The following changes since commit 23919ddfd56135cad3cb468a8f54d5a595f024f4: Merge remote-tracking branch 'remotes/aperard/tags/pull-xen-20190827' into staging (2019-08-27 15:52:36 +0100) are available in the Git repository at: https://github.com/XanClic/qemu.git tags/pull-block-2019-08-27 for you to fetch changes up to bb043c056cffcc2f3ce88bfdaf2e76e455c09e2c: iotests: Unify cache mode quoting (2019-08-27 19:48:44 +0200) ---------------------------------------------------------------- Block patches: - qemu-io now accepts a file to read a write pattern from - Ensure that raw files have their first block allocated so we can probe the O_DIRECT alignment if necessary - Various fixes ---------------------------------------------------------------- Denis Plotnikov (1): qemu-io: add pattern file for write command Max Reitz (7): iotests: Fix _filter_img_create() vmdk: Use bdrv_dirname() for relative extent paths iotests: Keep testing broken relative extent paths vmdk: Reject invalid compressed writes iotests: Disable broken streamOptimized tests iotests: Disable 110 for vmdk.twoGbMaxExtentSparse iotests: Disable 126 for flat vmdk subformats Nir Soffer (3): block: posix: Always allocate the first block iotests: Test allocate_first_block() with O_DIRECT iotests: Unify cache mode quoting Stefan Hajnoczi (1): file-posix: fix request_alignment typo Thomas Huth (2): iotests: Check for enabled drivers before testing them tests/check-block: Skip iotests when sanitizers are enabled Vladimir Sementsov-Ogievskiy (1): block: fix permission update in bdrv_replace_node block.c | 5 +- block/file-posix.c | 53 +++++++++- block/vmdk.c | 64 ++++++++---- qemu-io-cmds.c | 99 +++++++++++++++++-- tests/check-block.sh | 5 + tests/qemu-iotests/002 | 1 + tests/qemu-iotests/003 | 1 + tests/qemu-iotests/005 | 3 +- tests/qemu-iotests/009 | 1 + tests/qemu-iotests/010 | 1 + tests/qemu-iotests/011 | 1 + tests/qemu-iotests/017 | 3 +- tests/qemu-iotests/018 | 3 +- tests/qemu-iotests/019 | 3 +- tests/qemu-iotests/020 | 3 +- tests/qemu-iotests/026 | 4 +- tests/qemu-iotests/027 | 1 + tests/qemu-iotests/032 | 1 + tests/qemu-iotests/033 | 1 + tests/qemu-iotests/034 | 3 +- tests/qemu-iotests/037 | 3 +- tests/qemu-iotests/039 | 4 +- tests/qemu-iotests/052 | 2 +- tests/qemu-iotests/059 | 34 ++++++- tests/qemu-iotests/059.out | 26 +++-- tests/qemu-iotests/063 | 3 +- tests/qemu-iotests/071 | 1 + tests/qemu-iotests/072 | 1 + tests/qemu-iotests/081 | 4 +- tests/qemu-iotests/091 | 4 +- tests/qemu-iotests/099 | 1 + tests/qemu-iotests/105 | 3 +- tests/qemu-iotests/110 | 3 +- tests/qemu-iotests/120 | 1 + tests/qemu-iotests/126 | 2 + tests/qemu-iotests/{150.out => 150.out.qcow2} | 0 tests/qemu-iotests/150.out.raw | 12 +++ tests/qemu-iotests/162 | 4 +- tests/qemu-iotests/175 | 47 +++++++-- tests/qemu-iotests/175.out | 16 ++- tests/qemu-iotests/178.out.qcow2 | 4 +- tests/qemu-iotests/184 | 1 + tests/qemu-iotests/186 | 1 + tests/qemu-iotests/197 | 1 + tests/qemu-iotests/215 | 1 + tests/qemu-iotests/221.out | 12 ++- tests/qemu-iotests/251 | 1 + tests/qemu-iotests/253.out | 12 ++- tests/qemu-iotests/common.filter | 4 +- tests/qemu-iotests/common.rc | 14 +++ 50 files changed, 391 insertions(+), 87 deletions(-) rename tests/qemu-iotests/{150.out => 150.out.qcow2} (100%) create mode 100644 tests/qemu-iotests/150.out.raw -- 2.21.0 From: Denis Plotnikov <dplotnikov@virtuozzo.com> The patch allows to provide a pattern file for write command. There was no similar ability before. Signed-off-by: Denis Plotnikov <dplotnikov@virtuozzo.com> Message-id: 20190820164616.4072-1-dplotnikov@virtuozzo.com Reviewed-by: Eric Blake <eblake@redhat.com> [mreitz: Keep optstring in alphabetical order] Signed-off-by: Max Reitz <mreitz@redhat.com> --- qemu-io-cmds.c | 99 +++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 93 insertions(+), 6 deletions(-) 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 void qemu_io_free(void *p) qemu_vfree(p); } +/* + * qemu_io_alloc_from_file() + * + * Allocates the buffer and populates it with the content of the given file + * up to @len bytes. If the file length is less than @len, then the buffer + * is populated with the file content cyclically. + * + * @blk - the block backend where the buffer content is going to be written to + * @len - the buffer length + * @file_name - the file to read the content from + * + * Returns: the buffer pointer on success + * NULL on error + */ +static void *qemu_io_alloc_from_file(BlockBackend *blk, size_t len, + const char *file_name) +{ + char *buf, *buf_origin; + FILE *f = fopen(file_name, "r"); + int pattern_len; + + if (!f) { + perror(file_name); + return NULL; + } + + if (qemuio_misalign) { + len += MISALIGN_OFFSET; + } + + buf_origin = buf = blk_blockalign(blk, len); + + if (qemuio_misalign) { + buf_origin += MISALIGN_OFFSET; + buf += MISALIGN_OFFSET; + len -= MISALIGN_OFFSET; + } + + pattern_len = fread(buf_origin, 1, len, f); + + if (ferror(f)) { + perror(file_name); + goto error; + } + + if (pattern_len == 0) { + fprintf(stderr, "%s: file is empty\n", file_name); + goto error; + } + + fclose(f); + + if (len > pattern_len) { + len -= pattern_len; + buf += pattern_len; + + while (len > 0) { + size_t len_to_copy = MIN(pattern_len, len); + + memcpy(buf, buf_origin, len_to_copy); + + len -= len_to_copy; + buf += len_to_copy; + } + } + + return buf_origin; + +error: + qemu_io_free(buf_origin); + return NULL; +} + static void dump_buffer(const void *buffer, int64_t offset, int64_t len) { uint64_t i; @@ -XXX,XX +XXX,XX @@ static void write_help(void) " -n, -- with -z, don't allow slow fallback\n" " -p, -- ignored for backwards compatibility\n" " -P, -- use different pattern to fill file\n" +" -s, -- use a pattern file to fill the write buffer\n" " -C, -- report statistics in a machine parsable format\n" " -q, -- quiet mode, do not show I/O statistics\n" " -u, -- with -z, allow unmapping\n" @@ -XXX,XX +XXX,XX @@ static const cmdinfo_t write_cmd = { .perm = BLK_PERM_WRITE, .argmin = 2, .argmax = -1, - .args = "[-bcCfnquz] [-P pattern] off len", + .args = "[-bcCfnquz] [-P pattern | -s source_file] off len", .oneline = "writes a number of bytes at a specified offset", .help = write_help, }; @@ -XXX,XX +XXX,XX @@ static int write_f(BlockBackend *blk, int argc, char **argv) { struct timespec t1, t2; bool Cflag = false, qflag = false, bflag = false; - bool Pflag = false, zflag = false, cflag = false; + bool Pflag = false, zflag = false, cflag = false, sflag = false; int flags = 0; int c, cnt, ret; char *buf = NULL; @@ -XXX,XX +XXX,XX @@ static int write_f(BlockBackend *blk, int argc, char **argv) /* Some compilers get confused and warn if this is not initialized. */ int64_t total = 0; int pattern = 0xcd; + const char *file_name = NULL; - while ((c = getopt(argc, argv, "bcCfnpP:quz")) != -1) { + while ((c = getopt(argc, argv, "bcCfnpP:qs:uz")) != -1) { switch (c) { case 'b': bflag = true; @@ -XXX,XX +XXX,XX @@ static int write_f(BlockBackend *blk, int argc, char **argv) case 'q': qflag = true; break; + case 's': + sflag = true; + file_name = optarg; + break; case 'u': flags |= BDRV_REQ_MAY_UNMAP; break; @@ -XXX,XX +XXX,XX @@ static int write_f(BlockBackend *blk, int argc, char **argv) return -EINVAL; } - if (zflag && Pflag) { - printf("-z and -P cannot be specified at the same time\n"); + if (zflag + Pflag + sflag > 1) { + printf("Only one of -z, -P, and -s " + "can be specified at the same time\n"); return -EINVAL; } @@ -XXX,XX +XXX,XX @@ static int write_f(BlockBackend *blk, int argc, char **argv) } if (!zflag) { - buf = qemu_io_alloc(blk, count, pattern); + if (sflag) { + buf = qemu_io_alloc_from_file(blk, count, file_name); + if (!buf) { + return -EINVAL; + } + } else { + buf = qemu_io_alloc(blk, count, pattern); + } } clock_gettime(CLOCK_MONOTONIC, &t1); -- 2.21.0 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> It's wrong to OR shared permissions. It may lead to crash on further permission updates. Also, no needs to consider previously calculated permissions, as at this point we already bind all new parents and bdrv_get_cumulative_perm result is enough. So fix the bug by just set permissions by bdrv_get_cumulative_perm result. Bug was introduced in long ago 234ac1a9025, in 2.9. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-id: 20190824100740.61635-1-vsementsov@virtuozzo.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- block.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/block.c b/block.c index XXXXXXX..XXXXXXX 100644 --- a/block.c +++ b/block.c @@ -XXX,XX +XXX,XX @@ void bdrv_replace_node(BlockDriverState *from, BlockDriverState *to, { BdrvChild *c, *next; GSList *list = NULL, *p; - uint64_t old_perm, old_shared; uint64_t perm = 0, shared = BLK_PERM_ALL; int ret; @@ -XXX,XX +XXX,XX @@ void bdrv_replace_node(BlockDriverState *from, BlockDriverState *to, bdrv_unref(from); } - bdrv_get_cumulative_perm(to, &old_perm, &old_shared); - bdrv_set_perm(to, old_perm | perm, old_shared | shared); + bdrv_get_cumulative_perm(to, &perm, &shared); + bdrv_set_perm(to, perm, shared); out: g_slist_free(list); -- 2.21.0 From: Nir Soffer <nirsof@gmail.com> When creating an image with preallocation "off" or "falloc", the first block of the image is typically not allocated. When using Gluster storage backed by XFS filesystem, reading this block using direct I/O succeeds regardless of request length, fooling alignment detection. In this case we fallback to a safe value (4096) instead of the optimal value (512), which may lead to unneeded data copying when aligning requests. Allocating the first block avoids the fallback. Since we allocate the first block even with preallocation=off, we no longer create images with zero disk size: $ ./qemu-img create -f raw test.raw 1g Formatting 'test.raw', fmt=raw size=1073741824 $ ls -lhs test.raw 4.0K -rw-r--r--. 1 nsoffer nsoffer 1.0G Aug 16 23:48 test.raw And converting the image requires additional cluster: $ ./qemu-img measure -f raw -O qcow2 test.raw required size: 458752 fully allocated size: 1074135040 When using format like vmdk with multiple files per image, we allocate one block per file: $ ./qemu-img create -f vmdk -o subformat=twoGbMaxExtentFlat test.vmdk 4g Formatting 'test.vmdk', fmt=vmdk size=4294967296 compat6=off hwversion=undefined subformat=twoGbMaxExtentFlat $ ls -lhs test*.vmdk 4.0K -rw-r--r--. 1 nsoffer nsoffer 2.0G Aug 27 03:23 test-f001.vmdk 4.0K -rw-r--r--. 1 nsoffer nsoffer 2.0G Aug 27 03:23 test-f002.vmdk 4.0K -rw-r--r--. 1 nsoffer nsoffer 353 Aug 27 03:23 test.vmdk I did quick performance test for copying disks with qemu-img convert to new raw target image to Gluster storage with sector size of 512 bytes: for i in $(seq 10); do rm -f dst.raw sleep 10 time ./qemu-img convert -f raw -O raw -t none -T none src.raw dst.raw done Here is a table comparing the total time spent: Type Before(s) After(s) Diff(%) --------------------------------------- real 530.028 469.123 -11.4 user 17.204 10.768 -37.4 sys 17.881 7.011 -60.7 We can see very clear improvement in CPU usage. Signed-off-by: Nir Soffer <nsoffer@redhat.com> Message-id: 20190827010528.8818-2-nsoffer@redhat.com Reviewed-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/file-posix.c | 51 +++++++++++++++++++ tests/qemu-iotests/059.out | 2 +- tests/qemu-iotests/{150.out => 150.out.qcow2} | 0 tests/qemu-iotests/150.out.raw | 12 +++++ tests/qemu-iotests/175 | 19 ++++--- tests/qemu-iotests/175.out | 8 +-- tests/qemu-iotests/178.out.qcow2 | 4 +- tests/qemu-iotests/221.out | 12 +++-- tests/qemu-iotests/253.out | 12 +++-- 9 files changed, 99 insertions(+), 21 deletions(-) rename tests/qemu-iotests/{150.out => 150.out.qcow2} (100%) create mode 100644 tests/qemu-iotests/150.out.raw 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 handle_aiocb_discard(void *opaque) return ret; } +/* + * Help alignment probing by allocating the first block. + * + * When reading with direct I/O from unallocated area on Gluster backed by XFS, + * reading succeeds regardless of request length. In this case we fallback to + * safe alignment which is not optimal. Allocating the first block avoids this + * fallback. + * + * fd may be opened with O_DIRECT, but we don't know the buffer alignment or + * request alignment, so we use safe values. + * + * Returns: 0 on success, -errno on failure. Since this is an optimization, + * caller may ignore failures. + */ +static int allocate_first_block(int fd, size_t max_size) +{ + size_t write_size = (max_size < MAX_BLOCKSIZE) + ? BDRV_SECTOR_SIZE + : MAX_BLOCKSIZE; + size_t max_align = MAX(MAX_BLOCKSIZE, getpagesize()); + void *buf; + ssize_t n; + int ret; + + buf = qemu_memalign(max_align, write_size); + memset(buf, 0, write_size); + + do { + n = pwrite(fd, buf, write_size, 0); + } while (n == -1 && errno == EINTR); + + ret = (n == -1) ? -errno : 0; + + qemu_vfree(buf); + return ret; +} + static int handle_aiocb_truncate(void *opaque) { RawPosixAIOData *aiocb = opaque; @@ -XXX,XX +XXX,XX @@ static int handle_aiocb_truncate(void *opaque) /* posix_fallocate() doesn't set errno. */ error_setg_errno(errp, -result, "Could not preallocate new data"); + } else if (current_length == 0) { + /* + * posix_fallocate() uses fallocate() if the filesystem + * supports it, or fallback to manually writing zeroes. If + * fallocate() was used, unaligned reads from the fallocated + * area in raw_probe_alignment() will succeed, hence we need to + * allocate the first block. + * + * Optimize future alignment probing; ignore failures. + */ + allocate_first_block(fd, offset); } } else { result = 0; @@ -XXX,XX +XXX,XX @@ static int handle_aiocb_truncate(void *opaque) if (ftruncate(fd, offset) != 0) { result = -errno; error_setg_errno(errp, -result, "Could not resize file"); + } else if (current_length == 0 && offset > current_length) { + /* Optimize future alignment probing; ignore failures. */ + allocate_first_block(fd, offset); } return result; default: diff --git a/tests/qemu-iotests/059.out b/tests/qemu-iotests/059.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/059.out +++ b/tests/qemu-iotests/059.out @@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1073741824000 subformat=twoGbMax image: TEST_DIR/t.vmdk file format: vmdk virtual size: 0.977 TiB (1073741824000 bytes) -disk size: 16 KiB +disk size: 1.97 MiB Format specific information: cid: XXXXXXXX parent cid: XXXXXXXX diff --git a/tests/qemu-iotests/150.out b/tests/qemu-iotests/150.out.qcow2 similarity index 100% rename from tests/qemu-iotests/150.out rename to tests/qemu-iotests/150.out.qcow2 diff --git a/tests/qemu-iotests/150.out.raw b/tests/qemu-iotests/150.out.raw new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/150.out.raw @@ -XXX,XX +XXX,XX @@ +QA output created by 150 + +=== Mapping sparse conversion === + +Offset Length File +0 0x1000 TEST_DIR/t.IMGFMT + +=== Mapping non-sparse conversion === + +Offset Length File +0 0x100000 TEST_DIR/t.IMGFMT +*** done diff --git a/tests/qemu-iotests/175 b/tests/qemu-iotests/175 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/175 +++ b/tests/qemu-iotests/175 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 # the file size. This function hides the resulting difference in the # stat -c '%b' output. # Parameter 1: Number of blocks an empty file occupies -# Parameter 2: Image size in bytes +# Parameter 2: Minimal number of blocks in an image +# Parameter 3: Image size in bytes _filter_blocks() { extra_blocks=$1 - img_size=$2 + min_blocks=$2 + img_size=$3 - sed -e "s/blocks=$extra_blocks\\(\$\\|[^0-9]\\)/nothing allocated/" \ - -e "s/blocks=$((extra_blocks + img_size / 512))\\(\$\\|[^0-9]\\)/everything allocated/" + sed -e "s/blocks=$min_blocks\\(\$\\|[^0-9]\\)/min allocation/" \ + -e "s/blocks=$((extra_blocks + img_size / 512))\\(\$\\|[^0-9]\\)/max allocation/" } # get standard environment, filters and checks @@ -XXX,XX +XXX,XX @@ size=$((1 * 1024 * 1024)) touch "$TEST_DIR/empty" extra_blocks=$(stat -c '%b' "$TEST_DIR/empty") +# We always write the first byte; check how many blocks this filesystem +# allocates to match empty image alloation. +printf "\0" > "$TEST_DIR/empty" +min_blocks=$(stat -c '%b' "$TEST_DIR/empty") + echo echo "== creating image with default preallocation ==" _make_test_img $size | _filter_imgfmt -stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $size +stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $size for mode in off full falloc; do echo echo "== creating image with preallocation $mode ==" IMGOPTS=preallocation=$mode _make_test_img $size | _filter_imgfmt - stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $size + stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $size done # success, all done diff --git a/tests/qemu-iotests/175.out b/tests/qemu-iotests/175.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/175.out +++ b/tests/qemu-iotests/175.out @@ -XXX,XX +XXX,XX @@ QA output created by 175 == creating image with default preallocation == Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576 -size=1048576, nothing allocated +size=1048576, min allocation == creating image with preallocation off == Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576 preallocation=off -size=1048576, nothing allocated +size=1048576, min allocation == creating image with preallocation full == Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576 preallocation=full -size=1048576, everything allocated +size=1048576, max allocation == creating image with preallocation falloc == Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576 preallocation=falloc -size=1048576, everything allocated +size=1048576, max allocation *** done diff --git a/tests/qemu-iotests/178.out.qcow2 b/tests/qemu-iotests/178.out.qcow2 index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/178.out.qcow2 +++ b/tests/qemu-iotests/178.out.qcow2 @@ -XXX,XX +XXX,XX @@ converted image file size in bytes: 196608 == raw input image with data (human) == Formatting 'TEST_DIR/t.qcow2', fmt=IMGFMT size=1073741824 -required size: 393216 +required size: 458752 fully allocated size: 1074135040 wrote 512/512 bytes at offset 512 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) @@ -XXX,XX +XXX,XX @@ converted image file size in bytes: 196608 Formatting 'TEST_DIR/t.qcow2', fmt=IMGFMT size=1073741824 { - "required": 393216, + "required": 458752, "fully-allocated": 1074135040 } wrote 512/512 bytes at offset 512 diff --git a/tests/qemu-iotests/221.out b/tests/qemu-iotests/221.