preallocate filter
PATCH v4 3/10 block/io: split out bdrv_find_conflicting_request
In reply to Vladimir Sementsov-OgievskiyTo be reused in separate.
--- block/io.c | 71 +++++++++++++++++++++++++++++++----------------------- 1 file changed, 41 insertions(+), 30 deletions(-)
block/io.c +41 -30
@@ -727,43 +727,54 @@ static bool tracked_request_overlaps(BdrvTrackedRequest *req,return true; } +/* Called with self->bs->reqs_lock held */ +static BdrvTrackedRequest * +bdrv_find_conflicting_request (BdrvTrackedRequest *self) +{ +BdrvTrackedRequest *req; + + QLIST_FOREACH(req, &self->bs->tracked_requests, list) { +if (req == self || (!req->serialising && !self->serialising)) { +continue ; + } +if (tracked_request_overlaps (req, self->overlap_offset, + self->overlap_bytes)) + { +/* +* Hitting this means there was a reentrant request, for +* example, a block driver issuing nested requests. This must +* never happen since it means deadlock. +*/ +assert (qemu_coroutine_self () != req->co); + +/* +* If the request 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 (!req->waiting_for) { +return req; + } + } + } + +return NULL ; +} + static bool coroutine_fnbdrv_wait_serialising_requests_locked (BlockDriverState *bs,BdrvTrackedRequest *self) {BdrvTrackedRequest *req; -bool retry;bool waited = false; -do { - retry = false; - QLIST_FOREACH(req, &bs->tracked_requests, list) { -if (req == self || (!req->serialising && !self->serialising)) { -continue ; - } -if (tracked_request_overlaps (req, self->overlap_offset, - self->overlap_bytes)) - { -/* Hitting this means there was a reentrant request, for -* example, a block driver issuing nested requests. This must -* never happen since it means deadlock. -*/ -assert (qemu_coroutine_self () != req->co); - -/* If the request 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 (!req->waiting_for) { - self->waiting_for = req; -qemu_co_queue_wait (&req->wait_queue, &bs->reqs_lock); - self->waiting_for =NULL ; - retry = true; - waited = true; -break ; - } - } - } - }while (retry); +while ((req =bdrv_find_conflicting_request (self))) { + self->waiting_for = req; +qemu_co_queue_wait (&req->wait_queue, &bs->reqs_lock); + self->waiting_for =NULL ; + waited = true; + } +return waited; }
Signature · 1 line
2.21.3
v4 cover preallocate filter
Hi all!
Here is a filter, which does preallocation on write.
In Virtuozzo we have to deal with some custom distributed storage solution, where allocation is relatively expensive operation. We have to workaround it in Qemu, so here is a new filter.
v4:
01-04: add r-bs
05: add coroutine_fn tag
06: use QEMU_LOCK_GUARD and fix reqs_lock leak
07: grammar
08-10: add r-b
I decided to not resend patches 11-12, for which I don't know how they can be applied universally. I think I'll keep them downstream only for now, and probably drop in future if we can insert filter in libvirt.
Vladimir Sementsov-Ogievskiy (10):
block: simplify comment to BDRV_REQ_SERIALISING block/io.c: drop assertion on double waiting for request serialisation block/io: split out bdrv_find_conflicting_request block/io: bdrv_wait_serialising_requests_locked: drop extra bs arg
block: bdrv_mark_request_serialising: split non-waiting function
block: introduce BDRV_REQ_NO_WAIT flag
block: introduce preallocate filter
iotests.py: add verify_o_direct helper
iotests.py: add filter_img_check
iotests: add 298 to test new preallocate filter driver
docs/system/qemu-block-drivers.rst.inc | 26 +++ qapi/block-core.json | 20 +- include/block/block.h | 20 +- include/block/block_int.h | 3 +- block/file-posix.c | 2 +- block/io.c | 130 ++++++----- block/preallocate.c | 291 +++++++++++++++++++++++++ block/Makefile.objs | 1 + tests/qemu-iotests/298 | 50 +++++ tests/qemu-iotests/298.out | 6 + tests/qemu-iotests/group | 1 + tests/qemu-iotests/iotests.py | 10 + 12 files changed, 492 insertions(+), 68 deletions(-) create mode 100644 block/preallocate.c create mode 100644 tests/qemu-iotests/298 create mode 100644 tests/qemu-iotests/298.out
Signature · 1 line
2.21.3
PATCH v4 4/10 block/io: bdrv_wait_serialising_requests_locked: drop extra bs arg
bs is linked in req, so no needs to pass it separately. Most of tracked-requests API doesn't have bs argument. Actually, after this patch only tracked_request_begin has it, but it's for purpose.
