 1 + tests/qemu-iotests/common.rc | 25 ++++++++++++++++++++ 4 files changed, 50 insertions(+), 49 deletions(-) diff --git a/tests/qemu-iotests/common b/tests/qemu-iotests/common index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common +++ b/tests/qemu-iotests/common @@ -XXX,XX +XXX,XX @@ set_prog_path() fi } +if [ -z "$TEST_DIR" ]; then + TEST_DIR=`pwd`/scratch +fi + +if [ ! -e "$TEST_DIR" ]; then + mkdir "$TEST_DIR" +fi + diff="diff -u" verbose=false debug=false @@ -XXX,XX +XXX,XX @@ if [ "$IMGFMT" == "luks" ] && ! (echo "$IMGOPTS" | grep "iter-time=" > /dev/null IMGOPTS=$(_optstr_add "$IMGOPTS" "iter-time=10") fi +if [ -z "$SAMPLE_IMG_DIR" ]; then + SAMPLE_IMG_DIR="$source_iotests/sample_images" +fi + +export TEST_DIR +export SAMPLE_IMG_DIR + if [ -s $tmp.list ] then # found some valid test numbers ... this is good @@ -XXX,XX +XXX,XX @@ if [ -x "$build_iotests/socket_scm_helper" ] then export SOCKET_SCM_HELPER="$build_iotests/socket_scm_helper" fi + +default_machine=$($QEMU_PROG -machine help | sed -n '/(default)/ s/ .*//p') +default_alias_machine=$($QEMU_PROG -machine help | \ + sed -n "/(alias of $default_machine)/ { s/ .*//p; q; }") +if [[ "$default_alias_machine" ]]; then + default_machine="$default_alias_machine" +fi + +export QEMU_DEFAULT_MACHINE="$default_machine" diff --git a/tests/qemu-iotests/common.config b/tests/qemu-iotests/common.config index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.config +++ b/tests/qemu-iotests/common.config @@ -XXX,XX +XXX,XX @@ arch=`uname -m` export PWD=`pwd` -export _QEMU_HANDLE=0 - # make sure we have a standard umask umask 022 @@ -XXX,XX +XXX,XX @@ _optstr_add() fi } -QEMU_IMG_EXTRA_ARGS= -if [ "$IMGOPTSSYNTAX" = "true" ]; then - QEMU_IMG_EXTRA_ARGS="--image-opts $QEMU_IMG_EXTRA_ARGS" - if [ -n "$IMGKEYSECRET" ]; then - QEMU_IMG_EXTRA_ARGS="--object secret,id=keysec0,data=$IMGKEYSECRET $QEMU_IMG_EXTRA_ARGS" - fi -fi -export QEMU_IMG_EXTRA_ARGS - - -default_machine=$($QEMU_PROG -machine help | sed -n '/(default)/ s/ .*//p') -default_alias_machine=$($QEMU_PROG -machine help | \ - sed -n "/(alias of $default_machine)/ { s/ .*//p; q; }") -if [[ "$default_alias_machine" ]]; then - default_machine="$default_alias_machine" -fi - -export QEMU_DEFAULT_MACHINE="$default_machine" - -if [ -z "$TEST_DIR" ]; then - TEST_DIR=`pwd`/scratch -fi - -QEMU_TEST_DIR="${TEST_DIR}" - -if [ ! -e "$TEST_DIR" ]; then - mkdir "$TEST_DIR" -fi - -if [ ! -d "$TEST_DIR" ]; then - echo "common.config: Error: \$TEST_DIR ($TEST_DIR) is not a directory" - exit 1 -fi - -export TEST_DIR - -if [ -z "$SAMPLE_IMG_DIR" ]; then - SAMPLE_IMG_DIR="$source_iotests/sample_images" -fi - -if [ ! -d "$SAMPLE_IMG_DIR" ]; then - echo "common.config: Error: \$SAMPLE_IMG_DIR ($SAMPLE_IMG_DIR) is not a directory" - exit 1 -fi - -export SAMPLE_IMG_DIR - # make sure this script returns success true diff --git a/tests/qemu-iotests/common.qemu b/tests/qemu-iotests/common.qemu index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.qemu +++ b/tests/qemu-iotests/common.qemu @@ -XXX,XX +XXX,XX @@ QEMU_FIFO_IN="${QEMU_TEST_DIR}/qmp-in-$$" QEMU_FIFO_OUT="${QEMU_TEST_DIR}/qmp-out-$$" QEMU_HANDLE=0 +export _QEMU_HANDLE=0 # If bash version is >= 4.1, these will be overwritten and dynamic # file descriptor values assigned. diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.rc +++ b/tests/qemu-iotests/common.rc @@ -XXX,XX +XXX,XX @@ export QEMU_VXHS=_qemu_vxhs_wrapper if [ "$IMGOPTSSYNTAX" = "true" ]; then DRIVER="driver=$IMGFMT" + QEMU_IMG_EXTRA_ARGS="--image-opts $QEMU_IMG_EXTRA_ARGS" + if [ -n "$IMGKEYSECRET" ]; then + QEMU_IMG_EXTRA_ARGS="--object secret,id=keysec0,data=$IMGKEYSECRET $QEMU_IMG_EXTRA_ARGS" + fi if [ "$IMGFMT" = "luks" ]; then DRIVER="$DRIVER,key-secret=keysec0" fi @@ -XXX,XX +XXX,XX @@ if [ "$IMGOPTSSYNTAX" = "true" ]; then TEST_IMG="$DRIVER,file.driver=$IMGPROTO,file.filename=$TEST_DIR/t.$IMGFMT" fi else + QEMU_IMG_EXTRA_ARGS= if [ "$IMGPROTO" = "file" ]; then TEST_IMG=$TEST_DIR/t.$IMGFMT elif [ "$IMGPROTO" = "nbd" ]; then @@ -XXX,XX +XXX,XX @@ else fi ORIG_TEST_IMG="$TEST_IMG" +if [ -z "$TEST_DIR" ]; then + TEST_DIR=`pwd`/scratch +fi + +QEMU_TEST_DIR="${TEST_DIR}" + +if [ ! -e "$TEST_DIR" ]; then + mkdir "$TEST_DIR" +fi + +if [ ! -d "$TEST_DIR" ]; then + echo "common.config: Error: \$TEST_DIR ($TEST_DIR) is not a directory" + exit 1 +fi + +if [ ! -d "$SAMPLE_IMG_DIR" ]; then + echo "common.config: Error: \$SAMPLE_IMG_DIR ($SAMPLE_IMG_DIR) is not a directory" + exit 1 +fi + _use_sample_img() { SAMPLE_IMG_FILE="${1%\.bz2}" -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> The variable is used in "common" but defined only after the file is sourced. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/check | 2 -- tests/qemu-iotests/common | 2 ++ 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -XXX,XX +XXX,XX @@ fi TIMESTAMP_FILE=check.time-$IMGPROTO-$IMGFMT -tmp="${TEST_DIR}"/$$ - _wallclock() { date "+%H %M %S" | awk '{ print $1*3600 + $2*60 + $3 }' diff --git a/tests/qemu-iotests/common b/tests/qemu-iotests/common index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common +++ b/tests/qemu-iotests/common @@ -XXX,XX +XXX,XX @@ sortme=false expunge=true have_test_arg=false cachemode=false + +tmp="${TEST_DIR}"/$$ rm -f $tmp.list $tmp.tmp $tmp.sed export IMGFMT=raw -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> The variable is almost unused, and one of the two uses is actually uninitialized. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/check | 5 +---- tests/qemu-iotests/common.rc | 2 +- 2 files changed, 2 insertions(+), 5 deletions(-) diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -XXX,XX +XXX,XX @@ interrupt=true # by default don't output timestamps timestamp=${TIMESTAMP:=false} -# generic initialization -iam=check - _init_error() { - echo "$iam: $1" >&2 + echo "check: $1" >&2 exit 1 } diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.rc +++ b/tests/qemu-iotests/common.rc @@ -XXX,XX +XXX,XX @@ poke_file() if ! . ./common.config then - echo "$iam: failed to source common.config" + echo "$0: failed to source common.config" exit 1 fi -- 2.13.6 From: Paolo Bonzini <pbonzini@redhat.com> "check" is full of qemu-iotests--specific details. Separating it from "common" does not make much sense anymore. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/check | 533 +++++++++++++++++++++++++++++++++++++++++++- tests/qemu-iotests/common | 552 ---------------------------------------------- 2 files changed, 531 insertions(+), 554 deletions(-) delete mode 100644 tests/qemu-iotests/common diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -XXX,XX +XXX,XX @@ then _init_error "failed to source common.config" fi -# we need common -. "$source_iotests/common" +_full_imgfmt_details() +{ + if [ -n "$IMGOPTS" ]; then + echo "$IMGFMT ($IMGOPTS)" + else + echo "$IMGFMT" + fi +} + +_full_platform_details() +{ + os=`uname -s` + host=`hostname -s` + kernel=`uname -r` + platform=`uname -m` + echo "$os/$platform $host $kernel" +} + +# $1 = prog to look for +set_prog_path() +{ + p=`command -v $1 2> /dev/null` + if [ -n "$p" -a -x "$p" ]; then + realpath -- "$(type -p "$p")" + else + return 1 + fi +} + +if [ -z "$TEST_DIR" ]; then + TEST_DIR=`pwd`/scratch +fi + +if [ ! -e "$TEST_DIR" ]; then + mkdir "$TEST_DIR" +fi + +diff="diff -u" +verbose=false +debug=false +group=false +xgroup=false +imgopts=false +showme=false +sortme=false +expunge=true +have_test_arg=false +cachemode=false + +tmp="${TEST_DIR}"/$$ +rm -f $tmp.list $tmp.tmp $tmp.sed + +export IMGFMT=raw +export IMGFMT_GENERIC=true +export IMGPROTO=file +export IMGOPTS="" +export CACHEMODE="writeback" +export QEMU_IO_OPTIONS="" +export QEMU_IO_OPTIONS_NO_FMT="" +export CACHEMODE_IS_DEFAULT=true +export QEMU_OPTIONS="-nodefaults -machine accel=qtest" +export VALGRIND_QEMU= +export IMGKEYSECRET= +export IMGOPTSSYNTAX=false + +for r +do + + if $group + then + # arg after -g + group_list=`sed -n <"$source_iotests/group" -e 's/$/ /' -e "/^[0-9][0-9][0-9].* $r /"'{ +s/ .*//p +}'` + if [ -z "$group_list" ] + then + echo "Group \"$r\" is empty or not defined?" + exit 1 + fi + [ ! -s $tmp.list ] && touch $tmp.list + for t in $group_list + do + if grep -s "^$t\$" $tmp.list >/dev/null + then + : + else + echo "$t" >>$tmp.list + fi + done + group=false + continue + + elif $xgroup + then + # arg after -x + # Populate $tmp.list with all tests + awk '/^[0-9]{3,}/ {print $1}' "${source_iotests}/group" > $tmp.list 2>/dev/null + group_list=`sed -n <"$source_iotests/group" -e 's/$/ /' -e "/^[0-9][0-9][0-9].* $r /"'{ +s/ .*//p +}'` + if [ -z "$group_list" ] + then + echo "Group \"$r\" is empty or not defined?" + exit 1 + fi + numsed=0 + rm -f $tmp.sed + for t in $group_list + do + if [ $numsed -gt 100 ] + then + sed -f $tmp.sed <$tmp.list >$tmp.tmp + mv $tmp.tmp $tmp.list + numsed=0 + rm -f $tmp.sed + fi + echo "/^$t\$/d" >>$tmp.sed + numsed=`expr $numsed + 1` + done + sed -f $tmp.sed <$tmp.list >$tmp.tmp + mv $tmp.tmp $tmp.list + xgroup=false + continue + + elif $imgopts + then + IMGOPTS="$r" + imgopts=false + continue + elif $cachemode + then + CACHEMODE="$r" + CACHEMODE_IS_DEFAULT=false + cachemode=false + continue + fi + + xpand=true + case "$r" + in + + -\? | -h | --help) # usage + echo "Usage: $0 [options] [testlist]"' + +common options + -v verbose + -d debug + +image format options + -raw test raw (default) + -bochs test bochs + -cloop test cloop + -parallels test parallels + -qcow test qcow + -qcow2 test qcow2 + -qed test qed + -vdi test vdi + -vpc test vpc + -vhdx test vhdx + -vmdk test vmdk + -luks test luks + +image protocol options + -file test file (default) + -rbd test rbd + -sheepdog test sheepdog + -nbd test nbd + -ssh test ssh + -nfs test nfs + -vxhs test vxhs + +other options + -xdiff graphical mode diff + -nocache use O_DIRECT on backing file + -misalign misalign memory allocations + -n show me, do not run tests + -o options -o options to pass to qemu-img create/convert + -T output timestamps + -c mode cache mode + +testlist options + -g group[,group...] include tests from these groups + -x group[,group...] exclude tests from these groups + NNN include test NNN + NNN-NNN include test range (eg. 012-021) +' + exit 0 + ;; + + -raw) + IMGFMT=raw + xpand=false + ;; + + -bochs) + IMGFMT=bochs + IMGFMT_GENERIC=false + xpand=false + ;; + + -cloop) + IMGFMT=cloop + IMGFMT_GENERIC=false + xpand=false + ;; + + -parallels) + IMGFMT=parallels + IMGFMT_GENERIC=false + xpand=false + ;; + + -qcow) + IMGFMT=qcow + xpand=false + ;; + + -qcow2) + IMGFMT=qcow2 + xpand=false + ;; + + -luks) + IMGOPTSSYNTAX=true + IMGFMT=luks + IMGKEYSECRET=123456 + xpand=false + ;; + + -qed) + IMGFMT=qed + xpand=false + ;; + + -vdi) + IMGFMT=vdi + xpand=false + ;; + + -vmdk) + IMGFMT=vmdk + xpand=false + ;; + + -vpc) + IMGFMT=vpc + xpand=false + ;; + + -vhdx) + IMGFMT=vhdx + xpand=false + ;; + + -file) + IMGPROTO=file + xpand=false + ;; + + -rbd) + IMGPROTO=rbd + xpand=false + ;; + + -sheepdog) + IMGPROTO=sheepdog + xpand=false + ;; + + -nbd) + IMGPROTO=nbd + xpand=false + ;; + + -vxhs) + IMGPROTO=vxhs + xpand=false + ;; + + -ssh) + IMGPROTO=ssh + xpand=false + ;; + + -nfs) + IMGPROTO=nfs + xpand=false + ;; + + -nocache) + CACHEMODE="none" + CACHEMODE_IS_DEFAULT=false + xpand=false + ;; + + -misalign) + QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS --misalign" + xpand=false + ;; + + -valgrind) + VALGRIND_QEMU='y' + xpand=false + ;; + + -g) # -g group ... pick from group file + group=true + xpand=false + ;; + + -xdiff) # graphical diff mode + xpand=false + + if [ ! -z "$DISPLAY" ] + then + command -v xdiff >/dev/null 2>&1 && diff=xdiff + command -v gdiff >/dev/null 2>&1 && diff=gdiff + command -v tkdiff >/dev/null 2>&1 && diff=tkdiff + command -v xxdiff >/dev/null 2>&1 && diff=xxdiff + fi + ;; + + -n) # show me, don't do it + showme=true + xpand=false + ;; + -o) + imgopts=true + xpand=false + ;; + -c) + cachemode=true + xpand=false + ;; + -T) # turn on timestamp output + timestamp=true + xpand=false + ;; + + -v) + verbose=true + xpand=false + ;; + -d) + debug=true + xpand=false + ;; + -x) # -x group ... exclude from group file + xgroup=true + xpand=false + ;; + '[0-9][0-9][0-9] [0-9][0-9][0-9][0-9]') + echo "No tests?" + status=1 + exit $status + ;; + + [0-9]*-[0-9]*) + eval `echo $r | sed -e 's/^/start=/' -e 's/-/ end=/'` + ;; + + [0-9]*-) + eval `echo $r | sed -e 's/^/start=/' -e 's/-//'` + end=`echo [0-9][0-9][0-9] [0-9][0-9][0-9][0-9] | sed -e 's/\[0-9]//g' -e 's/ *$//' -e 's/.* //'` + if [ -z "$end" ] + then + echo "No tests in range \"$r\"?" + status=1 + exit $status + fi + ;; + + *) + start=$r + end=$r + ;; + + esac + + # get rid of leading 0s as can be interpreted as octal + start=`echo $start | sed 's/^0*//'` + end=`echo $end | sed 's/^0*//'` + + if $xpand + then + have_test_arg=true + awk </dev/null ' +BEGIN { for (t='$start'; t<='$end'; t++) printf "%03d\n",t }' \ + | while read id + do + if grep -s "^$id " "$source_iotests/group" >/dev/null + then + # in group file ... OK + echo $id >>$tmp.list + else + if [ -f expunged ] && $expunge && egrep "^$id([ ]|\$)" expunged >/dev/null + then + # expunged ... will be reported, but not run, later + echo $id >>$tmp.list + else + # oops + if [ "$start" == "$end" -a "$id" == "$end" ] + then + echo "$id - unknown test" + exit 1 + else + echo "$id - unknown test, ignored" + fi + fi + fi + done || exit 1 + fi + +done + +# Set qemu-io cache mode with $CACHEMODE we have +QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS --cache $CACHEMODE" + +QEMU_IO_OPTIONS_NO_FMT="$QEMU_IO_OPTIONS" +if [ "$IMGOPTSSYNTAX" != "true" ]; then + QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS -f $IMGFMT" +fi + +# Set default options for qemu-img create -o if they were not specified +if [ "$IMGFMT" == "qcow2" ] && ! (echo "$IMGOPTS" | grep "compat=" > /dev/null); then + IMGOPTS=$(_optstr_add "$IMGOPTS" "compat=1.1") +fi +if [ "$IMGFMT" == "luks" ] && ! (echo "$IMGOPTS" | grep "iter-time=" > /dev/null); then + IMGOPTS=$(_optstr_add "$IMGOPTS" "iter-time=10") +fi + +if [ -z "$SAMPLE_IMG_DIR" ]; then + SAMPLE_IMG_DIR="$source_iotests/sample_images" +fi + +export TEST_DIR +export SAMPLE_IMG_DIR + +if [ -s $tmp.list ] +then + # found some valid test numbers ... this is good + : +else + if $have_test_arg + then + # had test numbers, but none in group file ... do nothing + touch $tmp.list + else + # no test numbers, do everything from group file + sed -n -e '/^[0-9][0-9][0-9]*/s/[ ].*//p' <"$source_iotests/group" >$tmp.list + fi +fi + +# should be sort -n, but this did not work for Linux when this +# was ported from IRIX +# +list=`sort $tmp.list` +rm -f $tmp.list $tmp.tmp $tmp.sed + +if [ -z "$QEMU_PROG" ] +then + if [ -x "$build_iotests/qemu" ]; then + export QEMU_PROG="$build_iotests/qemu" + elif [ -x "$build_root/$arch-softmmu/qemu-system-$arch" ]; then + export QEMU_PROG="$build_root/$arch-softmmu/qemu-system-$arch" + else + pushd "$build_root" > /dev/null + for binary in *-softmmu/qemu-system-* + do + if [ -x "$binary" ] + then + export QEMU_PROG="$build_root/$binary" + break + fi + done + popd > /dev/null + [ "$QEMU_PROG" = "" ] && _init_error "qemu not found" + fi +fi +export QEMU_PROG=$(realpath -- "$(type -p "$QEMU_PROG")") + +if [ -z "$QEMU_IMG_PROG" ]; then + if [ -x "$build_iotests/qemu-img" ]; then + export QEMU_IMG_PROG="$build_iotests/qemu-img" + elif [ -x "$build_root/qemu-img" ]; then + export QEMU_IMG_PROG="$build_root/qemu-img" + else + _init_error "qemu-img not found" + fi +fi +export QEMU_IMG_PROG=$(realpath -- "$(type -p "$QEMU_IMG_PROG")") + +if [ -z "$QEMU_IO_PROG" ]; then + if [ -x "$build_iotests/qemu-io" ]; then + export QEMU_IO_PROG="$build_iotests/qemu-io" + elif [ -x "$build_root/qemu-io" ]; then + export QEMU_IO_PROG="$build_root/qemu-io" + else + _init_error "qemu-io not found" + fi +fi +export QEMU_IO_PROG=$(realpath -- "$(type -p "$QEMU_IO_PROG")") + +if [ -z $QEMU_NBD_PROG ]; then + if [ -x "$build_iotests/qemu-nbd" ]; then + export QEMU_NBD_PROG="$build_iotests/qemu-nbd" + elif [ -x "$build_root/qemu-nbd" ]; then + export QEMU_NBD_PROG="$build_root/qemu-nbd" + else + _init_error "qemu-nbd not found" + fi +fi +export QEMU_NBD_PROG=$(realpath -- "$(type -p "$QEMU_NBD_PROG")") + +if [ -z "$QEMU_VXHS_PROG" ]; then + export QEMU_VXHS_PROG="`set_prog_path qnio_server`" +fi + +if [ -x "$build_iotests/socket_scm_helper" ] +then + export SOCKET_SCM_HELPER="$build_iotests/socket_scm_helper" +fi + +default_machine=$($QEMU_PROG -machine help | sed -n '/(default)/ s/ .*//p') +default_alias_machine=$($QEMU_PROG -machine help | \ + sed -n "/(alias of $default_machine)/ { s/ .*//p; q; }") +if [[ "$default_alias_machine" ]]; then + default_machine="$default_alias_machine" +fi + +export QEMU_DEFAULT_MACHINE="$default_machine" TIMESTAMP_FILE=check.time-$IMGPROTO-$IMGFMT diff --git a/tests/qemu-iotests/common b/tests/qemu-iotests/common deleted file mode 100644 index XXXXXXX..XXXXXXX --- a/tests/qemu-iotests/common +++ /dev/null @@ -XXX,XX +XXX,XX @@ -#!/bin/bash -# -# Copyright (C) 2009 Red Hat, Inc. -# Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. -# -# This program is free software; you can redistribute it and/or -# modify it under the terms of the GNU General Public License as -# published by the Free Software Foundation. -# -# This program is distributed in the hope that it would be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program. If not, see <http://www.gnu.org/licenses/>. -# -# -# common procedures for QA scripts -# - -_full_imgfmt_details() -{ - if [ -n "$IMGOPTS" ]; then - echo "$IMGFMT ($IMGOPTS)" - else - echo "$IMGFMT" - fi -} - -_full_platform_details() -{ - os=`uname -s` - host=`hostname -s` - kernel=`uname -r` - platform=`uname -m` - echo "$os/$platform $host $kernel" -} - -# $1 = prog to look for -set_prog_path() -{ - p=`command -v $1 2> /dev/null` - if [ -n "$p" -a -x "$p" ]; then - realpath -- "$(type -p "$p")" - else - return 1 - fi -} - -if [ -z "$TEST_DIR" ]; then - TEST_DIR=`pwd`/scratch -fi - -if [ ! -e "$TEST_DIR" ]; then - mkdir "$TEST_DIR" -fi - -diff="diff -u" -verbose=false -debug=false -group=false -xgroup=false -imgopts=false -showme=false -sortme=false -expunge=true -have_test_arg=false -cachemode=false - -tmp="${TEST_DIR}"/$$ -rm -f $tmp.list $tmp.tmp $tmp.sed - -export IMGFMT=raw -export IMGFMT_GENERIC=true -export IMGPROTO=file -export IMGOPTS="" -export CACHEMODE="writeback" -export QEMU_IO_OPTIONS="" -export QEMU_IO_OPTIONS_NO_FMT="" -export CACHEMODE_IS_DEFAULT=true -export QEMU_OPTIONS="-nodefaults -machine accel=qtest" -export VALGRIND_QEMU= -export IMGKEYSECRET= -export IMGOPTSSYNTAX=false - -for r -do - - if $group - then - # arg after -g - group_list=`sed -n <"$source_iotests/group" -e 's/$/ /' -e "/^[0-9][0-9][0-9].* $r /"'{ -s/ .*//p -}'` - if [ -z "$group_list" ] - then - echo "Group \"$r\" is empty or not defined?" - exit 1 - fi - [ ! -s $tmp.list ] && touch $tmp.list - for t in $group_list - do - if grep -s "^$t\$" $tmp.list >/dev/null - then - : - else - echo "$t" >>$tmp.list - fi - done - group=false - continue - - elif $xgroup - then - # arg after -x - # Populate $tmp.list with all tests - awk '/^[0-9]{3,}/ {print $1}' "${source_iotests}/group" > $tmp.list 2>/dev/null - group_list=`sed -n <"$source_iotests/group" -e 's/$/ /' -e "/^[0-9][0-9][0-9].* $r /"'{ -s/ .*//p -}'` - if [ -z "$group_list" ] - then - echo "Group \"$r\" is empty or not defined?" - exit 1 - fi - numsed=0 - rm -f $tmp.sed - for t in $group_list - do - if [ $numsed -gt 100 ] - then - sed -f $tmp.sed <$tmp.list >$tmp.tmp - mv $tmp.tmp $tmp.list - numsed=0 - rm -f $tmp.sed - fi - echo "/^$t\$/d" >>$tmp.sed - numsed=`expr $numsed + 1` - done - sed -f $tmp.sed <$tmp.list >$tmp.tmp - mv $tmp.tmp $tmp.list - xgroup=false - continue - - elif $imgopts - then - IMGOPTS="$r" - imgopts=false - continue - elif $cachemode - then - CACHEMODE="$r" - CACHEMODE_IS_DEFAULT=false - cachemode=false - continue - fi - - xpand=true - case "$r" - in - - -\? | -h | --help) # usage - echo "Usage: $0 [options] [testlist]"' - -common options - -v verbose - -d debug - -image format options - -raw test raw (default) - -bochs test bochs - -cloop test cloop - -parallels test parallels - -qcow test qcow - -qcow2 test qcow2 - -qed test qed - -vdi test vdi - -vpc test vpc - -vhdx test vhdx - -vmdk test vmdk - -luks test luks - -image protocol options - -file test file (default) - -rbd test rbd - -sheepdog test sheepdog - -nbd test nbd - -ssh test ssh - -nfs test nfs - -vxhs test vxhs - -other options - -xdiff graphical mode diff - -nocache use O_DIRECT on backing file - -misalign misalign memory allocations - -n show me, do not run tests - -o options -o options to pass to qemu-img create/convert - -T output timestamps - -c mode cache mode - -testlist options - -g group[,group...] include tests from these groups - -x group[,group...] exclude tests from these groups - NNN include test NNN - NNN-NNN include test range (eg. 012-021) -' - exit 0 - ;; - - -raw) - IMGFMT=raw - xpand=false - ;; - - -bochs) - IMGFMT=bochs - IMGFMT_GENERIC=false - xpand=false - ;; - - -cloop) - IMGFMT=cloop - IMGFMT_GENERIC=false - xpand=false - ;; - - -parallels) - IMGFMT=parallels - IMGFMT_GENERIC=false - xpand=false - ;; - - -qcow) - IMGFMT=qcow - xpand=false - ;; - - -qcow2) - IMGFMT=qcow2 - xpand=false - ;; - - -luks) - IMGOPTSSYNTAX=true - IMGFMT=luks - IMGKEYSECRET=123456 - xpand=false - ;; - - -qed) - IMGFMT=qed - xpand=false - ;; - - -vdi) - IMGFMT=vdi - xpand=false - ;; - - -vmdk) - IMGFMT=vmdk - xpand=false - ;; - - -vpc) - IMGFMT=vpc - xpand=false - ;; - - -vhdx) - IMGFMT=vhdx - xpand=false - ;; - - -file) - IMGPROTO=file - xpand=false - ;; - - -rbd) - IMGPROTO=rbd - xpand=false - ;; - - -sheepdog) - IMGPROTO=sheepdog - xpand=false - ;; - - -nbd) - IMGPROTO=nbd - xpand=false - ;; - - -vxhs) - IMGPROTO=vxhs - xpand=false - ;; - - -ssh) - IMGPROTO=ssh - xpand=false - ;; - - -nfs) - IMGPROTO=nfs - xpand=false - ;; - - -nocache) - CACHEMODE="none" - CACHEMODE_IS_DEFAULT=false - xpand=false - ;; - - -misalign) - QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS --misalign" - xpand=false - ;; - - -valgrind) - VALGRIND_QEMU='y' - xpand=false - ;; - - -g) # -g group ... pick from group file - group=true - xpand=false - ;; - - -xdiff) # graphical diff mode - xpand=false - - if [ ! -z "$DISPLAY" ] - then - command -v xdiff >/dev/null 2>&1 && diff=xdiff - command -v gdiff >/dev/null 2>&1 && diff=gdiff - command -v tkdiff >/dev/null 2>&1 && diff=tkdiff - command -v xxdiff >/dev/null 2>&1 && diff=xxdiff - fi - ;; - - -n) # show me, don't do it - showme=true - xpand=false - ;; - -o) - imgopts=true - xpand=false - ;; - -c) - cachemode=true - xpand=false - ;; - -T) # turn on timestamp output - timestamp=true - xpand=false - ;; - - -v) - verbose=true - xpand=false - ;; - -d) - debug=true - xpand=false - ;; - -x) # -x group ... exclude from group file - xgroup=true - xpand=false - ;; - '[0-9][0-9][0-9] [0-9][0-9][0-9][0-9]') - echo "No tests?" - status=1 - exit $status - ;; - - [0-9]*-[0-9]*) - eval `echo $r | sed -e 's/^/start=/' -e 's/-/ end=/'` - ;; - - [0-9]*-) - eval `echo $r | sed -e 's/^/start=/' -e 's/-//'` - end=`echo [0-9][0-9][0-9] [0-9][0-9][0-9][0-9] | sed -e 's/\[0-9]//g' -e 's/ *$//' -e 's/.* //'` - if [ -z "$end" ] - then - echo "No tests in range \"$r\"?" - status=1 - exit $status - fi - ;; - - *) - start=$r - end=$r - ;; - - esac - - # get rid of leading 0s as can be interpreted as octal - start=`echo $start | sed 's/^0*//'` - end=`echo $end | sed 's/^0*//'` - - if $xpand - then - have_test_arg=true - awk </dev/null ' -BEGIN { for (t='$start'; t<='$end'; t++) printf "%03d\n",t }' \ - | while read id - do - if grep -s "^$id " "$source_iotests/group" >/dev/null - then - # in group file ... OK - echo $id >>$tmp.list - else - if [ -f expunged ] && $expunge && egrep "^$id([ ]|\$)" expunged >/dev/null - then - # expunged ... will be reported, but not run, later - echo $id >>$tmp.list - else - # oops - if [ "$start" == "$end" -a "$id" == "$end" ] - then - echo "$id - unknown test" - exit 1 - else - echo "$id - unknown test, ignored" - fi - fi - fi - done || exit 1 - fi - -done - -# Set qemu-io cache mode with $CACHEMODE we have -QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS --cache $CACHEMODE" - -QEMU_IO_OPTIONS_NO_FMT="$QEMU_IO_OPTIONS" -if [ "$IMGOPTSSYNTAX" != "true" ]; then - QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS -f $IMGFMT" -fi - -# Set default options for qemu-img create -o if they were not specified -if [ "$IMGFMT" == "qcow2" ] && ! (echo "$IMGOPTS" | grep "compat=" > /dev/null); then - IMGOPTS=$(_optstr_add "$IMGOPTS" "compat=1.1") -fi -if [ "$IMGFMT" == "luks" ] && ! (echo "$IMGOPTS" | grep "iter-time=" > /dev/null); then - IMGOPTS=$(_optstr_add "$IMGOPTS" "iter-time=10") -fi - -if [ -z "$SAMPLE_IMG_DIR" ]; then - SAMPLE_IMG_DIR="$source_iotests/sample_images" -fi - -export TEST_DIR -export SAMPLE_IMG_DIR - -if [ -s $tmp.list ] -then - # found some valid test numbers ... this is good - : -else - if $have_test_arg - then - # had test numbers, but none in group file ... do nothing - touch $tmp.list - else - # no test numbers, do everything from group file - sed -n -e '/^[0-9][0-9][0-9]*/s/[ ].