Now that node level interface bdrv_truncate() supports passing request
flags to the block driver, expose this on the BlockBackend level, too.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200424125448.63318-4-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
mirror_wait_for_free_in_flight_slot() just picks a random operation to
wait for. However, a MirrorOp is already in s->ops_in_flight when
mirror_co_read() waits for free slots, so if not enough slots are
immediately available, an operation can end up waiting for itself, or
two or more operations can wait for each other to complete, which
results in a hang.
Fix this by adding a flag to MirrorOp that tells us if the request is
already in flight (and therefore occupies slots that it will later
free), and picking only such operations for waiting.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1794692
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200326153628.4869-3-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
This reverts commit 7e6c4ff792.
The fix was incomplete as it only protected against requests waiting for
themselves, but not against requests waiting for each other. We need a
different solution.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200326153628.4869-2-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
local_err is used again in mirror_exit_common() after
bdrv_set_backing_hd(), so we must zero it. Otherwise try to set
non-NULL local_err will crash.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20200324153630.11882-3-vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
There is no guarantee that we can still replace the node we want to
replace at the end of the mirror job. Double-check by calling
bdrv_recurse_can_replace().
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20200218103454.296704-12-mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
mirror_wait_for_free_in_flight_slot() just picks a random operation to
wait for. However, when mirror_co_read() waits for free slots, its
MirrorOp is already in s->ops_in_flight, so if not enough slots are
immediately available, an operation can end up waiting for itself to
complete, which results in a hang.
Fix this by passing the current MirrorOp and skipping this operation
when picking an operation to wait for.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1794692
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
If a coroutine is launched, but the coroutine pointer isn't stored
anywhere, debugging any problems inside the coroutine is quite hard.
Let's store the coroutine pointer of a mirror operation in MirrorOp to
have it available in the debugger.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
We have two drivers (iscsi and file-posix) that (in some cases) return
success from their .bdrv_co_truncate() implementation if the block
device is larger than the requested offset, but cannot be shrunk. Some
callers do not want that behavior, so this patch adds a new parameter
that they can use to turn off that behavior.
This patch just adds the parameter and lets the block/io.c and
block/block-backend.c functions pass it around. All other callers
always pass false and none of the implementations evaluate it, so that
this patch does not change existing behavior. Future patches take care
of that.
Suggested-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190918095144.955-5-mreitz@redhat.com
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
mirror_exit_common() may be called twice (if it is called from
mirror_prepare() and fails, it will be called from mirror_abort()
again).
In such a case, many of the pointers in the MirrorBlockJob object will
already be freed. This can be seen most reliably for s->target, which
is set to NULL (and then dereferenced by blk_bs()).
Cc: qemu-stable@nongnu.org
Fixes: 737efc1eda
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20191014153931.20699-2-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
This reverts commit 9adc1cb49a.
"mirror: Only mirror granularity-aligned chunks"
Since previous commit unaligned chunks are supported by
do_sync_target_write.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20191011090711.19940-6-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
Prior 9adc1cb49a do_sync_target_write had a bug: it reset aligned-up
region in the dirty bitmap, which means that we may not copy some bytes
and assume them copied, which actually leads to producing corrupted
target.
So 9adc1cb49a forced dirty bitmap granularity to be
request_alignment for mirror-top filter, so we are not working with
unaligned requests. However forcing large alignment obviously decreases
performance of unaligned requests.
This commit provides another solution for the problem: if unaligned
padding is already dirty, we can safely ignore it, as
1. It's dirty, it will be copied by mirror_iteration anyway
2. It's dirty, so skipping it now we don't increase dirtiness of the
bitmap and therefore don't damage "synchronicity" of the
write-blocking mirror.
If unaligned padding is not dirty, we just write it, no reason to touch
dirty bitmap if we succeed (on failure we'll set the whole region
ofcourse, but we loss "synchronicity" on failure anyway).
Note: we need to disable dirty_bitmap, otherwise we will not be able to
see in do_sync_target_write bitmap state before current operation. We
may of course check dirty bitmap before the operation in
bdrv_mirror_top_do_write and remember it, but we don't need active
dirty bitmap for write-blocking mirror anyway.
New code-path is unused until the following commit reverts
9adc1cb49a.
Suggested-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20191011090711.19940-5-vsementsov@virtuozzo.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
do_sync_target_write is called from bdrv_mirror_top_do_write after
write/discard operation, all inside active_write/active_write_settle
protecting us from mirror iteration. So the whole area is dirty for
sure, no reason to examine dirty bitmap.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20191011090711.19940-3-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
Add bs field to BdrvDirtyBitmap structure. Drop BlockDriverState
parameter from bitmap APIs where possible.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Message-id: 20190916141911.5255-3-vsementsov@virtuozzo.com
[Rebased on top of block-copy. --js]
Signed-off-by: John Snow <jsnow@redhat.com>
In job_finish_sync job_enter should be enough for a job to make some
progress and draining is a wrong tool for it. So use job_enter directly
here and drop job_drain with all related staff not used more.
Suggested-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Tested-by: John Snow <jsnow@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
bdrv_has_zero_init() only has meaning for newly created images or image
areas. If the mirror job itself did not create the image, it cannot
rely on bdrv_has_zero_init()'s result to carry any meaning.
This is the case for drive-mirror with mode=existing and always for
blockdev-mirror.