While being here, also add a comment about what "_locked" is.
--- block/io.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
block/io.c +5 -5
@@ -761,16 +761,16 @@ bdrv_find_conflicting_request(BdrvTrackedRequest *self)return NULL ; } +/* Called with self->bs->reqs_lock held */ static bool coroutine_fn -bdrv_wait_serialising_requests_locked (BlockDriverState *bs, -BdrvTrackedRequest *self) +bdrv_wait_serialising_requests_locked (BdrvTrackedRequest *self) {BdrvTrackedRequest *req;bool waited = false;while ((req =bdrv_find_conflicting_request (self))) { self->waiting_for = req; -qemu_co_queue_wait (&req->wait_queue, &bs->reqs_lock); +qemu_co_queue_wait (&req->wait_queue, &self->bs->reqs_lock); self->waiting_for =NULL ; waited = true; }
@@ -794,7 +794,7 @@ bool bdrv_mark_request_serialising(BdrvTrackedRequest *req, uint64_t align) req->overlap_offset =MIN (req->overlap_offset, overlap_offset); req->overlap_bytes =MAX (req->overlap_bytes, overlap_bytes); - waited =bdrv_wait_serialising_requests_locked (bs, req); + waited =bdrv_wait_serialising_requests_locked (req);qemu_co_mutex_unlock (&bs->reqs_lock);return waited; }
@@ -876,7 +876,7 @@ static bool coroutine_fn bdrv_wait_serialising_requests(BdrvTrackedRequest *self }qemu_co_mutex_lock (&bs->reqs_lock); - waited =bdrv_wait_serialising_requests_locked (bs, self); + waited =bdrv_wait_serialising_requests_locked (self);qemu_co_mutex_unlock (&bs->reqs_lock);return waited;
Signature · 1 line
2.21.3
PATCH v4 1/10 block: simplify comment to BDRV_REQ_SERIALISING
In reply to Vladimir Sementsov-Ogievskiy1. BDRV_REQ_NO_SERIALISING doesn't exist already, don't mention it.
2. We are going to add one more user of BDRV_REQ_SERIALISING, so comment about backup becomes a bit confusing here. The use case in backup is documented in block/backup.c, so let's just drop duplication here.
3. The fact that BDRV_REQ_SERIALISING is only for write requests is omitted. Add a note.
--- include/block/block.h | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-)
include/block/block.h +1 -10
@@ -53,16 +53,7 @@ typedef enum {* content. */ BDRV_REQ_WRITE_UNCHANGED =0x40 , -/* -* BDRV_REQ_SERIALISING forces request serialisation for writes. -* It is used to ensure that writes to the backing file of a backup process -* target cannot race with a read of the backup target that defers to the -* backing file. -* -* Note, that BDRV_REQ_SERIALISING is _not_ opposite in meaning to -* BDRV_REQ_NO_SERIALISING. A more descriptive name for the latter might be -* _DO_NOT_WAIT_FOR_SERIALISING, except that is too long. -*/ +/* Forces request serialisation. Use only with write requests. */ BDRV_REQ_SERIALISING = 0x80,/* Execute the request only if the operation can be offloaded or otherwise
Signature · 1 line
2.21.3
PATCH v4 5/10 block: bdrv_mark_request_serialising: split non-waiting function
In reply to Vladimir Sementsov-OgievskiyWe'll need a separate function, which will only "mark" request serialising with specified align but not wait for conflicting requests. So, it will be like old bdrv_mark_request_serialising(), before merging bdrv_wait_serialising_requests_locked() into it.
To reduce the possible mess, let's do the following:
Public function that does both marking and waiting will be called bdrv_make_request_serialising, and private function which will only "mark" will be called tracked_request_set_serialising().