*//p' <"$source_iotests/group" >$tmp.list - fi -fi - -# should be sort -n, but this did not work for Linux when this -# was ported from IRIX -# -list=`sort $tmp.list` -rm -f $tmp.list $tmp.tmp $tmp.sed - -if [ -z "$QEMU_PROG" ] -then - if [ -x "$build_iotests/qemu" ]; then - export QEMU_PROG="$build_iotests/qemu" - elif [ -x "$build_root/$arch-softmmu/qemu-system-$arch" ]; then - export QEMU_PROG="$build_root/$arch-softmmu/qemu-system-$arch" - else - pushd "$build_root" > /dev/null - for binary in *-softmmu/qemu-system-* - do - if [ -x "$binary" ] - then - export QEMU_PROG="$build_root/$binary" - break - fi - done - popd > /dev/null - [ "$QEMU_PROG" = "" ] && _init_error "qemu not found" - fi -fi -export QEMU_PROG=$(realpath -- "$(type -p "$QEMU_PROG")") - -if [ -z "$QEMU_IMG_PROG" ]; then - if [ -x "$build_iotests/qemu-img" ]; then - export QEMU_IMG_PROG="$build_iotests/qemu-img" - elif [ -x "$build_root/qemu-img" ]; then - export QEMU_IMG_PROG="$build_root/qemu-img" - else - _init_error "qemu-img not found" - fi -fi -export QEMU_IMG_PROG=$(realpath -- "$(type -p "$QEMU_IMG_PROG")") - -if [ -z "$QEMU_IO_PROG" ]; then - if [ -x "$build_iotests/qemu-io" ]; then - export QEMU_IO_PROG="$build_iotests/qemu-io" - elif [ -x "$build_root/qemu-io" ]; then - export QEMU_IO_PROG="$build_root/qemu-io" - else - _init_error "qemu-io not found" - fi -fi -export QEMU_IO_PROG=$(realpath -- "$(type -p "$QEMU_IO_PROG")") - -if [ -z $QEMU_NBD_PROG ]; then - if [ -x "$build_iotests/qemu-nbd" ]; then - export QEMU_NBD_PROG="$build_iotests/qemu-nbd" - elif [ -x "$build_root/qemu-nbd" ]; then - export QEMU_NBD_PROG="$build_root/qemu-nbd" - else - _init_error "qemu-nbd not found" - fi -fi -export QEMU_NBD_PROG=$(realpath -- "$(type -p "$QEMU_NBD_PROG")") - -if [ -z "$QEMU_VXHS_PROG" ]; then - export QEMU_VXHS_PROG="`set_prog_path qnio_server`" -fi - -if [ -x "$build_iotests/socket_scm_helper" ] -then - export SOCKET_SCM_HELPER="$build_iotests/socket_scm_helper" -fi - -default_machine=$($QEMU_PROG -machine help | sed -n '/(default)/ s/ .*//p') -default_alias_machine=$($QEMU_PROG -machine help | \ - sed -n "/(alias of $default_machine)/ { s/ .*//p; q; }") -if [[ "$default_alias_machine" ]]; then - default_machine="$default_alias_machine" -fi - -export QEMU_DEFAULT_MACHINE="$default_machine" -- 2.13.6 There is no good reason for bdrv_drop_intermediate() to know the active layer above the subchain it is operating on - even more so, because the assumption that there is a single active layer above it is not generally true. In order to prepare removal of the active parameter, use a BdrvChildRole callback to update the backing file string in the overlay image instead of directly calling bdrv_change_backing_file(). Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> --- include/block/block_int.h | 6 ++++++ block.c | 33 ++++++++++++++++++++++++++++----- 2 files changed, 34 insertions(+), 5 deletions(-) diff --git a/include/block/block_int.h b/include/block/block_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block_int.h +++ b/include/block/block_int.h @@ -XXX,XX +XXX,XX @@ struct BdrvChildRole { void (*attach)(BdrvChild *child); void (*detach)(BdrvChild *child); + + /* Notifies the parent that the filename of its child has changed (e.g. + * because the direct child was removed from the backing chain), so that it + * can update its reference. */ + int (*update_filename)(BdrvChild *child, BlockDriverState *new_base, + const char *filename, Error **errp); }; extern const BdrvChildRole child_file; diff --git a/block.c b/block.c index XXXXXXX..XXXXXXX 100644 --- a/block.c +++ b/block.c @@ -XXX,XX +XXX,XX @@ static void bdrv_backing_options(int *child_flags, QDict *child_options, *child_flags = flags; } +static int bdrv_backing_update_filename(BdrvChild *c, BlockDriverState *base, + const char *filename, Error **errp) +{ + BlockDriverState *parent = c->opaque; + int ret; + + ret = bdrv_change_backing_file(parent, filename, + base->drv ? base->drv->format_name : ""); + if (ret < 0) { + error_setg_errno(errp, ret, "Could not update backing file link"); + } + + return ret; +} + const BdrvChildRole child_backing = { .get_parent_desc = bdrv_child_get_parent_desc, .attach = bdrv_backing_attach, @@ -XXX,XX +XXX,XX @@ const BdrvChildRole child_backing = { .drained_begin = bdrv_child_cb_drained_begin, .drained_end = bdrv_child_cb_drained_end, .inactivate = bdrv_child_cb_inactivate, + .update_filename = bdrv_backing_update_filename, }; static int bdrv_open_flags(BlockDriverState *bs, int flags) @@ -XXX,XX +XXX,XX @@ int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top, Error *local_err = NULL; int ret = -EIO; + bdrv_ref(top); + if (!top->drv || !base->drv) { goto exit; } @@ -XXX,XX +XXX,XX @@ int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top, } /* success - we can delete the intermediate states, and link top->base */ - backing_file_str = backing_file_str ? backing_file_str : base->filename; - ret = bdrv_change_backing_file(new_top_bs, backing_file_str, - base->drv ? base->drv->format_name : ""); - if (ret) { - goto exit; + if (new_top_bs->backing->role->update_filename) { + backing_file_str = backing_file_str ? backing_file_str : base->filename; + ret = new_top_bs->backing->role->update_filename(new_top_bs->backing, + base, backing_file_str, + &local_err); + if (ret < 0) { + bdrv_set_backing_hd(new_top_bs, top, &error_abort); + goto exit; + } } bdrv_set_backing_hd(new_top_bs, base, &local_err); @@ -XXX,XX +XXX,XX @@ int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top, ret = 0; exit: + bdrv_unref(top); return ret; } -- 2.13.6 This changes the commit block job to support operation in a graph where there is more than a single active layer that references the top node. This involves inserting the commit filter node not only on the path between the given active node and the top node, but between the top node and all of its parents. On completion, bdrv_drop_intermediate() must consider all parents for updating the backing file link. These parents may be backing files themselves and as such read-only; reopen them temporarily if necessary. Previously this was achieved by the bdrv_reopen() calls in the commit block job that made overlay_bs read-write for the whole duration of the block job, even though write access is only needed on completion. Now that we consider all parents, overlay_bs is meaningless. It is left in place in this commit, but we'll remove it soon. Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block.c | 68 ++++++++++++++++++++++++++++++++++------------------------ block/commit.c | 2 +- 2 files changed, 41 insertions(+), 29 deletions(-) diff --git a/block.c b/block.c index XXXXXXX..XXXXXXX 100644 --- a/block.c +++ b/block.c @@ -XXX,XX +XXX,XX @@ static int bdrv_backing_update_filename(BdrvChild *c, BlockDriverState *base, const char *filename, Error **errp) { BlockDriverState *parent = c->opaque; + int orig_flags = bdrv_get_flags(parent); int ret; + if (!(orig_flags & BDRV_O_RDWR)) { + ret = bdrv_reopen(parent, orig_flags | BDRV_O_RDWR, errp); + if (ret < 0) { + return ret; + } + } + ret = bdrv_change_backing_file(parent, filename, base->drv ? base->drv->format_name : ""); if (ret < 0) { error_setg_errno(errp, ret, "Could not update backing file link"); } + if (!(orig_flags & BDRV_O_RDWR)) { + bdrv_reopen(parent, orig_flags, NULL); + } + return ret; } @@ -XXX,XX +XXX,XX @@ BlockDriverState *bdrv_find_base(BlockDriverState *bs) int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top, BlockDriverState *base, const char *backing_file_str) { - BlockDriverState *new_top_bs = NULL; + BdrvChild *c, *next; Error *local_err = NULL; int ret = -EIO; @@ -XXX,XX +XXX,XX @@ int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top, goto exit; } - new_top_bs = bdrv_find_overlay(active, top); - - if (new_top_bs == NULL) { - /* we could not find the image above 'top', this is an error */ - goto exit; - } - - /* special case of new_top_bs->backing->bs already pointing to base - nothing - * to do, no intermediate images */ - if (backing_bs(new_top_bs) == base) { - ret = 0; - goto exit; - } - /* Make sure that base is in the backing chain of top */ if (!bdrv_chain_contains(top, base)) { goto exit; } /* success - we can delete the intermediate states, and link top->base */ - if (new_top_bs->backing->role->update_filename) { - backing_file_str = backing_file_str ? backing_file_str : base->filename; - ret = new_top_bs->backing->role->update_filename(new_top_bs->backing, - base, backing_file_str, - &local_err); - if (ret < 0) { - bdrv_set_backing_hd(new_top_bs, top, &error_abort); + backing_file_str = backing_file_str ? backing_file_str : base->filename; + + QLIST_FOREACH_SAFE(c, &top->parents, next_parent, next) { + /* Check whether we are allowed to switch c from top to base */ + GSList *ignore_children = g_slist_prepend(NULL, c); + bdrv_check_update_perm(base, NULL, c->perm, c->shared_perm, + ignore_children, &local_err); + if (local_err) { + ret = -EPERM; + error_report_err(local_err); goto exit; } - } + g_slist_free(ignore_children); - bdrv_set_backing_hd(new_top_bs, base, &local_err); - if (local_err) { - ret = -EPERM; - error_report_err(local_err); - goto exit; + /* If so, update the backing file path in the image file */ + if (c->role->update_filename) { + ret = c->role->update_filename(c, base, backing_file_str, + &local_err); + if (ret < 0) { + bdrv_abort_perm_update(base); + error_report_err(local_err); + goto exit; + } + } + + /* Do the actual switch in the in-memory graph. + * Completes bdrv_check_update_perm() transaction internally. */ + bdrv_ref(base); + bdrv_replace_child(c, base); + bdrv_unref(top); } ret = 0; diff --git a/block/commit.c b/block/commit.c index XXXXXXX..XXXXXXX 100644 --- a/block/commit.c +++ b/block/commit.c @@ -XXX,XX +XXX,XX @@ void commit_start(const char *job_id, BlockDriverState *bs, error_propagate(errp, local_err); goto fail; } - bdrv_set_backing_hd(overlay_bs, commit_top_bs, &local_err); + bdrv_replace_node(top, commit_top_bs, &local_err); if (local_err) { bdrv_unref(commit_top_bs); commit_top_bs = NULL; -- 2.13.6 QMP responses to certain commands can become quite long, which doesn't only make reading them hard, but also means that the maximum line length in patch emails can be exceeded. Allow tests to switch to QMP pretty printing, which results in more, but shorter lines. We also need to make sure to keep indentation in the response for this to work as expected. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> --- tests/qemu-iotests/common.qemu | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/tests/qemu-iotests/common.qemu b/tests/qemu-iotests/common.qemu index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.qemu +++ b/tests/qemu-iotests/common.qemu @@ -XXX,XX +XXX,XX @@ function _timed_wait_for() shift QEMU_STATUS[$h]=0 - while read -t ${QEMU_COMM_TIMEOUT} resp <&${QEMU_OUT[$h]} + while IFS= read -t ${QEMU_COMM_TIMEOUT} resp <&${QEMU_OUT[$h]} do if [ -z "${silent}" ]; then echo "${resp}" | _filter_testdir | _filter_qemu \ | _filter_qemu_io | _filter_qmp | _filter_hmp fi - grep -q "${*}" < <(echo ${resp}) + grep -q "${*}" < <(echo "${resp}") if [ $? -eq 0 ]; then return fi @@ -XXX,XX +XXX,XX @@ function _send_qemu_cmd() # $qemu_comm_method: set this variable to 'monitor' (case insensitive) # to use the QEMU HMP monitor for communication. # Otherwise, the default of QMP is used. +# $qmp_pretty: Set this variable to 'y' to enable QMP pretty printing. # $keep_stderr: Set this variable to 'y' to keep QEMU's stderr output on stderr. # If this variable is empty, stderr will be redirected to stdout. # Returns: @@ -XXX,XX +XXX,XX @@ function _launch_qemu() comm="-monitor stdio" else local qemu_comm_method="qmp" - comm="-monitor none -qmp stdio" + if [ "$qmp_pretty" = "y" ]; then + comm="-monitor none -qmp-pretty stdio" + else + comm="-monitor none -qmp stdio" + fi fi fifo_out=${QEMU_FIFO_OUT}_${_QEMU_HANDLE} @@ -XXX,XX +XXX,XX @@ function _launch_qemu() then # Don't print response, since it has version information in it silent=yes _timed_wait_for ${_QEMU_HANDLE} "capabilities" + if [ "$qmp_pretty" = "y" ]; then + silent=yes _timed_wait_for ${_QEMU_HANDLE} "^}" + fi fi QEMU_HANDLE=${_QEMU_HANDLE} let _QEMU_HANDLE++ -- 2.13.6 Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/191 | 153 +++++++++ tests/qemu-iotests/191.out | 827 +++++++++++++++++++++++++++++++++++++++++++++ tests/qemu-iotests/group | 1 + 3 files changed, 981 insertions(+) create mode 100755 tests/qemu-iotests/191 create mode 100644 tests/qemu-iotests/191.out diff --git a/tests/qemu-iotests/191 b/tests/qemu-iotests/191 new file mode 100755 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/191 @@ -XXX,XX +XXX,XX @@ +#!/bin/bash +# +# Test commit block job where top has two parents +# +# Copyright (C) 2017 Red Hat, Inc. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see <http://www.gnu.org/licenses/>. +# + +# creator +owner=kwolf@redhat.com + +seq=`basename $0` +echo "QA output created by $seq" + +here=`pwd` +status=1 # failure is the default! + +MIG_SOCKET="${TEST_DIR}/migrate" + +_cleanup() +{ + rm -f "${TEST_IMG}.mid" + rm -f "${TEST_IMG}.ovl2" + rm -f "${TEST_IMG}.ovl3" + _cleanup_test_img + _cleanup_qemu +} +trap "_cleanup; exit \$status" 0 1 2 3 15 + +# get standard environment, filters and checks +. ./common.rc +. ./common.filter +. ./common.qemu + +_supported_fmt qcow2 +_unsupported_imgopts compat=0.10 +_supported_proto file +_supported_os Linux + +size=64M + +echo +echo === Preparing and starting VM === +echo + +TEST_IMG="${TEST_IMG}.base" _make_test_img $size +TEST_IMG="${TEST_IMG}.mid" _make_test_img -b "${TEST_IMG}.base" +_make_test_img -b "${TEST_IMG}.mid" +TEST_IMG="${TEST_IMG}.ovl2" _make_test_img -b "${TEST_IMG}.mid" + +$QEMU_IO -c 'write -P 0x55 1M 64k' "${TEST_IMG}.mid" | _filter_qemu_io + +qemu_comm_method="qmp" +qmp_pretty="y" + +_launch_qemu \ + -blockdev "driver=${IMGFMT},file.driver=file,file.filename=${TEST_IMG}.base,node-name=base" \ + -blockdev "driver=${IMGFMT},file.driver=file,file.filename=${TEST_IMG}.mid,node-name=mid,backing=base" \ + -blockdev "driver=${IMGFMT},file.driver=file,file.filename=${TEST_IMG},node-name=top,backing=mid" \ + -blockdev "driver=${IMGFMT},file.driver=file,file.filename=${TEST_IMG}.ovl2,node-name=top2,backing=mid" +h=$QEMU_HANDLE +_send_qemu_cmd $h "{ 'execute': 'qmp_capabilities' }" '^}' + +echo +echo === Perform commit job === +echo + +_send_qemu_cmd $h \ + "{ 'execute': 'block-commit', + 'arguments': { 'job-id': 'commit0', + 'device': 'top', + 'base':'$TEST_IMG.base', + 'top': '$TEST_IMG.mid' } }" \ + "BLOCK_JOB_COMPLETED" +_send_qemu_cmd $h "" "^}" + +echo +echo === Check that both top and top2 point to base now === +echo + +_send_qemu_cmd $h "{ 'execute': 'query-named-block-nodes' }" "^}" | + _filter_generated_node_ids + +_send_qemu_cmd $h "{ 'execute': 'quit' }" "^}" +wait=1 _cleanup_qemu + +_img_info +TEST_IMG="$TEST_IMG.ovl2" _img_info + + +echo +echo === Preparing and starting VM with -drive === +echo + +TEST_IMG="${TEST_IMG}.base" _make_test_img $size +TEST_IMG="${TEST_IMG}.mid" _make_test_img -b "${TEST_IMG}.base" +_make_test_img -b "${TEST_IMG}.mid" +TEST_IMG="${TEST_IMG}.ovl2" _make_test_img -b "${TEST_IMG}.mid" +TEST_IMG="${TEST_IMG}.ovl3" _make_test_img -b "${TEST_IMG}.ovl2" + +$QEMU_IO -c 'write -P 0x55 1M 64k' "${TEST_IMG}.mid" | _filter_qemu_io + +qemu_comm_method="qmp" +qmp_pretty="y" + +_launch_qemu \ + -drive "driver=${IMGFMT},file=${TEST_IMG},node-name=top,backing.node-name=mid" \ + -drive "driver=${IMGFMT},file=${TEST_IMG}.ovl3,node-name=top2,backing.backing=mid" +h=$QEMU_HANDLE +_send_qemu_cmd $h "{ 'execute': 'qmp_capabilities' }" '^}' + +echo +echo === Perform commit job === +echo + +_send_qemu_cmd $h \ + "{ 'execute': 'block-commit', + 'arguments': { 'job-id': 'commit0', + 'device': 'top', + 'base':'$TEST_IMG.base', + 'top': '$TEST_IMG.mid' } }" \ + "BLOCK_JOB_COMPLETED" +_send_qemu_cmd $h "" "^}" + +echo +echo === Check that both top and top2 point to base now === +echo + +_send_qemu_cmd $h "{ 'execute': 'query-named-block-nodes' }" "^}" | + _filter_generated_node_ids + +_send_qemu_cmd $h "{ 'execute': 'quit' }" "^}" +wait=1 _cleanup_qemu + +_img_info +TEST_IMG="$TEST_IMG.ovl2" _img_info + +# success, all done +echo "*** done" +rm -f $seq.full +status=0 diff --git a/tests/qemu-iotests/191.out b/tests/qemu-iotests/191.out new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/191.out @@ -XXX,XX +XXX,XX @@ +QA output created by 191 + +=== Preparing and starting VM === + +Formatting 'TEST_DIR/t.IMGFMT.base', fmt=IMGFMT size=67108864 +Formatting 'TEST_DIR/t.IMGFMT.mid', fmt=IMGFMT size=67108864 backing_file=TEST_DIR/t.IMGFMT.base +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 backing_file=TEST_DIR/t.IMGFMT.mid +Formatting 'TEST_DIR/t.IMGFMT.ovl2', fmt=IMGFMT size=67108864 backing_file=TEST_DIR/t.IMGFMT.mid +wrote 65536/65536 bytes at offset 1048576 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +{ + "return": { + } +} + +=== Perform commit job === + +{ + "return": { + } +} +{ + "timestamp": { + "seconds": TIMESTAMP, + "microseconds": TIMESTAMP + }, + "event": "BLOCK_JOB_COMPLETED", + "data": { + "device": "commit0", + "len": 67108864, + "offset": 67108864, + "speed": 0, + "type": "commit" + } +} + +=== Check that both top and top2 point to base now === + +{ + "return": [ + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "backing-image": { + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.base", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "dirty-flag": false + }, + "backing-filename-format": "qcow2", + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.ovl2", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 200704, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "full-backing-filename": "TEST_DIR/t.qcow2.base", + "backing-filename": "TEST_DIR/t.qcow2.base", + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "top2", + "backing_file_depth": 1, + "drv": "qcow2", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "backing_file": "TEST_DIR/t.qcow2.base", + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.ovl2", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 197120, + "filename": "TEST_DIR/t.qcow2.ovl2", + "format": "file", + "actual-size": 200704, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "file", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.ovl2", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "backing-image": { + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.base", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "dirty-flag": false + }, + "backing-filename-format": "qcow2", + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 200704, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "full-backing-filename": "TEST_DIR/t.qcow2.base", + "backing-filename": "TEST_DIR/t.qcow2.base", + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "top", + "backing_file_depth": 1, + "drv": "qcow2", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "backing_file": "TEST_DIR/t.qcow2.base", + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 197120, + "filename": "TEST_DIR/t.qcow2", + "format": "file", + "actual-size": 200704, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "file", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "backing-image": { + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.base", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "dirty-flag": false + }, + "backing-filename-format": "qcow2", + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.mid", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "full-backing-filename": "TEST_DIR/t.qcow2.base", + "backing-filename": "TEST_DIR/t.qcow2.base", + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "mid", + "backing_file_depth": 1, + "drv": "qcow2", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "backing_file": "TEST_DIR/t.qcow2.base", + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.mid", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 393216, + "filename": "TEST_DIR/t.qcow2.mid", + "format": "file", + "actual-size": 397312, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "file", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.mid", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.base", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "base", + "backing_file_depth": 0, + "drv": "qcow2", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.base", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 393216, + "filename": "TEST_DIR/t.qcow2.base", + "format": "file", + "actual-size": 397312, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "file", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.base", + "encryption_key_missing": false + } + ] +} +{ + "return": { + } +} +{ + "timestamp": { + "seconds": TIMESTAMP, + "microseconds": TIMESTAMP + }, + "event": "SHUTDOWN", + "data": { + "guest": false + } +} +image: TEST_DIR/t.IMGFMT +file format: IMGFMT +virtual size: 64M (67108864 bytes) +cluster_size: 65536 +backing file: TEST_DIR/t.IMGFMT.base +backing file format: IMGFMT +image: TEST_DIR/t.IMGFMT.ovl2 +file format: IMGFMT +virtual size: 64M (67108864 bytes) +cluster_size: 65536 +backing file: TEST_DIR/t.IMGFMT.base +backing file format: IMGFMT + +=== Preparing and starting VM with -drive === + +Formatting 'TEST_DIR/t.IMGFMT.base', fmt=IMGFMT size=67108864 +Formatting 'TEST_DIR/t.IMGFMT.mid', fmt=IMGFMT size=67108864 backing_file=TEST_DIR/t.IMGFMT.base +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 backing_file=TEST_DIR/t.IMGFMT.mid +Formatting 'TEST_DIR/t.IMGFMT.ovl2', fmt=IMGFMT size=67108864 backing_file=TEST_DIR/t.IMGFMT.mid +Formatting 'TEST_DIR/t.IMGFMT.ovl3', fmt=IMGFMT size=67108864 backing_file=TEST_DIR/t.IMGFMT.ovl2 +wrote 65536/65536 bytes at offset 1048576 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +{ + "return": { + } +} + +=== Perform commit job === + +{ + "return": { + } +} +{ + "timestamp": { + "seconds": TIMESTAMP, + "microseconds": TIMESTAMP + }, + "event": "BLOCK_JOB_COMPLETED", + "data": { + "device": "commit0", + "len": 67108864, + "offset": 67108864, + "speed": 0, + "type": "commit" + } +} + +=== Check that both top and top2 point to base now === + +{ + "return": [ + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "backing-image": { + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.base", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "dirty-flag": false + }, + "backing-filename-format": "qcow2", + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.ovl2", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 200704, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "full-backing-filename": "TEST_DIR/t.qcow2.base", + "backing-filename": "TEST_DIR/t.qcow2.base", + "dirty-flag": false + }, + "iops_wr": 0, + "ro": true, + "node-name": "NODE_NAME", + "backing_file_depth": 1, + "drv": "qcow2", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "backing_file": "TEST_DIR/t.qcow2.base", + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.ovl2", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 197120, + "filename": "TEST_DIR/t.qcow2.ovl2", + "format": "file", + "actual-size": 200704, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": true, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "file", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.ovl2", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "backing-image": { + "backing-image": { + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.base", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "dirty-flag": false + }, + "backing-filename-format": "qcow2", + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.ovl2", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 200704, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "full-backing-filename": "TEST_DIR/t.qcow2.base", + "backing-filename": "TEST_DIR/t.qcow2.base", + "dirty-flag": false + }, + "backing-filename-format": "qcow2", + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.ovl3", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 200704, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "full-backing-filename": "TEST_DIR/t.qcow2.ovl2", + "backing-filename": "TEST_DIR/t.qcow2.ovl2", + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "top2", + "backing_file_depth": 2, + "drv": "qcow2", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "backing_file": "TEST_DIR/t.qcow2.ovl2", + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.ovl3", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 197120, + "filename": "TEST_DIR/t.qcow2.ovl3", + "format": "file", + "actual-size": 200704, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "file", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.ovl3", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.base", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": true, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "qcow2", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.base", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 393216, + "filename": "TEST_DIR/t.qcow2.base", + "format": "file", + "actual-size": 397312, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": true, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "file", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2.base", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "backing-image": { + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2.base", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 397312, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "dirty-flag": false + }, + "backing-filename-format": "qcow2", + "virtual-size": 67108864, + "filename": "TEST_DIR/t.qcow2", + "cluster-size": 65536, + "format": "qcow2", + "actual-size": 200704, + "format-specific": { + "type": "qcow2", + "data": { + "compat": "1.1", + "lazy-refcounts": false, + "refcount-bits": 16, + "corrupt": false + } + }, + "full-backing-filename": "TEST_DIR/t.qcow2.base", + "backing-filename": "TEST_DIR/t.qcow2.base", + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "top", + "backing_file_depth": 1, + "drv": "qcow2", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "backing_file": "TEST_DIR/t.qcow2.base", + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2", + "encryption_key_missing": false + }, + { + "iops_rd": 0, + "detect_zeroes": "off", + "image": { + "virtual-size": 197120, + "filename": "TEST_DIR/t.qcow2", + "format": "file", + "actual-size": 200704, + "dirty-flag": false + }, + "iops_wr": 0, + "ro": false, + "node-name": "NODE_NAME", + "backing_file_depth": 0, + "drv": "file", + "iops": 0, + "bps_wr": 0, + "write_threshold": 0, + "encrypted": false, + "bps": 0, + "bps_rd": 0, + "cache": { + "no-flush": false, + "direct": false, + "writeback": true + }, + "file": "TEST_DIR/t.qcow2", + "encryption_key_missing": false + } + ] +} +{ + "return": { + } +} +{ + "timestamp": { + "seconds": TIMESTAMP, + "microseconds": TIMESTAMP + }, + "event": "SHUTDOWN", + "data": { + "guest": false + } +} +image: TEST_DIR/t.IMGFMT +file format: IMGFMT +virtual size: 64M (67108864 bytes) +cluster_size: 65536 +backing file: TEST_DIR/t.IMGFMT.base +backing file format: IMGFMT +image: TEST_DIR/t.IMGFMT.ovl2 +file format: IMGFMT +virtual size: 64M (67108864 bytes) +cluster_size: 65536 +backing file: TEST_DIR/t.IMGFMT.base +backing file format: IMGFMT +*** done diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/group +++ b/tests/qemu-iotests/group @@ -XXX,XX +XXX,XX @@ 188 rw auto quick 189 rw auto 190 rw auto quick +191 rw auto 192 rw auto quick 194 rw auto migration quick 195 rw auto quick -- 2.13.6 We don't need to make any assumptions about the graph layout above the top node of the commit operation any more. Remove the use of bdrv_find_overlay() and related variables from the commit job code. bdrv_drop_intermediate() doesn't use the 'active' parameter any more, so we can just drop it. The overlay node was previously added to the block job to get a BLK_PERM_GRAPH_MOD. We really need to respect those permissions in bdrv_drop_intermediate() now, but as long as we haven't figured out yet how BLK_PERM_GRAPH_MOD is actually supposed to work, just leave a TODO comment there. With this change, it is now possible to perform another block job on an overlay node without conflicts. qemu-iotests 030 is changed accordingly. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> --- include/block/block.h | 3 +-- block.c | 6 +++-- block/commit.c | 62 ++++++++++++-------------------------------------- tests/qemu-iotests/030 | 4 ---- 4 files changed, 20 insertions(+), 55 deletions(-) diff --git a/include/block/block.h b/include/block/block.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block.h +++ b/include/block/block.h @@ -XXX,XX +XXX,XX @@ int bdrv_commit(BlockDriverState *bs); int bdrv_change_backing_file(BlockDriverState *bs, const char *backing_file, const char *backing_fmt); void bdrv_register(BlockDriver *bdrv); -int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top, - BlockDriverState *base, +int bdrv_drop_intermediate(BlockDriverState *top, BlockDriverState *base, const char *backing_file_str); BlockDriverState *bdrv_find_overlay(BlockDriverState *active, BlockDriverState *bs); diff --git a/block.c b/block.c index XXXXXXX..XXXXXXX 100644 --- a/block.c +++ b/block.c @@ -XXX,XX +XXX,XX @@ BlockDriverState *bdrv_find_base(BlockDriverState *bs) * if active == top, that is considered an error * */ -int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top, - BlockDriverState *base, const char *backing_file_str) +int bdrv_drop_intermediate(BlockDriverState *top, BlockDriverState *base, + const char *backing_file_str) { BdrvChild *c, *next; Error *local_err = NULL; @@ -XXX,XX +XXX,XX @@ int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top, } /* success - we can delete the intermediate states, and link top->base */ + /* TODO Check graph modification op blockers (BLK_PERM_GRAPH_MOD) once + * we've figured out how they should work. */ backing_file_str = backing_file_str ? backing_file_str : base->filename; QLIST_FOREACH_SAFE(c, &top->parents, next_parent, next) { diff --git a/block/commit.c b/block/commit.c index XXXXXXX..XXXXXXX 100644 --- a/block/commit.c +++ b/block/commit.c @@ -XXX,XX +XXX,XX @@ enum { typedef struct CommitBlockJob { BlockJob common; RateLimit limit; - BlockDriverState *active; BlockDriverState *commit_top_bs; BlockBackend *top; BlockBackend *base; BlockdevOnError on_error; int base_flags; - int orig_overlay_flags; char *backing_file_str; } CommitBlockJob; @@ -XXX,XX +XXX,XX @@ static void commit_complete(BlockJob *job, void *opaque) { CommitBlockJob *s = container_of(job, CommitBlockJob, common); CommitCompleteData *data = opaque; - BlockDriverState *active = s->active; BlockDriverState *top = blk_bs(s->top); BlockDriverState *base = blk_bs(s->base); - BlockDriverState *overlay_bs = bdrv_find_overlay(active, s->commit_top_bs); + BlockDriverState *commit_top_bs = s->commit_top_bs; int ret = data->ret; bool remove_commit_top_bs = false; - /* Make sure overlay_bs and top stay around until bdrv_set_backing_hd() */ + /* Make sure commit_top_bs and top stay around until bdrv_replace_node() */ bdrv_ref(top); - if (overlay_bs) { - bdrv_ref(overlay_bs); - } + bdrv_ref(commit_top_bs); /* Remove base node parent that still uses BLK_PERM_WRITE/RESIZE before * the normal backing chain can be restored. */ @@ -XXX,XX +XXX,XX @@ static void commit_complete(BlockJob *job, void *opaque) if (!block_job_is_cancelled(&s->common) && ret == 0) { /* success */ - ret = bdrv_drop_intermediate(active, s->commit_top_bs, base, + ret = bdrv_drop_intermediate(s->commit_top_bs, base, s->backing_file_str); - } else if (overlay_bs) { + } else { /* XXX Can (or should) we somehow keep 'consistent read' blocked even * after the failed/cancelled commit job is gone? If we already wrote * something to base, the intermediate images aren't valid any more. */ @@ -XXX,XX +XXX,XX @@ static void commit_complete(BlockJob *job, void *opaque) if (s->base_flags != bdrv_get_flags(base)) { bdrv_reopen(base, s->base_flags, NULL); } - if (overlay_bs && s->orig_overlay_flags != bdrv_get_flags(overlay_bs)) { - bdrv_reopen(overlay_bs, s->orig_overlay_flags, NULL); - } g_free(s->backing_file_str); blk_unref(s->top); @@ -XXX,XX +XXX,XX @@ static void commit_complete(BlockJob *job, void *opaque) * filter driver from the backing chain. Do this as the final step so that * the 'consistent read' permission can be granted. */ if (remove_commit_top_bs) { - bdrv_set_backing_hd(overlay_bs, top, &error_abort); + bdrv_child_try_set_perm(commit_top_bs->backing, 0, BLK_PERM_ALL, + &error_abort); + bdrv_replace_node(commit_top_bs, backing_bs(commit_top_bs), + &error_abort); } - bdrv_unref(overlay_bs); + bdrv_unref(commit_top_bs); bdrv_unref(top); } @@ -XXX,XX +XXX,XX @@ void commit_start(const char *job_id, BlockDriverState *bs, { CommitBlockJob *s; BlockReopenQueue *reopen_queue = NULL; - int orig_overlay_flags; int orig_base_flags; BlockDriverState *iter; - BlockDriverState *overlay_bs; BlockDriverState *commit_top_bs = NULL; Error *local_err = NULL; int ret; @@ -XXX,XX +XXX,XX @@ void commit_start(const char *job_id, BlockDriverState *bs, return; } - overlay_bs = bdrv_find_overlay(bs, top); - - if (overlay_bs == NULL) { - error_setg(errp, "Could not find overlay image for %s:", top->filename); - return; - } - s = block_job_create(job_id, &commit_job_driver, bs, 0, BLK_PERM_ALL, speed, BLOCK_JOB_DEFAULT, NULL, NULL, errp); if (!s) { return; } - orig_base_flags = bdrv_get_flags(base); - orig_overlay_flags = bdrv_get_flags(overlay_bs); - - /* convert base & overlay_bs to r/w, if necessary */ + /* convert base to r/w, if necessary */ + orig_base_flags = bdrv_get_flags(base); if (!(orig_base_flags & BDRV_O_RDWR)) { reopen_queue = bdrv_reopen_queue(reopen_queue, base, NULL, orig_base_flags | BDRV_O_RDWR); } - if (!(orig_overlay_flags & BDRV_O_RDWR)) { - reopen_queue = bdrv_reopen_queue(reopen_queue, overlay_bs, NULL, - orig_overlay_flags | BDRV_O_RDWR); - } + if (reopen_queue) { bdrv_reopen_multiple(bdrv_get_aio_context(bs), reopen_queue, &local_err); if (local_err != NULL) { @@ -XXX,XX +XXX,XX @@ void commit_start(const char *job_id, BlockDriverState *bs, goto fail; } - /* overlay_bs must be blocked because it needs to be modified to - * update the backing image string. */ - ret = block_job_add_bdrv(&s->common, "overlay of top", overlay_bs, - BLK_PERM_GRAPH_MOD, BLK_PERM_ALL, errp); - if (ret < 0) { - goto fail; - } - s->base = blk_new(BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE | BLK_PERM_RESIZE, @@ -XXX,XX +XXX,XX @@ void commit_start(const char *job_id, BlockDriverState *bs, goto fail; } - s->active = bs; - - s->base_flags = orig_base_flags; - s->orig_overlay_flags = orig_overlay_flags; - + s->base_flags = orig_base_flags; s->backing_file_str = g_strdup(backing_file_str); - s->on_error = on_error; trace_commit_start(bs, base, top, s); @@ -XXX,XX +XXX,XX @@ fail: blk_unref(s->top); } if (commit_top_bs) { - bdrv_set_backing_hd(overlay_bs, top, &error_abort); + bdrv_replace_node(commit_top_bs, top, &error_abort); } block_job_early_fail(&s->common); } diff --git a/tests/qemu-iotests/030 b/tests/qemu-iotests/030 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/030 +++ b/tests/qemu-iotests/030 @@ -XXX,XX +XXX,XX @@ class TestParallelOps(iotests.QMPTestCase): result = self.vm.qmp('block-stream', device='node6', base=self.imgs[4], job_id='stream-node6-v2') self.assert_qmp(result, 'error/class', 'GenericError') - # This fails because block-commit needs to block node6, the overlay of the 'top' image - result = self.vm.qmp('block-stream', device='node7', base=self.imgs[5], job_id='stream-node6-v3') - self.assert_qmp(result, 'error/class', 'GenericError') - # This fails because block-commit currently blocks the active layer even if it's not used result = self.vm.qmp('block-stream', device='drive0', base=self.imgs[5], job_id='stream-drive0') self.assert_qmp(result, 'error/class', 'GenericError') -- 2.13.6 From: Eric Blake <eblake@redhat.com> Make it easier to enable copy-on-read during iotests, by exposing a new bool option to main and open. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Jeff Cody <jcody@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- qemu-io.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/qemu-io.c b/qemu-io.c index XXXXXXX..XXXXXXX 100644 --- a/qemu-io.c +++ b/qemu-io.c @@ -XXX,XX +XXX,XX @@ static void open_help(void) " Opens a file for subsequent use by all of the other qemu-io commands.\n" " -r, -- open file read-only\n" " -s, -- use snapshot file\n" +" -C, -- use copy-on-read\n" " -n, -- disable host cache, short for -t none\n" " -U, -- force shared permissions\n" " -k, -- use kernel AIO implementation (on Linux only)\n" @@ -XXX,XX +XXX,XX @@ static const cmdinfo_t open_cmd = { .argmin = 1, .argmax = -1, .flags = CMD_NOFILE_OK, - .args = "[-rsnkU] [-t cache] [-d discard] [-o options] [path]", + .args = "[-rsCnkU] [-t cache] [-d discard] [-o options] [path]", .oneline = "open the file specified by path", .help = open_help, }; @@ -XXX,XX +XXX,XX @@ static int open_f(BlockBackend *blk, int argc, char **argv) QDict *opts; bool force_share = false; - while ((c = getopt(argc, argv, "snro:kt:d:U")) != -1) { + while ((c = getopt(argc, argv, "snCro:kt:d:U")) != -1) { switch (c) { case 's': flags |= BDRV_O_SNAPSHOT; @@ -XXX,XX +XXX,XX @@ static int open_f(BlockBackend *blk, int argc, char **argv) flags |= BDRV_O_NOCACHE; writethrough = false; break; + case 'C': + flags |= BDRV_O_COPY_ON_READ; + break; case 'r': readonly = 1; break; @@ -XXX,XX +XXX,XX @@ static void usage(const char *name) " -r, --read-only export read-only\n" " -s, --snapshot use snapshot file\n" " -n, --nocache disable host cache, short for -t none\n" +" -C, --copy-on-read enable copy-on-read\n" " -m, --misalign misalign allocations for O_DIRECT\n" " -k, --native-aio use kernel AIO implementation (on Linux only)\n" " -t, --cache=MODE use the given cache mode for the image\n" @@ -XXX,XX +XXX,XX @@ static QemuOptsList file_opts = { int main(int argc, char **argv) { int readonly = 0; - const char *sopt = "hVc:d:f:rsnmkt:T:U"; + const char *sopt = "hVc:d:f:rsnCmkt:T:U"; const struct option lopt[] = { { "help", no_argument, NULL, 'h' }, { "version", no_argument, NULL, 'V' }, @@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv) { "read-only", no_argument, NULL, 'r' }, { "snapshot", no_argument, NULL, 's' }, { "nocache", no_argument, NULL, 'n' }, + { "copy-on-read", no_argument, NULL, 'C' }, { "misalign", no_argument, NULL, 'm' }, { "native-aio", no_argument, NULL, 'k' }, { "discard", required_argument, NULL, 'd' }, @@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv) flags |= BDRV_O_NOCACHE; writethrough = false; break; + case 'C': + flags |= BDRV_O_COPY_ON_READ; + break; case 'd': if (bdrv_parse_discard_flags(optarg, &flags) < 0) { error_report("Invalid discard option: %s", optarg); -- 2.13.6 From: Eric Blake <eblake@redhat.com> Handle a 0-length block status request up front, with a uniform return value claiming the area is not allocated. Most callers don't pass a length of 0 to bdrv_get_block_status() and friends; but it definitely happens with a 0-length read when copy-on-read is enabled. While we could audit all callers to ensure that they never make a 0-length request, and then assert that fact, it was just as easy to fix things to always report success (as long as the callers are careful to not go into an infinite loop). However, we had inconsistent behavior on whether the status is reported as allocated or defers to the backing layer, depending on what callbacks the driver implements, and possibly wasting quite a few CPU cycles to get to that answer. Consistently reporting unallocated up front doesn't really hurt anything, and makes it easier both for callers (0-length requests now have well-defined behavior) and for drivers (drivers don't have to deal with 0-length requests). Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/io.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs, *pnum = 0; return BDRV_BLOCK_EOF; } + if (!nb_sectors) { + *pnum = 0; + return 0; + } n = total_sectors - sector_num; if (n < nb_sectors) { -- 2.13.6 From: Eric Blake <eblake@redhat.com> Executing qemu with a terminal as stdin will temporarily alter stty settings on that terminal (for example, disabling echo), because of how we run both the monitor and any multiplexing with guest input. Normally, qemu restores the original settings on exit; but if an iotest triggers qemu to abort in the middle, we can be left with the altered terminal setup. This can make life very annoying when debugging an iotest failure (not everyone remembers the trick of blind-typing 'stty sane' without echo, and some people prefer terminal settings that are slightly different than the defaults picked by 'stty sane'). It is possible to avoid qemu corrupting the terminal by not passing a terminal to qemu's stdin in the first place (as in, use './check ... </dev/null'), but that's extra typing to have to remember. But running 'exec </dev/null' in the harness seems like it might be too heavy of a hammer. So I instead went the the solution of saving and restoring the stty settings, only when the harness detects that it is run interactively. I tested this patch by forcing an allocation failure (I can't guarantee that this particular limit will work on all setups, but it shows the idea): $ (ulimit -S -v 500000; ./check -qcow2 1) Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/check | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -XXX,XX +XXX,XX @@ export VALGRIND_QEMU= export IMGKEYSECRET= export IMGOPTSSYNTAX=false +# Save current tty settings, since an aborting qemu call may leave things +# screwed up +STTY_RESTORE= +if test -t 0; then + STTY_RESTORE=$(stty -g) +fi + for r do @@ -XXX,XX +XXX,XX @@ END { if (NR > 0) { needwrap=false fi + if test -n "$STTY_RESTORE"; then + stty $STTY_RESTORE + fi rm -f "${TEST_DIR}"/*.out "${TEST_DIR}"/*.err "${TEST_DIR}"/*.time rm -f "${TEST_DIR}"/check.pid "${TEST_DIR}"/check.sts rm -f $tmp.* -- 2.13.6 From: Eric Blake <eblake@redhat.com> Make it possible to inject errors on writes performed during a read operation due to copy-on-read semantics. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Jeff Cody <jcody@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- qapi/block-core.json | 5 ++++- block/io.c | 1 + 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/qapi/block-core.json b/qapi/block-core.json index XXXXXXX..XXXXXXX 100644 --- a/qapi/block-core.json +++ b/qapi/block-core.json @@ -XXX,XX +XXX,XX @@ # # @l1_shrink_free_l2_clusters: discard the l2 tables. (since 2.11) # +# @cor_write: a write due to copy-on-read (since 2.11) +# # Since: 2.9 ## { 'enum': 'BlkdebugEvent', 'prefix': 'BLKDBG', @@ -XXX,XX +XXX,XX @@ 'flush_to_disk', 'pwritev_rmw_head', 'pwritev_rmw_after_head', 'pwritev_rmw_tail', 'pwritev_rmw_after_tail', 'pwritev', 'pwritev_zero', 'pwritev_done', 'empty_image_prepare', - 'l1_shrink_write_table', 'l1_shrink_free_l2_clusters' ] } + 'l1_shrink_write_table', 'l1_shrink_free_l2_clusters', + 'cor_write'] } ## # @BlkdebugInjectErrorOptions: diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_co_do_copy_on_readv(BdrvChild *child, goto err; } + bdrv_debug_event(bs, BLKDBG_COR_WRITE); if (drv->bdrv_co_pwrite_zeroes && buffer_is_zero(bounce_buffer, iov.iov_len)) { /* FIXME: Should we (perhaps conditionally) be setting -- 2.13.6 From: Eric Blake <eblake@redhat.com> Improve our braindead copy-on-read implementation. Pre-patch, we have multiple issues: - we create a bounce buffer and perform a write for the entire request, even if the active image already has 99% of the clusters occupied, and really only needs to copy-on-read the remaining 1% of the clusters - our bounce buffer was as large as the read request, and can needlessly exhaust our memory by using double the memory of the request size (the original request plus our bounce buffer), rather than a capped maximum overhead beyond the original - if a driver has a max_transfer limit, we are bypassing the normal code in bdrv_aligned_preadv() that fragments to that limit, and instead attempt to read the entire buffer from the driver in one go, which some drivers may assert on - a client can request a large request of nearly 2G such that rounding the request out to cluster boundaries results in a byte count larger than 2G. While this cannot exceed 32 bits, it DOES have some follow-on problems: -- the call to bdrv_driver_pread() can assert for exceeding BDRV_REQUEST_MAX_BYTES, if the driver is old and lacks .bdrv_co_preadv -- if the buffer is all zeroes, the subsequent call to bdrv_co_do_pwrite_zeroes is a no-op due to a negative size, which means we did not actually copy on read Fix all of these issues by breaking up the action into a loop, where each iteration is capped to sane limits. Also, querying the allocation status allows us to optimize: when data is already present in the active layer, we don't need to bounce. Note that the code has a telling comment that copy-on-read should probably be a filter driver rather than a bolt-on hack in io.c; but that remains a task for another day. CC: qemu-stable@nongnu.org Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/io.c | 120 +++++++++++++++++++++++++++++++++++++++++-------------------- 1 file changed, 82 insertions(+), 38 deletions(-) diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ #define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress */ +/* Maximum bounce buffer for copy-on-read and write zeroes, in bytes */ +#define MAX_BOUNCE_BUFFER (32768 << BDRV_SECTOR_BITS) + static int coroutine_fn bdrv_co_do_pwrite_zeroes(BlockDriverState *bs, int64_t offset, int bytes, BdrvRequestFlags flags); @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_co_do_copy_on_readv(BdrvChild *child, BlockDriver *drv = bs->drv; struct iovec iov; - QEMUIOVector bounce_qiov; + QEMUIOVector local_qiov; int64_t cluster_offset; unsigned int cluster_bytes; size_t skip_bytes; int ret; + int max_transfer = MIN_NON_ZERO(bs->bl.max_transfer, + BDRV_REQUEST_MAX_BYTES); + unsigned int progress = 0; /* FIXME We cannot require callers to have write permissions when all they * are doing is a read request. If we did things right, write permissions @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_co_do_copy_on_readv(BdrvChild *child, // assert(child->perm & (BLK_PERM_WRITE_UNCHANGED | BLK_PERM_WRITE)); /* Cover entire cluster so no additional backing file I/O is required when - * allocating cluster in the image file. + * allocating cluster in the image file. Note that this value may exceed + * BDRV_REQUEST_MAX_BYTES (even when the original read did not), which + * is one reason we loop rather than doing it all at once. */ bdrv_round_to_clusters(bs, offset, bytes, &cluster_offset, &cluster_bytes); + skip_bytes = offset - cluster_offset; trace_bdrv_co_do_copy_on_readv(bs, offset, bytes, cluster_offset, cluster_bytes); - iov.iov_len = cluster_bytes; - iov.iov_base = bounce_buffer = qemu_try_blockalign(bs, iov.iov_len); + bounce_buffer = qemu_try_blockalign(bs, + MIN(MIN(max_transfer, cluster_bytes), + MAX_BOUNCE_BUFFER)); if (bounce_buffer == NULL) { ret = -ENOMEM; goto err; } - qemu_iovec_init_external(&bounce_qiov, &iov, 1); + while (cluster_bytes) { + int64_t pnum; - ret = bdrv_driver_preadv(bs, cluster_offset, cluster_bytes, - &bounce_qiov, 0); - if (ret < 0) { - goto err; - } + ret = bdrv_is_allocated(bs, cluster_offset, + MIN(cluster_bytes, max_transfer), &pnum); + if (ret < 0) { + /* Safe to treat errors in querying allocation as if + * unallocated; we'll probably fail again soon on the + * read, but at least that will set a decent errno. + */ + pnum = MIN(cluster_bytes, max_transfer); + } - bdrv_debug_event(bs, BLKDBG_COR_WRITE); - if (drv->bdrv_co_pwrite_zeroes && - buffer_is_zero(bounce_buffer, iov.iov_len)) { - /* FIXME: Should we (perhaps conditionally) be setting - * BDRV_REQ_MAY_UNMAP, if it will allow for a sparser copy - * that still correctly reads as zero? */ - ret = bdrv_co_do_pwrite_zeroes(bs, cluster_offset, cluster_bytes, 0); - } else { - /* This does not change the data on the disk, it is not necessary - * to flush even in cache=writethrough mode. - */ - ret = bdrv_driver_pwritev(bs, cluster_offset, cluster_bytes, - &bounce_qiov, 0); - } + assert(skip_bytes < pnum); - if (ret < 0) { - /* It might be okay to ignore write errors for guest requests. If this - * is a deliberate copy-on-read then we don't want to ignore the error. - * Simply report it in all cases. - */ - goto err; - } + if (ret <= 0) { + /* Must copy-on-read; use the bounce buffer */ + iov.iov_base = bounce_buffer; + iov.iov_len = pnum = MIN(pnum, MAX_BOUNCE_BUFFER); + qemu_iovec_init_external(&local_qiov, &iov, 1); - skip_bytes = offset - cluster_offset; - qemu_iovec_from_buf(qiov, 0, bounce_buffer + skip_bytes, bytes); + ret = bdrv_driver_preadv(bs, cluster_offset, pnum, + &local_qiov, 0); + if (ret < 0) { + goto err; + } + + bdrv_debug_event(bs, BLKDBG_COR_WRITE); + if (drv->bdrv_co_pwrite_zeroes && + buffer_is_zero(bounce_buffer, pnum)) { + /* FIXME: Should we (perhaps conditionally) be setting + * BDRV_REQ_MAY_UNMAP, if it will allow for a sparser copy + * that still correctly reads as zero? */ + ret = bdrv_co_do_pwrite_zeroes(bs, cluster_offset, pnum, 0); + } else { + /* This does not change the data on the disk, it is not + * necessary to flush even in cache=writethrough mode. + */ + ret = bdrv_driver_pwritev(bs, cluster_offset, pnum, + &local_qiov, 0); + } + + if (ret < 0) { + /* It might be okay to ignore write errors for guest + * requests. If this is a deliberate copy-on-read + * then we don't want to ignore the error. Simply + * report it in all cases. + */ + goto err; + } + + qemu_iovec_from_buf(qiov, progress, bounce_buffer + skip_bytes, + pnum - skip_bytes); + } else { + /* Read directly into the destination */ + qemu_iovec_init(&local_qiov, qiov->niov); + qemu_iovec_concat(&local_qiov, qiov, progress, pnum - skip_bytes); + ret = bdrv_driver_preadv(bs, offset + progress, local_qiov.size, + &local_qiov, 0); + qemu_iovec_destroy(&local_qiov); + if (ret < 0) { + goto err; + } + } + + cluster_offset += pnum; + cluster_bytes -= pnum; + progress += pnum - skip_bytes; + skip_bytes = 0; + } + ret = 0; err: qemu_vfree(bounce_buffer); @@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_readv(BdrvChild *child, int64_t sector_num, return bdrv_co_do_readv(child, sector_num, nb_sectors, qiov, 0); } -/* Maximum buffer for write zeroes fallback, in bytes */ -#define MAX_WRITE_ZEROES_BOUNCE_BUFFER (32768 << BDRV_SECTOR_BITS) - static int coroutine_fn bdrv_co_do_pwrite_zeroes(BlockDriverState *bs, int64_t offset, int bytes, BdrvRequestFlags flags) { @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_co_do_pwrite_zeroes(BlockDriverState *bs, int max_write_zeroes = MIN_NON_ZERO(bs->bl.max_pwrite_zeroes, INT_MAX); int alignment = MAX(bs->bl.pwrite_zeroes_alignment, bs->bl.request_alignment); - int max_transfer = MIN_NON_ZERO(bs->bl.max_transfer, - MAX_WRITE_ZEROES_BOUNCE_BUFFER); + int max_transfer = MIN_NON_ZERO(bs->bl.max_transfer, MAX_BOUNCE_BUFFER); assert(alignment % bs->bl.request_alignment == 0); head = offset % alignment; -- 2.13.6 From: Eric Blake <eblake@redhat.com> Add a test for qcow2 copy-on-read behavior, including exposure for the just-fixed bugs. The copy-on-read behavior is always to a qcow2 image, but the test is careful to allow running with most image protocol/format combos as the backing file being copied from (luks being the exception, as it is harder to pass the right secret to all the right places). In fact, for './check nbd', this appears to be the first time we've had a qcow2 image wrapping NBD, requiring an additional line in _filter_img_create to match the similar line in _filter_img_info. Invoking blkdebug to prove we don't write too much took some effort to get working; and it requires that $TEST_WRAP (based on $TEST_DIR) not be subject to word splitting. We may decide later to have the entire iotests suite use relative rather than absolute names, to avoid problems inherited by the absolute name of $PWD or $TEST_DIR, at which point the sanity check in this commit could be simplified. This test requires at least 2G of consecutive memory to succeed; as such, it is prone to spurious failures, particularly on 32-bit machines under load. This situation is detected and triggers an early exit to skip the test, rather than a failure. To manually provoke this setup on a beefier machine, I used: $ (ulimit -S -v 1000000; ./check -qcow2 197) Signed-off-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/197 | 109 +++++++++++++++++++++++++++++++++++++++ tests/qemu-iotests/197.out | 26 ++++++++++ tests/qemu-iotests/common.filter | 1 + tests/qemu-iotests/group | 1 + 4 files changed, 137 insertions(+) create mode 100755 tests/qemu-iotests/197 create mode 100644 tests/qemu-iotests/197.out diff --git a/tests/qemu-iotests/197 b/tests/qemu-iotests/197 new file mode 100755 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/197 @@ -XXX,XX +XXX,XX @@ +#!