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/221.out +++ b/tests/qemu-iotests/221.out @@ -XXX,XX +XXX,XX @@ QA output created by 221 === Check mapping of unaligned raw image === Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=65537 -[{ "start": 0, "length": 66048, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] -[{ "start": 0, "length": 66048, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] +[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, +{ "start": 4096, "length": 61952, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] +[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, +{ "start": 4096, "length": 61952, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] wrote 1/1 bytes at offset 65536 1 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -[{ "start": 0, "length": 65536, "depth": 0, "zero": true, "data": false, "offset": OFFSET}, +[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, +{ "start": 4096, "length": 61440, "depth": 0, "zero": true, "data": false, "offset": OFFSET}, { "start": 65536, "length": 1, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, { "start": 65537, "length": 511, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] -[{ "start": 0, "length": 65536, "depth": 0, "zero": true, "data": false, "offset": OFFSET}, +[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, +{ "start": 4096, "length": 61440, "depth": 0, "zero": true, "data": false, "offset": OFFSET}, { "start": 65536, "length": 1, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, { "start": 65537, "length": 511, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] *** done diff --git a/tests/qemu-iotests/253.out b/tests/qemu-iotests/253.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/253.out +++ b/tests/qemu-iotests/253.out @@ -XXX,XX +XXX,XX @@ QA output created by 253 === Check mapping of unaligned raw image === Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048575 -[{ "start": 0, "length": 1048576, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] -[{ "start": 0, "length": 1048576, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] +[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, +{ "start": 4096, "length": 1044480, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] +[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, +{ "start": 4096, "length": 1044480, "depth": 0, "zero": true, "data": false, "offset": OFFSET}] wrote 65535/65535 bytes at offset 983040 63.999 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -[{ "start": 0, "length": 983040, "depth": 0, "zero": true, "data": false, "offset": OFFSET}, +[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, +{ "start": 4096, "length": 978944, "depth": 0, "zero": true, "data": false, "offset": OFFSET}, { "start": 983040, "length": 65536, "depth": 0, "zero": false, "data": true, "offset": OFFSET}] -[{ "start": 0, "length": 983040, "depth": 0, "zero": true, "data": false, "offset": OFFSET}, +[{ "start": 0, "length": 4096, "depth": 0, "zero": false, "data": true, "offset": OFFSET}, +{ "start": 4096, "length": 978944, "depth": 0, "zero": true, "data": false, "offset": OFFSET}, { "start": 983040, "length": 65536, "depth": 0, "zero": false, "data": true, "offset": OFFSET}] *** done -- 2.21.0 From: Nir Soffer <nirsof@gmail.com> Using block_resize we can test allocate_first_block() with file descriptor opened with O_DIRECT, ensuring that it works for any size larger than 4096 bytes. Testing smaller sizes is tricky as the result depends on the filesystem used for testing. For example on NFS any size will work since O_DIRECT does not require any alignment. Signed-off-by: Nir Soffer <nsoffer@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-id: 20190827010528.8818-3-nsoffer@redhat.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/175 | 28 ++++++++++++++++++++++++++++ tests/qemu-iotests/175.out | 8 ++++++++ 2 files changed, 36 insertions(+) diff --git a/tests/qemu-iotests/175 b/tests/qemu-iotests/175 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/175 +++ b/tests/qemu-iotests/175 @@ -XXX,XX +XXX,XX @@ _filter_blocks() -e "s/blocks=$((extra_blocks + img_size / 512))\\(\$\\|[^0-9]\\)/max allocation/" } +# Resize image using block_resize. +# Parameter 1: image path +# Parameter 2: new size +_block_resize() +{ + local path=$1 + local size=$2 + + $QEMU -qmp stdio -nographic -nodefaults \ + -blockdev file,node-name=file,filename=$path,cache.direct=on \ + <<EOF +{'execute': 'qmp_capabilities'} +{'execute': 'block_resize', 'arguments': {'node-name': 'file', 'size': $size}} +{'execute': 'quit'} +EOF +} + # get standard environment, filters and checks . ./common.rc . ./common.filter @@ -XXX,XX +XXX,XX @@ _supported_fmt raw _supported_proto file _supported_os Linux +_default_cache_mode none +_supported_cache_modes none directsync + size=$((1 * 1024 * 1024)) touch "$TEST_DIR/empty" @@ -XXX,XX +XXX,XX @@ for mode in off full falloc; do stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $size done +for new_size in 4096 1048576; do + echo + echo "== resize empty image with block_resize ==" + _make_test_img 0 | _filter_imgfmt + _block_resize $TEST_IMG $new_size >/dev/null + stat -c "size=%s, blocks=%b" $TEST_IMG | _filter_blocks $extra_blocks $min_blocks $new_size +done + # success, all done echo "*** done" rm -f $seq.full diff --git a/tests/qemu-iotests/175.out b/tests/qemu-iotests/175.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/175.out +++ b/tests/qemu-iotests/175.out @@ -XXX,XX +XXX,XX @@ size=1048576, max allocation == creating image with preallocation falloc == Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576 preallocation=falloc size=1048576, max allocation + +== resize empty image with block_resize == +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=0 +size=4096, min allocation + +== resize empty image with block_resize == +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=0 +size=1048576, min allocation *** done -- 2.21.0 fe646693acc changed qemu-img create's output so that it no longer prints single quotes around parameter values. The subformat and adapter_type filters in _filter_img_create() have never been adapted to that change. Fixes: fe646693acc13ac48b98435d14149ab04dc597bc Signed-off-by: Max Reitz <mreitz@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Message-id: 20190815153638.4600-2-mreitz@redhat.com Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/059.out | 16 ++++++++-------- tests/qemu-iotests/common.filter | 4 ++-- 2 files changed, 10 insertions(+), 10 deletions(-) diff --git a/tests/qemu-iotests/059.out b/tests/qemu-iotests/059.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/059.out +++ b/tests/qemu-iotests/059.out @@ -XXX,XX +XXX,XX @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 qemu-io: can't open device TEST_DIR/t.vmdk: L1 size too big === Testing monolithicFlat creation and opening === -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2147483648 subformat=monolithicFlat +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2147483648 image: TEST_DIR/t.IMGFMT file format: IMGFMT virtual size: 2 GiB (2147483648 bytes) === Testing monolithicFlat with zeroed_grain === qemu-img: TEST_DIR/t.IMGFMT: Flat image can't enable zeroed grain -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2147483648 subformat=monolithicFlat +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=2147483648 === Testing big twoGbMaxExtentFlat === -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1073741824000 subformat=twoGbMaxExtentFlat +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1073741824000 image: TEST_DIR/t.vmdk file format: vmdk virtual size: 0.977 TiB (1073741824000 bytes) @@ -XXX,XX +XXX,XX @@ Format specific information: qemu-img: Could not open 'TEST_DIR/t.IMGFMT': Invalid extent line: RW 12582912 VMFS "dummy.IMGFMT" 1 === Testing truncated sparse === -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=107374182400 subformat=monolithicSparse +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=107374182400 qemu-img: Could not open 'TEST_DIR/t.IMGFMT': File truncated, expecting at least 13172736 bytes === Converting to streamOptimized from image with small cluster size=== @@ -XXX,XX +XXX,XX @@ wrote 512/512 bytes at offset 10240 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) === Testing monolithicFlat with internally generated JSON file name === -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 subformat=monolithicFlat +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 qemu-io: can't open: Cannot use relative extent paths with VMDK descriptor file 'json:{"image": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}, "driver": "blkdebug", "inject-error.0.event": "read_aio"}' === Testing version 3 === @@ -XXX,XX +XXX,XX @@ read 512/512 bytes at offset 64931328 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) === Testing 4TB monolithicFlat creation and IO === -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=4398046511104 subformat=monolithicFlat +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=4398046511104 image: TEST_DIR/t.IMGFMT file format: IMGFMT virtual size: 4 TiB (4398046511104 bytes) @@ -XXX,XX +XXX,XX @@ read 1024/1024 bytes at offset 966367641600 1 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) === Testing qemu-img map on extents === -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=33285996544 subformat=monolithicSparse +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=33285996544 wrote 1024/1024 bytes at offset 65024 1 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) wrote 1024/1024 bytes at offset 2147483136 @@ -XXX,XX +XXX,XX @@ Offset Length Mapped to File 0 0x20000 0x3f0000 TEST_DIR/t.vmdk 0x7fff0000 0x20000 0x410000 TEST_DIR/t.vmdk 0x140000000 0x10000 0x430000 TEST_DIR/t.vmdk -Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=33285996544 subformat=twoGbMaxExtentSparse +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=33285996544 wrote 1024/1024 bytes at offset 65024 1 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) wrote 1024/1024 bytes at offset 2147483136 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() -e "s# compat6=\\(on\\|off\\)##g" \ -e "s# static=\\(on\\|off\\)##g" \ -e "s# zeroed_grain=\\(on\\|off\\)##g" \ - -e "s# subformat='[^']*'##g" \ - -e "s# adapter_type='[^']*'##g" \ + -e "s# subformat=[^ ]*##g" \ + -e "s# adapter_type=[^ ]*##g" \ -e "s# hwversion=[^ ]*##g" \ -e "s# lazy_refcounts=\\(on\\|off\\)##g" \ -e "s# block_size=[0-9]\\+##g" \ -- 2.21.0 This makes iotest 033 pass with e.g. subformat=monolithicFlat. It also turns a former error in 059 into success. Signed-off-by: Max Reitz <mreitz@redhat.com> Message-id: 20190815153638.4600-3-mreitz@redhat.com Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/vmdk.c | 54 ++++++++++++++++++++++++-------------- tests/qemu-iotests/059 | 7 +++-- tests/qemu-iotests/059.out | 4 ++- 3 files changed, 42 insertions(+), 23 deletions(-) diff --git a/block/vmdk.c b/block/vmdk.c index XXXXXXX..XXXXXXX 100644 --- a/block/vmdk.c +++ b/block/vmdk.c @@ -XXX,XX +XXX,XX @@ static const char *next_line(const char *s) } static int vmdk_parse_extents(const char *desc, BlockDriverState *bs, - const char *desc_file_path, QDict *options, - Error **errp) + QDict *options, Error **errp) { int ret; int matches; @@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs, const char *p, *np; int64_t sectors = 0; int64_t flat_offset; + char *desc_file_dir = NULL; char *extent_path; BdrvChild *extent_file; BDRVVmdkState *s = bs->opaque; @@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs, continue; } - if (!path_is_absolute(fname) && !path_has_protocol(fname) && - !desc_file_path[0]) - { - bdrv_refresh_filename(bs->file->bs); - error_setg(errp, "Cannot use relative extent paths with VMDK " - "descriptor file '%s'", bs->file->bs->filename); - return -EINVAL; - } + if (path_is_absolute(fname)) { + extent_path = g_strdup(fname); + } else { + if (!desc_file_dir) { + desc_file_dir = bdrv_dirname(bs->file->bs, errp); + if (!desc_file_dir) { + bdrv_refresh_filename(bs->file->bs); + error_prepend(errp, "Cannot use relative paths with VMDK " + "descriptor file '%s': ", + bs->file->bs->filename); + ret = -EINVAL; + goto out; + } + } - extent_path = path_combine(desc_file_path, fname); + extent_path = g_strconcat(desc_file_dir, fname, NULL); + } ret = snprintf(extent_opt_prefix, 32, "extents.%d", s->num_extents); assert(ret < 32); @@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs, g_free(extent_path); if (local_err) { error_propagate(errp, local_err); - return -EINVAL; + ret = -EINVAL; + goto out; } /* save to extents array */ @@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs, 0, 0, 0, 0, 0, &extent, errp); if (ret < 0) { bdrv_unref_child(bs, extent_file); - return ret; + goto out; } extent->flat_start_offset = flat_offset << 9; } else if (!strcmp(type, "SPARSE") || !strcmp(type, "VMFSSPARSE")) { @@ -XXX,XX +XXX,XX @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs, g_free(buf); if (ret) { bdrv_unref_child(bs, extent_file); - return ret; + goto out; } extent = &s->extents[s->num_extents - 1]; } else if (!strcmp(type, "SESPARSE")) { ret = vmdk_open_se_sparse(bs, extent_file, bs->open_flags, errp); if (ret) { bdrv_unref_child(bs, extent_file); - return ret; + goto out; } extent = &s->extents[s->num_extents - 1]; } else { error_setg(errp, "Unsupported extent type '%s'", type); bdrv_unref_child(bs, extent_file); - return -ENOTSUP; + ret = -ENOTSUP; + goto out; } extent->type = g_strdup(type); } - return 0; + + ret = 0; + goto out; invalid: np = next_line(p); @@ -XXX,XX +XXX,XX @@ invalid: np--; } error_setg(errp, "Invalid extent line: %.*s", (int)(np - p), p); - return -EINVAL; + ret = -EINVAL; + +out: + g_free(desc_file_dir); + return ret; } static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf, @@ -XXX,XX +XXX,XX @@ static int vmdk_open_desc_file(BlockDriverState *bs, int flags, char *buf, } s->create_type = g_strdup(ct); s->desc_offset = 0; - ret = vmdk_parse_extents(buf, bs, bs->file->bs->exact_filename, options, - errp); + ret = vmdk_parse_extents(buf, bs, options, errp); exit: return ret; } diff --git a/tests/qemu-iotests/059 b/tests/qemu-iotests/059 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/059 +++ b/tests/qemu-iotests/059 @@ -XXX,XX +XXX,XX @@ $QEMU_IMG convert -f qcow2 -O vmdk -o subformat=streamOptimized "$TEST_IMG.qcow2 echo echo "=== Testing monolithicFlat with internally generated JSON file name ===" +# Should work, because bdrv_dirname() works fine with blkdebug IMGOPTS="subformat=monolithicFlat" _make_test_img 64M -$QEMU_IO -c "open -o driver=$IMGFMT,file.driver=blkdebug,file.image.filename=$TEST_IMG,file.inject-error.0.event=read_aio" 2>&1 \ - | _filter_testdir | _filter_imgfmt +$QEMU_IO -c "open -o driver=$IMGFMT,file.driver=blkdebug,file.image.filename=$TEST_IMG,file.inject-error.0.event=read_aio" \ + -c info \ + 2>&1 \ + | _filter_testdir | _filter_imgfmt | _filter_img_info _cleanup_test_img echo diff --git a/tests/qemu-iotests/059.out b/tests/qemu-iotests/059.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/059.out +++ b/tests/qemu-iotests/059.out @@ -XXX,XX +XXX,XX @@ wrote 512/512 bytes at offset 10240 === Testing monolithicFlat with internally generated JSON file name === Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 -qemu-io: can't open: Cannot use relative extent paths with VMDK descriptor file 'json:{"image": {"driver": "file", "filename": "TEST_DIR/t.IMGFMT"}, "driver": "blkdebug", "inject-error.0.event": "read_aio"}' +format name: IMGFMT +cluster size: 0 bytes +vm state offset: 0 bytes === Testing version 3 === image: TEST_DIR/iotest-version3.IMGFMT -- 2.21.0 We had a test for a case where relative extent paths did not work, but unfortunately we just fixed the underlying problem, so it works now. This patch adds a new test case that still fails. Signed-off-by: Max Reitz <mreitz@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Message-id: 20190815153638.4600-4-mreitz@redhat.com Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/059 | 27 +++++++++++++++++++++++++++ tests/qemu-iotests/059.out | 4 ++++ 2 files changed, 31 insertions(+) diff --git a/tests/qemu-iotests/059 b/tests/qemu-iotests/059 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/059 +++ b/tests/qemu-iotests/059 @@ -XXX,XX +XXX,XX @@ $QEMU_IMG convert -f qcow2 -O vmdk -o subformat=streamOptimized "$TEST_IMG.qcow2 echo echo "=== Testing monolithicFlat with internally generated JSON file name ===" + +echo '--- blkdebug ---' # Should work, because bdrv_dirname() works fine with blkdebug IMGOPTS="subformat=monolithicFlat" _make_test_img 64M $QEMU_IO -c "open -o driver=$IMGFMT,file.driver=blkdebug,file.image.filename=$TEST_IMG,file.inject-error.0.event=read_aio" \ @@ -XXX,XX +XXX,XX @@ $QEMU_IO -c "open -o driver=$IMGFMT,file.driver=blkdebug,file.image.filename=$TE | _filter_testdir | _filter_imgfmt | _filter_img_info _cleanup_test_img +echo '--- quorum ---' +# Should not work, because bdrv_dirname() does not work with quorum +IMGOPTS="subformat=monolithicFlat" _make_test_img 64M +cp "$TEST_IMG" "$TEST_IMG.orig" + +filename="json:{ + \"driver\": \"$IMGFMT\", + \"file\": { + \"driver\": \"quorum\", + \"children\": [ { + \"driver\": \"file\", + \"filename\": \"$TEST_IMG\" + }, { + \"driver\": \"file\", + \"filename\": \"$TEST_IMG.orig\" + } ], + \"vote-threshold\": 1 + } }" + +filename=$(echo "$filename" | tr '\n' ' ' | sed -e 's/\s\+/ /g') +$QEMU_IMG info "$filename" 2>&1 \ + | sed -e "s/'json:[^']*'/\$QUORUM_FILE/g" \ + | _filter_testdir | _filter_imgfmt | _filter_img_info + + echo echo "=== Testing version 3 ===" _use_sample_img iotest-version3.vmdk.bz2 diff --git a/tests/qemu-iotests/059.out b/tests/qemu-iotests/059.