--- include/block/block_int.h | 3 ++- block/file-posix.c | 2 +- block/io.c | 35 +++++++++++++++++++++++------------ 3 files changed, 26 insertions(+), 14 deletions(-)
include/block/block_int.h +2 -1
@@ -1039,7 +1039,8 @@ extern unsigned int bdrv_drain_all_count;void bdrv_apply_subtree_drain (BdrvChild *child,BlockDriverState *new_parent);void bdrv_unapply_subtree_drain (BdrvChild *child,BlockDriverState *old_parent); -bool coroutine_fn bdrv_mark_request_serialising(BdrvTrackedRequest *req, uint64_t align); +bool coroutine_fn bdrv_make_request_serialising(BdrvTrackedRequest *req, +uint64_t align); BdrvTrackedRequest *coroutine_fn bdrv_co_get_self_request(BlockDriverState *bs);int get_tmp_filename (char *filename,int size);
block/file-posix.c +1 -1
@@ -2957,7 +2957,7 @@ raw_do_pwrite_zeroes(BlockDriverState *bs, int64_t offset, int bytes, req->bytes = end - req->offset; req->overlap_bytes = req->bytes; - bdrv_mark_request_serialising(req, bs->bl.request_alignment); + bdrv_make_request_serialising(req, bs->bl.request_alignment); }#endif
block/io.c +23 -12
@@ -778,15 +778,14 @@ bdrv_wait_serialising_requests_locked(BdrvTrackedRequest *self)return waited; } -bool bdrv_mark_request_serialising (BdrvTrackedRequest *req,uint64_t align) +/* Called with req->bs->reqs_lock held */ +static void tracked_request_set_serialising (BdrvTrackedRequest *req, +uint64_t align) { -BlockDriverState *bs = req->bs;int64_t overlap_offset = req->offset & ~(align -1 );uint64_t overlap_bytes =ROUND_UP (req->offset + req->bytes, align) - overlap_offset; -bool waited; -qemu_co_mutex_lock (&bs->reqs_lock);if (!req->serialising) {atomic_inc (&req->bs->serialising_in_flight); req->serialising = true;
@@ -794,9 +793,6 @@ bool bdrv_mark_request_serialising(BdrvTrackedRequest *req, uint64_t align) req->overlap_offset =MIN (req->overlap_offset, overlap_offset); req->overlap_bytes =MAX (req->overlap_bytes, overlap_bytes); - waited =bdrv_wait_serialising_requests_locked (req); -qemu_co_mutex_unlock (&bs->reqs_lock); -return waited; }/**
@@ -882,6 +878,21 @@ static bool coroutine_fn bdrv_wait_serialising_requests(BdrvTrackedRequest *selfreturn waited; } +bool coroutine_fn bdrv_make_request_serialising(BdrvTrackedRequest *req, +uint64_t align) +{ +bool waited; + +qemu_co_mutex_lock (&req->bs->reqs_lock); + +tracked_request_set_serialising (req, align); + waited =bdrv_wait_serialising_requests_locked (req); + +qemu_co_mutex_unlock (&req->bs->reqs_lock); + +return waited; +} +static int bdrv_check_byte_request (BlockDriverState *bs,int64_t offset,size_t size) {
@@ -1492,7 +1503,7 @@ static int coroutine_fn bdrv_aligned_preadv(BdrvChild *child,* with each other for the same cluster. For example, in copy-on-read * it ensures that the CoR read and write operations are atomic and * guest writes cannot interleave between them. */ -bdrv_mark_request_serialising (req,bdrv_get_cluster_size (bs)); +bdrv_make_request_serialising (req,bdrv_get_cluster_size (bs)); }else {bdrv_wait_serialising_requests (req); }
@@ -1903,7 +1914,7 @@ bdrv_co_write_req_prepare(BdrvChild *child, int64_t offset, uint64_t bytes,assert (!(flags & ~BDRV_REQ_MASK ));if (flags &BDRV_REQ_SERIALISING ) { -bdrv_mark_request_serialising (req,bdrv_get_cluster_size (bs)); +bdrv_make_request_serialising (req,bdrv_get_cluster_size (bs)); }else {bdrv_wait_serialising_requests (req); }
@@ -2069,7 +2080,7 @@ static int coroutine_fn bdrv_co_do_zero_pwritev(BdrvChild *child, padding =bdrv_init_padding (bs, offset, bytes, &pad);if (padding) { -bdrv_mark_request_serialising (req, align); +bdrv_make_request_serialising (req, align);bdrv_padding_rmw_read (child, req, &pad, true);
@@ -2183,7 +2194,7 @@ int coroutine_fn bdrv_co_pwritev_part(BdrvChild *child, }if (bdrv_pad_request (bs, &qiov, &qiov_offset, &offset, &bytes, &pad)) { -bdrv_mark_request_serialising (&req, align); +bdrv_make_request_serialising (&req, align);bdrv_padding_rmw_read (child, &req, &pad, false); }