/bin/bash +# +# Test case for copy-on-read into qcow2 +# +# Copyright (C) 2017 Red Hat, Inc. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see <http://www.gnu.org/licenses/>. +# + +# creator +owner=eblake@redhat.com + +seq="$(basename $0)" +echo "QA output created by $seq" + +here="$PWD" +status=1 # failure is the default! + +# get standard environment, filters and checks +. ./common.rc +. ./common.filter + +TEST_WRAP="$TEST_DIR/t.wrap.qcow2" +BLKDBG_CONF="$TEST_DIR/blkdebug.conf" + +# Sanity check: our use of blkdebug fails if $TEST_DIR contains spaces +# or other problems +case "$TEST_DIR" in + *[^-_a-zA-Z0-9/]*) + _notrun "Suspicious TEST_DIR='$TEST_DIR', cowardly refusing to run" ;; +esac + +_cleanup() +{ + _cleanup_test_img + rm -f "$BLKDBG_CONF" +} +trap "_cleanup; exit \$status" 0 1 2 3 15 + +# Test is supported for any backing file; but we force qcow2 for our wrapper. +_supported_fmt generic +_supported_proto generic +_supported_os Linux +# LUKS support may be possible, but it complicates things. +_unsupported_fmt luks + +echo +echo '=== Copy-on-read ===' +echo + +# Prep the images +_make_test_img 4G +$QEMU_IO -c "write -P 55 3G 1k" "$TEST_IMG" | _filter_qemu_io +IMGPROTO=file IMGFMT=qcow2 IMGOPTS= TEST_IMG_FILE="$TEST_WRAP" \ + _make_test_img -F "$IMGFMT" -b "$TEST_IMG" | _filter_img_create +$QEMU_IO -f qcow2 -c "write -z -u 1M 64k" "$TEST_WRAP" | _filter_qemu_io + +# Ensure that a read of two clusters, but where one is already allocated, +# does not re-write the allocated cluster +cat > "$BLKDBG_CONF" <<EOF +[inject-error] +event = "cor_write" +sector = "2048" +EOF +$QEMU_IO -c "open -C \ + -o driver=blkdebug,config=$BLKDBG_CONF,image.driver=qcow2 $TEST_WRAP" \ + -c "read -P 0 1M 128k" | _filter_qemu_io + +# Read the areas we want copied. A zero-length read should still be a +# no-op. The next read is under 2G, but aligned so that rounding to +# clusters copies more than 2G of zeroes. The final read will pick up +# the non-zero data in the same cluster. Since a 2G read may exhaust +# memory on some machines (particularly 32-bit), we skip the test if +# that fails due to memory pressure. +$QEMU_IO -f qcow2 -C -c "read 0 0" "$TEST_WRAP" | _filter_qemu_io +output=$($QEMU_IO -f qcow2 -C -c "read -P 0 1k $((2*1024*1024*1024 - 512))" \ + "$TEST_WRAP" 2>&1 | _filter_qemu_io) +case $output in + *allocate*) + _notrun "Insufficent memory to run test" ;; + *) printf '%s\n' "$output" ;; +esac +$QEMU_IO -f qcow2 -C -c "read -P 0 $((3*1024*1024*1024 + 1024)) 1k" \ + "$TEST_WRAP" | _filter_qemu_io + +# Copy-on-read is incompatible with read-only +$QEMU_IO -f qcow2 -C -r "$TEST_WRAP" 2>&1 | _filter_testdir + +# Break the backing chain, and show that images are identical, and that +# we properly copied over explicit zeros. +$QEMU_IMG rebase -u -b "" -f qcow2 "$TEST_WRAP" +$QEMU_IO -f qcow2 -c map "$TEST_WRAP" +_check_test_img +$QEMU_IMG compare -f $IMGFMT -F qcow2 "$TEST_IMG" "$TEST_WRAP" + +# success, all done +echo '*** done' +status=0 diff --git a/tests/qemu-iotests/197.out b/tests/qemu-iotests/197.out new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/197.out @@ -XXX,XX +XXX,XX @@ +QA output created by 197 + +=== Copy-on-read === + +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=4294967296 +wrote 1024/1024 bytes at offset 3221225472 +1 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +Formatting 'TEST_DIR/t.wrap.IMGFMT', fmt=IMGFMT size=4294967296 backing_file=TEST_DIR/t.IMGFMT backing_fmt=IMGFMT +wrote 65536/65536 bytes at offset 1048576 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +read 131072/131072 bytes at offset 1048576 +128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +read 0/0 bytes at offset 0 +0 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +read 2147483136/2147483136 bytes at offset 1024 +2 GiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +read 1024/1024 bytes at offset 3221226496 +1 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +can't open device TEST_DIR/t.wrap.qcow2: Can't use copy-on-read on read-only device +2 GiB (0x80010000) bytes allocated at offset 0 bytes (0x0) +1023.938 MiB (0x3fff0000) bytes not allocated at offset 2 GiB (0x80010000) +64 KiB (0x10000) bytes allocated at offset 3 GiB (0xc0000000) +1023.938 MiB (0x3fff0000) bytes not allocated at offset 3 GiB (0xc0010000) +No errors were found on the image. +Images are identical. +*** done diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/common.filter +++ b/tests/qemu-iotests/common.filter @@ -XXX,XX +XXX,XX @@ _filter_img_create() sed -e "s#$IMGPROTO:$TEST_DIR#TEST_DIR#g" \ -e "s#$TEST_DIR#TEST_DIR#g" \ -e "s#$IMGFMT#IMGFMT#g" \ + -e 's#nbd:127.0.0.1:10810#TEST_DIR/t.IMGFMT#g' \ -e "s# encryption=off##g" \ -e "s# cluster_size=[0-9]\\+##g" \ -e "s# table_size=[0-9]\\+##g" \ diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/group +++ b/tests/qemu-iotests/group @@ -XXX,XX +XXX,XX @@ 192 rw auto quick 194 rw auto migration quick 195 rw auto quick +197 rw auto quick -- 2.13.6 From: "Daniel P. Berrange" <berrange@redhat.com> Using 16KB bounce buffers creates a significant performance penalty for I/O to encrypted volumes on storage which high I/O latency (rotating rust & network drives), because it triggers lots of fairly small I/O operations. On tests with rotating rust, and cache=none|directsync, write speed increased from 2MiB/s to 32MiB/s, on a par with that achieved by the in-kernel luks driver. With other cache modes the in-kernel driver is still notably faster because it is able to report completion of the I/O request before any encryption is done, while the in-QEMU driver must encrypt the data before completion. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20170927125340.12360-2-berrange@redhat.com Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/crypto.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) diff --git a/block/crypto.c b/block/crypto.c index XXXXXXX..XXXXXXX 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -XXX,XX +XXX,XX @@ static void block_crypto_close(BlockDriverState *bs) } -#define BLOCK_CRYPTO_MAX_SECTORS 32 +/* + * 1 MB bounce buffer gives good performance / memory tradeoff + * when using cache=none|directsync. + */ +#define BLOCK_CRYPTO_MAX_IO_SIZE (1024 * 1024) static coroutine_fn int block_crypto_co_readv(BlockDriverState *bs, int64_t sector_num, @@ -XXX,XX +XXX,XX @@ block_crypto_co_readv(BlockDriverState *bs, int64_t sector_num, qemu_iovec_init(&hd_qiov, qiov->niov); - /* Bounce buffer so we have a linear mem region for - * entire sector. XXX optimize so we avoid bounce - * buffer in case that qiov->niov == 1 + /* Bounce buffer because we don't wish to expose cipher text + * in qiov which points to guest memory. */ cipher_data = - qemu_try_blockalign(bs->file->bs, MIN(BLOCK_CRYPTO_MAX_SECTORS * 512, + qemu_try_blockalign(bs->file->bs, MIN(BLOCK_CRYPTO_MAX_IO_SIZE, qiov->size)); if (cipher_data == NULL) { ret = -ENOMEM; @@ -XXX,XX +XXX,XX @@ block_crypto_co_readv(BlockDriverState *bs, int64_t sector_num, while (remaining_sectors) { cur_nr_sectors = remaining_sectors; - if (cur_nr_sectors > BLOCK_CRYPTO_MAX_SECTORS) { - cur_nr_sectors = BLOCK_CRYPTO_MAX_SECTORS; + if (cur_nr_sectors > (BLOCK_CRYPTO_MAX_IO_SIZE / 512)) { + cur_nr_sectors = (BLOCK_CRYPTO_MAX_IO_SIZE / 512); } qemu_iovec_reset(&hd_qiov); @@ -XXX,XX +XXX,XX @@ block_crypto_co_writev(BlockDriverState *bs, int64_t sector_num, qemu_iovec_init(&hd_qiov, qiov->niov); - /* Bounce buffer so we have a linear mem region for - * entire sector. XXX optimize so we avoid bounce - * buffer in case that qiov->niov == 1 + /* Bounce buffer because we're not permitted to touch + * contents of qiov - it points to guest memory. */ cipher_data = - qemu_try_blockalign(bs->file->bs, MIN(BLOCK_CRYPTO_MAX_SECTORS * 512, + qemu_try_blockalign(bs->file->bs, MIN(BLOCK_CRYPTO_MAX_IO_SIZE, qiov->size)); if (cipher_data == NULL) { ret = -ENOMEM; @@ -XXX,XX +XXX,XX @@ block_crypto_co_writev(BlockDriverState *bs, int64_t sector_num, while (remaining_sectors) { cur_nr_sectors = remaining_sectors; - if (cur_nr_sectors > BLOCK_CRYPTO_MAX_SECTORS) { - cur_nr_sectors = BLOCK_CRYPTO_MAX_SECTORS; + if (cur_nr_sectors > (BLOCK_CRYPTO_MAX_IO_SIZE / 512)) { + cur_nr_sectors = (BLOCK_CRYPTO_MAX_IO_SIZE / 512); } qemu_iovec_to_buf(qiov, bytes_done, -- 2.13.6 From: "Daniel P. Berrange" <berrange@redhat.com> While current encryption schemes all have a fixed sector size of 512 bytes, this is not guaranteed to be the case in future. Expose the sector size in the APIs so the block layer can remove assumptions about fixed 512 byte sectors. Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20170927125340.12360-3-berrange@redhat.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- crypto/blockpriv.h | 1 + include/crypto/block.h | 15 +++++++++++++++ crypto/block-luks.c | 6 ++++-- crypto/block-qcow.c | 1 + crypto/block.c | 6 ++++++ 5 files changed, 27 insertions(+), 2 deletions(-) diff --git a/crypto/blockpriv.h b/crypto/blockpriv.h index XXXXXXX..XXXXXXX 100644 --- a/crypto/blockpriv.h +++ b/crypto/blockpriv.h @@ -XXX,XX +XXX,XX @@ struct QCryptoBlock { QCryptoHashAlgorithm kdfhash; size_t niv; uint64_t payload_offset; /* In bytes */ + uint64_t sector_size; /* In bytes */ }; struct QCryptoBlockDriver { diff --git a/include/crypto/block.h b/include/crypto/block.h index XXXXXXX..XXXXXXX 100644 --- a/include/crypto/block.h +++ b/include/crypto/block.h @@ -XXX,XX +XXX,XX @@ QCryptoHashAlgorithm qcrypto_block_get_kdf_hash(QCryptoBlock *block); uint64_t qcrypto_block_get_payload_offset(QCryptoBlock *block); /** + * qcrypto_block_get_sector_size: + * @block: the block encryption object + * + * Get the size of sectors used for payload encryption. A new + * IV is used at the start of each sector. The encryption + * sector size is not required to match the sector size of the + * underlying storage. For example LUKS will always use a 512 + * byte sector size, even if the volume is on a disk with 4k + * sectors. + * + * Returns: the sector in bytes + */ +uint64_t qcrypto_block_get_sector_size(QCryptoBlock *block); + +/** * qcrypto_block_free: * @block: the block encryption object * diff --git a/crypto/block-luks.c b/crypto/block-luks.c index XXXXXXX..XXXXXXX 100644 --- a/crypto/block-luks.c +++ b/crypto/block-luks.c @@ -XXX,XX +XXX,XX @@ qcrypto_block_luks_open(QCryptoBlock *block, } } + block->sector_size = QCRYPTO_BLOCK_LUKS_SECTOR_SIZE; block->payload_offset = luks->header.payload_offset * - QCRYPTO_BLOCK_LUKS_SECTOR_SIZE; + block->sector_size; luks->cipher_alg = cipheralg; luks->cipher_mode = ciphermode; @@ -XXX,XX +XXX,XX @@ qcrypto_block_luks_create(QCryptoBlock *block, QCRYPTO_BLOCK_LUKS_SECTOR_SIZE)) * QCRYPTO_BLOCK_LUKS_NUM_KEY_SLOTS); + block->sector_size = QCRYPTO_BLOCK_LUKS_SECTOR_SIZE; block->payload_offset = luks->header.payload_offset * - QCRYPTO_BLOCK_LUKS_SECTOR_SIZE; + block->sector_size; /* Reserve header space to match payload offset */ initfunc(block, block->payload_offset, opaque, &local_err); diff --git a/crypto/block-qcow.c b/crypto/block-qcow.c index XXXXXXX..XXXXXXX 100644 --- a/crypto/block-qcow.c +++ b/crypto/block-qcow.c @@ -XXX,XX +XXX,XX @@ qcrypto_block_qcow_init(QCryptoBlock *block, goto fail; } + block->sector_size = QCRYPTO_BLOCK_QCOW_SECTOR_SIZE; block->payload_offset = 0; return 0; diff --git a/crypto/block.c b/crypto/block.c index XXXXXXX..XXXXXXX 100644 --- a/crypto/block.c +++ b/crypto/block.c @@ -XXX,XX +XXX,XX @@ uint64_t qcrypto_block_get_payload_offset(QCryptoBlock *block) } +uint64_t qcrypto_block_get_sector_size(QCryptoBlock *block) +{ + return block->sector_size; +} + + void qcrypto_block_free(QCryptoBlock *block) { if (!block) { -- 2.13.6 From: "Daniel P. Berrange" <berrange@redhat.com> The crypto APIs report the offset of the data payload as an uint64_t type, but the block driver is casting to size_t or ssize_t which will potentially truncate. Most of the block APIs use int64_t for offsets meanwhile, so even if using uint64_t in the crypto block driver we are still at risk of truncation. Change the block crypto driver to use uint64_t, but add asserts that the value is less than INT64_MAX. Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20170927125340.12360-4-berrange@redhat.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/crypto.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/block/crypto.c b/block/crypto.c index XXXXXXX..XXXXXXX 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -XXX,XX +XXX,XX @@ static int block_crypto_truncate(BlockDriverState *bs, int64_t offset, PreallocMode prealloc, Error **errp) { BlockCrypto *crypto = bs->opaque; - size_t payload_offset = + uint64_t payload_offset = qcrypto_block_get_payload_offset(crypto->block); + assert(payload_offset < (INT64_MAX - offset)); offset += payload_offset; @@ -XXX,XX +XXX,XX @@ block_crypto_co_readv(BlockDriverState *bs, int64_t sector_num, uint8_t *cipher_data = NULL; QEMUIOVector hd_qiov; int ret = 0; - size_t payload_offset = + uint64_t payload_offset = qcrypto_block_get_payload_offset(crypto->block) / 512; + assert(payload_offset < (INT64_MAX / 512)); qemu_iovec_init(&hd_qiov, qiov->niov); @@ -XXX,XX +XXX,XX @@ block_crypto_co_writev(BlockDriverState *bs, int64_t sector_num, uint8_t *cipher_data = NULL; QEMUIOVector hd_qiov; int ret = 0; - size_t payload_offset = + uint64_t payload_offset = qcrypto_block_get_payload_offset(crypto->block) / 512; + assert(payload_offset < (INT64_MAX / 512)); qemu_iovec_init(&hd_qiov, qiov->niov); @@ -XXX,XX +XXX,XX @@ static int64_t block_crypto_getlength(BlockDriverState *bs) BlockCrypto *crypto = bs->opaque; int64_t len = bdrv_getlength(bs->file->bs); - ssize_t offset = qcrypto_block_get_payload_offset(crypto->block); + uint64_t offset = qcrypto_block_get_payload_offset(crypto->block); + assert(offset < INT64_MAX); + assert(offset < len); len -= offset; -- 2.13.6 From: "Daniel P. Berrange" <berrange@redhat.com> Make the crypto driver implement the bdrv_co_preadv|pwritev callbacks, and also use bdrv_co_preadv|pwritev for I/O with the protocol driver beneath. This replaces sector based I/O with byte based I/O, and allows us to stop assuming the physical sector size matches the encryption sector size. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20170927125340.12360-5-berrange@redhat.com Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/crypto.c | 106 +++++++++++++++++++++++++++++---------------------------- 1 file changed, 54 insertions(+), 52 deletions(-) diff --git a/block/crypto.c b/block/crypto.c index XXXXXXX..XXXXXXX 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -XXX,XX +XXX,XX @@ static void block_crypto_close(BlockDriverState *bs) #define BLOCK_CRYPTO_MAX_IO_SIZE (1024 * 1024) static coroutine_fn int -block_crypto_co_readv(BlockDriverState *bs, int64_t sector_num, - int remaining_sectors, QEMUIOVector *qiov) +block_crypto_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes, + QEMUIOVector *qiov, int flags) { BlockCrypto *crypto = bs->opaque; - int cur_nr_sectors; /* number of sectors in current iteration */ + uint64_t cur_bytes; /* number of bytes in current iteration */ uint64_t bytes_done = 0; uint8_t *cipher_data = NULL; QEMUIOVector hd_qiov; int ret = 0; - uint64_t payload_offset = - qcrypto_block_get_payload_offset(crypto->block) / 512; - assert(payload_offset < (INT64_MAX / 512)); + uint64_t sector_size = qcrypto_block_get_sector_size(crypto->block); + uint64_t payload_offset = qcrypto_block_get_payload_offset(crypto->block); + uint64_t sector_num = offset / sector_size; + + assert(!flags); + assert(payload_offset < INT64_MAX); + assert(QEMU_IS_ALIGNED(offset, sector_size)); + assert(QEMU_IS_ALIGNED(bytes, sector_size)); qemu_iovec_init(&hd_qiov, qiov->niov); @@ -XXX,XX +XXX,XX @@ block_crypto_co_readv(BlockDriverState *bs, int64_t sector_num, goto cleanup; } - while (remaining_sectors) { - cur_nr_sectors = remaining_sectors; - - if (cur_nr_sectors > (BLOCK_CRYPTO_MAX_IO_SIZE / 512)) { - cur_nr_sectors = (BLOCK_CRYPTO_MAX_IO_SIZE / 512); - } + while (bytes) { + cur_bytes = MIN(bytes, BLOCK_CRYPTO_MAX_IO_SIZE); qemu_iovec_reset(&hd_qiov); - qemu_iovec_add(&hd_qiov, cipher_data, cur_nr_sectors * 512); + qemu_iovec_add(&hd_qiov, cipher_data, cur_bytes); - ret = bdrv_co_readv(bs->file, - payload_offset + sector_num, - cur_nr_sectors, &hd_qiov); + ret = bdrv_co_preadv(bs->file, payload_offset + offset + bytes_done, + cur_bytes, &hd_qiov, 0); if (ret < 0) { goto cleanup; } - if (qcrypto_block_decrypt(crypto->block, - sector_num, - cipher_data, cur_nr_sectors * 512, - NULL) < 0) { + if (qcrypto_block_decrypt(crypto->block, sector_num, cipher_data, + cur_bytes, NULL) < 0) { ret = -EIO; goto cleanup; } - qemu_iovec_from_buf(qiov, bytes_done, - cipher_data, cur_nr_sectors * 512); + qemu_iovec_from_buf(qiov, bytes_done, cipher_data, cur_bytes); - remaining_sectors -= cur_nr_sectors; - sector_num += cur_nr_sectors; - bytes_done += cur_nr_sectors * 512; + sector_num += cur_bytes / sector_size; + bytes -= cur_bytes; + bytes_done += cur_bytes; } cleanup: @@ -XXX,XX +XXX,XX @@ block_crypto_co_readv(BlockDriverState *bs, int64_t sector_num, static coroutine_fn int -block_crypto_co_writev(BlockDriverState *bs, int64_t sector_num, - int remaining_sectors, QEMUIOVector *qiov) +block_crypto_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes, + QEMUIOVector *qiov, int flags) { BlockCrypto *crypto = bs->opaque; - int cur_nr_sectors; /* number of sectors in current iteration */ + uint64_t cur_bytes; /* number of bytes in current iteration */ uint64_t bytes_done = 0; uint8_t *cipher_data = NULL; QEMUIOVector hd_qiov; int ret = 0; - uint64_t payload_offset = - qcrypto_block_get_payload_offset(crypto->block) / 512; - assert(payload_offset < (INT64_MAX / 512)); + uint64_t sector_size = qcrypto_block_get_sector_size(crypto->block); + uint64_t payload_offset = qcrypto_block_get_payload_offset(crypto->block); + uint64_t sector_num = offset / sector_size; + + assert(!flags); + assert(payload_offset < INT64_MAX); + assert(QEMU_IS_ALIGNED(offset, sector_size)); + assert(QEMU_IS_ALIGNED(bytes, sector_size)); qemu_iovec_init(&hd_qiov, qiov->niov); @@ -XXX,XX +XXX,XX @@ block_crypto_co_writev(BlockDriverState *bs, int64_t sector_num, goto cleanup; } - while (remaining_sectors) { - cur_nr_sectors = remaining_sectors; + while (bytes) { + cur_bytes = MIN(bytes, BLOCK_CRYPTO_MAX_IO_SIZE); - if (cur_nr_sectors > (BLOCK_CRYPTO_MAX_IO_SIZE / 512)) { - cur_nr_sectors = (BLOCK_CRYPTO_MAX_IO_SIZE / 512); - } - - qemu_iovec_to_buf(qiov, bytes_done, - cipher_data, cur_nr_sectors * 512); + qemu_iovec_to_buf(qiov, bytes_done, cipher_data, cur_bytes); - if (qcrypto_block_encrypt(crypto->block, - sector_num, - cipher_data, cur_nr_sectors * 512, - NULL) < 0) { + if (qcrypto_block_encrypt(crypto->block, sector_num, cipher_data, + cur_bytes, NULL) < 0) { ret = -EIO; goto cleanup; } qemu_iovec_reset(&hd_qiov); - qemu_iovec_add(&hd_qiov, cipher_data, cur_nr_sectors * 512); + qemu_iovec_add(&hd_qiov, cipher_data, cur_bytes); - ret = bdrv_co_writev(bs->file, - payload_offset + sector_num, - cur_nr_sectors, &hd_qiov); + ret = bdrv_co_pwritev(bs->file, payload_offset + offset + bytes_done, + cur_bytes, &hd_qiov, 0); if (ret < 0) { goto cleanup; } - remaining_sectors -= cur_nr_sectors; - sector_num += cur_nr_sectors; - bytes_done += cur_nr_sectors * 512; + sector_num += cur_bytes / sector_size; + bytes -= cur_bytes; + bytes_done += cur_bytes; } cleanup: @@ -XXX,XX +XXX,XX @@ block_crypto_co_writev(BlockDriverState *bs, int64_t sector_num, return ret; } +static void block_crypto_refresh_limits(BlockDriverState *bs, Error **errp) +{ + BlockCrypto *crypto = bs->opaque; + uint64_t sector_size = qcrypto_block_get_sector_size(crypto->block); + bs->bl.request_alignment = sector_size; /* No sub-sector I/O */ +} + static int64_t block_crypto_getlength(BlockDriverState *bs) { @@ -XXX,XX +XXX,XX @@ BlockDriver bdrv_crypto_luks = { .bdrv_truncate = block_crypto_truncate, .create_opts = &block_crypto_create_opts_luks, - .bdrv_co_readv = block_crypto_co_readv, - .bdrv_co_writev = block_crypto_co_writev, + .bdrv_refresh_limits = block_crypto_refresh_limits, + .bdrv_co_preadv = block_crypto_co_preadv, + .bdrv_co_pwritev = block_crypto_co_pwritev, .bdrv_getlength = block_crypto_getlength, .bdrv_get_info = block_crypto_get_info_luks, .bdrv_get_specific_info = block_crypto_get_specific_info_luks, -- 2.13.6 From: "Daniel P. Berrange" <berrange@redhat.com> Instead of sector offset, take the bytes offset when encrypting or decrypting data. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20170927125340.12360-6-berrange@redhat.com Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- crypto/blockpriv.h | 4 ++-- include/crypto/block.h | 14 ++++++++------ block/crypto.c | 12 ++++-------- block/qcow.c | 11 +++++++---- block/qcow2-cluster.c | 8 +++----- block/qcow2.c | 4 ++-- crypto/block-luks.c | 12 ++++++++---- crypto/block-qcow.c | 12 ++++++++---- crypto/block.c | 20 ++++++++++++++------ 9 files changed, 56 insertions(+), 41 deletions(-) diff --git a/crypto/blockpriv.h b/crypto/blockpriv.h index XXXXXXX..XXXXXXX 100644 --- a/crypto/blockpriv.h +++ b/crypto/blockpriv.h @@ -XXX,XX +XXX,XX @@ int qcrypto_block_decrypt_helper(QCryptoCipher *cipher, size_t niv, QCryptoIVGen *ivgen, int sectorsize, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp); @@ -XXX,XX +XXX,XX @@ int qcrypto_block_encrypt_helper(QCryptoCipher *cipher, size_t niv, QCryptoIVGen *ivgen, int sectorsize, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp); diff --git a/include/crypto/block.h b/include/crypto/block.h index XXXXXXX..XXXXXXX 100644 --- a/include/crypto/block.h +++ b/include/crypto/block.h @@ -XXX,XX +XXX,XX @@ QCryptoBlockInfo *qcrypto_block_get_info(QCryptoBlock *block, /** * @qcrypto_block_decrypt: * @block: the block encryption object - * @startsector: the sector from which @buf was read + * @offset: the position at which @iov was read * @buf: the buffer to decrypt * @len: the length of @buf in bytes * @errp: pointer to a NULL-initialized error object * * Decrypt @len bytes of cipher text in @buf, writing - * plain text back into @buf + * plain text back into @buf. @len and @offset must be + * a multiple of the encryption format sector size. * * Returns 0 on success, -1 on failure */ int qcrypto_block_decrypt(QCryptoBlock *block, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp); @@ -XXX,XX +XXX,XX @@ int qcrypto_block_decrypt(QCryptoBlock *block, /** * @qcrypto_block_encrypt: * @block: the block encryption object - * @startsector: the sector to which @buf will be written + * @offset: the position at which @iov will be written * @buf: the buffer to decrypt * @len: the length of @buf in bytes * @errp: pointer to a NULL-initialized error object * * Encrypt @len bytes of plain text in @buf, writing - * cipher text back into @buf + * cipher text back into @buf. @len and @offset must be + * a multiple of the encryption format sector size. * * Returns 0 on success, -1 on failure */ int qcrypto_block_encrypt(QCryptoBlock *block, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp); diff --git a/block/crypto.c b/block/crypto.c index XXXXXXX..XXXXXXX 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -XXX,XX +XXX,XX @@ block_crypto_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes, int ret = 0; uint64_t sector_size = qcrypto_block_get_sector_size(crypto->block); uint64_t payload_offset = qcrypto_block_get_payload_offset(crypto->block); - uint64_t sector_num = offset / sector_size; assert(!flags); assert(payload_offset < INT64_MAX); @@ -XXX,XX +XXX,XX @@ block_crypto_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes, goto cleanup; } - if (qcrypto_block_decrypt(crypto->block, sector_num, cipher_data, - cur_bytes, NULL) < 0) { + if (qcrypto_block_decrypt(crypto->block, offset + bytes_done, + cipher_data, cur_bytes, NULL) < 0) { ret = -EIO; goto cleanup; } qemu_iovec_from_buf(qiov, bytes_done, cipher_data, cur_bytes); - sector_num += cur_bytes / sector_size; bytes -= cur_bytes; bytes_done += cur_bytes; } @@ -XXX,XX +XXX,XX @@ block_crypto_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes, int ret = 0; uint64_t sector_size = qcrypto_block_get_sector_size(crypto->block); uint64_t payload_offset = qcrypto_block_get_payload_offset(crypto->block); - uint64_t sector_num = offset / sector_size; assert(!flags); assert(payload_offset < INT64_MAX); @@ -XXX,XX +XXX,XX @@ block_crypto_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes, qemu_iovec_to_buf(qiov, bytes_done, cipher_data, cur_bytes); - if (qcrypto_block_encrypt(crypto->block, sector_num, cipher_data, - cur_bytes, NULL) < 0) { + if (qcrypto_block_encrypt(crypto->block, offset + bytes_done, + cipher_data, cur_bytes, NULL) < 0) { ret = -EIO; goto cleanup; } @@ -XXX,XX +XXX,XX @@ block_crypto_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes, goto cleanup; } - sector_num += cur_bytes / sector_size; bytes -= cur_bytes; bytes_done += cur_bytes; } diff --git a/block/qcow.c b/block/qcow.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow.c +++ b/block/qcow.c @@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs, for(i = 0; i < s->cluster_sectors; i++) { if (i < n_start || i >= n_end) { memset(s->cluster_data, 0x00, 512); - if (qcrypto_block_encrypt(s->crypto, start_sect + i, + if (qcrypto_block_encrypt(s->crypto, + (start_sect + i) * + BDRV_SECTOR_SIZE, s->cluster_data, BDRV_SECTOR_SIZE, NULL) < 0) { @@ -XXX,XX +XXX,XX @@ static coroutine_fn int qcow_co_readv(BlockDriverState *bs, int64_t sector_num, } if (bs->encrypted) { assert(s->crypto); - if (qcrypto_block_decrypt(s->crypto, sector_num, buf, + if (qcrypto_block_decrypt(s->crypto, + sector_num * BDRV_SECTOR_SIZE, buf, n * BDRV_SECTOR_SIZE, NULL) < 0) { ret = -EIO; break; @@ -XXX,XX +XXX,XX @@ static coroutine_fn int qcow_co_writev(BlockDriverState *bs, int64_t sector_num, } if (bs->encrypted) { assert(s->crypto); - if (qcrypto_block_encrypt(s->crypto, sector_num, buf, - n * BDRV_SECTOR_SIZE, NULL) < 0) { + if (qcrypto_block_encrypt(s->crypto, sector_num * BDRV_SECTOR_SIZE, + buf, n * BDRV_SECTOR_SIZE, NULL) < 0) { ret = -EIO; break; } diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-cluster.c +++ b/block/qcow2-cluster.c @@ -XXX,XX +XXX,XX @@ static bool coroutine_fn do_perform_cow_encrypt(BlockDriverState *bs, { if (bytes && bs->encrypted) { BDRVQcow2State *s = bs->opaque; - int64_t sector = (s->crypt_physical_offset ? + int64_t offset = (s->crypt_physical_offset ? (cluster_offset + offset_in_cluster) : - (src_cluster_offset + offset_in_cluster)) - >> BDRV_SECTOR_BITS; + (src_cluster_offset + offset_in_cluster)); assert((offset_in_cluster & ~BDRV_SECTOR_MASK) == 0); assert((bytes & ~BDRV_SECTOR_MASK) == 0); assert(s->crypto); - if (qcrypto_block_encrypt(s->crypto, sector, buffer, - bytes, NULL) < 0) { + if (qcrypto_block_encrypt(s->crypto, offset, buffer, bytes, NULL) < 0) { return false; } } diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset, if (qcrypto_block_decrypt(s->crypto, (s->crypt_physical_offset ? cluster_offset + offset_in_cluster : - offset) >> BDRV_SECTOR_BITS, + offset), cluster_data, cur_bytes, NULL) < 0) { @@ -XXX,XX +XXX,XX @@ static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset, if (qcrypto_block_encrypt(s->crypto, (s->crypt_physical_offset ? cluster_offset + offset_in_cluster : - offset) >> BDRV_SECTOR_BITS, + offset), cluster_data, cur_bytes, NULL) < 0) { ret = -EIO; diff --git a/crypto/block-luks.c b/crypto/block-luks.c index XXXXXXX..XXXXXXX 100644 --- a/crypto/block-luks.c +++ b/crypto/block-luks.c @@ -XXX,XX +XXX,XX @@ static void qcrypto_block_luks_cleanup(QCryptoBlock *block) static int qcrypto_block_luks_decrypt(QCryptoBlock *block, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp) { + assert(QEMU_IS_ALIGNED(offset, QCRYPTO_BLOCK_LUKS_SECTOR_SIZE)); + assert(QEMU_IS_ALIGNED(len, QCRYPTO_BLOCK_LUKS_SECTOR_SIZE)); return qcrypto_block_decrypt_helper(block->cipher, block->niv, block->ivgen, QCRYPTO_BLOCK_LUKS_SECTOR_SIZE, - startsector, buf, len, errp); + offset, buf, len, errp); } static int qcrypto_block_luks_encrypt(QCryptoBlock *block, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp) { + assert(QEMU_IS_ALIGNED(offset, QCRYPTO_BLOCK_LUKS_SECTOR_SIZE)); + assert(QEMU_IS_ALIGNED(len, QCRYPTO_BLOCK_LUKS_SECTOR_SIZE)); return qcrypto_block_encrypt_helper(block->cipher, block->niv, block->ivgen, QCRYPTO_BLOCK_LUKS_SECTOR_SIZE, - startsector, buf, len, errp); + offset, buf, len, errp); } diff --git a/crypto/block-qcow.c b/crypto/block-qcow.c index XXXXXXX..XXXXXXX 100644 --- a/crypto/block-qcow.c +++ b/crypto/block-qcow.c @@ -XXX,XX +XXX,XX @@ qcrypto_block_qcow_cleanup(QCryptoBlock *block) static int qcrypto_block_qcow_decrypt(QCryptoBlock *block, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp) { + assert(QEMU_IS_ALIGNED(offset, QCRYPTO_BLOCK_QCOW_SECTOR_SIZE)); + assert(QEMU_IS_ALIGNED(len, QCRYPTO_BLOCK_QCOW_SECTOR_SIZE)); return qcrypto_block_decrypt_helper(block->cipher, block->niv, block->ivgen, QCRYPTO_BLOCK_QCOW_SECTOR_SIZE, - startsector, buf, len, errp); + offset, buf, len, errp); } static int qcrypto_block_qcow_encrypt(QCryptoBlock *block, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp) { + assert(QEMU_IS_ALIGNED(offset, QCRYPTO_BLOCK_QCOW_SECTOR_SIZE)); + assert(QEMU_IS_ALIGNED(len, QCRYPTO_BLOCK_QCOW_SECTOR_SIZE)); return qcrypto_block_encrypt_helper(block->cipher, block->niv, block->ivgen, QCRYPTO_BLOCK_QCOW_SECTOR_SIZE, - startsector, buf, len, errp); + offset, buf, len, errp); } diff --git a/crypto/block.c b/crypto/block.c index XXXXXXX..XXXXXXX 100644 --- a/crypto/block.c +++ b/crypto/block.c @@ -XXX,XX +XXX,XX @@ QCryptoBlockInfo *qcrypto_block_get_info(QCryptoBlock *block, int qcrypto_block_decrypt(QCryptoBlock *block, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp) { - return block->driver->decrypt(block, startsector, buf, len, errp); + return block->driver->decrypt(block, offset, buf, len, errp); } int qcrypto_block_encrypt(QCryptoBlock *block, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp) { - return block->driver->encrypt(block, startsector, buf, len, errp); + return block->driver->encrypt(block, offset, buf, len, errp); } @@ -XXX,XX +XXX,XX @@ int qcrypto_block_decrypt_helper(QCryptoCipher *cipher, size_t niv, QCryptoIVGen *ivgen, int sectorsize, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp) { uint8_t *iv; int ret = -1; + uint64_t startsector = offset / sectorsize; + + assert(QEMU_IS_ALIGNED(offset, sectorsize)); + assert(QEMU_IS_ALIGNED(len, sectorsize)); iv = niv ? g_new0(uint8_t, niv) : NULL; @@ -XXX,XX +XXX,XX @@ int qcrypto_block_encrypt_helper(QCryptoCipher *cipher, size_t niv, QCryptoIVGen *ivgen, int sectorsize, - uint64_t startsector, + uint64_t offset, uint8_t *buf, size_t len, Error **errp) { uint8_t *iv; int ret = -1; + uint64_t startsector = offset / sectorsize; + + assert(QEMU_IS_ALIGNED(offset, sectorsize)); + assert(QEMU_IS_ALIGNED(len, sectorsize)); iv = niv ? g_new0(uint8_t, niv) : NULL; -- 2.13.6 From: "Daniel P. Berrange" <berrange@redhat.com> The BDRV_REQ_FUA flag can trivially be allowed in the crypt driver as a passthrough to the underlying block driver. Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-id: 20170927125340.12360-7-berrange@redhat.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/crypto.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/block/crypto.c b/block/crypto.c index XXXXXXX..XXXXXXX 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -XXX,XX +XXX,XX @@ static int block_crypto_open_generic(QCryptoBlockFormat format, return -EINVAL; } + bs->supported_write_flags = BDRV_REQ_FUA & + bs->file->bs->supported_write_flags; + opts = qemu_opts_create(opts_spec, NULL, 0, &error_abort); qemu_opts_absorb_qdict(opts, options, &local_err); if (local_err) { @@ -XXX,XX +XXX,XX @@ block_crypto_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes, uint64_t sector_size = qcrypto_block_get_sector_size(crypto->block); uint64_t payload_offset = qcrypto_block_get_payload_offset(crypto->block); - assert(!flags); + assert(!(flags & ~BDRV_REQ_FUA)); assert(payload_offset < INT64_MAX); assert(QEMU_IS_ALIGNED(offset, sector_size)); assert(QEMU_IS_ALIGNED(bytes, sector_size)); @@ -XXX,XX +XXX,XX @@ block_crypto_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes, qemu_iovec_add(&hd_qiov, cipher_data, cur_bytes); ret = bdrv_co_pwritev(bs->file, payload_offset + offset + bytes_done, - cur_bytes, &hd_qiov, 0); + cur_bytes, &hd_qiov, flags); if (ret < 0) { goto cleanup; } -- 2.13.6 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Backing may be zero after failed bdrv_attach_child in bdrv_set_backing_hd, which leads to SIGSEGV. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-id: 20170928120300.58164-1-vsementsov@virtuozzo.com Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/mirror.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_mirror_top_pdiscard(BlockDriverState *bs, static void bdrv_mirror_top_refresh_filename(BlockDriverState *bs, QDict *opts) { + if (bs->backing == NULL) { + /* we can be here after failed bdrv_attach_child in + * bdrv_set_backing_hd */ + return; + } bdrv_refresh_filename(bs->backing->bs); pstrcpy(bs->exact_filename, sizeof(bs->exact_filename), bs->backing->bs->filename); -- 2.13.6 From: Max Reitz <mreitz@redhat.com> do_run_qemu() in iotest 195 first applies _filter_imgfmt when printing qemu's command line and _filter_testdir only afterwards. Therefore, if the image format is part of the test directory path, _filter_testdir will no longer apply and the actual output will differ from the reference output even in case of success. For example, TEST_DIR might be "/tmp/test-qcow2", in which case _filter_imgfmt first transforms this to "/tmp/test-IMGFMT" which is no longer recognized as the TEST_DIR by _filter_testdir. Fix this by not applying _filter_imgfmt in do_run_qemu() but in run_qemu() instead, and only after _filter_testdir. Signed-off-by: Max Reitz <mreitz@redhat.com> Message-id: 20170927211334.3988-1-mreitz@redhat.com Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/195 | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/tests/qemu-iotests/195 b/tests/qemu-iotests/195 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/195 +++ b/tests/qemu-iotests/195 @@ -XXX,XX +XXX,XX @@ _supported_os Linux function do_run_qemu() { - echo Testing: "$@" | _filter_imgfmt + echo Testing: "$@" $QEMU -nographic -qmp-pretty stdio -serial none "$@" echo } function run_qemu() { - do_run_qemu "$@" 2>&1 | _filter_testdir | _filter_qemu | _filter_qmp \ - | _filter_qemu_io | _filter_generated_node_ids + do_run_qemu "$@" 2>&1 | _filter_testdir | _filter_imgfmt | _filter_qemu \ + | _filter_qmp | _filter_qemu_io \ + | _filter_generated_node_ids } size=64M -- 2.13.6 From: Pavel Butsykin <pbutsykin@virtuozzo.com> Signed-off-by: Pavel Butsykin <pbutsykin@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-id: 20170929121613.25997-2-pbutsykin@virtuozzo.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/qcow2.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ static int qcow2_truncate(BlockDriverState *bs, int64_t offset, if (old_file_size < 0) { error_setg_errno(errp, -old_file_size, "Failed to inquire current file length"); - return ret; + return old_file_size; } nb_new_data_clusters = DIV_ROUND_UP(offset - old_length, @@ -XXX,XX +XXX,XX @@ static int qcow2_truncate(BlockDriverState *bs, int64_t offset, if (allocation_start < 0) { error_setg_errno(errp, -allocation_start, "Failed to resize refcount structures"); - return -allocation_start; + return allocation_start; } clusters_allocated = qcow2_alloc_clusters_at(bs, allocation_start, -- 2.13.6 From: Pavel Butsykin <pbutsykin@virtuozzo.com> Now after shrinking the image, at the end of the image file, there might be a tail that probably will never be used. So we can find the last used cluster and cut the tail. Signed-off-by: Pavel Butsykin <pbutsykin@virtuozzo.com> Reviewed-by: John Snow <jsnow@redhat.com> Message-id: 20170929121613.25997-3-pbutsykin@virtuozzo.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/qcow2.h | 1 + block/qcow2-refcount.c | 22 ++++++++++++++++++++++ block/qcow2.c | 23 +++++++++++++++++++++++ 3 files changed, 46 insertions(+) diff --git a/block/qcow2.h b/block/qcow2.h index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.h +++ b/block/qcow2.h @@ -XXX,XX +XXX,XX @@ int qcow2_change_refcount_order(BlockDriverState *bs, int refcount_order, BlockDriverAmendStatusCB *status_cb, void *cb_opaque, Error **errp); int qcow2_shrink_reftable(BlockDriverState *bs); +int64_t qcow2_get_last_cluster(BlockDriverState *bs, int64_t size); /* qcow2-cluster.c functions */ int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size, diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-refcount.c +++ b/block/qcow2-refcount.c @@ -XXX,XX +XXX,XX @@ out: g_free(reftable_tmp); return ret; } + +int64_t qcow2_get_last_cluster(BlockDriverState *bs, int64_t size) +{ + BDRVQcow2State *s = bs->opaque; + int64_t i; + + for (i = size_to_clusters(s, size) - 1; i >= 0; i--) { + uint64_t refcount; + int ret = qcow2_get_refcount(bs, i, &refcount); + if (ret < 0) { + fprintf(stderr, "Can't get refcount for cluster %" PRId64 ": %s\n", + i, strerror(-ret)); + return ret; + } + if (refcount > 0) { + return i; + } + } + qcow2_signal_corruption(bs, true, -1, -1, + "There are no references in the refcount table."); + return -EIO; +} diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ static int qcow2_truncate(BlockDriverState *bs, int64_t offset, new_l1_size = size_to_l1(s, offset); if (offset < old_length) { + int64_t last_cluster, old_file_size; if (prealloc != PREALLOC_MODE_OFF) { error_setg(errp, "Preallocation can't be used for shrinking an image"); @@ -XXX,XX +XXX,XX @@ static int qcow2_truncate(BlockDriverState *bs, int64_t offset, "Failed to discard unused refblocks"); return ret; } + + old_file_size = bdrv_getlength(bs->file->bs); + if (old_file_size < 0) { + error_setg_errno(errp, -old_file_size, + "Failed to inquire current file length"); + return old_file_size; + } + last_cluster = qcow2_get_last_cluster(bs, old_file_size); + if (last_cluster < 0) { + error_setg_errno(errp, -last_cluster, + "Failed to find the last cluster"); + return last_cluster; + } + if ((last_cluster + 1) * s->cluster_size < old_file_size) { + ret = bdrv_truncate(bs->file, (last_cluster + 1) * s->cluster_size, + PREALLOC_MODE_OFF, NULL); + if (ret < 0) { + warn_report("Failed to truncate the tail of the image: %s", + strerror(-ret)); + ret = 0; + } + } } else { ret = qcow2_grow_l1_table(bs, new_l1_size, true); if (ret < 0) { -- 2.13.6 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Backing may be zero after failed bdrv_append in mirror_start_job, which leads to SIGSEGV. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-id: 20170929152255.5431-1-vsementsov@virtuozzo.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/mirror.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_mirror_top_pwritev(BlockDriverState *bs, static int coroutine_fn bdrv_mirror_top_flush(BlockDriverState *bs) { + if (bs->backing == NULL) { + /* we can be here after failed bdrv_append in mirror_start_job */ + return 0; + } return bdrv_co_flush(bs->backing->bs); } -- 2.13.6 The following changes since commit 16aaacb307ed607b9780c12702c44f0fe52edc7e: Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20200430' into staging (2020-04-30 14:00:36 +0100) are available in the Git repository at: git://repo.or.cz/qemu/kevin.git tags/for-upstream for you to fetch changes up to eaae29ef89d498d0eac553c77b554f310a47f809: qemu-storage-daemon: Fix non-string --object properties (2020-04-30 17:51:07 +0200) ---------------------------------------------------------------- Block layer patches: - Fix resize (extending) of short overlays - nvme: introduce PMR support from NVMe 1.4 spec - qemu-storage-daemon: Fix non-string --object properties ---------------------------------------------------------------- Alberto Garcia (1): qcow2: Add incompatibility note between backing files and raw external data files Andrzej Jakowski (1): nvme: introduce PMR support from NVMe 1.4 spec Kevin Wolf (12): block: Add flags to BlockDriver.bdrv_co_truncate() block: Add flags to bdrv(_co)_truncate() block-backend: Add flags to blk_truncate() qcow2: Support BDRV_REQ_ZERO_WRITE for truncate raw-format: Support BDRV_REQ_ZERO_WRITE for truncate file-posix: Support BDRV_REQ_ZERO_WRITE for truncate block: truncate: Don't make backing file data visible iotests: Filter testfiles out in filter_img_info() iotests: Test committing to short backing file qcow2: Forward ZERO_WRITE flag for full preallocation qom: Factor out user_creatable_add_dict() qemu-storage-daemon: Fix non-string --object properties Paolo Bonzini (1): qemu-iotests: allow qcow2 external discarded clusters to contain stale data docs/interop/qcow2.txt | 3 + hw/block/nvme.h | 2 + include/block/block.h | 5 +- include/block/block_int.h | 10 +- include/block/nvme.h | 172 ++++++++++++++++++++++++++ include/qom/object_interfaces.h | 16 +++ include/sysemu/block-backend.h | 2 +- block.c | 3 +- block/block-backend.c | 4 +- block/commit.c | 4 +- block/crypto.c | 7 +- block/file-posix.c | 6 +- block/file-win32.c | 2 +- block/gluster.c | 1 + block/io.c | 43 ++++++- block/iscsi.c | 2 +- block/mirror.c | 2 +- block/nfs.c | 3 +- block/parallels.c | 6 +- block/qcow.c | 4 +- block/qcow2-cluster.c | 2 +- block/qcow2-refcount.c | 2 +- block/qcow2.c | 73 +++++++++-- block/qed.c | 3 +- block/raw-format.c | 6 +- block/rbd.c | 1 + block/sheepdog.c | 4 +- block/ssh.c | 2 +- block/vdi.c | 2 +- block/vhdx-log.c | 2 +- block/vhdx.c | 6 +- block/vmdk.c | 8 +- block/vpc.c | 2 +- blockdev.c | 2 +- hw/block/nvme.c | 109 ++++++++++++++++ qemu-img.c | 2 +- qemu-io-cmds.c | 2 +- qemu-storage-daemon.c | 4 +- qom/object_interfaces.c | 31 +++++ qom/qom-qmp-cmds.c | 24 +--- tests/test-block-iothread.c | 9 +- tests/qemu-iotests/iotests.py | 5 +- hw/block/Makefile.objs | 2 +- hw/block/trace-events | 4 + tests/qemu-iotests/244 | 10 +- tests/qemu-iotests/244.out | 9 +- tests/qemu-iotests/274 | 155 +++++++++++++++++++++++ tests/qemu-iotests/274.out | 268 ++++++++++++++++++++++++++++++++++++++++ tests/qemu-iotests/group | 1 + 49 files changed, 951 insertions(+), 96 deletions(-) create mode 100755 tests/qemu-iotests/274 create mode 100644 tests/qemu-iotests/274.out From: Alberto Garcia <berto@igalia.com> Backing files and raw external data files are mutually exclusive. The documentation of the raw external data bit (in autoclear_features) already indicates that, but we should also mention it on the other side. Suggested-by: Eric Blake <eblake@redhat.com> Signed-off-by: Alberto Garcia <berto@igalia.com> Message-Id: <20200410121816.8334-1-berto@igalia.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- docs/interop/qcow2.txt | 3 +++ 1 file changed, 3 insertions(+) diff --git a/docs/interop/qcow2.txt b/docs/interop/qcow2.txt index XXXXXXX..XXXXXXX 100644 --- a/docs/interop/qcow2.txt +++ b/docs/interop/qcow2.txt @@ -XXX,XX +XXX,XX @@ The first cluster of a qcow2 image contains the file header: is stored (NB: The string is not null terminated). 0 if the image doesn't have a backing file. + Note: backing files are incompatible with raw external data + files (auto-clear feature bit 1). + 16 - 19: backing_file_size Length of the backing file name in bytes. Must not be longer than 1023 bytes. Undefined if the image doesn't have -- 2.25.3 From: Paolo Bonzini <pbonzini@redhat.com> Test 244 checks the expected behavior of qcow2 external data files with respect to zero and discarded clusters. Filesystems however are free to ignore discard requests, and this seems to be the case for overlayfs. Relax the tests to skip checks on the external data file for discarded areas, which implies not using qemu-img compare in the data_file_raw=on case. This fixes docker tests on RHEL8. Cc: Kevin Wolf <kwolf@redhat.com> Cc: qemu-block@nongnu.org Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Message-Id: <20200409191006.24429-1-pbonzini@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/244 | 10 ++++++++-- tests/qemu-iotests/244.out | 9 ++++++--- 2 files changed, 14 insertions(+), 5 deletions(-) diff --git a/tests/qemu-iotests/244 b/tests/qemu-iotests/244 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/244 +++ b/tests/qemu-iotests/244 @@ -XXX,XX +XXX,XX @@ $QEMU_IO -c 'read -P 0 0 1M' \ echo $QEMU_IO -c 'read -P 0 0 1M' \ -c 'read -P 0x11 1M 1M' \ - -c 'read -P 0 2M 2M' \ -c 'read -P 0x11 4M 1M' \ -c 'read -P 0 5M 1M' \ -f raw "$TEST_IMG.data" | @@ -XXX,XX +XXX,XX @@ $QEMU_IO -c 'read -P 0 0 1M' \ -f $IMGFMT "$TEST_IMG" | _filter_qemu_io +# Discarded clusters are only marked as such in the qcow2 metadata, but +# they can contain stale data in the external data file. Instead, zero +# clusters must be zeroed in the external data file too. echo -$QEMU_IMG compare "$TEST_IMG" "$TEST_IMG.data" +$QEMU_IO -c 'read -P 0 0 1M' \ + -c 'read -P 0x11 1M 1M' \ + -c 'read -P 0 3M 3M' \ + -f raw "$TEST_IMG".data | + _filter_qemu_io echo -n "qcow2 file size after I/O: " du -b $TEST_IMG | cut -f1 diff --git a/tests/qemu-iotests/244.out b/tests/qemu-iotests/244.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/244.out +++ b/tests/qemu-iotests/244.out @@ -XXX,XX +XXX,XX @@ read 1048576/1048576 bytes at offset 0 1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) read 1048576/1048576 bytes at offset 1048576 1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -read 2097152/2097152 bytes at offset 2097152 -2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) read 1048576/1048576 bytes at offset 4194304 1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) read 1048576/1048576 bytes at offset 5242880 @@ -XXX,XX +XXX,XX @@ read 1048576/1048576 bytes at offset 1048576 read 4194304/4194304 bytes at offset 2097152 4 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -Images are identical. +read 1048576/1048576 bytes at offset 0 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +read 1048576/1048576 bytes at offset 1048576 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +read 3145728/3145728 bytes at offset 3145728 +3 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) qcow2 file size after I/O: 327680 === bdrv_co_block_status test for file and offset=0 === -- 2.25.3 This adds a new BdrvRequestFlags parameter to the .bdrv_co_truncate() driver callbacks, and a supported_truncate_flags field in BlockDriverState that allows drivers to advertise support for request flags in the context of truncate. For now, we always pass 0 and no drivers declare support for any flag. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-Id: <20200424125448.63318-2-kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/block_int.h | 10 +++++++++- block/crypto.c | 3 ++- block/file-posix.c | 2 +- block/file-win32.c | 2 +- block/gluster.c | 1 + block/io.c | 8 +++++++- block/iscsi.c | 2 +- block/nfs.c | 3 ++- block/qcow2.c | 2 +- block/qed.c | 1 + block/raw-format.c | 2 +- block/rbd.c | 1 + block/sheepdog.c | 4 ++-- block/ssh.c | 2 +- tests/test-block-iothread.c | 3 ++- 15 files changed, 33 insertions(+), 13 deletions(-) diff --git a/include/block/block_int.h b/include/block/block_int.