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/059.out +++ b/tests/qemu-iotests/059.out @@ -XXX,XX +XXX,XX @@ wrote 512/512 bytes at offset 10240 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) === Testing monolithicFlat with internally generated JSON file name === +--- blkdebug --- Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 format name: IMGFMT cluster size: 0 bytes vm state offset: 0 bytes +--- quorum --- +Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864 +qemu-img: Could not open $QUORUM_FILE: Cannot use relative paths with VMDK descriptor file $QUORUM_FILE: Cannot generate a base directory for quorum nodes === Testing version 3 === image: TEST_DIR/iotest-version3.IMGFMT -- 2.21.0 Compressed writes generally have to write full clusters, not just in theory but also in practice when it comes to vmdk's streamOptimized subformat. It currently is just silently broken for writes with non-zero in-cluster offsets: $ qemu-img create -f vmdk -o subformat=streamOptimized foo.vmdk 1M $ qemu-io -c 'write 4k 4k' -c 'read 4k 4k' foo.vmdk wrote 4096/4096 bytes at offset 4096 4 KiB, 1 ops; 00.01 sec (443.724 KiB/sec and 110.9309 ops/sec) read failed: Invalid argument (The technical reason is that vmdk_write_extent() just writes the incomplete compressed data actually to offset 4k. When reading the data, vmdk_read_extent() looks at offset 0 and finds the compressed data size to be 0, because that is what it reads from there. This yields an error.) For incomplete writes with zero in-cluster offsets, the error path when reading the rest of the cluster is a bit different, but the result is the same: $ qemu-img create -f vmdk -o subformat=streamOptimized foo.vmdk 1M $ qemu-io -c 'write 0k 4k' -c 'read 4k 4k' foo.vmdk wrote 4096/4096 bytes at offset 0 4 KiB, 1 ops; 00.01 sec (362.641 KiB/sec and 90.6603 ops/sec) read failed: Invalid argument (Here, vmdk_read_extent() finds the data and then sees that the uncompressed data is short.) It is better to reject invalid writes than to make the user believe they might have succeeded and then fail when trying to read it back. Signed-off-by: Max Reitz <mreitz@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Message-id: 20190815153638.4600-5-mreitz@redhat.com Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/vmdk.c | 10 ++++++++++ 1 file changed, 10 insertions(+) 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_write_extent(VmdkExtent *extent, int64_t cluster_offset, if (extent->compressed) { void *compressed_data; + /* Only whole clusters */ + if (offset_in_cluster || + n_bytes > (extent->cluster_sectors * SECTOR_SIZE) || + (n_bytes < (extent->cluster_sectors * SECTOR_SIZE) && + offset + n_bytes != extent->end_sector * SECTOR_SIZE)) + { + ret = -EINVAL; + goto out; + } + if (!extent->has_marker) { ret = -EINVAL; goto out; -- 2.21.0 streamOptimized does not support writes that do not span exactly one cluster. Furthermore, it cannot rewrite already allocated clusters. As such, many iotests do not work with it. Disable them. Signed-off-by: Max Reitz <mreitz@redhat.com> Message-id: 20190815153638.4600-6-mreitz@redhat.com Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/002 | 1 + tests/qemu-iotests/003 | 1 + tests/qemu-iotests/005 | 3 ++- tests/qemu-iotests/009 | 1 + tests/qemu-iotests/010 | 1 + tests/qemu-iotests/011 | 1 + tests/qemu-iotests/017 | 3 ++- tests/qemu-iotests/018 | 3 ++- tests/qemu-iotests/019 | 3 ++- tests/qemu-iotests/020 | 3 ++- tests/qemu-iotests/027 | 1 + tests/qemu-iotests/032 | 1 + tests/qemu-iotests/033 | 1 + tests/qemu-iotests/034 | 3 ++- tests/qemu-iotests/037 | 3 ++- tests/qemu-iotests/063 | 3 ++- tests/qemu-iotests/072 | 1 + tests/qemu-iotests/105 | 3 ++- tests/qemu-iotests/197 | 1 + tests/qemu-iotests/215 | 1 + tests/qemu-iotests/251 | 1 + 21 files changed, 30 insertions(+), 9 deletions(-) diff --git a/tests/qemu-iotests/002 b/tests/qemu-iotests/002 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/002 +++ b/tests/qemu-iotests/002 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt generic _supported_proto generic +_unsupported_imgopts "subformat=streamOptimized" size=128M diff --git a/tests/qemu-iotests/003 b/tests/qemu-iotests/003 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/003 +++ b/tests/qemu-iotests/003 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt generic _supported_proto generic +_unsupported_imgopts "subformat=streamOptimized" size=128M offset=67M diff --git a/tests/qemu-iotests/005 b/tests/qemu-iotests/005 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/005 +++ b/tests/qemu-iotests/005 @@ -XXX,XX +XXX,XX @@ _supported_fmt generic _supported_proto generic _supported_os Linux _unsupported_imgopts "subformat=twoGbMaxExtentFlat" \ - "subformat=twoGbMaxExtentSparse" + "subformat=twoGbMaxExtentSparse" \ + "subformat=streamOptimized" # vpc is limited to 127GB, so we can't test it here if [ "$IMGFMT" = "vpc" ]; then diff --git a/tests/qemu-iotests/009 b/tests/qemu-iotests/009 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/009 +++ b/tests/qemu-iotests/009 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt generic _supported_proto generic +_unsupported_imgopts "subformat=streamOptimized" size=6G diff --git a/tests/qemu-iotests/010 b/tests/qemu-iotests/010 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/010 +++ b/tests/qemu-iotests/010 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt generic _supported_proto generic +_unsupported_imgopts "subformat=streamOptimized" size=6G diff --git a/tests/qemu-iotests/011 b/tests/qemu-iotests/011 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/011 +++ b/tests/qemu-iotests/011 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt generic _supported_proto generic +_unsupported_imgopts "subformat=streamOptimized" size=6G diff --git a/tests/qemu-iotests/017 b/tests/qemu-iotests/017 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/017 +++ b/tests/qemu-iotests/017 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt qcow qcow2 vmdk qed _supported_proto generic _unsupported_proto vxhs -_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" +_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" \ + "subformat=streamOptimized" TEST_OFFSETS="0 4294967296" diff --git a/tests/qemu-iotests/018 b/tests/qemu-iotests/018 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/018 +++ b/tests/qemu-iotests/018 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt qcow qcow2 vmdk qed _supported_proto file _supported_os Linux -_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" +_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" \ + "streamOptimized" TEST_OFFSETS="0 4294967296" diff --git a/tests/qemu-iotests/019 b/tests/qemu-iotests/019 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/019 +++ b/tests/qemu-iotests/019 @@ -XXX,XX +XXX,XX @@ _supported_proto file _supported_os Linux _unsupported_imgopts "subformat=monolithicFlat" \ "subformat=twoGbMaxExtentFlat" \ - "subformat=twoGbMaxExtentSparse" + "subformat=twoGbMaxExtentSparse" \ + "subformat=streamOptimized" TEST_OFFSETS="0 4294967296" CLUSTER_SIZE=65536 diff --git a/tests/qemu-iotests/020 b/tests/qemu-iotests/020 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/020 +++ b/tests/qemu-iotests/020 @@ -XXX,XX +XXX,XX @@ _supported_fmt qcow qcow2 vmdk qed _supported_proto file _unsupported_imgopts "subformat=monolithicFlat" \ "subformat=twoGbMaxExtentFlat" \ - "subformat=twoGbMaxExtentSparse" + "subformat=twoGbMaxExtentSparse" \ + "subformat=streamOptimized" TEST_OFFSETS="0 4294967296" diff --git a/tests/qemu-iotests/027 b/tests/qemu-iotests/027 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/027 +++ b/tests/qemu-iotests/027 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt vmdk qcow qcow2 qed _supported_proto generic +_unsupported_imgopts "subformat=streamOptimized" size=128M diff --git a/tests/qemu-iotests/032 b/tests/qemu-iotests/032 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/032 +++ b/tests/qemu-iotests/032 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 # This works for any image format (though unlikely to segfault for raw) _supported_fmt generic _supported_proto generic +_unsupported_imgopts "subformat=streamOptimized" echo echo === Prepare image === diff --git a/tests/qemu-iotests/033 b/tests/qemu-iotests/033 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/033 +++ b/tests/qemu-iotests/033 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt generic _supported_proto generic +_unsupported_imgopts "subformat=streamOptimized" size=128M diff --git a/tests/qemu-iotests/034 b/tests/qemu-iotests/034 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/034 +++ b/tests/qemu-iotests/034 @@ -XXX,XX +XXX,XX @@ _supported_proto file _supported_os Linux _unsupported_imgopts "subformat=monolithicFlat" \ "subformat=twoGbMaxExtentFlat" \ - "subformat=twoGbMaxExtentSparse" + "subformat=twoGbMaxExtentSparse" \ + "subformat=streamOptimized" CLUSTER_SIZE=4k size=128M diff --git a/tests/qemu-iotests/037 b/tests/qemu-iotests/037 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/037 +++ b/tests/qemu-iotests/037 @@ -XXX,XX +XXX,XX @@ _supported_fmt qcow qcow2 vmdk qed _supported_proto file _unsupported_imgopts "subformat=monolithicFlat" \ "subformat=twoGbMaxExtentFlat" \ - "subformat=twoGbMaxExtentSparse" + "subformat=twoGbMaxExtentSparse" \ + "subformat=streamOptimized" CLUSTER_SIZE=4k size=128M diff --git a/tests/qemu-iotests/063 b/tests/qemu-iotests/063 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/063 +++ b/tests/qemu-iotests/063 @@ -XXX,XX +XXX,XX @@ _supported_fmt qcow qcow2 vmdk qed raw _supported_proto file _unsupported_imgopts "subformat=monolithicFlat" \ "subformat=twoGbMaxExtentFlat" \ - "subformat=twoGbMaxExtentSparse" + "subformat=twoGbMaxExtentSparse" \ + "subformat=streamOptimized" _make_test_img 4M diff --git a/tests/qemu-iotests/072 b/tests/qemu-iotests/072 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/072 +++ b/tests/qemu-iotests/072 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt vpc vmdk vhdx vdi qed qcow2 qcow _supported_proto file +_unsupported_imgopts "subformat=streamOptimized" IMG_SIZE=64M diff --git a/tests/qemu-iotests/105 b/tests/qemu-iotests/105 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/105 +++ b/tests/qemu-iotests/105 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt qcow2 vmdk vhdx qed _supported_proto generic _unsupported_imgopts "subformat=twoGbMaxExtentFlat" \ - "subformat=twoGbMaxExtentSparse" + "subformat=twoGbMaxExtentSparse" \ + "subformat=streamOptimized" echo echo "creating large image" diff --git a/tests/qemu-iotests/197 b/tests/qemu-iotests/197 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/197 +++ b/tests/qemu-iotests/197 @@ -XXX,XX +XXX,XX @@ _supported_fmt generic _supported_proto generic # LUKS support may be possible, but it complicates things. _unsupported_fmt luks +_unsupported_imgopts "subformat=streamOptimized" echo echo '=== Copy-on-read ===' diff --git a/tests/qemu-iotests/215 b/tests/qemu-iotests/215 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/215 +++ b/tests/qemu-iotests/215 @@ -XXX,XX +XXX,XX @@ _supported_fmt generic _supported_proto generic # LUKS support may be possible, but it complicates things. _unsupported_fmt luks +_unsupported_imgopts "subformat=streamOptimized" echo echo '=== Copy-on-read ===' diff --git a/tests/qemu-iotests/251 b/tests/qemu-iotests/251 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/251 +++ b/tests/qemu-iotests/251 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt generic _supported_proto file _supported_os Linux +_unsupported_imgopts "subformat=streamOptimized" if [ "$IMGOPTSSYNTAX" = "true" ]; then # We use json:{} filenames here, so we cannot work with additional options. -- 2.21.0 The error message for the test case where we have a quorum node for which no directory name can be generated is different: For twoGbMaxExtentSparse, it complains that it cannot open the extent file. For other (sub)formats, it just notes that it cannot determine the backing file path. Both are fine, but just disable twoGbMaxExtentSparse for simplicity's sake. Signed-off-by: Max Reitz <mreitz@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> Message-id: 20190815153638.4600-7-mreitz@redhat.com Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/110 | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/qemu-iotests/110 b/tests/qemu-iotests/110 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/110 +++ b/tests/qemu-iotests/110 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 # Any format supporting backing files _supported_fmt qed qcow qcow2 vmdk _supported_proto file -_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" +_unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" \ + "subformat=twoGbMaxExtentSparse" TEST_IMG_REL=$(basename "$TEST_IMG") -- 2.21.0 iotest 126 requires backing file support, which flat vmdks cannot offer. Skip this test for such subformats. Signed-off-by: Max Reitz <mreitz@redhat.com> Message-id: 20190815153638.4600-8-mreitz@redhat.com Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/126 | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/qemu-iotests/126 b/tests/qemu-iotests/126 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/126 +++ b/tests/qemu-iotests/126 @@ -XXX,XX +XXX,XX @@ status=1 # failure is the default! # Needs backing file support _supported_fmt qcow qcow2 qed vmdk +_unsupported_imgopts "subformat=monolithicFlat" \ + "subformat=twoGbMaxExtentFlat" # This is the default protocol (and we want to test the difference between # colons which separate a protocol prefix from the rest and colons which are # just part of the filename, so we cannot test protocols which require a prefix) -- 2.21.0 From: Stefan Hajnoczi <stefanha@redhat.com> Fixes: a6b257a08e3d72219f03e461a52152672fec0612 ("file-posix: Handle undetectable alignment") Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Message-id: 20190827101328.4062-1-stefanha@redhat.com Reviewed-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- block/file-posix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) 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 void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp) for (i = 0; i < ARRAY_SIZE(alignments); i++) { align = alignments[i]; if (raw_is_io_aligned(fd, buf + align, max_align)) { - /* Fallback to request_aligment. */ + /* Fallback to request_alignment. */ s->buf_align = (align != 1) ? align : bs->bl.request_alignment; break; } -- 2.21.0 From: Thomas Huth <thuth@redhat.com> It is possible to enable only a subset of the block drivers with the "--block-drv-rw-whitelist" option of the "configure" script. All other drivers are marked as unusable (or only included as read-only with the "--block-drv-ro-whitelist" option). If an iotest is now using such a disabled block driver, it is failing - which is bad, since at least the tests in the "auto" group should be able to deal with this situation. Thus let's introduce a "_require_drivers" function that can be used by the shell tests to check for the availability of certain drivers first, and marks the test as "not run" if one of the drivers is missing. This patch mainly targets the test in the "auto" group which should never fail in such a case, but also improves some of the other tests along the way. Note that we also assume that the "qcow2" and "file" drivers are always available - otherwise it does not make sense to run "make check-block" at all (which only tests with qcow2 by default). Signed-off-by: Thomas Huth <thuth@redhat.com> Message-id: 20190823133552.11680-1-thuth@redhat.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/071 | 1 + tests/qemu-iotests/081 | 4 +--- tests/qemu-iotests/099 | 1 + tests/qemu-iotests/120 | 1 + tests/qemu-iotests/162 | 4 +--- tests/qemu-iotests/184 | 1 + tests/qemu-iotests/186 | 1 + tests/qemu-iotests/common.rc | 14 ++++++++++++++ 8 files changed, 21 insertions(+), 6 deletions(-) diff --git a/tests/qemu-iotests/071 b/tests/qemu-iotests/071 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/071 +++ b/tests/qemu-iotests/071 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt qcow2 _supported_proto file +_require_drivers blkdebug blkverify do_run_qemu() { diff --git a/tests/qemu-iotests/081 b/tests/qemu-iotests/081 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/081 +++ b/tests/qemu-iotests/081 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt raw _supported_proto file _supported_os Linux +_require_drivers quorum do_run_qemu() { @@ -XXX,XX +XXX,XX @@ run_qemu() | _filter_qemu_io | _filter_generated_node_ids } -test_quorum=$($QEMU_IMG --help|grep quorum) -[ "$test_quorum" = "" ] && _supported_fmt quorum - quorum="driver=raw,file.driver=quorum,file.vote-threshold=2" quorum="$quorum,file.children.0.file.filename=$TEST_DIR/1.raw" quorum="$quorum,file.children.1.file.filename=$TEST_DIR/2.raw" diff --git a/tests/qemu-iotests/099 b/tests/qemu-iotests/099 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/099 +++ b/tests/qemu-iotests/099 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt qcow qcow2 qed vdi vhdx vmdk vpc _supported_proto file _supported_os Linux +_require_drivers blkdebug blkverify _unsupported_imgopts "subformat=monolithicFlat" "subformat=twoGbMaxExtentFlat" \ "subformat=twoGbMaxExtentSparse" diff --git a/tests/qemu-iotests/120 b/tests/qemu-iotests/120 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/120 +++ b/tests/qemu-iotests/120 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt generic _supported_proto file _unsupported_fmt luks +_require_drivers raw _make_test_img 64M diff --git a/tests/qemu-iotests/162 b/tests/qemu-iotests/162 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/162 +++ b/tests/qemu-iotests/162 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 . ./common.filter _supported_fmt generic - -test_ssh=$($QEMU_IMG --help | grep '^Supported formats:.* ssh\( \|$\)') -[ "$test_ssh" = "" ] && _notrun "ssh support required" +_require_drivers ssh echo echo '=== NBD ===' diff --git a/tests/qemu-iotests/184 b/tests/qemu-iotests/184 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/184 +++ b/tests/qemu-iotests/184 @@ -XXX,XX +XXX,XX @@ trap "exit \$status" 0 1 2 3 15 . ./common.filter _supported_os Linux +_require_drivers throttle do_run_qemu() { diff --git a/tests/qemu-iotests/186 b/tests/qemu-iotests/186 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/186 +++ b/tests/qemu-iotests/186 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt qcow2 _supported_proto file +_require_drivers null-co if [ "$QEMU_DEFAULT_MACHINE" != "pc" ]; then _notrun "Requires a PC 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 @@ _require_command() [ -x "$c" ] || _notrun "$1 utility required, skipped this test" } +# Check that a set of drivers has been whitelisted in the QEMU binary +# +_require_drivers() +{ + available=$($QEMU -drive format=help | \ + sed -e '/Supported formats:/!d' -e 's/Supported formats://') + for driver + do + if ! echo "$available" | grep -q " $driver\( \|$\)"; then + _notrun "$driver not available" + fi + done +} + # make sure this script returns success true -- 2.21.0 From: Thomas Huth <thuth@redhat.com> The sanitizers (especially the address sanitizer from Clang) are sometimes printing out warnings or false positives - this spoils the output of the iotests, causing some of the tests to fail. Thus let's skip the automatic iotests during "make check" when the user configured QEMU with --enable-sanitizers. Signed-off-by: Thomas Huth <thuth@redhat.com> Message-id: 20190823084203.29734-1-thuth@redhat.com Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/check-block.sh | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/tests/check-block.sh b/tests/check-block.sh index XXXXXXX..XXXXXXX 100755 --- a/tests/check-block.sh +++ b/tests/check-block.sh @@ -XXX,XX +XXX,XX @@ if grep -q "TARGET_GPROF=y" *-softmmu/config-target.mak 2>/dev/null ; then exit 0 fi +if grep -q "CFLAGS.