@@ -3347,7 +3358,7 @@ int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact,* new area, we need to make sure that no write requests are made to it * concurrently or they might be overwritten by preallocation. */ if (new_bytes) { -bdrv_mark_request_serialising (&req,1 ); +bdrv_make_request_serialising (&req,1 ); }if (bs->read_only) {error_setg (errp,"Image is read-only" );
Signature · 1 line
2.21.3
PATCH v4 2/10 block/io.c: drop assertion on double waiting for request serialisation
In reply to Vladimir Sementsov-OgievskiyThe comments states, that on misaligned request we should have already been waiting. But for bdrv_padding_rmw_read, we called bdrv_mark_request_serialising with align = request_alignment, and now we serialise with align = cluster_size. So we may have to wait again with larger alignment.
Note, that the only user of BDRV_REQ_SERIALISING is backup which issues cluster-aligned requests, so seems the assertion should not fire for now. But it's wrong anyway.
--- block/io.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-)
block/io.c +1 -10
@@ -1881,7 +1881,6 @@ bdrv_co_write_req_prepare(BdrvChild *child, int64_t offset, uint64_t bytes, BdrvTrackedRequest *req, int flags) {BlockDriverState *bs = child->bs; -bool waited;int64_t end_sector =DIV_ROUND_UP (offset + bytes,BDRV_SECTOR_SIZE );if (bs->read_only) {
@@ -1893,15 +1892,7 @@ bdrv_co_write_req_prepare(BdrvChild *child, int64_t offset, uint64_t bytes,assert (!(flags & ~BDRV_REQ_MASK ));if (flags &BDRV_REQ_SERIALISING ) { - waited =bdrv_mark_request_serialising (req,bdrv_get_cluster_size (bs)); -/* -* For a misaligned request we should have already waited earlier, -* because we come after bdrv_padding_rmw_read which must be called -* with the request already marked as serialising. -*/ -assert (!waited || - (req->offset == req->overlap_offset && - req->bytes == req->overlap_bytes)); +bdrv_mark_request_serialising (req,bdrv_get_cluster_size (bs)); }else {bdrv_wait_serialising_requests (req); }
Signature · 1 line
2.21.3
PATCH v4 6/10 block: introduce BDRV_REQ_NO_WAIT flag
In reply to Vladimir Sementsov-OgievskiyAdd flag to make serialising request no wait: if there are conflicting requests, just return error immediately. It's will be used in upcoming preallocate filter.
--- include/block/block.h | 9 ++++++++- block/io.c | 11 ++++++++++- 2 files changed, 18 insertions(+), 2 deletions(-)
include/block/block.h +8 -1
@@ -67,8 +67,15 @@ typedef enum {* written to qiov parameter which may be NULL. */ BDRV_REQ_PREFETCH =0x200 , + +/* +* If we need to wait for other requests, just fail immediately. Used +* only together with BDRV_REQ_SERIALISING. +*/ +BDRV_REQ_NO_WAIT =0x400 , +/* Mask of valid flags */ -BDRV_REQ_MASK =0x3ff , +BDRV_REQ_MASK =0x7ff , } BdrvRequestFlags; typedef struct BlockSizes {
block/io.c +10 -1
@@ -1912,9 +1912,18 @@ bdrv_co_write_req_prepare(BdrvChild *child, int64_t offset, uint64_t bytes,assert (!(bs->open_flags &BDRV_O_INACTIVE ));assert ((bs->open_flags &BDRV_O_NO_IO ) ==0 );assert (!(flags & ~BDRV_REQ_MASK )); +assert (!((flags &BDRV_REQ_NO_WAIT ) && !(flags &BDRV_REQ_SERIALISING )));if (flags &BDRV_REQ_SERIALISING ) { -bdrv_make_request_serialising (req,bdrv_get_cluster_size (bs)); +QEMU_LOCK_GUARD (&bs->reqs_lock); + +tracked_request_set_serialising (req,bdrv_get_cluster_size (bs)); + +if ((flags &BDRV_REQ_NO_WAIT ) &&bdrv_find_conflicting_request (req)) { +return -EBUSY ; + } + +bdrv_wait_serialising_requests_locked (req); }else {bdrv_wait_serialising_requests (req); }
Signature · 1 line
2.21.3
PATCH v4 7/10 block: introduce preallocate filter
In reply to Vladimir Sementsov-OgievskiyIt's intended to be inserted between format and protocol nodes to preallocate additional space (expanding protocol file) on writes crossing EOF. It improves performance for file-systems with slow allocation.