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block_int.h +++ b/include/block/block_int.h @@ -XXX,XX +XXX,XX @@ struct BlockDriver { */ int coroutine_fn (*bdrv_co_truncate)(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, - Error **errp); + BdrvRequestFlags flags, Error **errp); int64_t (*bdrv_getlength)(BlockDriverState *bs); bool has_variable_length; @@ -XXX,XX +XXX,XX @@ struct BlockDriverState { /* Flags honored during pwrite_zeroes (so far: BDRV_REQ_FUA, * BDRV_REQ_MAY_UNMAP, BDRV_REQ_WRITE_UNCHANGED) */ unsigned int supported_zero_flags; + /* + * Flags honoured during truncate (so far: BDRV_REQ_ZERO_WRITE). + * + * If BDRV_REQ_ZERO_WRITE is given, the truncate operation must make sure + * that any added space reads as all zeros. If this can't be guaranteed, + * the operation must fail. + */ + unsigned int supported_truncate_flags; /* the following member gives a name to every node on the bs graph. */ char node_name[32]; diff --git a/block/crypto.c b/block/crypto.c index XXXXXXX..XXXXXXX 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -XXX,XX +XXX,XX @@ static int block_crypto_co_create_generic(BlockDriverState *bs, static int coroutine_fn block_crypto_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp) + PreallocMode prealloc, BdrvRequestFlags flags, + Error **errp) { BlockCrypto *crypto = bs->opaque; uint64_t payload_offset = diff --git a/block/file-posix.c b/block/file-posix.c index XXXXXXX..XXXXXXX 100644 --- a/block/file-posix.c +++ b/block/file-posix.c @@ -XXX,XX +XXX,XX @@ raw_regular_truncate(BlockDriverState *bs, int fd, int64_t offset, static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, - Error **errp) + BdrvRequestFlags flags, Error **errp) { BDRVRawState *s = bs->opaque; struct stat st; diff --git a/block/file-win32.c b/block/file-win32.c index XXXXXXX..XXXXXXX 100644 --- a/block/file-win32.c +++ b/block/file-win32.c @@ -XXX,XX +XXX,XX @@ static void raw_close(BlockDriverState *bs) static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, - Error **errp) + BdrvRequestFlags flags, Error **errp) { BDRVRawState *s = bs->opaque; LONG low, high; diff --git a/block/gluster.c b/block/gluster.c index XXXXXXX..XXXXXXX 100644 --- a/block/gluster.c +++ b/block/gluster.c @@ -XXX,XX +XXX,XX @@ static coroutine_fn int qemu_gluster_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, + BdrvRequestFlags flags, Error **errp) { BDRVGlusterState *s = bs->opaque; diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact, BlockDriverState *bs = child->bs; BlockDriver *drv = bs->drv; BdrvTrackedRequest req; + BdrvRequestFlags flags = 0; int64_t old_size, new_bytes; int ret; @@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact, } if (drv->bdrv_co_truncate) { - ret = drv->bdrv_co_truncate(bs, offset, exact, prealloc, errp); + if (flags & ~bs->supported_truncate_flags) { + error_setg(errp, "Block driver does not support requested flags"); + ret = -ENOTSUP; + goto out; + } + ret = drv->bdrv_co_truncate(bs, offset, exact, prealloc, flags, errp); } else if (bs->file && drv->is_filter) { ret = bdrv_co_truncate(bs->file, offset, exact, prealloc, errp); } else { diff --git a/block/iscsi.c b/block/iscsi.c index XXXXXXX..XXXXXXX 100644 --- a/block/iscsi.c +++ b/block/iscsi.c @@ -XXX,XX +XXX,XX @@ static void iscsi_reopen_commit(BDRVReopenState *reopen_state) static int coroutine_fn iscsi_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, - Error **errp) + BdrvRequestFlags flags, Error **errp) { IscsiLun *iscsilun = bs->opaque; int64_t cur_length; diff --git a/block/nfs.c b/block/nfs.c index XXXXXXX..XXXXXXX 100644 --- a/block/nfs.c +++ b/block/nfs.c @@ -XXX,XX +XXX,XX @@ static int64_t nfs_get_allocated_file_size(BlockDriverState *bs) static int coroutine_fn nfs_file_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp) + PreallocMode prealloc, BdrvRequestFlags flags, + Error **errp) { NFSClient *client = bs->opaque; int ret; diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ fail: static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, - Error **errp) + BdrvRequestFlags flags, Error **errp) { BDRVQcow2State *s = bs->opaque; uint64_t old_length; diff --git a/block/qed.c b/block/qed.c index XXXXXXX..XXXXXXX 100644 --- a/block/qed.c +++ b/block/qed.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_qed_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, + BdrvRequestFlags flags, Error **errp) { BDRVQEDState *s = bs->opaque; diff --git a/block/raw-format.c b/block/raw-format.c index XXXXXXX..XXXXXXX 100644 --- a/block/raw-format.c +++ b/block/raw-format.c @@ -XXX,XX +XXX,XX @@ static void raw_refresh_limits(BlockDriverState *bs, Error **errp) static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, - Error **errp) + BdrvRequestFlags flags, Error **errp) { BDRVRawState *s = bs->opaque; diff --git a/block/rbd.c b/block/rbd.c index XXXXXXX..XXXXXXX 100644 --- a/block/rbd.c +++ b/block/rbd.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, + BdrvRequestFlags flags, Error **errp) { int r; diff --git a/block/sheepdog.c b/block/sheepdog.c index XXXXXXX..XXXXXXX 100644 --- a/block/sheepdog.c +++ b/block/sheepdog.c @@ -XXX,XX +XXX,XX @@ static int64_t sd_getlength(BlockDriverState *bs) static int coroutine_fn sd_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, - Error **errp) + BdrvRequestFlags flags, Error **errp) { BDRVSheepdogState *s = bs->opaque; int ret, fd; @@ -XXX,XX +XXX,XX @@ static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num, assert(!flags); if (offset > s->inode.vdi_size) { - ret = sd_co_truncate(bs, offset, false, PREALLOC_MODE_OFF, NULL); + ret = sd_co_truncate(bs, offset, false, PREALLOC_MODE_OFF, 0, NULL); if (ret < 0) { return ret; } diff --git a/block/ssh.c b/block/ssh.c index XXXXXXX..XXXXXXX 100644 --- a/block/ssh.c +++ b/block/ssh.c @@ -XXX,XX +XXX,XX @@ static int64_t ssh_getlength(BlockDriverState *bs) static int coroutine_fn ssh_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, PreallocMode prealloc, - Error **errp) + BdrvRequestFlags flags, Error **errp) { BDRVSSHState *s = bs->opaque; diff --git a/tests/test-block-iothread.c b/tests/test-block-iothread.c index XXXXXXX..XXXXXXX 100644 --- a/tests/test-block-iothread.c +++ b/tests/test-block-iothread.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_test_co_pdiscard(BlockDriverState *bs, static int coroutine_fn bdrv_test_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp) + PreallocMode prealloc, BdrvRequestFlags flags, + Error **errp) { return 0; } -- 2.25.3 Now that block drivers can support flags for .bdrv_co_truncate, expose the parameter in the node level interfaces bdrv_co_truncate() and bdrv_truncate(). Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-Id: <20200424125448.63318-3-kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/block/block.h | 5 +++-- block/block-backend.c | 2 +- block/crypto.c | 2 +- block/io.c | 12 +++++++----- block/parallels.c | 6 +++--- block/qcow.c | 4 ++-- block/qcow2-refcount.c | 2 +- block/qcow2.c | 15 +++++++++------ block/raw-format.c | 2 +- block/vhdx-log.c | 2 +- block/vhdx.c | 2 +- block/vmdk.c | 2 +- tests/test-block-iothread.c | 6 +++--- 13 files changed, 34 insertions(+), 28 deletions(-) diff --git a/include/block/block.h b/include/block/block.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block.h +++ b/include/block/block.h @@ -XXX,XX +XXX,XX @@ BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs, void bdrv_refresh_filename(BlockDriverState *bs); int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp); + PreallocMode prealloc, BdrvRequestFlags flags, + Error **errp); int bdrv_truncate(BdrvChild *child, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp); + PreallocMode prealloc, BdrvRequestFlags flags, Error **errp); int64_t bdrv_nb_sectors(BlockDriverState *bs); int64_t bdrv_getlength(BlockDriverState *bs); diff --git a/block/block-backend.c b/block/block-backend.c index XXXXXXX..XXXXXXX 100644 --- a/block/block-backend.c +++ b/block/block-backend.c @@ -XXX,XX +XXX,XX @@ int blk_truncate(BlockBackend *blk, int64_t offset, bool exact, return -ENOMEDIUM; } - return bdrv_truncate(blk->root, offset, exact, prealloc, errp); + return bdrv_truncate(blk->root, offset, exact, prealloc, 0, errp); } int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf, diff --git a/block/crypto.c b/block/crypto.c index XXXXXXX..XXXXXXX 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -XXX,XX +XXX,XX @@ block_crypto_co_truncate(BlockDriverState *bs, int64_t offset, bool exact, offset += payload_offset; - return bdrv_co_truncate(bs->file, offset, exact, prealloc, errp); + return bdrv_co_truncate(bs->file, offset, exact, prealloc, 0, errp); } static void block_crypto_close(BlockDriverState *bs) diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ static void bdrv_parent_cb_resize(BlockDriverState *bs) * 'offset' bytes in length. */ int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp) + PreallocMode prealloc, BdrvRequestFlags flags, + Error **errp) { BlockDriverState *bs = child->bs; BlockDriver *drv = bs->drv; BdrvTrackedRequest req; - BdrvRequestFlags flags = 0; int64_t old_size, new_bytes; int ret; @@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact, } ret = drv->bdrv_co_truncate(bs, offset, exact, prealloc, flags, errp); } else if (bs->file && drv->is_filter) { - ret = bdrv_co_truncate(bs->file, offset, exact, prealloc, errp); + ret = bdrv_co_truncate(bs->file, offset, exact, prealloc, flags, errp); } else { error_setg(errp, "Image format driver does not support resize"); ret = -ENOTSUP; @@ -XXX,XX +XXX,XX @@ typedef struct TruncateCo { int64_t offset; bool exact; PreallocMode prealloc; + BdrvRequestFlags flags; Error **errp; int ret; } TruncateCo; @@ -XXX,XX +XXX,XX @@ static void coroutine_fn bdrv_truncate_co_entry(void *opaque) { TruncateCo *tco = opaque; tco->ret = bdrv_co_truncate(tco->child, tco->offset, tco->exact, - tco->prealloc, tco->errp); + tco->prealloc, tco->flags, tco->errp); aio_wait_kick(); } int bdrv_truncate(BdrvChild *child, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp) + PreallocMode prealloc, BdrvRequestFlags flags, Error **errp) { Coroutine *co; TruncateCo tco = { @@ -XXX,XX +XXX,XX @@ int bdrv_truncate(BdrvChild *child, int64_t offset, bool exact, .offset = offset, .exact = exact, .prealloc = prealloc, + .flags = flags, .errp = errp, .ret = NOT_DONE, }; diff --git a/block/parallels.c b/block/parallels.c index XXXXXXX..XXXXXXX 100644 --- a/block/parallels.c +++ b/block/parallels.c @@ -XXX,XX +XXX,XX @@ static int64_t allocate_clusters(BlockDriverState *bs, int64_t sector_num, } else { ret = bdrv_truncate(bs->file, (s->data_end + space) << BDRV_SECTOR_BITS, - false, PREALLOC_MODE_OFF, NULL); + false, PREALLOC_MODE_OFF, 0, NULL); } if (ret < 0) { return ret; @@ -XXX,XX +XXX,XX @@ static int coroutine_fn parallels_co_check(BlockDriverState *bs, * That means we have to pass exact=true. */ ret = bdrv_truncate(bs->file, res->image_end_offset, true, - PREALLOC_MODE_OFF, &local_err); + PREALLOC_MODE_OFF, 0, &local_err); if (ret < 0) { error_report_err(local_err); res->check_errors++; @@ -XXX,XX +XXX,XX @@ static void parallels_close(BlockDriverState *bs) /* errors are ignored, so we might as well pass exact=true */ bdrv_truncate(bs->file, s->data_end << BDRV_SECTOR_BITS, true, - PREALLOC_MODE_OFF, NULL); + PREALLOC_MODE_OFF, 0, NULL); } g_free(s->bat_dirty_bmap); diff --git a/block/qcow.c b/block/qcow.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow.c +++ b/block/qcow.c @@ -XXX,XX +XXX,XX @@ static int get_cluster_offset(BlockDriverState *bs, return -E2BIG; } ret = bdrv_truncate(bs->file, cluster_offset + s->cluster_size, - false, PREALLOC_MODE_OFF, NULL); + false, PREALLOC_MODE_OFF, 0, NULL); if (ret < 0) { return ret; } @@ -XXX,XX +XXX,XX @@ static int qcow_make_empty(BlockDriverState *bs) l1_length) < 0) return -1; ret = bdrv_truncate(bs->file, s->l1_table_offset + l1_length, false, - PREALLOC_MODE_OFF, NULL); + PREALLOC_MODE_OFF, 0, NULL); if (ret < 0) return ret; diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-refcount.c +++ b/block/qcow2-refcount.c @@ -XXX,XX +XXX,XX @@ static int check_refblocks(BlockDriverState *bs, BdrvCheckResult *res, } ret = bdrv_truncate(bs->file, offset + s->cluster_size, false, - PREALLOC_MODE_OFF, &local_err); + PREALLOC_MODE_OFF, 0, &local_err); if (ret < 0) { error_report_err(local_err); goto resize_fail; diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn preallocate_co(BlockDriverState *bs, uint64_t offset, mode = PREALLOC_MODE_OFF; } ret = bdrv_co_truncate(s->data_file, host_offset + cur_bytes, false, - mode, errp); + mode, 0, errp); if (ret < 0) { return ret; } @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset, * always fulfilled, so there is no need to pass it on.) */ bdrv_co_truncate(bs->file, (last_cluster + 1) * s->cluster_size, - false, PREALLOC_MODE_OFF, &local_err); + false, PREALLOC_MODE_OFF, 0, &local_err); if (local_err) { warn_reportf_err(local_err, "Failed to truncate the tail of the image: "); @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset, * file should be resized to the exact target size, too, * so we pass @exact here. */ - ret = bdrv_co_truncate(s->data_file, offset, exact, prealloc, errp); + ret = bdrv_co_truncate(s->data_file, offset, exact, prealloc, 0, + errp); if (ret < 0) { goto fail; } @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset, new_file_size = allocation_start + nb_new_data_clusters * s->cluster_size; /* Image file grows, so @exact does not matter */ - ret = bdrv_co_truncate(bs->file, new_file_size, false, prealloc, errp); + ret = bdrv_co_truncate(bs->file, new_file_size, false, prealloc, 0, + errp); if (ret < 0) { error_prepend(errp, "Failed to resize underlying file: "); qcow2_free_clusters(bs, allocation_start, @@ -XXX,XX +XXX,XX @@ qcow2_co_pwritev_compressed_part(BlockDriverState *bs, if (len < 0) { return len; } - return bdrv_co_truncate(bs->file, len, false, PREALLOC_MODE_OFF, NULL); + return bdrv_co_truncate(bs->file, len, false, PREALLOC_MODE_OFF, 0, + NULL); } if (offset_into_cluster(s, offset)) { @@ -XXX,XX +XXX,XX @@ static int make_completely_empty(BlockDriverState *bs) } ret = bdrv_truncate(bs->file, (3 + l1_clusters) * s->cluster_size, false, - PREALLOC_MODE_OFF, &local_err); + PREALLOC_MODE_OFF, 0, &local_err); if (ret < 0) { error_report_err(local_err); goto fail; diff --git a/block/raw-format.c b/block/raw-format.c index XXXXXXX..XXXXXXX 100644 --- a/block/raw-format.c +++ b/block/raw-format.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset, s->size = offset; offset += s->offset; - return bdrv_co_truncate(bs->file, offset, exact, prealloc, errp); + return bdrv_co_truncate(bs->file, offset, exact, prealloc, 0, errp); } static void raw_eject(BlockDriverState *bs, bool eject_flag) diff --git a/block/vhdx-log.c b/block/vhdx-log.c index XXXXXXX..XXXXXXX 100644 --- a/block/vhdx-log.c +++ b/block/vhdx-log.c @@ -XXX,XX +XXX,XX @@ static int vhdx_log_flush(BlockDriverState *bs, BDRVVHDXState *s, goto exit; } ret = bdrv_truncate(bs->file, new_file_size, false, - PREALLOC_MODE_OFF, NULL); + PREALLOC_MODE_OFF, 0, NULL); if (ret < 0) { goto exit; } diff --git a/block/vhdx.c b/block/vhdx.c index XXXXXXX..XXXXXXX 100644 --- a/block/vhdx.c +++ b/block/vhdx.c @@ -XXX,XX +XXX,XX @@ static int vhdx_allocate_block(BlockDriverState *bs, BDRVVHDXState *s, } return bdrv_truncate(bs->file, *new_offset + s->block_size, false, - PREALLOC_MODE_OFF, NULL); + PREALLOC_MODE_OFF, 0, NULL); } /* 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 @@ vmdk_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset, } length = QEMU_ALIGN_UP(length, BDRV_SECTOR_SIZE); ret = bdrv_truncate(s->extents[i].file, length, false, - PREALLOC_MODE_OFF, NULL); + PREALLOC_MODE_OFF, 0, NULL); if (ret < 0) { return ret; } diff --git a/tests/test-block-iothread.c b/tests/test-block-iothread.c index XXXXXXX..XXXXXXX 100644 --- a/tests/test-block-iothread.c +++ b/tests/test-block-iothread.c @@ -XXX,XX +XXX,XX @@ static void test_sync_op_truncate(BdrvChild *c) int ret; /* Normal success path */ - ret = bdrv_truncate(c, 65536, false, PREALLOC_MODE_OFF, NULL); + ret = bdrv_truncate(c, 65536, false, PREALLOC_MODE_OFF, 0, NULL); g_assert_cmpint(ret, ==, 0); /* Early error: Negative offset */ - ret = bdrv_truncate(c, -2, false, PREALLOC_MODE_OFF, NULL); + ret = bdrv_truncate(c, -2, false, PREALLOC_MODE_OFF, 0, NULL); g_assert_cmpint(ret, ==, -EINVAL); /* Error: Read-only image */ c->bs->read_only = true; c->bs->open_flags &= ~BDRV_O_RDWR; - ret = bdrv_truncate(c, 65536, false, PREALLOC_MODE_OFF, NULL); + ret = bdrv_truncate(c, 65536, false, PREALLOC_MODE_OFF, 0, NULL); g_assert_cmpint(ret, ==, -EACCES); c->bs->read_only = false; -- 2.25.3 Now that node level interface bdrv_truncate() supports passing request flags to the block driver, expose this on the BlockBackend level, too. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-Id: <20200424125448.63318-4-kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/sysemu/block-backend.h | 2 +- block.c | 3 ++- block/block-backend.c | 4 ++-- block/commit.c | 4 ++-- block/crypto.c | 2 +- block/mirror.c | 2 +- block/qcow2.c | 4 ++-- block/qed.c | 2 +- block/vdi.c | 2 +- block/vhdx.c | 4 ++-- block/vmdk.c | 6 +++--- block/vpc.c | 2 +- blockdev.c | 2 +- qemu-img.c | 2 +- qemu-io-cmds.c | 2 +- 15 files changed, 22 insertions(+), 21 deletions(-) diff --git a/include/sysemu/block-backend.h b/include/sysemu/block-backend.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/block-backend.h +++ b/include/sysemu/block-backend.h @@ -XXX,XX +XXX,XX @@ int coroutine_fn blk_co_pwrite_zeroes(BlockBackend *blk, int64_t offset, int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf, int bytes); int blk_truncate(BlockBackend *blk, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp); + PreallocMode prealloc, BdrvRequestFlags flags, Error **errp); int blk_pdiscard(BlockBackend *blk, int64_t offset, int bytes); int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf, int64_t pos, int size); diff --git a/block.c b/block.c index XXXXXXX..XXXXXXX 100644 --- a/block.c +++ b/block.c @@ -XXX,XX +XXX,XX @@ static int64_t create_file_fallback_truncate(BlockBackend *blk, int64_t size; int ret; - ret = blk_truncate(blk, minimum_size, false, PREALLOC_MODE_OFF, &local_err); + ret = blk_truncate(blk, minimum_size, false, PREALLOC_MODE_OFF, 0, + &local_err); if (ret < 0 && ret != -ENOTSUP) { error_propagate(errp, local_err); return ret; diff --git a/block/block-backend.c b/block/block-backend.c index XXXXXXX..XXXXXXX 100644 --- a/block/block-backend.c +++ b/block/block-backend.c @@ -XXX,XX +XXX,XX @@ int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf, } int blk_truncate(BlockBackend *blk, int64_t offset, bool exact, - PreallocMode prealloc, Error **errp) + PreallocMode prealloc, BdrvRequestFlags flags, Error **errp) { if (!blk_is_available(blk)) { error_setg(errp, "No medium inserted"); return -ENOMEDIUM; } - return bdrv_truncate(blk->root, offset, exact, prealloc, 0, errp); + return bdrv_truncate(blk->root, offset, exact, prealloc, flags, errp); } int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf, diff --git a/block/commit.c b/block/commit.c index XXXXXXX..XXXXXXX 100644 --- a/block/commit.c +++ b/block/commit.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn commit_run(Job *job, Error **errp) } if (base_len < len) { - ret = blk_truncate(s->base, len, false, PREALLOC_MODE_OFF, NULL); + ret = blk_truncate(s->base, len, false, PREALLOC_MODE_OFF, 0, NULL); if (ret) { goto out; } @@ -XXX,XX +XXX,XX @@ int bdrv_commit(BlockDriverState *bs) * grow the backing file image if possible. If not possible, * we must return an error */ if (length > backing_length) { - ret = blk_truncate(backing, length, false, PREALLOC_MODE_OFF, + ret = blk_truncate(backing, length, false, PREALLOC_MODE_OFF, 0, &local_err); if (ret < 0) { error_report_err(local_err); diff --git a/block/crypto.c b/block/crypto.c index XXXXXXX..XXXXXXX 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -XXX,XX +XXX,XX @@ static ssize_t block_crypto_init_func(QCryptoBlock *block, * which will be used by the crypto header */ return blk_truncate(data->blk, data->size + headerlen, false, - data->prealloc, errp); + data->prealloc, 0, errp); } diff --git a/block/mirror.c b/block/mirror.c index XXXXXXX..XXXXXXX 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn mirror_run(Job *job, Error **errp) if (s->bdev_length > base_length) { ret = blk_truncate(s->target, s->bdev_length, false, - PREALLOC_MODE_OFF, NULL); + PREALLOC_MODE_OFF, 0, NULL); if (ret < 0) { goto immediate_exit; } diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ qcow2_co_create(BlockdevCreateOptions *create_options, Error **errp) /* Okay, now that we have a valid image, let's give it the right size */ ret = blk_truncate(blk, qcow2_opts->size, false, qcow2_opts->preallocation, - errp); + 0, errp); if (ret < 0) { error_prepend(errp, "Could not resize image: "); goto out; @@ -XXX,XX +XXX,XX @@ static int qcow2_amend_options(BlockDriverState *bs, QemuOpts *opts, * Amending image options should ensure that the image has * exactly the given new values, so pass exact=true here. */ - ret = blk_truncate(blk, new_size, true, PREALLOC_MODE_OFF, errp); + ret = blk_truncate(blk, new_size, true, PREALLOC_MODE_OFF, 0, errp); blk_unref(blk); if (ret < 0) { return ret; diff --git a/block/qed.c b/block/qed.c index XXXXXXX..XXXXXXX 100644 --- a/block/qed.c +++ b/block/qed.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_qed_co_create(BlockdevCreateOptions *opts, * The QED format associates file length with allocation status, * so a new file (which is empty) must have a length of 0. */ - ret = blk_truncate(blk, 0, true, PREALLOC_MODE_OFF, errp); + ret = blk_truncate(blk, 0, true, PREALLOC_MODE_OFF, 0, errp); if (ret < 0) { goto out; } diff --git a/block/vdi.c b/block/vdi.c index XXXXXXX..XXXXXXX 100644 --- a/block/vdi.c +++ b/block/vdi.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn vdi_co_do_create(BlockdevCreateOptions *create_options, if (image_type == VDI_TYPE_STATIC) { ret = blk_truncate(blk, offset + blocks * block_size, false, - PREALLOC_MODE_OFF, errp); + PREALLOC_MODE_OFF, 0, errp); if (ret < 0) { error_prepend(errp, "Failed to statically allocate file"); goto exit; diff --git a/block/vhdx.c b/block/vhdx.c index XXXXXXX..XXXXXXX 100644 --- a/block/vhdx.c +++ b/block/vhdx.c @@ -XXX,XX +XXX,XX @@ static int vhdx_create_bat(BlockBackend *blk, BDRVVHDXState *s, /* All zeroes, so we can just extend the file - the end of the BAT * is the furthest thing we have written yet */ ret = blk_truncate(blk, data_file_offset, false, PREALLOC_MODE_OFF, - errp); + 0, errp); if (ret < 0) { goto exit; } } else if (type == VHDX_TYPE_FIXED) { ret = blk_truncate(blk, data_file_offset + image_size, false, - PREALLOC_MODE_OFF, errp); + PREALLOC_MODE_OFF, 0, errp); if (ret < 0) { goto exit; } 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_init_extent(BlockBackend *blk, int gd_buf_size; if (flat) { - ret = blk_truncate(blk, filesize, false, PREALLOC_MODE_OFF, errp); + ret = blk_truncate(blk, filesize, false, PREALLOC_MODE_OFF, 0, errp); goto exit; } magic = cpu_to_be32(VMDK4_MAGIC); @@ -XXX,XX +XXX,XX @@ static int vmdk_init_extent(BlockBackend *blk, } ret = blk_truncate(blk, le64_to_cpu(header.grain_offset) << 9, false, - PREALLOC_MODE_OFF, errp); + PREALLOC_MODE_OFF, 0, errp); if (ret < 0) { goto exit; } @@ -XXX,XX +XXX,XX @@ static int coroutine_fn vmdk_co_do_create(int64_t size, /* bdrv_pwrite write padding zeros to align to sector, we don't need that * for description file */ if (desc_offset == 0) { - ret = blk_truncate(blk, desc_len, false, PREALLOC_MODE_OFF, errp); + ret = blk_truncate(blk, desc_len, false, PREALLOC_MODE_OFF, 0, errp); if (ret < 0) { goto exit; } diff --git a/block/vpc.c b/block/vpc.c index XXXXXXX..XXXXXXX 100644 --- a/block/vpc.c +++ b/block/vpc.c @@ -XXX,XX +XXX,XX @@ static int create_fixed_disk(BlockBackend *blk, uint8_t *buf, /* Add footer to total size */ total_size += HEADER_SIZE; - ret = blk_truncate(blk, total_size, false, PREALLOC_MODE_OFF, errp); + ret = blk_truncate(blk, total_size, false, PREALLOC_MODE_OFF, 0, errp); if (ret < 0) { return ret; } diff --git a/blockdev.c b/blockdev.c index XXXXXXX..XXXXXXX 100644 --- a/blockdev.c +++ b/blockdev.c @@ -XXX,XX +XXX,XX @@ void qmp_block_resize(bool has_device, const char *device, } bdrv_drained_begin(bs); - ret = blk_truncate(blk, size, false, PREALLOC_MODE_OFF, errp); + ret = blk_truncate(blk, size, false, PREALLOC_MODE_OFF, 0, errp); bdrv_drained_end(bs); out: diff --git a/qemu-img.c b/qemu-img.c index XXXXXXX..XXXXXXX 100644 --- a/qemu-img.c +++ b/qemu-img.c @@ -XXX,XX +XXX,XX @@ static int img_resize(int argc, char **argv) * resizing, so pass @exact=true. It is of no use to report * success when the image has not actually been resized. */ - ret = blk_truncate(blk, total_size, true, prealloc, &err); + ret = blk_truncate(blk, total_size, true, prealloc, 0, &err); if (!ret) { qprintf(quiet, "Image resized.