*-fsanitize" config-host.mak 2>/dev/null ; then + echo "Sanitizers are enabled ==> Not running the qemu-iotests." + exit 0 +fi + if [ -z "$(find . -name 'qemu-system-*' -print)" ]; then echo "No qemu-system binary available ==> Not running the qemu-iotests." exit 0 -- 2.21.0 From: Nir Soffer <nirsof@gmail.com> Quoting cache mode is not needed, and most tests use unquoted values. Unify all test to use the same style. Message-id: 20190827173432.7656-1-nsoffer@redhat.com Signed-off-by: Nir Soffer <nsoffer@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> --- tests/qemu-iotests/026 | 4 ++-- tests/qemu-iotests/039 | 4 ++-- tests/qemu-iotests/052 | 2 +- tests/qemu-iotests/091 | 4 ++-- 4 files changed, 7 insertions(+), 7 deletions(-) diff --git a/tests/qemu-iotests/026 b/tests/qemu-iotests/026 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/026 +++ b/tests/qemu-iotests/026 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 # Currently only qcow2 supports rebasing _supported_fmt qcow2 _supported_proto file -_default_cache_mode "writethrough" -_supported_cache_modes "writethrough" "none" +_default_cache_mode writethrough +_supported_cache_modes writethrough none # The refcount table tests expect a certain minimum width for refcount entries # (so that the refcount table actually needs to grow); that minimum is 16 bits, # being the default refcount entry width. diff --git a/tests/qemu-iotests/039 b/tests/qemu-iotests/039 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/039 +++ b/tests/qemu-iotests/039 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt qcow2 _supported_proto file _supported_os Linux -_default_cache_mode "writethrough" -_supported_cache_modes "writethrough" +_default_cache_mode writethrough +_supported_cache_modes writethrough size=128M diff --git a/tests/qemu-iotests/052 b/tests/qemu-iotests/052 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/052 +++ b/tests/qemu-iotests/052 @@ -XXX,XX +XXX,XX @@ _supported_fmt generic _supported_proto file # Don't do O_DIRECT on tmpfs -_supported_cache_modes "writeback" "writethrough" "unsafe" +_supported_cache_modes writeback writethrough unsafe size=128M _make_test_img $size diff --git a/tests/qemu-iotests/091 b/tests/qemu-iotests/091 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/091 +++ b/tests/qemu-iotests/091 @@ -XXX,XX +XXX,XX @@ trap "_cleanup; exit \$status" 0 1 2 3 15 _supported_fmt qcow2 _supported_proto file _supported_os Linux -_default_cache_mode "none" -_supported_cache_modes "writethrough" "none" "writeback" +_default_cache_mode none +_supported_cache_modes writethrough none writeback size=1G -- 2.21.0 The following changes since commit 0b6206b9c6825619cd721085fe082d7a0abc9af4: Merge remote-tracking branch 'remotes/rth-gitlab/tags/pull-tcg-20210914-4' into staging (2021-09-15 13:27:49 +0100) are available in the Git repository at: https://github.com/XanClic/qemu.git tags/pull-block-2021-09-15 for you to fetch changes up to 1899bf47375ad40555dcdff12ba49b4b8b82df38: qemu-img: Add -F shorthand to convert (2021-09-15 18:42:38 +0200) ---------------------------------------------------------------- Block patches: - Block-status cache for data regions - qcow2 optimization (when using subclusters) - iotests delinting, and let 297 (lint checker) cover named iotests - qcow2 check improvements - Added -F (target backing file format) option to qemu-img convert - Mirror job fix - Fix for when a migration is initiated while a backup job runs - Fix for uncached qemu-img convert to a volume with 4k sectors (for an unaligned image) - Minor gluster driver fix ---------------------------------------------------------------- Eric Blake (1): qemu-img: Add -F shorthand to convert Hanna Reitz (15): gluster: Align block-status tail block: Drop BDS comment regarding bdrv_append() block: block-status cache for data regions block: Clarify that @bytes is no limit on *pnum block/file-posix: Do not force-cap *pnum block/gluster: Do not force-cap *pnum block/iscsi: Do not force-cap *pnum iotests: Fix unspecified-encoding pylint warnings iotests: Fix use-{list,dict}-literal warnings iotests/297: Drop 169 and 199 from the skip list migrate-bitmaps-postcopy-test: Fix pylint warnings migrate-bitmaps-test: Fix pylint warnings mirror-top-perms: Fix AbnormalShutdown path iotests/297: Cover tests/ qemu-img: Allow target be aligned to sector size Stefano Garzarella (1): block/mirror: fix NULL pointer dereference in mirror_wait_on_conflicts() Vladimir Sementsov-Ogievskiy (15): tests: add migrate-during-backup block: bdrv_inactivate_recurse(): check for permissions and fix crash simplebench: add img_bench_templater.py qcow2: refactor handle_dependencies() loop body qcow2: handle_dependencies(): relax conflict detection qcow2-refcount: improve style of check_refcounts_l2() qcow2: compressed read: simplify cluster descriptor passing qcow2: introduce qcow2_parse_compressed_l2_entry() helper qcow2-refcount: introduce fix_l2_entry_by_zero() qcow2-refcount: fix_l2_entry_by_zero(): also zero L2 entry bitmap qcow2-refcount: check_refcounts_l2(): check l2_bitmap qcow2-refcount: check_refcounts_l2(): check reserved bits qcow2-refcount: improve style of check_refcounts_l1() qcow2-refcount: check_refcounts_l1(): check reserved bits qcow2-refcount: check_refblocks(): add separate message for reserved docs/tools/qemu-img.rst | 4 +- block/qcow2.h | 7 +- include/block/block_int.h | 61 +++- block.c | 88 +++++ block/file-posix.c | 7 +- block/gluster.c | 23 +- block/io.c | 68 +++- block/iscsi.c | 3 - block/mirror.c | 25 +- block/qcow2-cluster.c | 78 +++-- block/qcow2-refcount.c | 326 ++++++++++++------ block/qcow2.c | 13 +- qemu-img.c | 18 +- qemu-img-cmds.hx | 2 +- scripts/simplebench/img_bench_templater.py | 95 +++++ scripts/simplebench/table_templater.py | 62 ++++ tests/qemu-iotests/122 | 2 +- tests/qemu-iotests/271 | 5 +- tests/qemu-iotests/271.out | 4 +- tests/qemu-iotests/297 | 9 +- tests/qemu-iotests/iotests.py | 12 +- .../tests/migrate-bitmaps-postcopy-test | 13 +- tests/qemu-iotests/tests/migrate-bitmaps-test | 43 ++- .../qemu-iotests/tests/migrate-during-backup | 97 ++++++ .../tests/migrate-during-backup.out | 5 + tests/qemu-iotests/tests/mirror-top-perms | 2 +- 26 files changed, 855 insertions(+), 217 deletions(-) create mode 100755 scripts/simplebench/img_bench_templater.py create mode 100644 scripts/simplebench/table_templater.py create mode 100755 tests/qemu-iotests/tests/migrate-during-backup create mode 100644 tests/qemu-iotests/tests/migrate-during-backup.out -- 2.31.1 From: Max Reitz <mreitz@redhat.com> gluster's block-status implementation is basically a copy of that in block/file-posix.c, there is only one thing missing, and that is aligning trailing data extents to the request alignment (as added by commit 9c3db310ff0). Note that 9c3db310ff0 mentions that "there seems to be no other block driver that sets request_alignment and [...]", but while block/gluster.c does indeed not set request_alignment, block/io.c's bdrv_refresh_limits() will still default to an alignment of 512 because block/gluster.c does not provide a byte-aligned read function. Therefore, unaligned tails can conceivably occur, and so we should apply the change from 9c3db310ff0 to gluster's block-status implementation. Reported-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Max Reitz <mreitz@redhat.com> Message-Id: <20210805143603.59503-1-mreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/gluster.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) 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 int coroutine_fn qemu_gluster_co_block_status(BlockDriverState *bs, off_t data = 0, hole = 0; int ret = -EINVAL; + assert(QEMU_IS_ALIGNED(offset | bytes, bs->bl.request_alignment)); + if (!s->fd) { return ret; } @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qemu_gluster_co_block_status(BlockDriverState *bs, /* On a data extent, compute bytes to the end of the extent, * possibly including a partial sector at EOF. */ *pnum = MIN(bytes, hole - offset); + + /* + * We are not allowed to return partial sectors, though, so + * round up if necessary. + */ + if (!QEMU_IS_ALIGNED(*pnum, bs->bl.request_alignment)) { + int64_t file_length = qemu_gluster_getlength(bs); + if (file_length > 0) { + /* Ignore errors, this is just a safeguard */ + assert(hole == file_length); + } + *pnum = ROUND_UP(*pnum, bs->bl.request_alignment); + } + ret = BDRV_BLOCK_DATA; } else { /* On a hole, compute bytes to the beginning of the next extent. */ -- 2.31.1 There is a comment above the BDS definition stating care must be taken to consider handling newly added fields in bdrv_append(). Actually, this comment should have said "bdrv_swap()" as of 4ddc07cac (nine years ago), and in any case, bdrv_swap() was dropped in 8e419aefa (six years ago). So no such care is necessary anymore. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Message-Id: <20210812084148.14458-2-hreitz@redhat.com> --- include/block/block_int.h | 6 ------ 1 file changed, 6 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 BdrvChild { QLIST_ENTRY(BdrvChild) next_parent; }; -/* - * Note: the function bdrv_append() copies and swaps contents of - * BlockDriverStates, so if you add new fields to this struct, please - * inspect bdrv_append() to determine if the new fields need to be - * copied as well. - */ struct BlockDriverState { /* Protected by big QEMU lock or read-only after opening. No special * locking needed during I/O... -- 2.31.1 As we have attempted before (https://lists.gnu.org/archive/html/qemu-devel/2019-01/msg06451.html, "file-posix: Cache lseek result for data regions"; https://lists.nongnu.org/archive/html/qemu-block/2021-02/msg00934.html, "file-posix: Cache next hole"), this patch seeks to reduce the number of SEEK_DATA/HOLE operations the file-posix driver has to perform. The main difference is that this time it is implemented as part of the general block layer code. The problem we face is that on some filesystems or in some circumstances, SEEK_DATA/HOLE is unreasonably slow. Given the implementation is outside of qemu, there is little we can do about its performance. We have already introduced the want_zero parameter to bdrv_co_block_status() to reduce the number of SEEK_DATA/HOLE calls unless we really want zero information; but sometimes we do want that information, because for files that consist largely of zero areas, special-casing those areas can give large performance boosts. So the real problem is with files that consist largely of data, so that inquiring the block status does not gain us much performance, but where such an inquiry itself takes a lot of time. To address this, we want to cache data regions. Most of the time, when bad performance is reported, it is in places where the image is iterated over from start to end (qemu-img convert or the mirror job), so a simple yet effective solution is to cache only the current data region. (Note that only caching data regions but not zero regions means that returning false information from the cache is not catastrophic: Treating zeroes as data is fine. While we try to invalidate the cache on zero writes and discards, such incongruences may still occur when there are other processes writing to the image.) We only use the cache for nodes without children (i.e. protocol nodes), because that is where the problem is: Drivers that rely on block-status implementations outside of qemu (e.g. SEEK_DATA/HOLE). Resolves: https://gitlab.com/qemu-project/qemu/-/issues/307 Signed-off-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210812084148.14458-3-hreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> [hreitz: Added `local_file == bs` assertion, as suggested by Vladimir] Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- include/block/block_int.h | 50 ++++++++++++++++++++++++ block.c | 80 +++++++++++++++++++++++++++++++++++++++ block/io.c | 68 +++++++++++++++++++++++++++++++-- 3 files changed, 195 insertions(+), 3 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 @@ #include "qemu/hbitmap.h" #include "block/snapshot.h" #include "qemu/throttle.h" +#include "qemu/rcu.h" #define BLOCK_FLAG_LAZY_REFCOUNTS 8 @@ -XXX,XX +XXX,XX @@ struct BdrvChild { QLIST_ENTRY(BdrvChild) next_parent; }; +/* + * Allows bdrv_co_block_status() to cache one data region for a + * protocol node. + * + * @valid: Whether the cache is valid (should be accessed with atomic + * functions so this can be reset by RCU readers) + * @data_start: Offset where we know (or strongly assume) is data + * @data_end: Offset where the data region ends (which is not necessarily + * the start of a zeroed region) + */ +typedef struct BdrvBlockStatusCache { + struct rcu_head rcu; + + bool valid; + int64_t data_start; + int64_t data_end; +} BdrvBlockStatusCache; + struct BlockDriverState { /* Protected by big QEMU lock or read-only after opening. No special * locking needed during I/O... @@ -XXX,XX +XXX,XX @@ struct BlockDriverState { /* BdrvChild links to this node may never be frozen */ bool never_freeze; + + /* Lock for block-status cache RCU writers */ + CoMutex bsc_modify_lock; + /* Always non-NULL, but must only be dereferenced under an RCU read guard */ + BdrvBlockStatusCache *block_status_cache; }; struct BlockBackendRootState { @@ -XXX,XX +XXX,XX @@ static inline BlockDriverState *bdrv_primary_bs(BlockDriverState *bs) */ void bdrv_drain_all_end_quiesce(BlockDriverState *bs); +/** + * Check whether the given offset is in the cached block-status data + * region. + * + * If it is, and @pnum is not NULL, *pnum is set to + * `bsc.data_end - offset`, i.e. how many bytes, starting from + * @offset, are data (according to the cache). + * Otherwise, *pnum is not touched. + */ +bool bdrv_bsc_is_data(BlockDriverState *bs, int64_t offset, int64_t *pnum); + +/** + * If [offset, offset + bytes) overlaps with the currently cached + * block-status region, invalidate the cache. + * + * (To be used by I/O paths that cause data regions to be zero or + * holes.) + */ +void bdrv_bsc_invalidate_range(BlockDriverState *bs, + int64_t offset, int64_t bytes); + +/** + * Mark the range [offset, offset + bytes) as a data region. + */ +void bdrv_bsc_fill(BlockDriverState *bs, int64_t offset, int64_t bytes); + #endif /* BLOCK_INT_H */ diff --git a/block.c b/block.c index XXXXXXX..XXXXXXX 100644 --- a/block.c +++ b/block.c @@ -XXX,XX +XXX,XX @@ #include "qemu/timer.h" #include "qemu/cutils.h" #include "qemu/id.h" +#include "qemu/range.h" +#include "qemu/rcu.h" #include "block/coroutines.h" #ifdef CONFIG_BSD @@ -XXX,XX +XXX,XX @@ BlockDriverState *bdrv_new(void) qemu_co_queue_init(&bs->flush_queue); + qemu_co_mutex_init(&bs->bsc_modify_lock); + bs->block_status_cache = g_new0(BdrvBlockStatusCache, 1); + for (i = 0; i < bdrv_drain_all_count; i++) { bdrv_drained_begin(bs); } @@ -XXX,XX +XXX,XX @@ static void bdrv_close(BlockDriverState *bs) bs->explicit_options = NULL; qobject_unref(bs->full_open_options); bs->full_open_options = NULL; + g_free(bs->block_status_cache); + bs->block_status_cache = NULL; bdrv_release_named_dirty_bitmaps(bs); assert(QLIST_EMPTY(&bs->dirty_bitmaps)); @@ -XXX,XX +XXX,XX @@ BlockDriverState *bdrv_backing_chain_next(BlockDriverState *bs) { return bdrv_skip_filters(bdrv_cow_bs(bdrv_skip_filters(bs))); } + +/** + * Check whether [offset, offset + bytes) overlaps with the cached + * block-status data region. + * + * If so, and @pnum is not NULL, set *pnum to `bsc.data_end - offset`, + * which is what bdrv_bsc_is_data()'s interface needs. + * Otherwise, *pnum is not touched. + */ +static bool bdrv_bsc_range_overlaps_locked(BlockDriverState *bs, + int64_t offset, int64_t bytes, + int64_t *pnum) +{ + BdrvBlockStatusCache *bsc = qatomic_rcu_read(&bs->block_status_cache); + bool overlaps; + + overlaps = + qatomic_read(&bsc->valid) && + ranges_overlap(offset, bytes, bsc->data_start, + bsc->data_end - bsc->data_start); + + if (overlaps && pnum) { + *pnum = bsc->data_end - offset; + } + + return overlaps; +} + +/** + * See block_int.h for this function's documentation. + */ +bool bdrv_bsc_is_data(BlockDriverState *bs, int64_t offset, int64_t *pnum) +{ + RCU_READ_LOCK_GUARD(); + + return bdrv_bsc_range_overlaps_locked(bs, offset, 1, pnum); +} + +/** + * See block_int.h for this function's documentation. + */ +void bdrv_bsc_invalidate_range(BlockDriverState *bs, + int64_t offset, int64_t bytes) +{ + RCU_READ_LOCK_GUARD(); + + if (bdrv_bsc_range_overlaps_locked(bs, offset, bytes, NULL)) { + qatomic_set(&bs->block_status_cache->valid, false); + } +} + +/** + * See block_int.h for this function's documentation. + */ +void bdrv_bsc_fill(BlockDriverState *bs, int64_t offset, int64_t bytes) +{ + BdrvBlockStatusCache *new_bsc = g_new(BdrvBlockStatusCache, 1); + BdrvBlockStatusCache *old_bsc; + + *new_bsc = (BdrvBlockStatusCache) { + .valid = true, + .data_start = offset, + .data_end = offset + bytes, + }; + + QEMU_LOCK_GUARD(&bs->bsc_modify_lock); + + old_bsc = qatomic_rcu_read(&bs->block_status_cache); + qatomic_rcu_set(&bs->block_status_cache, new_bsc); + if (old_bsc) { + g_free_rcu(old_bsc, rcu); + } +} 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_pwrite_zeroes(BlockDriverState *bs, return -ENOTSUP; } + /* Invalidate the cached block-status data range if this write overlaps */ + bdrv_bsc_invalidate_range(bs, offset, bytes); + assert(alignment % bs->bl.request_alignment == 0); head = offset % alignment; tail = (offset + bytes) % alignment; @@ -XXX,XX +XXX,XX @@ static int coroutine_fn bdrv_co_block_status(BlockDriverState *bs, aligned_bytes = ROUND_UP(offset + bytes, align) - aligned_offset; if (bs->drv->bdrv_co_block_status) { - ret = bs->drv->bdrv_co_block_status(bs, want_zero, aligned_offset, - aligned_bytes, pnum, &local_map, - &local_file); + /* + * Use the block-status cache only for protocol nodes: Format + * drivers are generally quick to inquire the status, but