--- docs/system/qemu-block-drivers.rst.inc | 26 +++ qapi/block-core.json | 20 +- block/preallocate.c | 291 +++++++++++++++++++++++++ block/Makefile.objs | 1 + 4 files changed, 337 insertions(+), 1 deletion(-) create mode 100644 block/preallocate.c
docs/system/qemu-block-drivers.rst.inc +26 -0
@@ -952,3 +952,29 @@ on host and see if there are locks held by the QEMU process on the image file. More than one byte could be locked by the QEMU instance, each byte of which reflects a particular permission that is acquired or protected by the running block driver. + +Filter drivers +~~~~~~~~~~~~~~ + +QEMU supports several filter drivers, which don't store any data, but do some +additional tasks, hooking io requests. + +.. program:: filter-drivers +.. option:: preallocate + + Preallocate filter driver is intended to be inserted between format + and protocol nodes and does preallocation of some additional space + (expanding the protocol file) on write. It may be used for + file-systems with slow allocation. + + Supported options: + + .. program:: preallocate + .. option:: prealloc-align + + On preallocation, align file length to this number, default 1M. + + .. program:: preallocate + .. option:: prealloc-size + + How much to preallocate, default 128M.
qapi/block-core.json +19 -1
@@ -2805,7 +2805,7 @@ 'cloop', 'compress', 'copy-on-read', 'dmg', 'file', 'ftp', 'ftps', 'gluster', 'host_cdrom', 'host_device', 'http', 'https', 'iscsi', 'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels', - 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd', + 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd', { 'name': 'replication', 'if': 'defined(CONFIG_REPLICATION)' }, 'sheepdog', 'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
@@ -3074,6 +3074,23 @@ 'data': { 'aes': 'QCryptoBlockOptionsQCow', 'luks': 'QCryptoBlockOptionsLUKS'} } +## +# @BlockdevOptionsPreallocate: +# +# Filter driver intended to be inserted between format and protocol node +# and do preallocation in protocol node on write. +# +# @prealloc-align: on preallocation, align file length to this number, +# default 1048576 (1M) +# +# @prealloc-size: how much to preallocate, default 134217728 (128M) +# +# Since: 5.2 +## +{ 'struct': 'BlockdevOptionsPreallocate', + 'base': 'BlockdevOptionsGenericFormat', + 'data': { '*prealloc-align': 'int', '*prealloc-size': 'int' } } + ## # @BlockdevOptionsQcow2: #
@@ -3979,6 +3996,7 @@ 'null-co': 'BlockdevOptionsNull', 'nvme': 'BlockdevOptionsNVMe', 'parallels': 'BlockdevOptionsGenericFormat', + 'preallocate':'BlockdevOptionsPreallocate', 'qcow2': 'BlockdevOptionsQcow2', 'qcow': 'BlockdevOptionsQcow', 'qed': 'BlockdevOptionsGenericCOWFormat',
block/preallocate.c new file +291 -0
291 lines not on this page · raw message
block/Makefile.objs +1 -0
@@ -44,6 +44,7 @@ block-obj-y += crypto.o block-obj-y += aio_task.o block-obj-y += backup-top.o block-obj-y += filter-compress.o +block-obj-y += preallocate.o common-obj-y += monitor/ block-obj-y += monitor/
Signature · 1 line
2.21.3
PATCH v4 8/10 iotests.py: add verify_o_direct helper
In reply to Vladimir Sementsov-OgievskiyAdd python notrun-helper similar to _check_o_direct for bash tests. To be used in the following commit.