\n"); } else { diff --git a/qemu-io-cmds.c b/qemu-io-cmds.c index XXXXXXX..XXXXXXX 100644 --- a/qemu-io-cmds.c +++ b/qemu-io-cmds.c @@ -XXX,XX +XXX,XX @@ static int truncate_f(BlockBackend *blk, int argc, char **argv) * exact=true. It is better to err on the "emit more errors" side * than to be overly permissive. */ - ret = blk_truncate(blk, offset, true, PREALLOC_MODE_OFF, &local_err); + ret = blk_truncate(blk, offset, true, PREALLOC_MODE_OFF, 0, &local_err); if (ret < 0) { error_report_err(local_err); return ret; -- 2.25.3 If BDRV_REQ_ZERO_WRITE is set and we're extending the image, calling qcow2_cluster_zeroize() with flags=0 does the right thing: It doesn't undo any previous preallocation, but just adds the zero flag to all relevant L2 entries. If an external data file is in use, a write_zeroes request to the data file is made instead. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-Id: <20200424125448.63318-5-kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/qcow2-cluster.c | 2 +- block/qcow2.c | 34 ++++++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+), 1 deletion(-) diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-cluster.c +++ b/block/qcow2-cluster.c @@ -XXX,XX +XXX,XX @@ int qcow2_cluster_zeroize(BlockDriverState *bs, uint64_t offset, /* Caller must pass aligned values, except at image end */ assert(QEMU_IS_ALIGNED(offset, s->cluster_size)); assert(QEMU_IS_ALIGNED(end_offset, s->cluster_size) || - end_offset == bs->total_sectors << BDRV_SECTOR_BITS); + end_offset >= bs->total_sectors << BDRV_SECTOR_BITS); /* The zero flag is only supported by version 3 and newer */ if (s->qcow_version < 3) { diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qcow2_do_open(BlockDriverState *bs, QDict *options, bs->supported_zero_flags = header.version >= 3 ? BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK : 0; + bs->supported_truncate_flags = BDRV_REQ_ZERO_WRITE; /* Repair image if dirty */ if (!(flags & (BDRV_O_CHECK | BDRV_O_INACTIVE)) && !bs->read_only && @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset, g_assert_not_reached(); } + if ((flags & BDRV_REQ_ZERO_WRITE) && offset > old_length) { + uint64_t zero_start = QEMU_ALIGN_UP(old_length, s->cluster_size); + + /* + * Use zero clusters as much as we can. qcow2_cluster_zeroize() + * requires a cluster-aligned start. The end may be unaligned if it is + * at the end of the image (which it is here). + */ + ret = qcow2_cluster_zeroize(bs, zero_start, offset - zero_start, 0); + if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to zero out new clusters"); + goto fail; + } + + /* Write explicit zeros for the unaligned head */ + if (zero_start > old_length) { + uint64_t len = zero_start - old_length; + uint8_t *buf = qemu_blockalign0(bs, len); + QEMUIOVector qiov; + qemu_iovec_init_buf(&qiov, buf, len); + + qemu_co_mutex_unlock(&s->lock); + ret = qcow2_co_pwritev_part(bs, old_length, len, &qiov, 0, 0); + qemu_co_mutex_lock(&s->lock); + + qemu_vfree(buf); + if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to zero out the new area"); + goto fail; + } + } + } + if (prealloc != PREALLOC_MODE_OFF) { /* Flush metadata before actually changing the image size */ ret = qcow2_write_caches(bs); -- 2.25.3 The raw format driver can simply forward the flag and let its bs->file child take care of actually providing the zeros. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20200424125448.63318-6-kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/raw-format.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/block/raw-format.c b/block/raw-format.c index XXXXXXX..XXXXXXX 100644 --- a/block/raw-format.c +++ b/block/raw-format.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset, s->size = offset; offset += s->offset; - return bdrv_co_truncate(bs->file, offset, exact, prealloc, 0, errp); + return bdrv_co_truncate(bs->file, offset, exact, prealloc, flags, errp); } static void raw_eject(BlockDriverState *bs, bool eject_flag) @@ -XXX,XX +XXX,XX @@ static int raw_open(BlockDriverState *bs, QDict *options, int flags, bs->supported_zero_flags = BDRV_REQ_WRITE_UNCHANGED | ((BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK) & bs->file->bs->supported_zero_flags); + bs->supported_truncate_flags = bs->file->bs->supported_truncate_flags & + BDRV_REQ_ZERO_WRITE; if (bs->probed && !bdrv_is_read_only(bs)) { bdrv_refresh_filename(bs->file->bs); -- 2.25.3 For regular files, we always get BDRV_REQ_ZERO_WRITE behaviour from the OS, so we can advertise the flag and just ignore it. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-Id: <20200424125448.63318-7-kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/file-posix.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/block/file-posix.c b/block/file-posix.c index XXXXXXX..XXXXXXX 100644 --- a/block/file-posix.c +++ b/block/file-posix.c @@ -XXX,XX +XXX,XX @@ static int raw_open_common(BlockDriverState *bs, QDict *options, #endif bs->supported_zero_flags = BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK; + if (S_ISREG(st.st_mode)) { + /* When extending regular files, we get zeros from the OS */ + bs->supported_truncate_flags = BDRV_REQ_ZERO_WRITE; + } ret = 0; fail: if (filename && (bdrv_flags & BDRV_O_TEMPORARY)) { -- 2.25.3 When extending the size of an image that has a backing file larger than its old size, make sure that the backing file data doesn't become visible in the guest, but the added area is properly zeroed out. Consider the following scenario where the overlay is shorter than its backing file: base.qcow2: AAAAAAAA overlay.qcow2: BBBB When resizing (extending) overlay.qcow2, the new blocks should not stay unallocated and make the additional As from base.qcow2 visible like before this patch, but zeros should be read. A similar case happens with the various variants of a commit job when an intermediate file is short (- for unallocated): base.qcow2: A-A-AAAA mid.qcow2: BB-B top.qcow2: C--C--C- After commit top.qcow2 to mid.qcow2, the following happens: mid.qcow2: CB-C00C0 (correct result) mid.qcow2: CB-C--C- (before this fix) Without the fix, blocks that previously read as zeros on top.qcow2 suddenly turn into A. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20200424125448.63318-8-kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/io.c | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact, goto out; } + /* + * If the image has a backing file that is large enough that it would + * provide data for the new area, we cannot leave it unallocated because + * then the backing file content would become visible. Instead, zero-fill + * the new area. + * + * Note that if the image has a backing file, but was opened without the + * backing file, taking care of keeping things consistent with that backing + * file is the user's responsibility. + */ + if (new_bytes && bs->backing) { + int64_t backing_len; + + backing_len = bdrv_getlength(backing_bs(bs)); + if (backing_len < 0) { + ret = backing_len; + error_setg_errno(errp, -ret, "Could not get backing file size"); + goto out; + } + + if (backing_len > old_size) { + flags |= BDRV_REQ_ZERO_WRITE; + } + } + if (drv->bdrv_co_truncate) { if (flags & ~bs->supported_truncate_flags) { error_setg(errp, "Block driver does not support requested flags"); -- 2.25.3 We want to keep TEST_IMG for the full path of the main test image, but filter_testfiles() must be called for other test images before replacing other things like the image format because the test directory path could contain the format as a substring. Insert a filter_testfiles() call between both. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20200424125448.63318-9-kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/iotests.py | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/iotests.py +++ b/tests/qemu-iotests/iotests.py @@ -XXX,XX +XXX,XX @@ def filter_img_info(output, filename): for line in output.split('\n'): if 'disk size' in line or 'actual-size' in line: continue - line = line.replace(filename, 'TEST_IMG') \ - .replace(imgfmt, 'IMGFMT') + line = line.replace(filename, 'TEST_IMG') + line = filter_testfiles(line) + line = line.replace(imgfmt, 'IMGFMT') line = re.sub('iters: [0-9]+', 'iters: XXX', line) line = re.sub('uuid: [-a-f0-9]+', 'uuid: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX', line) line = re.sub('cid: [0-9]+', 'cid: XXXXXXXXXX', line) -- 2.25.3 Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-Id: <20200424125448.63318-10-kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- tests/qemu-iotests/274 | 155 +++++++++++++++++++++ tests/qemu-iotests/274.out | 268 +++++++++++++++++++++++++++++++++++++ tests/qemu-iotests/group | 1 + 3 files changed, 424 insertions(+) create mode 100755 tests/qemu-iotests/274 create mode 100644 tests/qemu-iotests/274.out diff --git a/tests/qemu-iotests/274 b/tests/qemu-iotests/274 new file mode 100755 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/274 @@ -XXX,XX +XXX,XX @@ +#!/usr/bin/env python3 +# +# Copyright (C) 2019 Red Hat, Inc. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see <http://www.gnu.org/licenses/>. +# +# Creator/Owner: Kevin Wolf <kwolf@redhat.com> +# +# Some tests for short backing files and short overlays + +import iotests + +iotests.verify_image_format(supported_fmts=['qcow2']) +iotests.verify_platform(['linux']) + +size_short = 1 * 1024 * 1024 +size_long = 2 * 1024 * 1024 +size_diff = size_long - size_short + +def create_chain() -> None: + iotests.qemu_img_log('create', '-f', iotests.imgfmt, base, + str(size_long)) + iotests.qemu_img_log('create', '-f', iotests.imgfmt, '-b', base, mid, + str(size_short)) + iotests.qemu_img_log('create', '-f', iotests.imgfmt, '-b', mid, top, + str(size_long)) + + iotests.qemu_io_log('-c', 'write -P 1 0 %d' % size_long, base) + +def create_vm() -> iotests.VM: + vm = iotests.VM() + vm.add_blockdev('file,filename=%s,node-name=base-file' % base) + vm.add_blockdev('%s,file=base-file,node-name=base' % iotests.imgfmt) + vm.add_blockdev('file,filename=%s,node-name=mid-file' % mid) + vm.add_blockdev('%s,file=mid-file,node-name=mid,backing=base' + % iotests.imgfmt) + vm.add_drive(top, 'backing=mid,node-name=top') + return vm + +with iotests.FilePath('base') as base, \ + iotests.FilePath('mid') as mid, \ + iotests.FilePath('top') as top: + + iotests.log('== Commit tests ==') + + create_chain() + + iotests.log('=== Check visible data ===') + + iotests.qemu_io_log('-c', 'read -P 1 0 %d' % size_short, top) + iotests.qemu_io_log('-c', 'read -P 0 %d %d' % (size_short, size_diff), top) + + iotests.log('=== Checking allocation status ===') + + iotests.qemu_io_log('-c', 'alloc 0 %d' % size_short, + '-c', 'alloc %d %d' % (size_short, size_diff), + base) + + iotests.qemu_io_log('-c', 'alloc 0 %d' % size_short, + '-c', 'alloc %d %d' % (size_short, size_diff), + mid) + + iotests.qemu_io_log('-c', 'alloc 0 %d' % size_short, + '-c', 'alloc %d %d' % (size_short, size_diff), + top) + + iotests.log('=== Checking map ===') + + iotests.qemu_img_log('map', '--output=json', base) + iotests.qemu_img_log('map', '--output=human', base) + iotests.qemu_img_log('map', '--output=json', mid) + iotests.qemu_img_log('map', '--output=human', mid) + iotests.qemu_img_log('map', '--output=json', top) + iotests.qemu_img_log('map', '--output=human', top) + + iotests.log('=== Testing qemu-img commit (top -> mid) ===') + + iotests.qemu_img_log('commit', top) + iotests.img_info_log(mid) + iotests.qemu_io_log('-c', 'read -P 1 0 %d' % size_short, mid) + iotests.qemu_io_log('-c', 'read -P 0 %d %d' % (size_short, size_diff), mid) + + iotests.log('=== Testing HMP commit (top -> mid) ===') + + create_chain() + with create_vm() as vm: + vm.launch() + vm.qmp_log('human-monitor-command', command_line='commit drive0') + + iotests.img_info_log(mid) + iotests.qemu_io_log('-c', 'read -P 1 0 %d' % size_short, mid) + iotests.qemu_io_log('-c', 'read -P 0 %d %d' % (size_short, size_diff), mid) + + iotests.log('=== Testing QMP active commit (top -> mid) ===') + + create_chain() + with create_vm() as vm: + vm.launch() + vm.qmp_log('block-commit', device='top', base_node='mid', + job_id='job0', auto_dismiss=False) + vm.run_job('job0', wait=5) + + iotests.img_info_log(mid) + iotests.qemu_io_log('-c', 'read -P 1 0 %d' % size_short, mid) + iotests.qemu_io_log('-c', 'read -P 0 %d %d' % (size_short, size_diff), mid) + + + iotests.log('== Resize tests ==') + + # Use different sizes for different allocation modes: + # + # We want to have at least one test where 32 bit truncation in the size of + # the overlapping area becomes visible. This is covered by the + # prealloc='off' case (1G to 6G is an overlap of 5G). + # + # However, we can only do this for modes that don't preallocate data + # because otherwise we might run out of space on the test host. + # + # We also want to test some unaligned combinations. + for (prealloc, base_size, top_size_old, top_size_new, off) in [ + ('off', '6G', '1G', '8G', '5G'), + ('metadata', '32G', '30G', '33G', '31G'), + ('falloc', '10M', '5M', '15M', '9M'), + ('full', '16M', '8M', '12M', '11M'), + ('off', '384k', '253k', '512k', '253k'), + ('off', '400k', '256k', '512k', '336k'), + ('off', '512k', '256k', '500k', '436k')]: + + iotests.log('=== preallocation=%s ===' % prealloc) + iotests.qemu_img_log('create', '-f', iotests.imgfmt, base, base_size) + iotests.qemu_img_log('create', '-f', iotests.imgfmt, '-b', base, top, + top_size_old) + iotests.qemu_io_log('-c', 'write -P 1 %s 64k' % off, base) + + # After this, top_size_old to base_size should be allocated/zeroed. + # + # In theory, leaving base_size to top_size_new unallocated would be + # correct, but in practice, if we zero out anything, we zero out + # everything up to top_size_new. + iotests.qemu_img_log('resize', '-f', iotests.imgfmt, + '--preallocation', prealloc, top, top_size_new) + iotests.qemu_io_log('-c', 'read -P 0 %s 64k' % off, top) + iotests.qemu_io_log('-c', 'map', top) + iotests.qemu_img_log('map', '--output=json', top) diff --git a/tests/qemu-iotests/274.out b/tests/qemu-iotests/274.out new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/274.out @@ -XXX,XX +XXX,XX @@ +== Commit tests == +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=2097152 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-mid', fmt=qcow2 size=1048576 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=2097152 backing_file=TEST_DIR/PID-mid cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 2097152/2097152 bytes at offset 0 +2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +=== Check visible data === +read 1048576/1048576 bytes at offset 0 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +read 1048576/1048576 bytes at offset 1048576 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +=== Checking allocation status === +1048576/1048576 bytes allocated at offset 0 bytes +1048576/1048576 bytes allocated at offset 1 MiB + +0/1048576 bytes allocated at offset 0 bytes +0/0 bytes allocated at offset 1 MiB + +0/1048576 bytes allocated at offset 0 bytes +0/1048576 bytes allocated at offset 1 MiB + +=== Checking map === +[{ "start": 0, "length": 2097152, "depth": 0, "zero": false, "data": true, "offset": 327680}] + +Offset Length Mapped to File +0 0x200000 0x50000 TEST_DIR/PID-base + +[{ "start": 0, "length": 1048576, "depth": 1, "zero": false, "data": true, "offset": 327680}] + +Offset Length Mapped to File +0 0x100000 0x50000 TEST_DIR/PID-base + +[{ "start": 0, "length": 1048576, "depth": 2, "zero": false, "data": true, "offset": 327680}, +{ "start": 1048576, "length": 1048576, "depth": 0, "zero": true, "data": false}] + +Offset Length Mapped to File +0 0x100000 0x50000 TEST_DIR/PID-base + +=== Testing qemu-img commit (top -> mid) === +Image committed. + +image: TEST_IMG +file format: IMGFMT +virtual size: 2 MiB (2097152 bytes) +cluster_size: 65536 +backing file: TEST_DIR/PID-base +Format specific information: + compat: 1.1 + lazy refcounts: false + refcount bits: 16 + corrupt: false + +read 1048576/1048576 bytes at offset 0 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +read 1048576/1048576 bytes at offset 1048576 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +=== Testing HMP commit (top -> mid) === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=2097152 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-mid', fmt=qcow2 size=1048576 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=2097152 backing_file=TEST_DIR/PID-mid cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 2097152/2097152 bytes at offset 0 +2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +{"execute": "human-monitor-command", "arguments": {"command-line": "commit drive0"}} +{"return": ""} +image: TEST_IMG +file format: IMGFMT +virtual size: 2 MiB (2097152 bytes) +cluster_size: 65536 +backing file: TEST_DIR/PID-base +Format specific information: + compat: 1.1 + lazy refcounts: false + refcount bits: 16 + corrupt: false + +read 1048576/1048576 bytes at offset 0 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +read 1048576/1048576 bytes at offset 1048576 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +=== Testing QMP active commit (top -> mid) === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=2097152 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-mid', fmt=qcow2 size=1048576 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=2097152 backing_file=TEST_DIR/PID-mid cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 2097152/2097152 bytes at offset 0 +2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +{"execute": "block-commit", "arguments": {"auto-dismiss": false, "base-node": "mid", "device": "top", "job-id": "job0"}} +{"return": {}} +{"execute": "job-complete", "arguments": {"id": "job0"}} +{"return": {}} +{"data": {"device": "job0", "len": 0, "offset": 0, "speed": 0, "type": "commit"}, "event": "BLOCK_JOB_READY", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}} +{"data": {"device": "job0", "len": 0, "offset": 0, "speed": 0, "type": "commit"}, "event": "BLOCK_JOB_COMPLETED", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}} +{"execute": "job-dismiss", "arguments": {"id": "job0"}} +{"return": {}} +image: TEST_IMG +file format: IMGFMT +virtual size: 2 MiB (2097152 bytes) +cluster_size: 65536 +backing file: TEST_DIR/PID-base +Format specific information: + compat: 1.1 + lazy refcounts: false + refcount bits: 16 + corrupt: false + +read 1048576/1048576 bytes at offset 0 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +read 1048576/1048576 bytes at offset 1048576 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +== Resize tests == +=== preallocation=off === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=6442450944 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=1073741824 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 65536/65536 bytes at offset 5368709120 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +Image resized. + +read 65536/65536 bytes at offset 5368709120 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +1 GiB (0x40000000) bytes not allocated at offset 0 bytes (0x0) +7 GiB (0x1c0000000) bytes allocated at offset 1 GiB (0x40000000) + +[{ "start": 0, "length": 1073741824, "depth": 1, "zero": true, "data": false}, +{ "start": 1073741824, "length": 7516192768, "depth": 0, "zero": true, "data": false}] + +=== preallocation=metadata === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=34359738368 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=32212254720 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 65536/65536 bytes at offset 33285996544 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +Image resized. + +read 65536/65536 bytes at offset 33285996544 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +30 GiB (0x780000000) bytes not allocated at offset 0 bytes (0x0) +3 GiB (0xc0000000) bytes allocated at offset 30 GiB (0x780000000) + +[{ "start": 0, "length": 32212254720, "depth": 1, "zero": true, "data": false}, +{ "start": 32212254720, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 327680}, +{ "start": 32749125632, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 537264128}, +{ "start": 33285996544, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 1074200576}, +{ "start": 33822867456, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 1611137024}, +{ "start": 34359738368, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 2148139008}, +{ "start": 34896609280, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 2685075456}] + +=== preallocation=falloc === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=10485760 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=5242880 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 65536/65536 bytes at offset 9437184 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +Image resized. + +read 65536/65536 bytes at offset 9437184 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +5 MiB (0x500000) bytes not allocated at offset 0 bytes (0x0) +10 MiB (0xa00000) bytes allocated at offset 5 MiB (0x500000) + +[{ "start": 0, "length": 5242880, "depth": 1, "zero": true, "data": false}, +{ "start": 5242880, "length": 10485760, "depth": 0, "zero": true, "data": false, "offset": 327680}] + +=== preallocation=full === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=16777216 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=8388608 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 65536/65536 bytes at offset 11534336 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +Image resized. + +read 65536/65536 bytes at offset 11534336 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +8 MiB (0x800000) bytes not allocated at offset 0 bytes (0x0) +4 MiB (0x400000) bytes allocated at offset 8 MiB (0x800000) + +[{ "start": 0, "length": 8388608, "depth": 1, "zero": true, "data": false}, +{ "start": 8388608, "length": 4194304, "depth": 0, "zero": true, "data": false, "offset": 327680}] + +=== preallocation=off === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=393216 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=259072 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 65536/65536 bytes at offset 259072 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +Image resized. + +read 65536/65536 bytes at offset 259072 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +192 KiB (0x30000) bytes not allocated at offset 0 bytes (0x0) +320 KiB (0x50000) bytes allocated at offset 192 KiB (0x30000) + +[{ "start": 0, "length": 196608, "depth": 1, "zero": true, "data": false}, +{ "start": 196608, "length": 65536, "depth": 0, "zero": false, "data": true, "offset": 327680}, +{ "start": 262144, "length": 262144, "depth": 0, "zero": true, "data": false}] + +=== preallocation=off === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=409600 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=262144 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 65536/65536 bytes at offset 344064 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +Image resized. + +read 65536/65536 bytes at offset 344064 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +256 KiB (0x40000) bytes not allocated at offset 0 bytes (0x0) +256 KiB (0x40000) bytes allocated at offset 256 KiB (0x40000) + +[{ "start": 0, "length": 262144, "depth": 1, "zero": true, "data": false}, +{ "start": 262144, "length": 262144, "depth": 0, "zero": true, "data": false}] + +=== preallocation=off === +Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=524288 cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=262144 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16 + +wrote 65536/65536 bytes at offset 446464 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +Image resized. + +read 65536/65536 bytes at offset 446464 +64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +256 KiB (0x40000) bytes not allocated at offset 0 bytes (0x0) +244 KiB (0x3d000) bytes allocated at offset 256 KiB (0x40000) + +[{ "start": 0, "length": 262144, "depth": 1, "zero": true, "data": false}, +{ "start": 262144, "length": 249856, "depth": 0, "zero": true, "data": false}] + diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/group +++ b/tests/qemu-iotests/group @@ -XXX,XX +XXX,XX @@ 270 rw backing quick 272 rw 273 backing quick +274 rw backing 277 rw quick 279 rw backing quick 280 rw migration quick -- 2.25.3 The BDRV_REQ_ZERO_WRITE is currently implemented in a way that first the image is possibly preallocated and then the zero flag is added to all clusters. This means that a copy-on-write operation may be needed when writing to these clusters, despite having used preallocation, negating one of the major benefits of preallocation. Instead, try to forward the BDRV_REQ_ZERO_WRITE to the protocol driver, and if the protocol driver can ensure that the new area reads as zeros, we can skip setting the zero flag in the qcow2 layer. Unfortunately, the same approach doesn't work for metadata preallocation, so we'll still set the zero flag there. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-Id: <20200424142701.67053-1-kwolf@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- block/qcow2.c | 22 +++++++++++++++++++--- tests/qemu-iotests/274.out | 4 ++-- 2 files changed, 21 insertions(+), 5 deletions(-) diff --git a/block/qcow2.c b/block/qcow2.