protocol + * drivers often need to get information from outside of qemu, so + * we do not have control over the actual implementation. There + * have been cases where inquiring the status took an unreasonably + * long time, and we can do nothing in qemu to fix it. + * This is especially problematic for images with large data areas, + * because finding the few holes in them and giving them special + * treatment does not gain much performance. Therefore, we try to + * cache the last-identified data region. + * + * Second, limiting ourselves to protocol nodes allows us to assume + * the block status for data regions to be DATA | OFFSET_VALID, and + * that the host offset is the same as the guest offset. + * + * Note that it is possible that external writers zero parts of + * the cached regions without the cache being invalidated, and so + * we may report zeroes as data. This is not catastrophic, + * however, because reporting zeroes as data is fine. + */ + if (QLIST_EMPTY(&bs->children) && + bdrv_bsc_is_data(bs, aligned_offset, pnum)) + { + ret = BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID; + local_file = bs; + local_map = aligned_offset; + } else { + ret = bs->drv->bdrv_co_block_status(bs, want_zero, aligned_offset, + aligned_bytes, pnum, &local_map, + &local_file); + + /* + * Note that checking QLIST_EMPTY(&bs->children) is also done when + * the cache is queried above. Technically, we do not need to check + * it here; the worst that can happen is that we fill the cache for + * non-protocol nodes, and then it is never used. However, filling + * the cache requires an RCU update, so double check here to avoid + * such an update if possible. + */ + if (ret == (BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID) && + QLIST_EMPTY(&bs->children)) + { + /* + * When a protocol driver reports BLOCK_OFFSET_VALID, the + * returned local_map value must be the same as the offset we + * have passed (aligned_offset), and local_bs must be the node + * itself. + * Assert this, because we follow this rule when reading from + * the cache (see the `local_file = bs` and + * `local_map = aligned_offset` assignments above), and the + * result the cache delivers must be the same as the driver + * would deliver. + */ + assert(local_file == bs); + assert(local_map == aligned_offset); + bdrv_bsc_fill(bs, aligned_offset, *pnum); + } + } } else { /* Default code for filters */ @@ -XXX,XX +XXX,XX @@ int coroutine_fn bdrv_co_pdiscard(BdrvChild *child, int64_t offset, return 0; } + /* Invalidate the cached block-status data range if this discard overlaps */ + bdrv_bsc_invalidate_range(bs, offset, bytes); + /* Discard is advisory, but some devices track and coalesce * unaligned requests, so we must pass everything down rather than * round here. Still, most devices will just silently ignore -- 2.31.1 .bdrv_co_block_status() implementations are free to return a *pnum that exceeds @bytes, because bdrv_co_block_status() in block/io.c will clamp *pnum as necessary. On the other hand, if drivers' implementations return values for *pnum that are as large as possible, our recently introduced block-status cache will become more effective. So, make a note in block_int.h that @bytes is no upper limit for *pnum. Suggested-by: Eric Blake <eblake@redhat.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20210812084148.14458-4-hreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> --- include/block/block_int.h | 9 +++++++++ 1 file changed, 9 insertions(+) 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 { * clamped to bdrv_getlength() and aligned to request_alignment, * as well as non-NULL pnum, map, and file; in turn, the driver * must return an error or set pnum to an aligned non-zero value. + * + * Note that @bytes is just a hint on how big of a region the + * caller wants to inspect. It is not a limit on *pnum. + * Implementations are free to return larger values of *pnum if + * doing so does not incur a performance penalty. + * + * block/io.c's bdrv_co_block_status() will utilize an unclamped + * *pnum value for the block-status cache on protocol nodes, prior + * to clamping *pnum for return to its caller. */ int coroutine_fn (*bdrv_co_block_status)(BlockDriverState *bs, bool want_zero, int64_t offset, int64_t bytes, int64_t *pnum, -- 2.31.1 bdrv_co_block_status() does it for us, we do not need to do it here. The advantage of not capping *pnum is that bdrv_co_block_status() can cache larger data regions than requested by its caller. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Message-Id: <20210812084148.14458-5-hreitz@redhat.com> --- block/file-posix.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) 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 find_allocation(BlockDriverState *bs, off_t start, * the specified offset) that are known to be in the same * allocated/unallocated state. * - * 'bytes' is the max value 'pnum' should be set to. + * 'bytes' is a soft cap for 'pnum'. If the information is free, 'pnum' may + * well exceed it. */ static int coroutine_fn raw_co_block_status(BlockDriverState *bs, bool want_zero, @@ -XXX,XX +XXX,XX @@ static int coroutine_fn raw_co_block_status(BlockDriverState *bs, } else if (data == offset) { /* On a data extent, compute bytes to the end of the extent, * possibly including a partial sector at EOF. */ - *pnum = MIN(bytes, hole - offset); + *pnum = hole - offset; /* * We are not allowed to return partial sectors, though, so @@ -XXX,XX +XXX,XX @@ static int coroutine_fn raw_co_block_status(BlockDriverState *bs, } else { /* On a hole, compute bytes to the beginning of the next extent. */ assert(hole == offset); - *pnum = MIN(bytes, data - offset); + *pnum = data - offset; ret = BDRV_BLOCK_ZERO; } *map = offset; -- 2.31.1 bdrv_co_block_status() does it for us, we do not need to do it here. The advantage of not capping *pnum is that bdrv_co_block_status() can cache larger data regions than requested by its caller. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Message-Id: <20210812084148.14458-6-hreitz@redhat.com> --- block/gluster.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) 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 @@ exit: * the specified offset) that are known to be in the same * allocated/unallocated state. * - * 'bytes' is the max value 'pnum' should be set to. + * 'bytes' is a soft cap for 'pnum'. If the information is free, 'pnum' may + * well exceed it. * * (Based on raw_co_block_status() from file-posix.c.) */ @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qemu_gluster_co_block_status(BlockDriverState *bs, } else if (data == offset) { /* On a data extent, compute bytes to the end of the extent, * possibly including a partial sector at EOF. */ - *pnum = MIN(bytes, hole - offset); + *pnum = hole - offset; /* * We are not allowed to return partial sectors, though, so @@ -XXX,XX +XXX,XX @@ static int coroutine_fn qemu_gluster_co_block_status(BlockDriverState *bs, } else { /* On a hole, compute bytes to the beginning of the next extent. */ assert(hole == offset); - *pnum = MIN(bytes, data - offset); + *pnum = data - offset; ret = BDRV_BLOCK_ZERO; } -- 2.31.1 bdrv_co_block_status() does it for us, we do not need to do it here. The advantage of not capping *pnum is that bdrv_co_block_status() can cache larger data regions than requested by its caller. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Message-Id: <20210812084148.14458-7-hreitz@redhat.com> --- block/iscsi.c | 3 --- 1 file changed, 3 deletions(-) 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 @@ retry: iscsi_allocmap_set_allocated(iscsilun, offset, *pnum); } - if (*pnum > bytes) { - *pnum = bytes; - } out_unlock: qemu_mutex_unlock(&iscsilun->mutex); g_free(iTask.err_str); -- 2.31.1 As of recently, pylint complains when `open()` calls are missing an `encoding=` specified. Everything we have should be UTF-8 (and in fact, everything should be UTF-8, period (exceptions apply)), so use that. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210824153540.177128-2-hreitz@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: John Snow <jsnow@redhat.com> --- tests/qemu-iotests/297 | 2 +- tests/qemu-iotests/iotests.py | 8 +++++--- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/tests/qemu-iotests/297 b/tests/qemu-iotests/297 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/297 +++ b/tests/qemu-iotests/297 @@ -XXX,XX +XXX,XX @@ def is_python_file(filename): if filename.endswith('.py'): return True - with open(filename) as f: + with open(filename, encoding='utf-8') as f: try: first_line = f.readline() return re.match('^#!.*python', first_line) is not None 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 _post_shutdown(self) -> None: return valgrind_filename = f"{test_dir}/{self._popen.pid}.valgrind" if self.exitcode() == 99: - with open(valgrind_filename) as f: + with open(valgrind_filename, encoding='utf-8') as f: print(f.read()) else: os.remove(valgrind_filename) @@ -XXX,XX +XXX,XX @@ def notrun(reason): # Each test in qemu-iotests has a number ("seq") seq = os.path.basename(sys.argv[0]) - with open('%s/%s.notrun' % (output_dir, seq), 'w') as outfile: + with open('%s/%s.notrun' % (output_dir, seq), 'w', encoding='utf-8') \ + as outfile: outfile.write(reason + '\n') logger.warning("%s not run: %s", seq, reason) sys.exit(0) @@ -XXX,XX +XXX,XX @@ def case_notrun(reason): # Each test in qemu-iotests has a number ("seq") seq = os.path.basename(sys.argv[0]) - with open('%s/%s.casenotrun' % (output_dir, seq), 'a') as outfile: + with open('%s/%s.casenotrun' % (output_dir, seq), 'a', encoding='utf-8') \ + as outfile: outfile.write(' [case not run] ' + reason + '\n') def _verify_image_format(supported_fmts: Sequence[str] = (), -- 2.31.1 pylint proposes using `[]` instead of `list()` and `{}` instead of `dict()`, because it is faster. That seems simple enough, so heed its advice. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210824153540.177128-3-hreitz@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> --- tests/qemu-iotests/iotests.py | 4 ++-- 1 file changed, 2 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 hmp_qemu_io(self, drive: str, cmd: str, def flatten_qmp_object(self, obj, output=None, basestr=''): if output is None: - output = dict() + output = {} if isinstance(obj, list): for i, item in enumerate(obj): self.flatten_qmp_object(item, output, basestr + str(i) + '.') @@ -XXX,XX +XXX,XX @@ def flatten_qmp_object(self, obj, output=None, basestr=''): def qmp_to_opts(self, obj): obj = self.flatten_qmp_object(obj) - output_list = list() + output_list = [] for key in obj: output_list += [key + '=' + obj[key]] return ','.join(output_list) -- 2.31.1 169 and 199 have been renamed and moved to tests/ (commit a44be0334be: "iotests: rename and move 169 and 199 tests"), so we can drop them from the skip list. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Reviewed-by: Willian Rampazzo <willianr@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Message-Id: <20210902094017.32902-2-hreitz@redhat.com> --- tests/qemu-iotests/297 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/qemu-iotests/297 b/tests/qemu-iotests/297 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/297 +++ b/tests/qemu-iotests/297 @@ -XXX,XX +XXX,XX @@ import iotests SKIP_FILES = ( '030', '040', '041', '044', '045', '055', '056', '057', '065', '093', '096', '118', '124', '132', '136', '139', '147', '148', '149', - '151', '152', '155', '163', '165', '169', '194', '196', '199', '202', + '151', '152', '155', '163', '165', '194', '196', '202', '203', '205', '206', '207', '208', '210', '211', '212', '213', '216', '218', '219', '224', '228', '234', '235', '236', '237', '238', '240', '242', '245', '246', '248', '255', '256', '257', '258', '260', -- 2.31.1 pylint complains that discards1_sha256 and all_discards_sha256 are first set in non-__init__ methods. These variables are not really class-variables anyway, so let them instead be returned by start_postcopy(), thus silencing pylint. Suggested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20210902094017.32902-3-hreitz@redhat.com> --- .../tests/migrate-bitmaps-postcopy-test | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/tests/qemu-iotests/tests/migrate-bitmaps-postcopy-test b/tests/qemu-iotests/tests/migrate-bitmaps-postcopy-test index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/tests/migrate-bitmaps-postcopy-test +++ b/tests/qemu-iotests/tests/migrate-bitmaps-postcopy-test @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapPostcopyMigration(iotests.QMPTestCase): result = self.vm_a.qmp('x-debug-block-dirty-bitmap-sha256', node='drive0', name='bitmap0') - self.discards1_sha256 = result['return']['sha256'] + discards1_sha256 = result['return']['sha256'] # Check, that updating the bitmap by discards works - assert self.discards1_sha256 != empty_sha256 + assert discards1_sha256 != empty_sha256 # We want to calculate resulting sha256. Do it in bitmap0, so, disable # other bitmaps @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapPostcopyMigration(iotests.QMPTestCase): result = self.vm_a.qmp('x-debug-block-dirty-bitmap-sha256', node='drive0', name='bitmap0') - self.all_discards_sha256 = result['return']['sha256'] + all_discards_sha256 = result['return']['sha256'] # Now, enable some bitmaps, to be updated during migration for i in range(2, nb_bitmaps, 2): @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapPostcopyMigration(iotests.QMPTestCase): event_resume = self.vm_b.event_wait('RESUME') self.vm_b_events.append(event_resume) - return event_resume + return (event_resume, discards1_sha256, all_discards_sha256) def test_postcopy_success(self): - event_resume = self.start_postcopy() + event_resume, discards1_sha256, all_discards_sha256 = \ + self.start_postcopy() # enabled bitmaps should be updated apply_discards(self.vm_b, discards2) @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapPostcopyMigration(iotests.QMPTestCase): for i in range(0, nb_bitmaps, 5): result = self.vm_b.qmp('x-debug-block-dirty-bitmap-sha256', node='drive0', name='bitmap{}'.format(i)) - sha = self.discards1_sha256 if i % 2 else self.all_discards_sha256 + sha = discards1_sha256 if i % 2 else all_discards_sha256 self.assert_qmp(result, 'return/sha256', sha) def test_early_shutdown_destination(self): -- 2.31.1 There are a couple of things pylint takes issue with: - The "time" import is unused - The import order (iotests should come last) - get_bitmap_hash() doesn't use @self and so should be a function - Semicolons at the end of some lines - Parentheses after "if" - Some lines are too long (80 characters instead of 79) - inject_test_case()'s @name parameter shadows a top-level @name variable - "lambda self: mc(self)" were equivalent to just "mc", but in inject_test_case(), it is not equivalent, so add a comment and disable the warning locally - Always put two empty lines after a function - f'exec: cat > /dev/null' does not need to be an f-string Fix them. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210902094017.32902-4-hreitz@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> --- tests/qemu-iotests/tests/migrate-bitmaps-test | 43 +++++++++++-------- 1 file changed, 25 insertions(+), 18 deletions(-) diff --git a/tests/qemu-iotests/tests/migrate-bitmaps-test b/tests/qemu-iotests/tests/migrate-bitmaps-test index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/tests/migrate-bitmaps-test +++ b/tests/qemu-iotests/tests/migrate-bitmaps-test @@ -XXX,XX +XXX,XX @@ # import os -import iotests -import time import itertools import operator import re +import iotests from iotests import qemu_img, qemu_img_create, Timeout @@ -XXX,XX +XXX,XX @@ mig_cmd = 'exec: cat > ' + mig_file incoming_cmd = 'exec: cat ' + mig_file +def get_bitmap_hash(vm): + result = vm.qmp('x-debug-block-dirty-bitmap-sha256', + node='drive0', name='bitmap0') + return result['return']['sha256'] + + class TestDirtyBitmapMigration(iotests.QMPTestCase): def tearDown(self): self.vm_a.shutdown() @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapMigration(iotests.QMPTestCase): params['persistent'] = True result = vm.qmp('block-dirty-bitmap-add', **params) - self.assert_qmp(result, 'return', {}); - - def get_bitmap_hash(self, vm): - result = vm.qmp('x-debug-block-dirty-bitmap-sha256', - node='drive0', name='bitmap0') - return result['return']['sha256'] + self.assert_qmp(result, 'return', {}) def check_bitmap(self, vm, sha256): result = vm.qmp('x-debug-block-dirty-bitmap-sha256', node='drive0', name='bitmap0') if sha256: - self.assert_qmp(result, 'return/sha256', sha256); + self.assert_qmp(result, 'return/sha256', sha256) else: self.assert_qmp(result, 'error/desc', - "Dirty bitmap 'bitmap0' not found"); + "Dirty bitmap 'bitmap0' not found") def do_test_migration_resume_source(self, persistent, migrate_bitmaps): granularity = 512 @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapMigration(iotests.QMPTestCase): self.add_bitmap(self.vm_a, granularity, persistent) for r in regions: self.vm_a.hmp_qemu_io('drive0', 'write %d %d' % r) - sha256 = self.get_bitmap_hash(self.vm_a) + sha256 = get_bitmap_hash(self.vm_a) result = self.vm_a.qmp('migrate', uri=mig_cmd) while True: @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapMigration(iotests.QMPTestCase): break while True: result = self.vm_a.qmp('query-status') - if (result['return']['status'] == 'postmigrate'): + if result['return']['status'] == 'postmigrate': break # test that bitmap is still here @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapMigration(iotests.QMPTestCase): self.add_bitmap(self.vm_a, granularity, persistent) for r in regions: self.vm_a.hmp_qemu_io('drive0', 'write %d %d' % r) - sha256 = self.get_bitmap_hash(self.vm_a) + sha256 = get_bitmap_hash(self.vm_a) if pre_shutdown: self.vm_a.shutdown() @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapMigration(iotests.QMPTestCase): self.check_bitmap(self.vm_b, sha256 if persistent else False) -def inject_test_case(klass, name, method, *args, **kwargs): +def inject_test_case(klass, suffix, method, *args, **kwargs): mc = operator.methodcaller(method, *args, **kwargs) - setattr(klass, 'test_' + method + name, lambda self: mc(self)) + # We want to add a function attribute to `klass`, so that it is + # correctly converted to a method on instantiation. The + # methodcaller object `mc` is a callable, not a function, so we + # need the lambda to turn it into a function. + # pylint: disable=unnecessary-lambda + setattr(klass, 