--- tests/qemu-iotests/iotests.py | 6 ++++++ 1 file changed, 6 insertions(+)
tests/qemu-iotests/iotests.py +6 -0
@@ -1083,6 +1083,12 @@ def _verify_aio_mode(supported_aio_modes: Sequence[str] = ()) -> None: if supported_aio_modes and (aiomode not in supported_aio_modes): notrun('not suitable for this aio mode: %s' % aiomode) +def verify_o_direct() -> None: + with FilePath('test_o_direct') as f: + qemu_img_create('-f', 'raw', f, '1M') + if 'O_DIRECT' in qemu_io('-f', 'raw', '-t', 'none', '-c', 'quit', f): + notrun(f'file system at {test_dir} does not support O_DIRECT') + def supports_quorum(): return 'quorum' in qemu_img_pipe('--help')
Signature · 1 line
2.21.3
PATCH v4 10/10 iotests: add 298 to test new preallocate filter driver
In reply to Vladimir Sementsov-Ogievskiy--- tests/qemu-iotests/298 | 50 ++++++++++++++++++++++++++++++++++++++ tests/qemu-iotests/298.out | 6 +++++ tests/qemu-iotests/group | 1 + 3 files changed, 57 insertions(+) create mode 100644 tests/qemu-iotests/298 create mode 100644 tests/qemu-iotests/298.out
tests/qemu-iotests/298 new file +50 -0
@@ -0,0 +1,50 @@ +#!/usr/bin/env python3 +# +# Test for preallocate filter +# +# Copyright (c) 2020 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 log + +iotests.script_initialize(supported_fmts=['qcow2']) +iotests.verify_o_direct() + +size = 10 * 1024 * 1024 +disk = iotests.file_path('disk') + +iotests.qemu_img_create('-f', iotests.imgfmt, disk, str(size)) + +opts = f'driver={iotests.imgfmt},' \ + f'file.driver=preallocate,file.file.filename={disk}' +p = iotests.QemuIoInteractive('--image-opts', '-t', 'none', opts) + +log(p.cmd('write 0 1M'), filters=[iotests.filter_qemu_io]) +p.cmd('flush') + +if os.path.getsize(disk) > 100 * 1024 * 1024: + log('file in progress is big, preallocation works') + +p.close() + +if os.path.getsize(disk) < 10 * 1024 * 1024: + log('file is small after close') + +# Check that there are no leaks. +log(iotests.qemu_img_pipe('check', '-f', 'qcow2', disk), + filters=[iotests.filter_img_check])
tests/qemu-iotests/298.out new file +6 -0
@@ -0,0 +1,6 @@ +wrote 1048576/1048576 bytes at offset 0 +1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) + +file in progress is big, preallocation works +file is small after close +No errors were found on the image.
tests/qemu-iotests/group +1 -0
@@ -306,6 +306,7 @@ 295 rw 296 rw 297 meta +298 auto quick 299 auto quick 301 backing quick 302 quick
Signature · 1 line
2.21.3
PATCH v4 9/10 iotests.py: add filter_img_check
In reply to Vladimir Sementsov-OgievskiyAdd analog of bash _filter_qemu_img_check to python framework.
--- tests/qemu-iotests/iotests.py | 4 ++++ 1 file changed, 4 insertions(+)
tests/qemu-iotests/iotests.py +4 -0
@@ -402,6 +402,10 @@ def filter_img_info(output, filename): lines.append(line) return '\n'.join(lines) +def filter_img_check(msg): + msg = re.sub(r'.*allocated.*fragmented.*compressed clusters', '', msg) + return re.sub(r'Image end offset: [0-9]+', '', msg).strip() + def filter_imgfmt(msg): return msg.replace(imgfmt, 'IMGFMT')
Signature · 1 line
2.21.3
Patchew URL: https://patchew.org/QEMU/[email protected]/
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series failed the docker-quick@centos7 build test. Please find the testing commands and their output below. If you have Docker installed, you can probably reproduce it locally.