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset, /* Allocate the data area */ new_file_size = allocation_start + nb_new_data_clusters * s->cluster_size; - /* Image file grows, so @exact does not matter */ - ret = bdrv_co_truncate(bs->file, new_file_size, false, prealloc, 0, - errp); + /* + * Image file grows, so @exact does not matter. + * + * If we need to zero out the new area, try first whether the protocol + * driver can already take care of this. + */ + if (flags & BDRV_REQ_ZERO_WRITE) { + ret = bdrv_co_truncate(bs->file, new_file_size, false, prealloc, + BDRV_REQ_ZERO_WRITE, NULL); + if (ret >= 0) { + flags &= ~BDRV_REQ_ZERO_WRITE; + } + } else { + ret = -1; + } + if (ret < 0) { + ret = bdrv_co_truncate(bs->file, new_file_size, false, prealloc, 0, + errp); + } if (ret < 0) { error_prepend(errp, "Failed to resize underlying file: "); qcow2_free_clusters(bs, allocation_start, diff --git a/tests/qemu-iotests/274.out b/tests/qemu-iotests/274.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/274.out +++ b/tests/qemu-iotests/274.out @@ -XXX,XX +XXX,XX @@ read 65536/65536 bytes at offset 9437184 10 MiB (0xa00000) bytes allocated at offset 5 MiB (0x500000) [{ "start": 0, "length": 5242880, "depth": 1, "zero": true, "data": false}, -{ "start": 5242880, "length": 10485760, "depth": 0, "zero": true, "data": false, "offset": 327680}] +{ "start": 5242880, "length": 10485760, "depth": 0, "zero": false, "data": true, "offset": 327680}] === preallocation=full === Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=16777216 cluster_size=65536 lazy_refcounts=off refcount_bits=16 @@ -XXX,XX +XXX,XX @@ read 65536/65536 bytes at offset 11534336 4 MiB (0x400000) bytes allocated at offset 8 MiB (0x800000) [{ "start": 0, "length": 8388608, "depth": 1, "zero": true, "data": false}, -{ "start": 8388608, "length": 4194304, "depth": 0, "zero": true, "data": false, "offset": 327680}] +{ "start": 8388608, "length": 4194304, "depth": 0, "zero": false, "data": true, "offset": 327680}] === preallocation=off === Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=393216 cluster_size=65536 lazy_refcounts=off refcount_bits=16 -- 2.25.3 From: Andrzej Jakowski <andrzej.jakowski@linux.intel.com> This patch introduces support for PMR that has been defined as part of NVMe 1.4 spec. User can now specify a pmrdev option that should point to HostMemoryBackend. pmrdev memory region will subsequently be exposed as PCI BAR 2 in emulated NVMe device. Guest OS can perform mmio read and writes to the PMR region that will stay persistent across system reboot. Signed-off-by: Andrzej Jakowski <andrzej.jakowski@linux.intel.com> Reviewed-by: Klaus Jensen <k.jensen@samsung.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Message-Id: <20200330164656.9348-1-andrzej.jakowski@linux.intel.com> Reviewed-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- hw/block/nvme.h | 2 + include/block/nvme.h | 172 +++++++++++++++++++++++++++++++++++++++++ hw/block/nvme.c | 109 ++++++++++++++++++++++++++ hw/block/Makefile.objs | 2 +- hw/block/trace-events | 4 + 5 files changed, 288 insertions(+), 1 deletion(-) diff --git a/hw/block/nvme.h b/hw/block/nvme.h index XXXXXXX..XXXXXXX 100644 --- a/hw/block/nvme.h +++ b/hw/block/nvme.h @@ -XXX,XX +XXX,XX @@ typedef struct NvmeCtrl { uint64_t timestamp_set_qemu_clock_ms; /* QEMU clock time */ char *serial; + HostMemoryBackend *pmrdev; + NvmeNamespace *namespaces; NvmeSQueue **sq; NvmeCQueue **cq; diff --git a/include/block/nvme.h b/include/block/nvme.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/nvme.h +++ b/include/block/nvme.h @@ -XXX,XX +XXX,XX @@ typedef struct NvmeBar { uint64_t acq; uint32_t cmbloc; uint32_t cmbsz; + uint8_t padding[3520]; /* not used by QEMU */ + uint32_t pmrcap; + uint32_t pmrctl; + uint32_t pmrsts; + uint32_t pmrebs; + uint32_t pmrswtp; + uint32_t pmrmsc; } NvmeBar; enum NvmeCapShift { @@ -XXX,XX +XXX,XX @@ enum NvmeCapShift { CAP_CSS_SHIFT = 37, CAP_MPSMIN_SHIFT = 48, CAP_MPSMAX_SHIFT = 52, + CAP_PMR_SHIFT = 56, }; enum NvmeCapMask { @@ -XXX,XX +XXX,XX @@ enum NvmeCapMask { CAP_CSS_MASK = 0xff, CAP_MPSMIN_MASK = 0xf, CAP_MPSMAX_MASK = 0xf, + CAP_PMR_MASK = 0x1, }; #define NVME_CAP_MQES(cap) (((cap) >> CAP_MQES_SHIFT) & CAP_MQES_MASK) @@ -XXX,XX +XXX,XX @@ enum NvmeCapMask { << CAP_MPSMIN_SHIFT) #define NVME_CAP_SET_MPSMAX(cap, val) (cap |= (uint64_t)(val & CAP_MPSMAX_MASK)\ << CAP_MPSMAX_SHIFT) +#define NVME_CAP_SET_PMRS(cap, val) (cap |= (uint64_t)(val & CAP_PMR_MASK)\ + << CAP_PMR_SHIFT) enum NvmeCcShift { CC_EN_SHIFT = 0, @@ -XXX,XX +XXX,XX @@ enum NvmeCmbszMask { #define NVME_CMBSZ_GETSIZE(cmbsz) \ (NVME_CMBSZ_SZ(cmbsz) * (1 << (12 + 4 * NVME_CMBSZ_SZU(cmbsz)))) +enum NvmePmrcapShift { + PMRCAP_RDS_SHIFT = 3, + PMRCAP_WDS_SHIFT = 4, + PMRCAP_BIR_SHIFT = 5, + PMRCAP_PMRTU_SHIFT = 8, + PMRCAP_PMRWBM_SHIFT = 10, + PMRCAP_PMRTO_SHIFT = 16, + PMRCAP_CMSS_SHIFT = 24, +}; + +enum NvmePmrcapMask { + PMRCAP_RDS_MASK = 0x1, + PMRCAP_WDS_MASK = 0x1, + PMRCAP_BIR_MASK = 0x7, + PMRCAP_PMRTU_MASK = 0x3, + PMRCAP_PMRWBM_MASK = 0xf, + PMRCAP_PMRTO_MASK = 0xff, + PMRCAP_CMSS_MASK = 0x1, +}; + +#define NVME_PMRCAP_RDS(pmrcap) \ + ((pmrcap >> PMRCAP_RDS_SHIFT) & PMRCAP_RDS_MASK) +#define NVME_PMRCAP_WDS(pmrcap) \ + ((pmrcap >> PMRCAP_WDS_SHIFT) & PMRCAP_WDS_MASK) +#define NVME_PMRCAP_BIR(pmrcap) \ + ((pmrcap >> PMRCAP_BIR_SHIFT) & PMRCAP_BIR_MASK) +#define NVME_PMRCAP_PMRTU(pmrcap) \ + ((pmrcap >> PMRCAP_PMRTU_SHIFT) & PMRCAP_PMRTU_MASK) +#define NVME_PMRCAP_PMRWBM(pmrcap) \ + ((pmrcap >> PMRCAP_PMRWBM_SHIFT) & PMRCAP_PMRWBM_MASK) +#define NVME_PMRCAP_PMRTO(pmrcap) \ + ((pmrcap >> PMRCAP_PMRTO_SHIFT) & PMRCAP_PMRTO_MASK) +#define NVME_PMRCAP_CMSS(pmrcap) \ + ((pmrcap >> PMRCAP_CMSS_SHIFT) & PMRCAP_CMSS_MASK) + +#define NVME_PMRCAP_SET_RDS(pmrcap, val) \ + (pmrcap |= (uint64_t)(val & PMRCAP_RDS_MASK) << PMRCAP_RDS_SHIFT) +#define NVME_PMRCAP_SET_WDS(pmrcap, val) \ + (pmrcap |= (uint64_t)(val & PMRCAP_WDS_MASK) << PMRCAP_WDS_SHIFT) +#define NVME_PMRCAP_SET_BIR(pmrcap, val) \ + (pmrcap |= (uint64_t)(val & PMRCAP_BIR_MASK) << PMRCAP_BIR_SHIFT) +#define NVME_PMRCAP_SET_PMRTU(pmrcap, val) \ + (pmrcap |= (uint64_t)(val & PMRCAP_PMRTU_MASK) << PMRCAP_PMRTU_SHIFT) +#define NVME_PMRCAP_SET_PMRWBM(pmrcap, val) \ + (pmrcap |= (uint64_t)(val & PMRCAP_PMRWBM_MASK) << PMRCAP_PMRWBM_SHIFT) +#define NVME_PMRCAP_SET_PMRTO(pmrcap, val) \ + (pmrcap |= (uint64_t)(val & PMRCAP_PMRTO_MASK) << PMRCAP_PMRTO_SHIFT) +#define NVME_PMRCAP_SET_CMSS(pmrcap, val) \ + (pmrcap |= (uint64_t)(val & PMRCAP_CMSS_MASK) << PMRCAP_CMSS_SHIFT) + +enum NvmePmrctlShift { + PMRCTL_EN_SHIFT = 0, +}; + +enum NvmePmrctlMask { + PMRCTL_EN_MASK = 0x1, +}; + +#define NVME_PMRCTL_EN(pmrctl) ((pmrctl >> PMRCTL_EN_SHIFT) & PMRCTL_EN_MASK) + +#define NVME_PMRCTL_SET_EN(pmrctl, val) \ + (pmrctl |= (uint64_t)(val & PMRCTL_EN_MASK) << PMRCTL_EN_SHIFT) + +enum NvmePmrstsShift { + PMRSTS_ERR_SHIFT = 0, + PMRSTS_NRDY_SHIFT = 8, + PMRSTS_HSTS_SHIFT = 9, + PMRSTS_CBAI_SHIFT = 12, +}; + +enum NvmePmrstsMask { + PMRSTS_ERR_MASK = 0xff, + PMRSTS_NRDY_MASK = 0x1, + PMRSTS_HSTS_MASK = 0x7, + PMRSTS_CBAI_MASK = 0x1, +}; + +#define NVME_PMRSTS_ERR(pmrsts) \ + ((pmrsts >> PMRSTS_ERR_SHIFT) & PMRSTS_ERR_MASK) +#define NVME_PMRSTS_NRDY(pmrsts) \ + ((pmrsts >> PMRSTS_NRDY_SHIFT) & PMRSTS_NRDY_MASK) +#define NVME_PMRSTS_HSTS(pmrsts) \ + ((pmrsts >> PMRSTS_HSTS_SHIFT) & PMRSTS_HSTS_MASK) +#define NVME_PMRSTS_CBAI(pmrsts) \ + ((pmrsts >> PMRSTS_CBAI_SHIFT) & PMRSTS_CBAI_MASK) + +#define NVME_PMRSTS_SET_ERR(pmrsts, val) \ + (pmrsts |= (uint64_t)(val & PMRSTS_ERR_MASK) << PMRSTS_ERR_SHIFT) +#define NVME_PMRSTS_SET_NRDY(pmrsts, val) \ + (pmrsts |= (uint64_t)(val & PMRSTS_NRDY_MASK) << PMRSTS_NRDY_SHIFT) +#define NVME_PMRSTS_SET_HSTS(pmrsts, val) \ + (pmrsts |= (uint64_t)(val & PMRSTS_HSTS_MASK) << PMRSTS_HSTS_SHIFT) +#define NVME_PMRSTS_SET_CBAI(pmrsts, val) \ + (pmrsts |= (uint64_t)(val & PMRSTS_CBAI_MASK) << PMRSTS_CBAI_SHIFT) + +enum NvmePmrebsShift { + PMREBS_PMRSZU_SHIFT = 0, + PMREBS_RBB_SHIFT = 4, + PMREBS_PMRWBZ_SHIFT = 8, +}; + +enum NvmePmrebsMask { + PMREBS_PMRSZU_MASK = 0xf, + PMREBS_RBB_MASK = 0x1, + PMREBS_PMRWBZ_MASK = 0xffffff, +}; + +#define NVME_PMREBS_PMRSZU(pmrebs) \ + ((pmrebs >> PMREBS_PMRSZU_SHIFT) & PMREBS_PMRSZU_MASK) +#define NVME_PMREBS_RBB(pmrebs) \ + ((pmrebs >> PMREBS_RBB_SHIFT) & PMREBS_RBB_MASK) +#define NVME_PMREBS_PMRWBZ(pmrebs) \ + ((pmrebs >> PMREBS_PMRWBZ_SHIFT) & PMREBS_PMRWBZ_MASK) + +#define NVME_PMREBS_SET_PMRSZU(pmrebs, val) \ + (pmrebs |= (uint64_t)(val & PMREBS_PMRSZU_MASK) << PMREBS_PMRSZU_SHIFT) +#define NVME_PMREBS_SET_RBB(pmrebs, val) \ + (pmrebs |= (uint64_t)(val & PMREBS_RBB_MASK) << PMREBS_RBB_SHIFT) +#define NVME_PMREBS_SET_PMRWBZ(pmrebs, val) \ + (pmrebs |= (uint64_t)(val & PMREBS_PMRWBZ_MASK) << PMREBS_PMRWBZ_SHIFT) + +enum NvmePmrswtpShift { + PMRSWTP_PMRSWTU_SHIFT = 0, + PMRSWTP_PMRSWTV_SHIFT = 8, +}; + +enum NvmePmrswtpMask { + PMRSWTP_PMRSWTU_MASK = 0xf, + PMRSWTP_PMRSWTV_MASK = 0xffffff, +}; + +#define NVME_PMRSWTP_PMRSWTU(pmrswtp) \ + ((pmrswtp >> PMRSWTP_PMRSWTU_SHIFT) & PMRSWTP_PMRSWTU_MASK) +#define NVME_PMRSWTP_PMRSWTV(pmrswtp) \ + ((pmrswtp >> PMRSWTP_PMRSWTV_SHIFT) & PMRSWTP_PMRSWTV_MASK) + +#define NVME_PMRSWTP_SET_PMRSWTU(pmrswtp, val) \ + (pmrswtp |= (uint64_t)(val & PMRSWTP_PMRSWTU_MASK) << PMRSWTP_PMRSWTU_SHIFT) +#define NVME_PMRSWTP_SET_PMRSWTV(pmrswtp, val) \ + (pmrswtp |= (uint64_t)(val & PMRSWTP_PMRSWTV_MASK) << PMRSWTP_PMRSWTV_SHIFT) + +enum NvmePmrmscShift { + PMRMSC_CMSE_SHIFT = 1, + PMRMSC_CBA_SHIFT = 12, +}; + +enum NvmePmrmscMask { + PMRMSC_CMSE_MASK = 0x1, + PMRMSC_CBA_MASK = 0xfffffffffffff, +}; + +#define NVME_PMRMSC_CMSE(pmrmsc) \ + ((pmrmsc >> PMRMSC_CMSE_SHIFT) & PMRMSC_CMSE_MASK) +#define NVME_PMRMSC_CBA(pmrmsc) \ + ((pmrmsc >> PMRMSC_CBA_SHIFT) & PMRMSC_CBA_MASK) + +#define NVME_PMRMSC_SET_CMSE(pmrmsc, val) \ + (pmrmsc |= (uint64_t)(val & PMRMSC_CMSE_MASK) << PMRMSC_CMSE_SHIFT) +#define NVME_PMRMSC_SET_CBA(pmrmsc, val) \ + (pmrmsc |= (uint64_t)(val & PMRMSC_CBA_MASK) << PMRMSC_CBA_SHIFT) + typedef struct NvmeCmd { uint8_t opcode; uint8_t fuse; 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 @@ * -drive file=<file>,if=none,id=<drive_id> * -device nvme,drive=<drive_id>,serial=<serial>,id=<id[optional]>, \ * cmb_size_mb=<cmb_size_mb[optional]>, \ + * [pmrdev=<mem_backend_file_id>,] \ * num_queues=<N[optional]> * * Note cmb_size_mb denotes size of CMB in MB. CMB is assumed to be at * offset 0 in BAR2 and supports only WDS, RDS and SQS for now. + * + * cmb_size_mb= and pmrdev= options are mutually exclusive due to limitation + * in available BAR's. cmb_size_mb= will take precedence over pmrdev= when + * both provided. + * Enabling pmr emulation can be achieved by pointing to memory-backend-file. + * For example: + * -object memory-backend-file,id=<mem_id>,share=on,mem-path=<file_path>, \ + * size=<size> .... -device nvme,...,pmrdev=<mem_id> */ #include "qemu/osdep.h" @@ -XXX,XX +XXX,XX @@ #include "sysemu/sysemu.h" #include "qapi/error.h" #include "qapi/visitor.h" +#include "sysemu/hostmem.h" #include "sysemu/block-backend.h" +#include "exec/ram_addr.h" #include "qemu/log.h" #include "qemu/module.h" @@ -XXX,XX +XXX,XX @@ static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, uint64_t data, NVME_GUEST_ERR(nvme_ub_mmiowr_cmbsz_readonly, "invalid write to read only CMBSZ, ignored"); return; + case 0xE00: /* PMRCAP */ + NVME_GUEST_ERR(nvme_ub_mmiowr_pmrcap_readonly, + "invalid write to PMRCAP register, ignored"); + return; + case 0xE04: /* TODO PMRCTL */ + break; + case 0xE08: /* PMRSTS */ + NVME_GUEST_ERR(nvme_ub_mmiowr_pmrsts_readonly, + "invalid write to PMRSTS register, ignored"); + return; + case 0xE0C: /* PMREBS */ + NVME_GUEST_ERR(nvme_ub_mmiowr_pmrebs_readonly, + "invalid write to PMREBS register, ignored"); + return; + case 0xE10: /* PMRSWTP */ + NVME_GUEST_ERR(nvme_ub_mmiowr_pmrswtp_readonly, + "invalid write to PMRSWTP register, ignored"); + return; + case 0xE14: /* TODO PMRMSC */ + break; default: NVME_GUEST_ERR(nvme_ub_mmiowr_invalid, "invalid MMIO write," @@ -XXX,XX +XXX,XX @@ static uint64_t nvme_mmio_read(void *opaque, hwaddr addr, unsigned size) } if (addr < sizeof(n->bar)) { + /* + * When PMRWBM bit 1 is set then read from + * from PMRSTS should ensure prior writes + * made it to persistent media + */ + if (addr == 0xE08 && + (NVME_PMRCAP_PMRWBM(n->bar.pmrcap) & 0x02)) { + qemu_ram_writeback(n->pmrdev->mr.ram_block, + 0, n->pmrdev->size); + } memcpy(&val, ptr + addr, size); } else { NVME_GUEST_ERR(nvme_ub_mmiord_invalid_ofs, @@ -XXX,XX +XXX,XX @@ static void nvme_realize(PCIDevice *pci_dev, Error **errp) error_setg(errp, "serial property not set"); return; } + + if (!n->cmb_size_mb && n->pmrdev) { + if (host_memory_backend_is_mapped(n->pmrdev)) { + char *path = object_get_canonical_path_component(OBJECT(n->pmrdev)); + error_setg(errp, "can't use already busy memdev: %s", path); + g_free(path); + return; + } + + if (!is_power_of_2(n->pmrdev->size)) { + error_setg(errp, "pmr backend size needs to be power of 2 in size"); + return; + } + + host_memory_backend_set_mapped(n->pmrdev, true); + } + blkconf_blocksizes(&n->conf); if (!blkconf_apply_backend_options(&n->conf, blk_is_read_only(n->conf.blk), false, errp)) { @@ -XXX,XX +XXX,XX @@ static void nvme_realize(PCIDevice *pci_dev, Error **errp) PCI_BASE_ADDRESS_SPACE_MEMORY | PCI_BASE_ADDRESS_MEM_TYPE_64 | PCI_BASE_ADDRESS_MEM_PREFETCH, &n->ctrl_mem); + } else if (n->pmrdev) { + /* Controller Capabilities register */ + NVME_CAP_SET_PMRS(n->bar.cap, 1); + + /* PMR Capabities register */ + n->bar.pmrcap = 0; + NVME_PMRCAP_SET_RDS(n->bar.pmrcap, 0); + NVME_PMRCAP_SET_WDS(n->bar.pmrcap, 0); + NVME_PMRCAP_SET_BIR(n->bar.pmrcap, 2); + NVME_PMRCAP_SET_PMRTU(n->bar.pmrcap, 0); + /* Turn on bit 1 support */ + NVME_PMRCAP_SET_PMRWBM(n->bar.pmrcap, 0x02); + NVME_PMRCAP_SET_PMRTO(n->bar.pmrcap, 0); + NVME_PMRCAP_SET_CMSS(n->bar.pmrcap, 0); + + /* PMR Control register */ + n->bar.pmrctl = 0; + NVME_PMRCTL_SET_EN(n->bar.pmrctl, 0); + + /* PMR Status register */ + n->bar.pmrsts = 0; + NVME_PMRSTS_SET_ERR(n->bar.pmrsts, 0); + NVME_PMRSTS_SET_NRDY(n->bar.pmrsts, 0); + NVME_PMRSTS_SET_HSTS(n->bar.pmrsts, 0); + NVME_PMRSTS_SET_CBAI(n->bar.pmrsts, 0); + + /* PMR Elasticity Buffer Size register */ + n->bar.pmrebs = 0; + NVME_PMREBS_SET_PMRSZU(n->bar.pmrebs, 0); + NVME_PMREBS_SET_RBB(n->bar.pmrebs, 0); + NVME_PMREBS_SET_PMRWBZ(n->bar.pmrebs, 0); + + /* PMR Sustained Write Throughput register */ + n->bar.pmrswtp = 0; + NVME_PMRSWTP_SET_PMRSWTU(n->bar.pmrswtp, 0); + NVME_PMRSWTP_SET_PMRSWTV(n->bar.pmrswtp, 0); + + /* PMR Memory Space Control register */ + n->bar.pmrmsc = 0; + NVME_PMRMSC_SET_CMSE(n->bar.pmrmsc, 0); + NVME_PMRMSC_SET_CBA(n->bar.pmrmsc, 0); + + pci_register_bar(pci_dev, NVME_PMRCAP_BIR(n->bar.pmrcap), + PCI_BASE_ADDRESS_SPACE_MEMORY | PCI_BASE_ADDRESS_MEM_TYPE_64 | + PCI_BASE_ADDRESS_MEM_PREFETCH, &n->pmrdev->mr); } for (i = 0; i < n->num_namespaces; i++) { @@ -XXX,XX +XXX,XX @@ static void nvme_exit(PCIDevice *pci_dev) if (n->cmb_size_mb) { g_free(n->cmbuf); } + + if (n->pmrdev) { + host_memory_backend_set_mapped(n->pmrdev, false); + } msix_uninit_exclusive_bar(pci_dev); } static Property nvme_props[] = { DEFINE_BLOCK_PROPERTIES(NvmeCtrl, conf), + DEFINE_PROP_LINK("pmrdev", NvmeCtrl, pmrdev, TYPE_MEMORY_BACKEND, + HostMemoryBackend *), DEFINE_PROP_STRING("serial", NvmeCtrl, serial), DEFINE_PROP_UINT32("cmb_size_mb", NvmeCtrl, cmb_size_mb, 0), DEFINE_PROP_UINT32("num_queues", NvmeCtrl, num_queues, 64), diff --git a/hw/block/Makefile.objs b/hw/block/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/block/Makefile.objs +++ b/hw/block/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_PFLASH_CFI02) += pflash_cfi02.o common-obj-$(CONFIG_XEN) += xen-block.o common-obj-$(CONFIG_ECC) += ecc.o common-obj-$(CONFIG_ONENAND) += onenand.o -common-obj-$(CONFIG_NVME_PCI) += nvme.o common-obj-$(CONFIG_SWIM) += swim.o common-obj-$(CONFIG_SH4) += tc58128.o obj-$(CONFIG_VIRTIO_BLK) += virtio-blk.o obj-$(CONFIG_VHOST_USER_BLK) += vhost-user-blk.o +obj-$(CONFIG_NVME_PCI) += nvme.o obj-y += dataplane/ diff --git a/hw/block/trace-events b/hw/block/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/block/trace-events +++ b/hw/block/trace-events @@ -XXX,XX +XXX,XX @@ nvme_ub_mmiowr_ssreset_w1c_unsupported(void) "attempted to W1C CSTS.NSSRO but CA nvme_ub_mmiowr_ssreset_unsupported(void) "attempted NVM subsystem reset but CAP.NSSRS is zero (not supported)" nvme_ub_mmiowr_cmbloc_reserved(void) "invalid write to reserved CMBLOC when CMBSZ is zero, ignored" nvme_ub_mmiowr_cmbsz_readonly(void) "invalid write to read only CMBSZ, ignored" +nvme_ub_mmiowr_pmrcap_readonly(void) "invalid write to read only PMRCAP, ignored" +nvme_ub_mmiowr_pmrsts_readonly(void) "invalid write to read only PMRSTS, ignored" +nvme_ub_mmiowr_pmrebs_readonly(void) "invalid write to read only PMREBS, ignored" +nvme_ub_mmiowr_pmrswtp_readonly(void) "invalid write to read only PMRSWTP, ignored" nvme_ub_mmiowr_invalid(uint64_t offset, uint64_t data) "invalid MMIO write, offset=0x%"PRIx64", data=0x%"PRIx64"" nvme_ub_mmiord_misaligned32(uint64_t offset) "MMIO read not 32-bit aligned, offset=0x%"PRIx64"" nvme_ub_mmiord_toosmall(uint64_t offset) "MMIO read smaller than 32-bits, offset=0x%"PRIx64"" -- 2.25.3 The QMP handler qmp_object_add() and the implementation of --object in qemu-storage-daemon can share most of the code. Currently, qemu-storage-daemon calls qmp_object_add(), but this is not correct because different visitors need to be used. As a first step towards a fix, make qmp_object_add() a wrapper around a new function user_creatable_add_dict() that can get an additional parameter. The handling of "props" is only required for compatibility and not required for the qemu-storage-daemon command line, so it stays in qmp_object_add(). Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/qom/object_interfaces.h | 12 ++++++++++++ qom/object_interfaces.c | 27 +++++++++++++++++++++++++++ qom/qom-qmp-cmds.c | 24 +----------------------- 3 files changed, 40 insertions(+), 23 deletions(-) diff --git a/include/qom/object_interfaces.h b/include/qom/object_interfaces.h index XXXXXXX..XXXXXXX 100644 --- a/include/qom/object_interfaces.h +++ b/include/qom/object_interfaces.h @@ -XXX,XX +XXX,XX @@ Object *user_creatable_add_type(const char *type, const char *id, const QDict *qdict, Visitor *v, Error **errp); +/** + * user_creatable_add_dict: + * @qdict: the object definition + * @errp: if an error occurs, a pointer to an area to store the error + * + * Create an instance of the user creatable object that is defined by + * @qdict. The object type is taken from the QDict key 'qom-type', its + * ID from the key 'id'. The remaining entries in @qdict are used to + * initialize the object properties. + */ +void user_creatable_add_dict(QDict *qdict, Error **errp); + /** * user_creatable_add_opts: * @opts: the object definition diff --git a/qom/object_interfaces.c b/qom/object_interfaces.c index XXXXXXX..XXXXXXX 100644 --- a/qom/object_interfaces.c +++ b/qom/object_interfaces.c @@ -XXX,XX +XXX,XX @@ #include "qapi/qmp/qerror.h" #include "qapi/qmp/qjson.h" #include "qapi/qmp/qstring.h" +#include "qapi/qobject-input-visitor.h" #include "qom/object_interfaces.h" #include "qemu/help_option.h" #include "qemu/module.h" @@ -XXX,XX +XXX,XX @@ out: return obj; } +void user_creatable_add_dict(QDict *qdict, Error **errp) +{ + Visitor *v; + Object *obj; + g_autofree char *type = NULL; + g_autofree char *id = NULL; + + type = g_strdup(qdict_get_try_str(qdict, "qom-type")); + if (!type) { + error_setg(errp, QERR_MISSING_PARAMETER, "qom-type"); + return; + } + qdict_del(qdict, "qom-type"); + + id = g_strdup(qdict_get_try_str(qdict, "id")); + if (!id) { + error_setg(errp, QERR_MISSING_PARAMETER, "id"); + return; + } + qdict_del(qdict, "id"); + + v = qobject_input_visitor_new(QOBJECT(qdict)); + obj = user_creatable_add_type(type, id, qdict, v, errp); + visit_free(v); + object_unref(obj); +} Object *user_creatable_add_opts(QemuOpts *opts, Error **errp) { diff --git a/qom/qom-qmp-cmds.c b/qom/qom-qmp-cmds.c index XXXXXXX..XXXXXXX 100644 --- a/qom/qom-qmp-cmds.c +++ b/qom/qom-qmp-cmds.c @@ -XXX,XX +XXX,XX @@ #include "qapi/qapi-commands-qom.h" #include "qapi/qmp/qdict.h" #include "qapi/qmp/qerror.h" -#include "qapi/qobject-input-visitor.h" #include "qemu/cutils.h" #include "qom/object_interfaces.h" #include "qom/qom-qobject.h" @@ -XXX,XX +XXX,XX @@ void qmp_object_add(QDict *qdict, QObject **ret_data, Error **errp) { QObject *props; QDict *pdict; - Visitor *v; - Object *obj; - g_autofree char *type = NULL; - g_autofree char *id = NULL; - - type = g_strdup(qdict_get_try_str(qdict, "qom-type")); - if (!type) { - error_setg(errp, QERR_MISSING_PARAMETER, "qom-type"); - return; - } - qdict_del(qdict, "qom-type"); - - id = g_strdup(qdict_get_try_str(qdict, "id")); - if (!id) { - error_setg(errp, QERR_MISSING_PARAMETER, "id"); - return; - } - qdict_del(qdict, "id"); props = qdict_get(qdict, "props"); if (props) { @@ -XXX,XX +XXX,XX @@ void qmp_object_add(QDict *qdict, QObject **ret_data, Error **errp) qobject_unref(pdict); } - v = qobject_input_visitor_new(QOBJECT(qdict)); - obj = user_creatable_add_type(type, id, qdict, v, errp); - visit_free(v); - object_unref(obj); + user_creatable_add_dict(qdict, errp); } void qmp_object_del(const char *id, Error **errp) -- 2.25.3 After processing the option string with the keyval parser, we get a QDict that contains only strings. This QDict must be fed to a keyval visitor which converts the strings into the right data types. qmp_object_add(), however, uses the normal QObject input visitor, which expects a QDict where all properties already have the QType that matches the data type required by the QOM object type. Change the --object implementation in qemu-storage-daemon so that it doesn't call qmp_object_add(), but calls user_creatable_add_dict() directly instead and pass it a new keyval boolean that decides which visitor must be used. Reported-by: Coiby Xu <coiby.xu@gmail.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- include/qom/object_interfaces.h | 6 +++++- qemu-storage-daemon.c | 4 +--- qom/object_interfaces.c | 8 ++++++-- qom/qom-qmp-cmds.c | 2 +- 4 files changed, 13 insertions(+), 7 deletions(-) diff --git a/include/qom/object_interfaces.h b/include/qom/object_interfaces.h index XXXXXXX..XXXXXXX 100644 --- a/include/qom/object_interfaces.h +++ b/include/qom/object_interfaces.h @@ -XXX,XX +XXX,XX @@ Object *user_creatable_add_type(const char *type, const char *id, /** * user_creatable_add_dict: * @qdict: the object definition + * @keyval: if true, use a keyval visitor for processing @qdict (i.e. + * assume that all @qdict values are strings); otherwise, use + * the normal QObject visitor (i.e. assume all @qdict values + * have the QType expected by the QOM object type) * @errp: if an error occurs, a pointer to an area to store the error * * Create an instance of the user creatable object that is defined by @@ -XXX,XX +XXX,XX @@ Object *user_creatable_add_type(const char *type, const char *id, * ID from the key 'id'. The remaining entries in @qdict are used to * initialize the object properties. */ -void user_creatable_add_dict(QDict *qdict, Error **errp); +void user_creatable_add_dict(QDict *qdict, bool keyval, Error **errp); /** * user_creatable_add_opts: diff --git a/qemu-storage-daemon.c b/qemu-storage-daemon.c index XXXXXXX..XXXXXXX 100644 --- a/qemu-storage-daemon.c +++ b/qemu-storage-daemon.c @@ -XXX,XX +XXX,XX @@ static void process_options(int argc, char *argv[]) QemuOpts *opts; const char *type; QDict *args; - QObject *ret_data = NULL; /* FIXME The keyval parser rejects 'help' arguments, so we must * unconditionall try QemuOpts first. */ @@ -XXX,XX +XXX,XX @@ static void process_options(int argc, char *argv[]) qemu_opts_del(opts); args = keyval_parse(optarg, "qom-type", &error_fatal); - qmp_object_add(args, &ret_data, &error_fatal); + user_creatable_add_dict(args, true, &error_fatal); qobject_unref(args); - qobject_unref(ret_data); break; } default: diff --git a/qom/object_interfaces.c b/qom/object_interfaces.c index XXXXXXX..XXXXXXX 100644 --- a/qom/object_interfaces.c +++ b/qom/object_interfaces.c @@ -XXX,XX +XXX,XX @@ out: return obj; } -void user_creatable_add_dict(QDict *qdict, Error **errp) +void user_creatable_add_dict(QDict *qdict, bool keyval, Error **errp) { Visitor *v; Object *obj; @@ -XXX,XX +XXX,XX @@ void user_creatable_add_dict(QDict *qdict, Error **errp) } qdict_del(qdict, "id"); - v = qobject_input_visitor_new(QOBJECT(qdict)); + if (keyval) { + v = qobject_input_visitor_new_keyval(QOBJECT(qdict)); + } else { + v = qobject_input_visitor_new(QOBJECT(qdict)); + } obj = user_creatable_add_type(type, id, qdict, v, errp); visit_free(v); object_unref(obj); diff --git a/qom/qom-qmp-cmds.c b/qom/qom-qmp-cmds.c index XXXXXXX..XXXXXXX 100644 --- a/qom/qom-qmp-cmds.c +++ b/qom/qom-qmp-cmds.c @@ -XXX,XX +XXX,XX @@ void qmp_object_add(QDict *qdict, QObject **ret_data, Error **errp) qobject_unref(pdict); } - user_creatable_add_dict(qdict, errp); + user_creatable_add_dict(qdict, false, errp); } void qmp_object_del(const char *id, Error **errp) -- 2.25.3