'test_' + method + suffix, lambda self: mc(self)) + for cmb in list(itertools.product((True, False), repeat=5)): name = ('_' if cmb[0] else '_not_') + 'persistent_' name += ('_' if cmb[1] else '_not_') + 'migbitmap_' name += '_online' if cmb[2] else '_offline' name += '_shared' if cmb[3] else '_nonshared' - if (cmb[4]): + if cmb[4]: name += '__pre_shutdown' inject_test_case(TestDirtyBitmapMigration, name, 'do_test_migration', @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapBackingMigration(iotests.QMPTestCase): self.assert_qmp(result, 'return', {}) # Check that the bitmaps are there - for node in self.vm.qmp('query-named-block-nodes', flat=True)['return']: + nodes = self.vm.qmp('query-named-block-nodes', flat=True)['return'] + for node in nodes: if 'node0' in node['node-name']: self.assert_qmp(node, 'dirty-bitmaps[0]/name', 'bmap0') @@ -XXX,XX +XXX,XX @@ class TestDirtyBitmapBackingMigration(iotests.QMPTestCase): """ Continue the source after migration. """ - result = self.vm.qmp('migrate', uri=f'exec: cat > /dev/null') + result = self.vm.qmp('migrate', uri='exec: cat > /dev/null') self.assert_qmp(result, 'return', {}) with Timeout(10, 'Migration timeout'): -- 2.31.1 The AbnormalShutdown exception class is not in qemu.machine, but in qemu.machine.machine. (qemu.machine.AbnormalShutdown was enough for Python to find it in order to run this test, but pylint complains about it.) Signed-off-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210902094017.32902-5-hreitz@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> --- tests/qemu-iotests/tests/mirror-top-perms | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/qemu-iotests/tests/mirror-top-perms b/tests/qemu-iotests/tests/mirror-top-perms index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/tests/mirror-top-perms +++ b/tests/qemu-iotests/tests/mirror-top-perms @@ -XXX,XX +XXX,XX @@ class TestMirrorTopPerms(iotests.QMPTestCase): def tearDown(self): try: self.vm.shutdown() - except qemu.machine.AbnormalShutdown: + except qemu.machine.machine.AbnormalShutdown: pass if self.vm_b is not None: -- 2.31.1 297 so far does not check the named tests, which reside in the tests/ directory (i.e. full path tests/qemu-iotests/tests). Fix it. Thanks to the previous two commits, all named tests pass its scrutiny, so we do not have to add anything to SKIP_FILES. Signed-off-by: Hanna Reitz <hreitz@redhat.com> Reviewed-by: Willian Rampazzo <willianr@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Message-Id: <20210902094017.32902-6-hreitz@redhat.com> --- tests/qemu-iotests/297 | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/tests/qemu-iotests/297 b/tests/qemu-iotests/297 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/297 +++ b/tests/qemu-iotests/297 @@ -XXX,XX +XXX,XX @@ def is_python_file(filename): def run_linters(): - files = [filename for filename in (set(os.listdir('.')) - set(SKIP_FILES)) - if is_python_file(filename)] + named_tests = [f'tests/{entry}' for entry in os.listdir('tests')] + check_tests = set(os.listdir('.') + named_tests) - set(SKIP_FILES) + files = [filename for filename in check_tests if is_python_file(filename)] iotests.logger.debug('Files to be checked:') iotests.logger.debug(', '.join(sorted(files))) -- 2.31.1 From: Stefano Garzarella <sgarzare@redhat.com> In mirror_iteration() we call mirror_wait_on_conflicts() with `self` parameter set to NULL. Starting from commit d44dae1a7c we dereference `self` pointer in mirror_wait_on_conflicts() without checks if it is not NULL. Backtrace: Program terminated with signal SIGSEGV, Segmentation fault. #0 mirror_wait_on_conflicts (self=0x0, s=<optimized out>, offset=<optimized out>, bytes=<optimized out>) at ../block/mirror.c:172 172 self->waiting_for_op = op; [Current thread is 1 (Thread 0x7f0908931ec0 (LWP 380249))] (gdb) bt #0 mirror_wait_on_conflicts (self=0x0, s=<optimized out>, offset=<optimized out>, bytes=<optimized out>) at ../block/mirror.c:172 #1 0x00005610c5d9d631 in mirror_run (job=0x5610c76a2c00, errp=<optimized out>) at ../block/mirror.c:491 #2 0x00005610c5d58726 in job_co_entry (opaque=0x5610c76a2c00) at ../job.c:917 #3 0x00005610c5f046c6 in coroutine_trampoline (i0=<optimized out>, i1=<optimized out>) at ../util/coroutine-ucontext.c:173 #4 0x00007f0909975820 in ?? () at ../sysdeps/unix/sysv/linux/x86_64/__start_context.S:91 from /usr/lib64/libc.so.6 Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2001404 Fixes: d44dae1a7c ("block/mirror: fix active mirror dead-lock in mirror_wait_on_conflicts") Signed-off-by: Stefano Garzarella <sgarzare@redhat.com> Message-Id: <20210910124533.288318-1-sgarzare@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/mirror.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 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 coroutine_fn mirror_wait_on_conflicts(MirrorOp *self, if (ranges_overlap(self_start_chunk, self_nb_chunks, op_start_chunk, op_nb_chunks)) { - /* - * If the operation is already (indirectly) waiting for us, or - * will wait for us as soon as it wakes up, then just go on - * (instead of producing a deadlock in the former case). - */ - if (op->waiting_for_op) { - continue; + if (self) { + /* + * If the operation is already (indirectly) waiting for us, + * or will wait for us as soon as it wakes up, then just go + * on (instead of producing a deadlock in the former case). + */ + if (op->waiting_for_op) { + continue; + } + + self->waiting_for_op = op; } - self->waiting_for_op = op; qemu_co_queue_wait(&op->waiting_requests, NULL); - self->waiting_for_op = NULL; + + if (self) { + self->waiting_for_op = NULL; + } + break; } } -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Add a simple test which tries to run migration during backup. bdrv_inactivate_all() should fail. But due to bug (see next commit with fix) it doesn't, nodes are inactivated and continued backup crashes on assertion "assert(!(bs->open_flags & BDRV_O_INACTIVE));" in bdrv_co_write_req_prepare(). Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20210911120027.8063-2-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- .../qemu-iotests/tests/migrate-during-backup | 97 +++++++++++++++++++ .../tests/migrate-during-backup.out | 5 + 2 files changed, 102 insertions(+) create mode 100755 tests/qemu-iotests/tests/migrate-during-backup create mode 100644 tests/qemu-iotests/tests/migrate-during-backup.out diff --git a/tests/qemu-iotests/tests/migrate-during-backup b/tests/qemu-iotests/tests/migrate-during-backup new file mode 100755 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/tests/migrate-during-backup @@ -XXX,XX +XXX,XX @@ +#!/usr/bin/env python3 +# group: migration disabled +# +# Copyright (c) 2021 Virtuozzo International GmbH +# +# 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/>. +# + +import os +import iotests +from iotests import qemu_img_create, qemu_io + + +disk_a = os.path.join(iotests.test_dir, 'disk_a') +disk_b = os.path.join(iotests.test_dir, 'disk_b') +size = '1M' +mig_file = os.path.join(iotests.test_dir, 'mig_file') +mig_cmd = 'exec: cat > ' + mig_file + + +class TestMigrateDuringBackup(iotests.QMPTestCase): + def tearDown(self): + self.vm.shutdown() + os.remove(disk_a) + os.remove(disk_b) + os.remove(mig_file) + + def setUp(self): + qemu_img_create('-f', iotests.imgfmt, disk_a, size) + qemu_img_create('-f', iotests.imgfmt, disk_b, size) + qemu_io('-c', f'write 0 {size}', disk_a) + + self.vm = iotests.VM().add_drive(disk_a) + self.vm.launch() + result = self.vm.qmp('blockdev-add', { + 'node-name': 'target', + 'driver': iotests.imgfmt, + 'file': { + 'driver': 'file', + 'filename': disk_b + } + }) + self.assert_qmp(result, 'return', {}) + + def test_migrate(self): + result = self.vm.qmp('blockdev-backup', device='drive0', + target='target', sync='full', + speed=1, x_perf={ + 'max-workers': 1, + 'max-chunk': 64 * 1024 + }) + self.assert_qmp(result, 'return', {}) + + result = self.vm.qmp('job-pause', id='drive0') + self.assert_qmp(result, 'return', {}) + + result = self.vm.qmp('migrate-set-capabilities', + capabilities=[{'capability': 'events', + 'state': True}]) + self.assert_qmp(result, 'return', {}) + result = self.vm.qmp('migrate', uri=mig_cmd) + self.assert_qmp(result, 'return', {}) + + e = self.vm.events_wait((('MIGRATION', + {'data': {'status': 'completed'}}), + ('MIGRATION', + {'data': {'status': 'failed'}}))) + + # Don't assert that e is 'failed' now: this way we'll miss + # possible crash when backup continues :) + + result = self.vm.qmp('block-job-set-speed', device='drive0', + speed=0) + self.assert_qmp(result, 'return', {}) + result = self.vm.qmp('job-resume', id='drive0') + self.assert_qmp(result, 'return', {}) + + # For future: if something changes so that both migration + # and backup pass, let's not miss that moment, as it may + # be a bug as well as improvement. + self.assert_qmp(e, 'data/status', 'failed') + + +if __name__ == '__main__': + iotests.main(supported_fmts=['qcow2'], + supported_protocols=['file']) diff --git a/tests/qemu-iotests/tests/migrate-during-backup.out b/tests/qemu-iotests/tests/migrate-during-backup.out new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/tests/qemu-iotests/tests/migrate-during-backup.out @@ -XXX,XX +XXX,XX @@ +. +---------------------------------------------------------------------- +Ran 1 tests + +OK -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> We must not inactivate child when parent has write permissions on it. Calling .bdrv_inactivate() doesn't help: actually only qcow2 has this handler and it is used to flush caches, not for permission manipulations. So, let's simply check cumulative parent permissions before inactivating the node. This commit fixes a crash when we do migration during backup: prior to the commit nothing prevents all nodes inactivation at migration finish and following backup write to the target crashes on assertion "assert(!(bs->open_flags & BDRV_O_INACTIVE));" in bdrv_co_write_req_prepare(). After the commit, we rely on the fact that copy-before-write filter keeps write permission on target node to be able to write to it. So inactivation fails and migration fails as expected. Corresponding test now passes, so, enable it. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210911120027.8063-3-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block.c | 8 ++++++++ tests/qemu-iotests/tests/migrate-during-backup | 2 +- 2 files changed, 9 insertions(+), 1 deletion(-) 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_inactivate_recurse(BlockDriverState *bs) { BdrvChild *child, *parent; int ret; + uint64_t cumulative_perms, cumulative_shared_perms; if (!bs->drv) { return -ENOMEDIUM; @@ -XXX,XX +XXX,XX @@ static int bdrv_inactivate_recurse(BlockDriverState *bs) } } + bdrv_get_cumulative_perm(bs, &cumulative_perms, + &cumulative_shared_perms); + if (cumulative_perms & (BLK_PERM_WRITE | BLK_PERM_WRITE_UNCHANGED)) { + /* Our inactive parents still need write access. Inactivation failed. */ + return -EPERM; + } + bs->open_flags |= BDRV_O_INACTIVE; /* diff --git a/tests/qemu-iotests/tests/migrate-during-backup b/tests/qemu-iotests/tests/migrate-during-backup index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/tests/migrate-during-backup +++ b/tests/qemu-iotests/tests/migrate-during-backup @@ -XXX,XX +XXX,XX @@ #!/usr/bin/env python3 -# group: migration disabled +# group: migration # # Copyright (c) 2021 Virtuozzo International GmbH # -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Add simple grammar-parsing template benchmark. New tool consume test template written in bash with some special grammar injections and produces multiple tests, run them and finally print a performance comparison table of different tests produced from one template. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20210824101517.59802-2-vsementsov@virtuozzo.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- scripts/simplebench/img_bench_templater.py | 95 ++++++++++++++++++++++ scripts/simplebench/table_templater.py | 62 ++++++++++++++ 2 files changed, 157 insertions(+) create mode 100755 scripts/simplebench/img_bench_templater.py create mode 100644 scripts/simplebench/table_templater.py diff --git a/scripts/simplebench/img_bench_templater.py b/scripts/simplebench/img_bench_templater.py new file mode 100755 index XXXXXXX..XXXXXXX --- /dev/null +++ b/scripts/simplebench/img_bench_templater.py @@ -XXX,XX +XXX,XX @@ +#!/usr/bin/env python3 +# +# Process img-bench test templates +# +# Copyright (c) 2021 Virtuozzo International GmbH. +# +# 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/>. +# + + +import sys +import subprocess +import re +import json + +import simplebench +from results_to_text import results_to_text +from table_templater import Templater + + +def bench_func(env, case): + test = templater.gen(env['data'], case['data']) + + p = subprocess.run(test, shell=True, stdout=subprocess.PIPE, + stderr=subprocess.STDOUT, universal_newlines=True) + + if p.returncode == 0: + try: + m = re.search(r'Run completed in (\d+.\d+) seconds.', p.stdout) + return {'seconds': float(m.group(1))} + except Exception: + return {'error': f'failed to parse qemu-img output: {p.stdout}'} + else: + return {'error': f'qemu-img failed: {p.returncode}: {p.stdout}'} + + +if __name__ == '__main__': + if len(sys.argv) > 1: + print(""" +Usage: img_bench_templater.py < path/to/test-template.sh + +This script generates performance tests from a test template (example below), +runs them, and displays the results in a table. The template is read from +stdin. It must be written in bash and end with a `qemu-img bench` invocation +(whose result is parsed to get the test instance’s result). + +Use the following syntax in the template to create the various different test +instances: + + column templating: {var1|var2|...} - test will use different values in + different columns. You may use several {} constructions in the test, in this + case product of all choice-sets will be used. + + row templating: [var1|var2|...] - similar thing to define rows (test-cases) + +Test template example: + +Assume you want to compare two qemu-img binaries, called qemu-img-old and +qemu-img-new in your build directory in two test-cases with 4K writes and 64K +writes. The template may look like this: + +qemu_img=/path/to/qemu/build/qemu-img-{old|new} +$qemu_img create -f qcow2 /ssd/x.qcow2 1G +$qemu_img bench -c 100 -d 8 [-s 4K|-s 64K] -w -t none -n /ssd/x.qcow2 + +When passing this to stdin of img_bench_templater.py, the resulting comparison +table will contain two columns (for two binaries) and two rows (for two +test-cases). + +In addition to displaying the results, script also stores results in JSON +format into results.json file in current directory. +""") + sys.exit() + + templater = Templater(sys.stdin.read()) + + envs = [{'id': ' / '.join(x), 'data': x} for x in templater.columns] + cases = [{'id': ' / '.join(x), 'data': x} for x in templater.rows] + + result = simplebench.bench(bench_func, envs, cases, count=5, + initial_run=False) + print(results_to_text(result)) + with open('results.json', 'w') as f: + json.dump(result, f, indent=4) diff --git a/scripts/simplebench/table_templater.py b/scripts/simplebench/table_templater.py new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/scripts/simplebench/table_templater.py @@ -XXX,XX +XXX,XX @@ +# Parser for test templates +# +# Copyright (c) 2021 Virtuozzo International GmbH. +# +# 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/>. +# + +import itertools +from lark import Lark + +grammar = """ +start: ( text | column_switch | row_switch )+ + +column_switch: "{" text ["|" text]+ "}" +row_switch: "[" text ["|" text]+ "]" +text: /[^|{}\[\]]+/ +""" + +parser = Lark(grammar) + +class Templater: + def __init__(self, template): + self.tree = parser.parse(template) + + c_switches = [] + r_switches = [] + for x in self.tree.children: + if x.data == 'column_switch': + c_switches.append([el.children[0].value for el in x.children]) + elif x.data == 'row_switch': + r_switches.append([el.children[0].value for el in x.children]) + + self.columns = list(itertools.product(*c_switches)) + self.rows = list(itertools.product(*r_switches)) + + def gen(self, column, row): + i = 0 + j = 0 + result = [] + + for x in self.tree.children: + if x.data == 'text': + result.append(x.children[0].value) + elif x.data == 'column_switch': + result.append(column[i]) + i += 1 + elif x.data == 'row_switch': + result.append(row[j]) + j += 1 + + return ''.join(result) -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> No logic change, just prepare for the following commit. While being here do also small grammar fix in a comment. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210824101517.59802-3-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2-cluster.c | 49 ++++++++++++++++++++++++------------------- 1 file changed, 28 insertions(+), 21 deletions(-) diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-cluster.c +++ b/block/qcow2-cluster.c @@ -XXX,XX +XXX,XX @@ static int handle_dependencies(BlockDriverState *bs, uint64_t guest_offset, if (end <= old_start || start >= old_end) { /* No intersection */ - } else { - if (start < old_start) { - /* Stop at the start of a running allocation */ - bytes = old_start - start; - } else { - bytes = 0; - } + continue; + } - /* Stop if already an l2meta exists. After yielding, it wouldn't - * be valid any more, so we'd have to clean up the old L2Metas - * and deal with requests depending on them before starting to - * gather new ones. Not worth the trouble. */ - if (bytes == 0 && *m) { - *cur_bytes = 0; - return 0; - } + /* Conflict */ - if (bytes == 0) { - /* Wait for the dependency to complete. We need to recheck - * the free/allocated clusters when we continue. */ - qemu_co_queue_wait(&old_alloc->dependent_requests, &s->lock); - return -EAGAIN; - } + if (start < old_start) { + /* Stop at the start of a running allocation */ + bytes = old_start - start; + } else { + bytes = 0; + } + + /* + * Stop if an l2meta already exists. After yielding, it wouldn't + * be valid any more, so we'd have to clean up the old L2Metas + * and deal with requests depending on them before starting to + * gather new ones. Not worth the trouble. + */ + if (bytes == 0 && *m) { + *cur_bytes = 0; + return 0; + } + + if (bytes == 0) { + /* + * Wait for the dependency to complete. We need to recheck + * the free/allocated clusters when we continue. + */ + qemu_co_queue_wait(&old_alloc->dependent_requests, &s->lock); + return -EAGAIN; } } -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> There is no conflict and no dependency if we have parallel writes to different subclusters of one cluster when the cluster itself is already allocated. So, relax extra dependency. Measure performance: First, prepare build/qemu-img-old and build/qemu-img-new images. cd scripts/simplebench ./img_bench_templater.py Paste the following to stdin of running script: qemu_img=../../build/qemu-img-{old|new} $qemu_img create -f qcow2 -o extended_l2=on /ssd/x.qcow2 1G $qemu_img bench -c 100000 -d 8 [-s 2K|-s 2K -o 512|-s $((1024*2+512))] \ -w -t none -n /ssd/x.qcow2 The result: All results are in seconds ------------------ --------- --------- old new -s 2K 6.7 ± 15% 6.2 ± 12% -7% -s 2K -o 512 13 ± 3% 11 ± 5% -16% -s $((1024*2+512)) 9.5 ± 4% 8.4 -12% ------------------ --------- --------- So small writes are more independent now and that helps to keep deeper io queue which improves performance. 