=== TEST SCRIPT BEGIN ===
#!/bin/bash
make docker-image-centos7 V=1 NETWORK=1
time make docker-test-quick@centos7 SHOW_ENV=1 J=14 NETWORK=1 === TEST SCRIPT END ===
TEST check-unit: tests/test-char
Unexpected error in object_property_try_add() at /tmp/qemu-test/src/qom/object.c:1181:
attempt to add duplicate property 'serial-id' to object (type 'container')
ERROR test-char - too few tests run (expected 38, got 9)
make: *** [check-unit] Error 1
make: *** Waiting for unfinished jobs....
TEST iotest-qcow2: 018
TEST iotest-qcow2: 019
---
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['sudo', '-n', 'docker', 'run', '--label', 'com.qemu.instance.uuid=a80f0516394e4d83b19a19e329899c17', '-u', '1003', '--security-opt', 'seccomp=unconfined', '--rm', '-e', 'TARGET_LIST=', '-e', 'EXTRA_CONFIGURE_OPTS=', '-e', 'V=', '-e', 'J=14', '-e', 'DEBUG=', '-e', 'SHOW_ENV=1', '-e', 'CCACHE_DIR=/var/tmp/ccache', '-v', '/home/patchew2/.cache/qemu-docker-ccache:/var/tmp/ccache:z', '-v', '/var/tmp/patchew-tester-tmp-n_a2v5bj/src/docker-src.2020-08-20-15.00.31.12987:/var/tmp/qemu:z,ro', 'qemu/centos7', '/var/tmp/qemu/run', 'test-quick']' returned non-zero exit status 2. filter=--filter=label=com.qemu.instance.uuid=a80f0516394e4d83b19a19e329899c17
make[1]: *** [docker-run] Error 1
make[1]: Leaving directory `/var/tmp/patchew-tester-tmp-n_a2v5bj/src'
make: *** [docker-run-test-quick@centos7] Error 2
real 16m13.105s
user 0m8.886s
The full log is available at
http://patchew.org/logs/[email protected]/testing.docker-quick@centos7/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]
Quoted 36 lines · 20.08.2020 22:16, [email protected] wrote:
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series failed the docker-quick@centos7 build test. Please find the testing commands and their output below. If you have Docker installed, you can probably reproduce it locally.
=== TEST SCRIPT BEGIN ===
#!/bin/bash
make docker-image-centos7 V=1 NETWORK=1
time make docker-test-quick@centos7 SHOW_ENV=1 J=14 NETWORK=1 === TEST SCRIPT END ===TEST check-unit: tests/test-char
Unexpected error inobject_property_try_add()at/tmp/qemu-test/src/qom/object.c:1181:
attempt to add duplicate property 'serial-id' to object (type 'container')
ERROR test-char - too few tests run (expected 38, got 9)
make: *** [check-unit] Error 1
make: *** Waiting for unfinished jobs....
TEST iotest-qcow2: 018
TEST iotest-qcow2: 019
---
raiseCalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['sudo', '-n', 'docker', 'run', '--label', 'com.qemu.instance.uuid=a80f0516394e4d83b19a19e329899c17', '-u', '1003', '--security-opt', 'seccomp=unconfined', '--rm', '-e', 'TARGET_LIST=', '-e', 'EXTRA_CONFIGURE_OPTS=', '-e', 'V=', '-e', 'J=14', '-e', 'DEBUG=', '-e', 'SHOW_ENV=1', '-e', 'CCACHE_DIR=/var/tmp/ccache', '-v', '/home/patchew2/.cache/qemu-docker-ccache:/var/tmp/ccache:z', '-v', '/var/tmp/patchew-tester-tmp-n_a2v5bj/src/docker-src.2020-08-20-15.00.31.12987:/var/tmp/qemu:z,ro', 'qemu/centos7', '/var/tmp/qemu/run', 'test-quick']' returned non-zero exit status 2. filter=--filter=label=com.qemu.instance.uuid=a80f0516394e4d83b19a19e329899c17
make[1]: *** [docker-run] Error 1
make[1]: Leaving directory `/var/tmp/patchew-tester-tmp-n_a2v5bj/src'
make: *** [docker-run-test-quick@centos7] Error 2real 16m13.105s
user 0m8.886s
Quoted 6 lines
The full log is available at
http://patchew.org/logs/[email protected]/testing.docker-quick@centos7/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]
Seems something unrelated
Signature · 2 lines
Best regards, Vladimir
Re: [PATCH v4 08/10] iotests.py: add verify_o_direct helper
In reply to Vladimir Sementsov-OgievskiyQuoted 15 lines · Vladimir Sementsov-Ogievskiy
Add python notrun-helper similar to
_check_o_directfor bash tests. To be used in the following commit.Signed-off-by Vladimir Sementsov-Ogievskiy <[email protected]>Reviewed-by Stefan Hajnoczi <[email protected]>--- tests/qemu-iotests/iotests.py | 6 ++++++ 1 file changed, 6 insertions(+)tests/qemu-iotests/iotests.py +0
-0
tests/qemu-iotests/iotests.py def _verify_aio_mode(supported_aio_modes: Sequence[str] = ()) -> None: · line 1090
Quoted 8 lines
tests/qemu-iotests/iotests.py +5
-0@@ -1083,6 +1083,12 @@ def _verify_aio_mode(supported_aio_modes: Sequence[str] = ()) -> None: if supported_aio_modes and (aiomode not in supported_aio_modes): notrun('not suitable for this aio mode: %s' % aiomode) +def verify_o_direct() -> None: + with FilePath('test_o_direct') as f: + qemu_img_create('-f', 'raw', f, '1M') + if 'O_DIRECT' in qemu_io('-f', 'raw', '-t', 'none', '-c', 'quit', f): + notrun(f'file system at {test_dir} does not support O_DIRECT')
Why not:
with FilePath('test_o_direct') as f:
try:
fd = os.open(f, os.O_DIRECT | os.O_CREAT | os.O_RDWR)
except OSError as e:
if e.errno != errno.EINVAL:
raise
notrun(...)
else:
os.close(fd)
More verbose, but the intent is more clear, and we do not depend on the output of qemu-io which is not a public API. For example if someone improves qemu-io to fail with:
Direct I/O is not supported
It would break the tests using this helper.
Nir
Quoted 8 lines
tests/qemu-iotests/iotests.py +1
-1+ def supports_quorum(): return 'quorum' in qemu_img_pipe('--help') --2.21.3
Re: [PATCH v4 08/10] iotests.py: add verify_o_direct helper
Quoted 38 lines · Nir Soffer
Quoted 15 lines · Vladimir Sementsov-Ogievskiy
Add python notrun-helper similar to
_check_o_directfor bash tests. To be used in the following commit.Signed-off-by Vladimir Sementsov-Ogievskiy <[email protected]>Reviewed-by Stefan Hajnoczi <[email protected]>--- tests/qemu-iotests/iotests.py | 6 ++++++ 1 file changed, 6 insertions(+)tests/qemu-iotests/iotests.py +0
-0tests/qemu-iotests/iotests.py def _verify_aio_mode(supported_aio_modes: Sequence[str] = ()) -> None: · line 1090
Quoted 8 lines
tests/qemu-iotests/iotests.py +5
-0@@ -1083,6 +1083,12 @@ def _verify_aio_mode(supported_aio_modes: Sequence[str] = ()) -> None: if supported_aio_modes and (aiomode not in supported_aio_modes): notrun('not suitable for this aio mode: %s' % aiomode) +def verify_o_direct() -> None: + with FilePath('test_o_direct') as f: + qemu_img_create('-f', 'raw', f, '1M') + if 'O_DIRECT' in qemu_io('-f', 'raw', '-t', 'none', '-c', 'quit', f): + notrun(f'file system at {test_dir} does not support O_DIRECT')Why not:
with
FilePath('test_o_direct')as f:
try:
fd =os.open(f, os.O_DIRECT | os.O_CREAT | os.O_RDWR)
except OSError as e:
if e.errno != errno.EINVAL:
raise
notrun(...)
else:
os.close(fd)
Quoted 8 lines
More verbose, but the intent is more clear, and we do not depend on the output of qemu-io which is not a public API. For example if someone improves qemu-io to fail with:
Direct I/O is not supported
It would break the tests using this helper.
Agree, that's better. Will use it, thanks!
Quoted 10 lines
Quoted 9 lines
tests/qemu-iotests/iotests.py +1
-1+ def supports_quorum(): return 'quorum' in qemu_img_pipe('--help') --2.21.3
Signature · 2 lines
Best regards, Vladimir