271 iotest output becomes racy for three allocation in one cluster. Second and third writes may finish in different order. Second and third requests don't depend on each other any more. Still they both depend on first request anyway. Filter out second and third write offsets to cover both possible outputs. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Message-Id: <20210824101517.59802-4-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> [hreitz: s/ an / and /] Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2-cluster.c | 11 +++++++++++ tests/qemu-iotests/271 | 5 ++++- tests/qemu-iotests/271.out | 4 ++-- 3 files changed, 17 insertions(+), 3 deletions(-) diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-cluster.c +++ b/block/qcow2-cluster.c @@ -XXX,XX +XXX,XX @@ static int handle_dependencies(BlockDriverState *bs, uint64_t guest_offset, continue; } + if (old_alloc->keep_old_clusters && + (end <= l2meta_cow_start(old_alloc) || + start >= l2meta_cow_end(old_alloc))) + { + /* + * Clusters intersect but COW areas don't. And cluster itself is + * already allocated. So, there is no actual conflict. + */ + continue; + } + /* Conflict */ if (start < old_start) { diff --git a/tests/qemu-iotests/271 b/tests/qemu-iotests/271 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/271 +++ b/tests/qemu-iotests/271 @@ -XXX,XX +XXX,XX @@ EOF } _make_test_img -o extended_l2=on 1M -_concurrent_io | $QEMU_IO | _filter_qemu_io +# Second and third writes in _concurrent_io() are independent and may finish in +# different order. So, filter offset out to match both possible variants. +_concurrent_io | $QEMU_IO | _filter_qemu_io | \ + $SED -e 's/\(20480\|40960\)/OFFSET/' _concurrent_verify | $QEMU_IO | _filter_qemu_io # success, all done diff --git a/tests/qemu-iotests/271.out b/tests/qemu-iotests/271.out index XXXXXXX..XXXXXXX 100644 --- a/tests/qemu-iotests/271.out +++ b/tests/qemu-iotests/271.out @@ -XXX,XX +XXX,XX @@ blkdebug: Suspended request 'A' blkdebug: Resuming request 'A' wrote 2048/2048 bytes at offset 30720 2 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -wrote 2048/2048 bytes at offset 20480 +wrote 2048/2048 bytes at offset OFFSET 2 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -wrote 2048/2048 bytes at offset 40960 +wrote 2048/2048 bytes at offset OFFSET 2 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) *** done -- 2.31.1 We cannot write to images opened with O_DIRECT unless we allow them to be resized so they are aligned to the sector size: Since 9c60a5d1978, bdrv_node_refresh_perm() ensures that for nodes whose length is not aligned to the request alignment and where someone has taken a WRITE permission, the RESIZE permission is taken, too). Let qemu-img convert pass the BDRV_O_RESIZE flag (which causes blk_new_open() to take the RESIZE permission) when using cache=none for the target, so that when writing to it, it can be aligned to the target sector size. Without this patch, an error is returned: $ qemu-img convert -f raw -O raw -t none foo.img /mnt/tmp/foo.img qemu-img: Could not open '/mnt/tmp/foo.img': Cannot get 'write' permission without 'resize': Image size is not a multiple of request alignment Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1994266 Signed-off-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210819101200.64235-1-hreitz@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> --- qemu-img.c | 8 ++++++++ 1 file changed, 8 insertions(+) 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_convert(int argc, char **argv) goto out; } + if (flags & BDRV_O_NOCACHE) { + /* + * If we open the target with O_DIRECT, it may be necessary to + * extend its size to align to the physical sector size. + */ + flags |= BDRV_O_RESIZE; + } + if (skip_create) { s.target = img_open(tgt_image_opts, out_filename, out_fmt, flags, writethrough, s.quiet, false); -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> - don't use same name for size in bytes and in entries - use g_autofree for l2_table - add whitespace - fix block comment style Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-2-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2-refcount.c | 47 +++++++++++++++++++++--------------------- 1 file changed, 24 insertions(+), 23 deletions(-) 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_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, int flags, BdrvCheckMode fix, bool active) { BDRVQcow2State *s = bs->opaque; - uint64_t *l2_table, l2_entry; + uint64_t l2_entry; uint64_t next_contiguous_offset = 0; - int i, l2_size, nb_csectors, ret; + int i, nb_csectors, ret; + size_t l2_size_bytes = s->l2_size * l2_entry_size(s); + g_autofree uint64_t *l2_table = g_malloc(l2_size_bytes); /* Read L2 table from disk */ - l2_size = s->l2_size * l2_entry_size(s); - l2_table = g_malloc(l2_size); - - ret = bdrv_pread(bs->file, l2_offset, l2_table, l2_size); + ret = bdrv_pread(bs->file, l2_offset, l2_table, l2_size_bytes); if (ret < 0) { fprintf(stderr, "ERROR: I/O error in check_refcounts_l2\n"); res->check_errors++; - goto fail; + return ret; } /* Do the actual checks */ - for(i = 0; i < s->l2_size; i++) { + for (i = 0; i < s->l2_size; i++) { l2_entry = get_l2_entry(s, l2_table, i); switch (qcow2_get_cluster_type(bs, l2_entry)) { @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, l2_entry & QCOW2_COMPRESSED_SECTOR_MASK, nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE); if (ret < 0) { - goto fail; + return ret; } if (flags & CHECK_FRAG_INFO) { res->bfi.allocated_clusters++; res->bfi.compressed_clusters++; - /* Compressed clusters are fragmented by nature. Since they + /* + * Compressed clusters are fragmented by nature. Since they * take up sub-sector space but we only have sector granularity * I/O we need to re-read the same sectors even for adjacent * compressed clusters. @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, if (ret < 0) { fprintf(stderr, "ERROR: Overlap check failed\n"); res->check_errors++; - /* Something is seriously wrong, so abort checking - * this L2 table */ - goto fail; + /* + * Something is seriously wrong, so abort checking + * this L2 table. + */ + return ret; } ret = bdrv_pwrite_sync(bs->file, l2e_offset, @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, fprintf(stderr, "ERROR: Failed to overwrite L2 " "table entry: %s\n", strerror(-ret)); res->check_errors++; - /* Do not abort, continue checking the rest of this - * L2 table's entries */ + /* + * Do not abort, continue checking the rest of this + * L2 table's entries. + */ } else { res->corruptions--; res->corruptions_fixed++; - /* Skip marking the cluster as used - * (it is unused now) */ + /* + * Skip marking the cluster as used + * (it is unused now). + */ continue; } } @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, refcount_table_size, offset, s->cluster_size); if (ret < 0) { - goto fail; + return ret; } } break; @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, } } - g_free(l2_table); return 0; - -fail: - g_free(l2_table); - return ret; } /* -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Let's pass the whole L2 entry and not bother with L2E_COMPRESSED_OFFSET_SIZE_MASK. It also helps further refactoring that adds generic qcow2_parse_compressed_l2_entry() helper. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-3-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2.h | 1 - block/qcow2-cluster.c | 5 ++--- block/qcow2.c | 12 +++++++----- 3 files changed, 9 insertions(+), 9 deletions(-) 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 @@ typedef enum QCow2MetadataOverlap { #define L1E_OFFSET_MASK 0x00fffffffffffe00ULL #define L2E_OFFSET_MASK 0x00fffffffffffe00ULL -#define L2E_COMPRESSED_OFFSET_SIZE_MASK 0x3fffffffffffffffULL #define REFT_OFFSET_MASK 0xfffffffffffffe00ULL diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c index XXXXXXX..XXXXXXX 100644 --- a/block/qcow2-cluster.c +++ b/block/qcow2-cluster.c @@ -XXX,XX +XXX,XX @@ static int coroutine_fn do_perform_cow_write(BlockDriverState *bs, * offset needs to be aligned to a cluster boundary. * * If the cluster is unallocated then *host_offset will be 0. - * If the cluster is compressed then *host_offset will contain the - * complete compressed cluster descriptor. + * If the cluster is compressed then *host_offset will contain the l2 entry. * * On entry, *bytes is the maximum number of contiguous bytes starting at * offset that we are interested in. @@ -XXX,XX +XXX,XX @@ int qcow2_get_host_offset(BlockDriverState *bs, uint64_t offset, ret = -EIO; goto fail; } - *host_offset = l2_entry & L2E_COMPRESSED_OFFSET_SIZE_MASK; + *host_offset = l2_entry; break; case QCOW2_SUBCLUSTER_ZERO_PLAIN: case QCOW2_SUBCLUSTER_UNALLOCATED_PLAIN: 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 @@ typedef struct { static int coroutine_fn qcow2_co_preadv_compressed(BlockDriverState *bs, - uint64_t cluster_descriptor, + uint64_t l2_entry, uint64_t offset, uint64_t bytes, QEMUIOVector *qiov, @@ -XXX,XX +XXX,XX @@ typedef struct Qcow2AioTask { BlockDriverState *bs; QCow2SubclusterType subcluster_type; /* only for read */ - uint64_t host_offset; /* or full descriptor in compressed clusters */ + uint64_t host_offset; /* or l2_entry for compressed read */ uint64_t offset; uint64_t bytes; QEMUIOVector *qiov; @@ -XXX,XX +XXX,XX @@ qcow2_co_pwritev_compressed_part(BlockDriverState *bs, static int coroutine_fn qcow2_co_preadv_compressed(BlockDriverState *bs, - uint64_t cluster_descriptor, + uint64_t l2_entry, uint64_t offset, uint64_t bytes, QEMUIOVector *qiov, @@ -XXX,XX +XXX,XX @@ qcow2_co_preadv_compressed(BlockDriverState *bs, uint8_t *buf, *out_buf; int offset_in_cluster = offset_into_cluster(s, offset); - coffset = cluster_descriptor & s->cluster_offset_mask; - nb_csectors = ((cluster_descriptor >> s->csize_shift) & s->csize_mask) + 1; + assert(qcow2_get_cluster_type(bs, l2_entry) == QCOW2_CLUSTER_COMPRESSED); + + coffset = l2_entry & s->cluster_offset_mask; + nb_csectors = ((l2_entry >> s->csize_shift) & s->csize_mask) + 1; csize = nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE - (coffset & ~QCOW2_COMPRESSED_SECTOR_MASK); -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Add helper to parse compressed l2_entry and use it everywhere instead of open-coding. Note, that in most places we move to precise coffset/csize instead of sector-aligned. Still it should work good enough for updating refcounts. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-4-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2.h | 3 ++- block/qcow2-cluster.c | 15 +++++++++++++++ block/qcow2-refcount.c | 36 +++++++++++++++++------------------- block/qcow2.c | 9 ++------- 4 files changed, 36 insertions(+), 27 deletions(-) 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 @@ /* Defined in the qcow2 spec (compressed cluster descriptor) */ #define QCOW2_COMPRESSED_SECTOR_SIZE 512U -#define QCOW2_COMPRESSED_SECTOR_MASK (~(QCOW2_COMPRESSED_SECTOR_SIZE - 1ULL)) /* Must be at least 2 to cover COW */ #define MIN_L2_CACHE_SIZE 2 /* cache entries */ @@ -XXX,XX +XXX,XX @@ int qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs, uint64_t offset, int compressed_size, uint64_t *host_offset); +void qcow2_parse_compressed_l2_entry(BlockDriverState *bs, uint64_t l2_entry, + uint64_t *coffset, int *csize); int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m); void qcow2_alloc_cluster_abort(BlockDriverState *bs, QCowL2Meta *m); 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 @@ fail: g_free(l1_table); return ret; } + +void qcow2_parse_compressed_l2_entry(BlockDriverState *bs, uint64_t l2_entry, + uint64_t *coffset, int *csize) +{ + BDRVQcow2State *s = bs->opaque; + int nb_csectors; + + assert(qcow2_get_cluster_type(bs, l2_entry) == QCOW2_CLUSTER_COMPRESSED); + + *coffset = l2_entry & s->cluster_offset_mask; + + nb_csectors = ((l2_entry >> s->csize_shift) & s->csize_mask) + 1; + *csize = nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE - + (*coffset & (QCOW2_COMPRESSED_SECTOR_SIZE - 1)); +} 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 @@ void qcow2_free_any_cluster(BlockDriverState *bs, uint64_t l2_entry, switch (ctype) { case QCOW2_CLUSTER_COMPRESSED: { - int64_t offset = (l2_entry & s->cluster_offset_mask) - & QCOW2_COMPRESSED_SECTOR_MASK; - int size = QCOW2_COMPRESSED_SECTOR_SIZE * - (((l2_entry >> s->csize_shift) & s->csize_mask) + 1); - qcow2_free_clusters(bs, offset, size, type); + uint64_t coffset; + int csize; + + qcow2_parse_compressed_l2_entry(bs, l2_entry, &coffset, &csize); + qcow2_free_clusters(bs, coffset, csize, type); } break; case QCOW2_CLUSTER_NORMAL: @@ -XXX,XX +XXX,XX @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs, bool l1_allocated = false; int64_t old_entry, old_l2_offset; unsigned slice, slice_size2, n_slices; - int i, j, l1_modified = 0, nb_csectors; + int i, j, l1_modified = 0; int ret; assert(addend >= -1 && addend <= 1); @@ -XXX,XX +XXX,XX @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs, switch (qcow2_get_cluster_type(bs, entry)) { case QCOW2_CLUSTER_COMPRESSED: - nb_csectors = ((entry >> s->csize_shift) & - s->csize_mask) + 1; if (addend != 0) { - uint64_t coffset = (entry & s->cluster_offset_mask) - & QCOW2_COMPRESSED_SECTOR_MASK; + uint64_t coffset; + int csize; + + qcow2_parse_compressed_l2_entry(bs, entry, + &coffset, &csize); ret = update_refcount( - bs, coffset, - nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE, + bs, coffset, csize, abs(addend), addend < 0, QCOW2_DISCARD_SNAPSHOT); if (ret < 0) { @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, BDRVQcow2State *s = bs->opaque; uint64_t l2_entry; uint64_t next_contiguous_offset = 0; - int i, nb_csectors, ret; + int i, ret; size_t l2_size_bytes = s->l2_size * l2_entry_size(s); g_autofree uint64_t *l2_table = g_malloc(l2_size_bytes); @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, /* Do the actual checks */ for (i = 0; i < s->l2_size; i++) { + uint64_t coffset; + int csize; l2_entry = get_l2_entry(s, l2_table, i); switch (qcow2_get_cluster_type(bs, l2_entry)) { @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, } /* Mark cluster as used */ - nb_csectors = ((l2_entry >> s->csize_shift) & - s->csize_mask) + 1; - l2_entry &= s->cluster_offset_mask; + qcow2_parse_compressed_l2_entry(bs, l2_entry, &coffset, &csize); ret = qcow2_inc_refcounts_imrt( - bs, res, refcount_table, refcount_table_size, - l2_entry & QCOW2_COMPRESSED_SECTOR_MASK, - nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE); + bs, res, refcount_table, refcount_table_size, coffset, csize); if (ret < 0) { return 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 @@ qcow2_co_preadv_compressed(BlockDriverState *bs, size_t qiov_offset) { BDRVQcow2State *s = bs->opaque; - int ret = 0, csize, nb_csectors; + int ret = 0, csize; uint64_t coffset; uint8_t *buf, *out_buf; int offset_in_cluster = offset_into_cluster(s, offset); - assert(qcow2_get_cluster_type(bs, l2_entry) == QCOW2_CLUSTER_COMPRESSED); - - coffset = l2_entry & s->cluster_offset_mask; - nb_csectors = ((l2_entry >> s->csize_shift) & s->csize_mask) + 1; - csize = nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE - - (coffset & ~QCOW2_COMPRESSED_SECTOR_MASK); + qcow2_parse_compressed_l2_entry(bs, l2_entry, &coffset, &csize); buf = g_try_malloc(csize); if (!buf) { -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Split fix_l2_entry_by_zero() out of check_refcounts_l2() to be reused in further patch. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-5-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2-refcount.c | 87 +++++++++++++++++++++++++++++------------- 1 file changed, 60 insertions(+), 27 deletions(-) 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 @@ enum { CHECK_FRAG_INFO = 0x2, /* update BlockFragInfo counters */ }; +/* + * Fix L2 entry by making it QCOW2_CLUSTER_ZERO_PLAIN. + * + * This function decrements res->corruptions on success, so the caller is + * responsible to increment res->corruptions prior to the call. + * + * On failure in-memory @l2_table may be modified. + */ +static int fix_l2_entry_by_zero(BlockDriverState *bs, BdrvCheckResult *res, + uint64_t l2_offset, + uint64_t *l2_table, int l2_index, bool active, + bool *metadata_overlap) +{ + BDRVQcow2State *s = bs->opaque; + int ret; + int idx = l2_index * (l2_entry_size(s) / sizeof(uint64_t)); + uint64_t l2e_offset = l2_offset + (uint64_t)l2_index * l2_entry_size(s); + int ign = active ? QCOW2_OL_ACTIVE_L2 : QCOW2_OL_INACTIVE_L2; + uint64_t l2_entry = has_subclusters(s) ? 0 : QCOW_OFLAG_ZERO; + + set_l2_entry(s, l2_table, l2_index, l2_entry); + ret = qcow2_pre_write_overlap_check(bs, ign, l2e_offset, l2_entry_size(s), + false); + if (metadata_overlap) { + *metadata_overlap = ret < 0; + } + if (ret < 0) { + fprintf(stderr, "ERROR: Overlap check failed\n"); + goto fail; + } + + ret = bdrv_pwrite_sync(bs->file, l2e_offset, &l2_table[idx], + l2_entry_size(s)); + if (ret < 0) { + fprintf(stderr, "ERROR: Failed to overwrite L2 " + "table entry: %s\n", strerror(-ret)); + goto fail; + } + + res->corruptions--; + res->corruptions_fixed++; + return 0; + +fail: + res->check_errors++; + return ret; +} + /* * Increases the refcount in the given refcount table for the all clusters * referenced in the L2 table. While doing so, performs some checks on L2 @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, int i, ret; size_t l2_size_bytes = s->l2_size * l2_entry_size(s); g_autofree uint64_t *l2_table = g_malloc(l2_size_bytes); + bool metadata_overlap; /* Read L2 table from disk */ ret = bdrv_pread(bs->file, l2_offset, l2_table, l2_size_bytes); @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, fix & BDRV_FIX_ERRORS ? "Repairing" : "ERROR", offset); if (fix & BDRV_FIX_ERRORS) { - int idx = i * (l2_entry_size(s) / sizeof(uint64_t)); - uint64_t l2e_offset = - l2_offset + (uint64_t)i * l2_entry_size(s); - int ign = active ? QCOW2_OL_ACTIVE_L2 : - QCOW2_OL_INACTIVE_L2; - - l2_entry = has_subclusters(s) ? 0 : QCOW_OFLAG_ZERO; - set_l2_entry(s, l2_table, i, l2_entry); - ret = qcow2_pre_write_overlap_check(bs, ign, - l2e_offset, l2_entry_size(s), false); - if (ret < 0) { - fprintf(stderr, "ERROR: Overlap check failed\n"); - res->check_errors++; + ret = fix_l2_entry_by_zero(bs, res, l2_offset, + l2_table, i, active, + &metadata_overlap); + if (metadata_overlap) { /* * Something is seriously wrong, so abort checking * this L2 table. @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, return ret; } - ret = bdrv_pwrite_sync(bs->file, l2e_offset, - &l2_table[idx], - l2_entry_size(s)); - if (ret < 0) { - fprintf(stderr, "ERROR: Failed to overwrite L2 " - "table entry: %s\n", strerror(-ret)); - res->check_errors++; - /* - * Do not abort, continue checking the rest of this - * L2 table's entries. - */ - } else { - res->corruptions--; - res->corruptions_fixed++; + if (ret == 0) { /* * Skip marking the cluster as used * (it is unused now). */ continue; } + + /* + * Failed to fix. + * Do not abort, continue checking the rest of this + * L2 table's entries. + */ } } else { fprintf(stderr, "ERROR offset=%" PRIx64 ": Data cluster is " -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> We'll reuse the function to fix wrong L2 entry bitmap. Support it now. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-6-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2-refcount.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) 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 @@ enum { }; /* - * Fix L2 entry by making it QCOW2_CLUSTER_ZERO_PLAIN. + * Fix L2 entry by making it QCOW2_CLUSTER_ZERO_PLAIN (or making all its present + * subclusters QCOW2_SUBCLUSTER_ZERO_PLAIN). * * This function decrements res->corruptions on success, so the caller is * responsible to increment res->corruptions prior to the call. @@ -XXX,XX +XXX,XX @@ static int fix_l2_entry_by_zero(BlockDriverState *bs, BdrvCheckResult *res, int idx = l2_index * (l2_entry_size(s) / sizeof(uint64_t)); uint64_t l2e_offset = l2_offset + (uint64_t)l2_index * l2_entry_size(s); int ign = active ? QCOW2_OL_ACTIVE_L2 : QCOW2_OL_INACTIVE_L2; - uint64_t l2_entry = has_subclusters(s) ? 0 : QCOW_OFLAG_ZERO; - set_l2_entry(s, l2_table, l2_index, l2_entry); + if (has_subclusters(s)) { + uint64_t l2_bitmap = get_l2_bitmap(s, l2_table, l2_index); + + /* Allocated subclusters become zero */ + l2_bitmap |= l2_bitmap << 32; + l2_bitmap &= QCOW_L2_BITMAP_ALL_ZEROES; + + set_l2_bitmap(s, l2_table, l2_index, l2_bitmap); + set_l2_entry(s, l2_table, l2_index, 0); + } else { + set_l2_entry(s, l2_table, l2_index, QCOW_OFLAG_ZERO); + } + ret = qcow2_pre_write_overlap_check(bs, ign, l2e_offset, l2_entry_size(s), false); if (metadata_overlap) { -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Check subcluster bitmap of the l2 entry for different types of clusters: - for compressed it must be zero - for allocated check consistency of two parts of the bitmap - for unallocated all subclusters should be unallocated (or zero-plain) Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Tested-by: Kirill Tkhai <ktkhai@virtuozzo.com> Message-Id: <20210914122454.141075-7-vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2-refcount.c | 28 ++++++++++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-) 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_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, int flags, BdrvCheckMode fix, bool active) { BDRVQcow2State *s = bs->opaque; - uint64_t l2_entry; + uint64_t l2_entry, l2_bitmap; uint64_t next_contiguous_offset = 0; int i, ret; size_t l2_size_bytes = s->l2_size * l2_entry_size(s); @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, uint64_t coffset; int csize; l2_entry = get_l2_entry(s, l2_table, i); + l2_bitmap = get_l2_bitmap(s, l2_table, i); switch (qcow2_get_cluster_type(bs, l2_entry)) { case QCOW2_CLUSTER_COMPRESSED: @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, break; } + if (l2_bitmap) { + fprintf(stderr, "ERROR compressed cluster %d with non-zero " + "subcluster allocation bitmap, entry=0x%" PRIx64 "\n", + i, l2_entry); + res->corruptions++; + break; + } + /* Mark cluster as used */ qcow2_parse_compressed_l2_entry(bs, l2_entry, &coffset, &csize); ret = qcow2_inc_refcounts_imrt( @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, { uint64_t offset = l2_entry & L2E_OFFSET_MASK; + if ((l2_bitmap >> 32) & l2_bitmap) { + res->corruptions++; + fprintf(stderr, "ERROR offset=%" PRIx64 ": Allocated " + "cluster has corrupted subcluster allocation bitmap\n", + offset); + } + /* Correct offsets are cluster aligned */ if (offset_into_cluster(s, offset)) { bool contains_data; res->corruptions++; if (has_subclusters(s)) { - uint64_t l2_bitmap = get_l2_bitmap(s, l2_table, i); contains_data = (l2_bitmap & QCOW_L2_BITMAP_ALL_ALLOC); } else { contains_data = !(l2_entry & QCOW_OFLAG_ZERO); @@ -XXX,XX +XXX,XX @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, } case QCOW2_CLUSTER_ZERO_PLAIN: + /* Impossible when image has subclusters */ + assert(!l2_bitmap); + break; + case QCOW2_CLUSTER_UNALLOCATED: + if (l2_bitmap & QCOW_L2_BITMAP_ALL_ALLOC) { + res->corruptions++; + fprintf(stderr, "ERROR: Unallocated " + "cluster has non-zero subcluster allocation map\n"); + } break; default: -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Tested-by: Kirill Tkhai <ktkhai@virtuozzo.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-8-vsementsov@virtuozzo.com> [hreitz: Separated `type` declaration from statements] Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2.h | 1 + block/qcow2-refcount.c | 14 +++++++++++++- 2 files changed, 14 insertions(+), 1 deletion(-) 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 @@ typedef enum QCow2MetadataOverlap { #define L1E_OFFSET_MASK 0x00fffffffffffe00ULL #define L2E_OFFSET_MASK 0x00fffffffffffe00ULL +#define L2E_STD_RESERVED_MASK 0x3f000000000001feULL #define REFT_OFFSET_MASK 0xfffffffffffffe00ULL 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_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res, for (i = 0; i < s->l2_size; i++) { uint64_t coffset; int csize; + QCow2ClusterType type; + l2_entry = get_l2_entry(s, l2_table, i); l2_bitmap = get_l2_bitmap(s, l2_table, i); + type = qcow2_get_cluster_type(bs, l2_entry); + + if (type != QCOW2_CLUSTER_COMPRESSED) { + /* Check reserved bits of Standard Cluster Descriptor */ + if (l2_entry & L2E_STD_RESERVED_MASK) { + fprintf(stderr, "ERROR found l2 entry with reserved bits set: " + "%" PRIx64 "\n", l2_entry); + res->corruptions++; + } + } - switch (qcow2_get_cluster_type(bs, l2_entry)) { + switch (type) { case QCOW2_CLUSTER_COMPRESSED: /* Compressed clusters don't have QCOW_OFLAG_COPIED */ if (l2_entry & QCOW_OFLAG_COPIED) { -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> - use g_autofree for l1_table - better name for size in bytes variable - reduce code blocks nesting - whitespaces, braces, newlines Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-9-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2-refcount.c | 98 +++++++++++++++++++++--------------------- 1 file changed, 50 insertions(+), 48 deletions(-) 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_refcounts_l1(BlockDriverState *bs, int flags, BdrvCheckMode fix, bool active) { BDRVQcow2State *s = bs->opaque; - uint64_t *l1_table = NULL, l2_offset, l1_size2; + size_t l1_size_bytes = l1_size * L1E_SIZE; + g_autofree uint64_t *l1_table = NULL; + uint64_t l2_offset; int i, ret; - l1_size2 = l1_size * L1E_SIZE; + if (!l1_size) { + return 0; + } /* Mark L1 table as used */ ret = qcow2_inc_refcounts_imrt(bs, res, refcount_table, refcount_table_size, - l1_table_offset, l1_size2); + l1_table_offset, l1_size_bytes); if (ret < 0) { - goto fail; + return ret; + } + + l1_table = g_try_malloc(l1_size_bytes); + if (l1_table == NULL) { + res->check_errors++; + return -ENOMEM; } /* Read L1 table entries from disk */ - if (l1_size2 > 0) { - l1_table = g_try_malloc(l1_size2); - if (l1_table == NULL) { - ret = -ENOMEM; - res->check_errors++; - goto fail; - } - ret = bdrv_pread(bs->file, l1_table_offset, l1_table, l1_size2); - if (ret < 0) { - fprintf(stderr, "ERROR: I/O error in check_refcounts_l1\n"); - res->check_errors++; - goto fail; - } - for(i = 0;i < l1_size; i++) - be64_to_cpus(&l1_table[i]); + ret = bdrv_pread(bs->file, l1_table_offset, l1_table, l1_size_bytes); + if (ret < 0) { + fprintf(stderr, "ERROR: I/O error in check_refcounts_l1\n"); + res->check_errors++; + return ret; + } + + for (i = 0; i < l1_size; i++) { + be64_to_cpus(&l1_table[i]); } /* Do the actual checks */ - for(i = 0; i < l1_size; i++) { - l2_offset = l1_table[i]; - if (l2_offset) { - /* Mark L2 table as used */ - l2_offset &= L1E_OFFSET_MASK; - ret = qcow2_inc_refcounts_imrt(bs, res, - refcount_table, refcount_table_size, - l2_offset, s->cluster_size); - if (ret < 0) { - goto fail; - } + for (i = 0; i < l1_size; i++) { + if (!l1_table[i]) { + continue; + } - /* L2 tables are cluster aligned */ - if (offset_into_cluster(s, l2_offset)) { - fprintf(stderr, "ERROR l2_offset=%" PRIx64 ": Table is not " - "cluster aligned; L1 entry corrupted\n", l2_offset); - res->corruptions++; - } + l2_offset = l1_table[i] & L1E_OFFSET_MASK; - /* Process and check L2 entries */ - ret = check_refcounts_l2(bs, res, refcount_table, - refcount_table_size, l2_offset, flags, - fix, active); - if (ret < 0) { - goto fail; - } + /* Mark L2 table as used */ + ret = qcow2_inc_refcounts_imrt(bs, res, + refcount_table, refcount_table_size, + l2_offset, s->cluster_size); + if (ret < 0) { + return ret; + } + + /* L2 tables are cluster aligned */ + if (offset_into_cluster(s, l2_offset)) { + fprintf(stderr, "ERROR l2_offset=%" PRIx64 ": Table is not " + "cluster aligned; L1 entry corrupted\n", l2_offset); + res->corruptions++; + } + + /* Process and check L2 entries */ + ret = check_refcounts_l2(bs, res, refcount_table, + refcount_table_size, l2_offset, flags, + fix, active); + if (ret < 0) { + return ret; } } - g_free(l1_table); - return 0; -fail: - g_free(l1_table); - return ret; + return 0; } /* -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Tested-by: Kirill Tkhai <ktkhai@virtuozzo.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-10-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2.h | 1 + block/qcow2-refcount.c | 6 ++++++ 2 files changed, 7 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 @@ typedef enum QCow2MetadataOverlap { (QCOW2_OL_CACHED | QCOW2_OL_INACTIVE_L2) #define L1E_OFFSET_MASK 0x00fffffffffffe00ULL +#define L1E_RESERVED_MASK 0x7f000000000001ffULL #define L2E_OFFSET_MASK 0x00fffffffffffe00ULL #define L2E_STD_RESERVED_MASK 0x3f000000000001feULL 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_refcounts_l1(BlockDriverState *bs, continue; } + if (l1_table[i] & L1E_RESERVED_MASK) { + fprintf(stderr, "ERROR found L1 entry with reserved bits set: " + "%" PRIx64 "\n", l1_table[i]); + res->corruptions++; + } + l2_offset = l1_table[i] & L1E_OFFSET_MASK; /* Mark L2 table as used */ -- 2.31.1 From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Split checking for reserved bits out of aligned offset check. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Eric Blake <eblake@redhat.com> Tested-by: Kirill Tkhai <ktkhai@virtuozzo.com> Reviewed-by: Hanna Reitz <hreitz@redhat.com> Message-Id: <20210914122454.141075-11-vsementsov@virtuozzo.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- block/qcow2.h | 1 + block/qcow2-refcount.c | 10 +++++++++- 2 files changed, 10 insertions(+), 1 deletion(-) 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 @@ typedef enum QCow2MetadataOverlap { #define L2E_STD_RESERVED_MASK 0x3f000000000001feULL #define REFT_OFFSET_MASK 0xfffffffffffffe00ULL +#define REFT_RESERVED_MASK 0x1ffULL #define INV_OFFSET (-1ULL) 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, for(i = 0; i < s->refcount_table_size; i++) { uint64_t offset, cluster; - offset = s->refcount_table[i]; + offset = s->refcount_table[i] & REFT_OFFSET_MASK; cluster = offset >> s->cluster_bits; + if (s->refcount_table[i] & REFT_RESERVED_MASK) { + fprintf(stderr, "ERROR refcount table entry %" PRId64 " has " + "reserved bits set\n", i); + res->corruptions++; + *rebuild = true; + continue; + } + /* Refcount blocks are cluster aligned */ if (offset_into_cluster(s, offset)) { fprintf(stderr, "ERROR refcount block %" PRId64 " is not " -- 2.31.1 From: Eric Blake <eblake@redhat.com> Although we have long supported 'qemu-img convert -o backing_file=foo,backing_fmt=bar', the fact that we have a shortcut -B for backing_file but none for backing_fmt has made it more likely that users accidentally run into: qemu-img: warning: Deprecated use of backing file without explicit backing format when using -B instead of -o. For similarity with other qemu-img commands, such as create and compare, add '-F $fmt' as the shorthand for '-o backing_fmt=$fmt'. Update iotest 122 for coverage of both spellings. Signed-off-by: Eric Blake <eblake@redhat.com> Message-Id: <20210913131735.1948339-1-eblake@redhat.com> Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by: Hanna Reitz <hreitz@redhat.com> --- docs/tools/qemu-img.rst | 4 ++-- qemu-img.c | 10 +++++++--- qemu-img-cmds.hx | 2 +- tests/qemu-iotests/122 | 2 +- 4 files changed, 11 insertions(+), 7 deletions(-) diff --git a/docs/tools/qemu-img.rst b/docs/tools/qemu-img.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/tools/qemu-img.rst +++ b/docs/tools/qemu-img.rst @@ -XXX,XX +XXX,XX @@ Command description: 4 Error on reading data -.. option:: convert [--object OBJECTDEF] [--image-opts] [--target-image-opts] [--target-is-zero] [--bitmaps [--skip-broken-bitmaps]] [-U] [-C] [-c] [-p] [-q] [-n] [-f FMT] [-t CACHE] [-T SRC_CACHE] [-O OUTPUT_FMT] [-B BACKING_FILE] [-o OPTIONS] [-l SNAPSHOT_PARAM] [-S SPARSE_SIZE] [-r RATE_LIMIT] [-m NUM_COROUTINES] [-W] FILENAME [FILENAME2 [...]] OUTPUT_FILENAME +.. option:: convert [--object OBJECTDEF] [--image-opts] [--target-image-opts] [--target-is-zero] [--bitmaps [--skip-broken-bitmaps]] [-U] [-C] [-c] [-p] [-q] [-n] [-f FMT] [-t CACHE] [-T SRC_CACHE] [-O OUTPUT_FMT] [-B BACKING_FILE [-F backing_fmt]] [-o OPTIONS] [-l SNAPSHOT_PARAM] [-S SPARSE_SIZE] [-r RATE_LIMIT] [-m NUM_COROUTINES] [-W] FILENAME [FILENAME2 [...]] OUTPUT_FILENAME Convert the disk image *FILENAME* or a snapshot *SNAPSHOT_PARAM* to disk image *OUTPUT_FILENAME* using format *OUTPUT_FMT*. It can @@ -XXX,XX +XXX,XX @@ Command description: You can use the *BACKING_FILE* option to force the output image to be created as a copy on write image of the specified base image; the *BACKING_FILE* should have the same content as the input's base image, - however the path, image format, etc may differ. + however the path, image format (as given by *BACKING_FMT*), etc may differ. If a relative path name is given, the backing file is looked up relative to the directory containing *OUTPUT_FILENAME*. 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_convert(int argc, char **argv) int c, bs_i, flags, src_flags = BDRV_O_NO_SHARE; const char *fmt = NULL, *out_fmt = NULL, *cache = "unsafe", *src_cache = BDRV_DEFAULT_CACHE, *out_baseimg = NULL, - *out_filename, *out_baseimg_param, *snapshot_name = NULL; + *out_filename, *out_baseimg_param, *snapshot_name = NULL, + *backing_fmt = NULL; BlockDriver *drv = NULL, *proto_drv = NULL; BlockDriverInfo bdi; BlockDriverState *out_bs; @@ -XXX,XX +XXX,XX @@ static int img_convert(int argc, char **argv) {"skip-broken-bitmaps", no_argument, 0, OPTION_SKIP_BROKEN}, {0, 0, 0, 0} }; - c = getopt_long(argc, argv, ":hf:O:B:Cco:l:S:pt:T:qnm:WUr:", + c = getopt_long(argc, argv, ":hf:O:B:CcF:o:l:S:pt:T:qnm:WUr:", long_options, NULL); if (c == -1) { break; @@ -XXX,XX +XXX,XX @@ static int img_convert(int argc, char **argv) case 'c': s.compressed = true; break; + case 'F': + backing_fmt = optarg; + break; case 'o': if (accumulate_options(&options, optarg) < 0) { goto fail_getopt; @@ -XXX,XX +XXX,XX @@ static int img_convert(int argc, char **argv) qemu_opt_set_number(opts, BLOCK_OPT_SIZE, s.total_sectors * BDRV_SECTOR_SIZE, &error_abort); - ret = add_old_style_options(out_fmt, opts, out_baseimg, NULL); + ret = add_old_style_options(out_fmt, opts, out_baseimg, backing_fmt); if (ret < 0) { goto out; } diff --git a/qemu-img-cmds.hx b/qemu-img-cmds.hx index XXXXXXX..XXXXXXX 100644 --- a/qemu-img-cmds.hx +++ b/qemu-img-cmds.hx @@ -XXX,XX +XXX,XX @@ SRST ERST DEF("convert", img_convert, - "convert [--object objectdef] [--image-opts] [--target-image-opts] [--target-is-zero] [--bitmaps] [-U] [-C] [-c] [-p] [-q] [-n] [-f fmt] [-t cache] [-T src_cache] [-O output_fmt] [-B backing_file] [-o options] [-l snapshot_param] [-S sparse_size] [-r rate_limit] [-m num_coroutines] [-W] [--salvage] filename [filename2 [...]] output_filename") + "convert [--object objectdef] [--image-opts] [--target-image-opts] [--target-is-zero] [--bitmaps] [-U] [-C] [-c] [-p] [-q] [-n] [-f fmt] [-t cache] [-T src_cache] [-O output_fmt] [-B backing_file [-F backing_fmt]] [-o options] [-l snapshot_param] [-S sparse_size] [-r rate_limit] [-m num_coroutines] [-W] [--salvage] filename [filename2 [...]] output_filename") SRST .. option:: convert [--object OBJECTDEF] [--image-opts] [--target-image-opts] [--target-is-zero] [--bitmaps] [-U] [-C] [-c] [-p] [-q] [-n] [-f FMT] [-t CACHE] [-T SRC_CACHE] [-O OUTPUT_FMT] [-B BACKING_FILE] [-o OPTIONS] [-l SNAPSHOT_PARAM] [-S SPARSE_SIZE] [-r RATE_LIMIT] [-m NUM_COROUTINES] [-W] [--salvage] FILENAME [FILENAME2 [...]] OUTPUT_FILENAME ERST diff --git a/tests/qemu-iotests/122 b/tests/qemu-iotests/122 index XXXXXXX..XXXXXXX 100755 --- a/tests/qemu-iotests/122 +++ b/tests/qemu-iotests/122 @@ -XXX,XX +XXX,XX @@ echo _make_test_img -b "$TEST_IMG".base -F $IMGFMT $QEMU_IO -c "write -P 0 0 3M" "$TEST_IMG" 2>&1 | _filter_qemu_io | _filter_testdir -$QEMU_IMG convert -O $IMGFMT -B "$TEST_IMG".base -o backing_fmt=$IMGFMT \ +$QEMU_IMG convert -O $IMGFMT -B "$TEST_IMG".base -F $IMGFMT \ "$TEST_IMG" "$TEST_IMG".orig $QEMU_IO -c "read -P 0 0 3M" "$TEST_IMG".orig 2>&1 | _filter_qemu_io | _filter_testdir $QEMU_IMG convert -O $IMGFMT -c -B "$TEST_IMG".base -o backing_fmt=$IMGFMT \ -- 2.31.1