Note that we only have to zero-initialize the target with sync=full,
because other modes actually do not promise that the target will contain
the same data as the source after the job -- sync=top only promises to
copy anything allocated in the top layer, and sync=none will only copy
new I/O. (Which is how mirror has always handled it.)
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190724171239.8764-3-mreitz@redhat.com
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Add a public interface for get. While we're at it,
rename "bdrv_get_dirty_bitmap_locked" to "bdrv_dirty_bitmap_get_locked".
(There are more functions to rename to the bdrv_dirty_bitmap_VERB form,
but they will wait until the conclusion of this series.)
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190709232550.10724-11-jsnow@redhat.com
Signed-off-by: John Snow <jsnow@redhat.com>
We don't need or want a new sync mode for simple differences in
semantics. Create a new mode simply named "BITMAP" that is designed to
make use of the new Bitmap Sync Mode field.
Because the only bitmap sync mode is 'on-success', this adds no new
functionality to the backup job (yet). The old incremental backup mode
is maintained as a syntactic sugar for sync=bitmap, mode=on-success.
Add all of the plumbing necessary to support this new instruction.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190709232550.10724-6-jsnow@redhat.com
Signed-off-by: John Snow <jsnow@redhat.com>
This fixes devices like IDE that can still start new requests from I/O
handlers in the CPU thread while the block backend is drained.
The basic assumption is that in a drain section, no new requests should
be allowed through a BlockBackend (blk_drained_begin/end don't exist,
we get drain sections only on the node level). However, there are two
special cases where requests should not be queued:
1. Block jobs: We already make sure that block jobs are paused in a
drain section, so they won't start new requests. However, if the
drain_begin is called on the job's BlockBackend first, it can happen
that we deadlock because the job stays busy until it reaches a pause
point - which it can't if its requests aren't processed any more.
The proper solution here would be to make all requests through the
job's filter node instead of using a BlockBackend. For now, just
disabling request queuing on the job BlockBackend is simpler.
2. In test cases where making requests through bdrv_* would be
cumbersome because we'd need a BdrvChild. As we already got the
functionality to disable request queuing from 1., use it in tests,
too, for convenience.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
mirror_top_bs is currently implicitly drained through its connection to
the source or the target node. However, the drain section for target_bs
ends early after moving mirror_top_bs from src to target_bs, so that
requests can already be restarted while mirror_top_bs is still present
in the chain, but has dropped all permissions and therefore runs into an
assertion failure like this:
qemu-system-x86_64: block/io.c:1634: bdrv_co_write_req_prepare:
Assertion `child->perm & BLK_PERM_WRITE' failed.
Keep mirror_top_bs drained until all graph changes have completed.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
In write-blocking mode, all writes to the top node directly go to the
target. We must only mirror chunks of data that are aligned to the
job's granularity, because that is how the dirty bitmap works.
Therefore, the request alignment for writes must be the job's
granularity (in write-blocking mode).
Unfortunately, this forces all reads and writes to have the same
granularity (we only need this alignment for writes to the target, not
the source), but that is something to be fixed another time.
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190805153308.2657-1-mreitz@redhat.com
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Fixes: d06107ade0
Signed-off-by: Max Reitz <mreitz@redhat.com>
The commit and the mirror block job must be able to drop their filter
node at any point. However, this will not be possible if any of the
BdrvChild links to them is frozen. Therefore, we need to prevent them
from ever becoming frozen.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Message-id: 20190703172813.6868-2-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
This patch is used in the 'block/stream: introduce a bottom node'
that is following. Instead of the base node, the caller may pass
the node that has the base as its backing image to the function
bdrv_is_allocated_above() with a new parameter include_base = true
and get rid of the dependency on the base that may change during
commit/stream parallel jobs. Now, if the specified base is not
found in the backing image chain, the QEMU will abort.
Suggested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Message-id: 1559152576-281803-2-git-send-email-andrey.shinkevich@virtuozzo.com
[mreitz: Squashed in the following as a rebase on conflicting patches:]
Message-id: e3cf99ae-62e9-8b6e-5a06-d3c8b9363b85@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
We cannot use bdrv_child_try_set_perm() to give up all restrictions on
the child edge, and still have bdrv_mirror_top_child_perm() request
BLK_PERM_WRITE. Fix this by making bdrv_mirror_top_child_perm() return
0/BLK_PERM_ALL when we want to give up all permissions, and replacing
bdrv_child_try_set_perm() by bdrv_child_refresh_perms().
The bdrv_child_try_set_perm() before removing the node with
bdrv_replace_node() is then unnecessary. No permissions have changed
since the previous invocation of bdrv_child_try_set_perm().
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are going to remove bs->job pointer. Drop it's usage in replication
code. Additionally we have to return job pointer from some mirror APIs.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The mirror and commit block jobs use bdrv_set_aio_context() to move
their filter node into the right AioContext before hooking it up in the
graph. Similarly, bdrv_open_backing_file() explicitly moves the backing
file node into the right AioContext first.
This isn't necessary any more, they get automatically moved into the
right context now when attaching them.
However, in the case of bdrv_open_backing_file() with a node reference,
it's actually not only unnecessary, but even wrong: The unchecked
bdrv_set_aio_context() changes the AioContext of the child node even if
other parents require it to retain the old context. So this is not only
a simplification, but a bug fix, too.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1684342
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
This adds a new parameter to blk_new() which requires its callers to
declare from which AioContext this BlockBackend is going to be used (or
the locks of which AioContext need to be taken anyway).
The given context is only stored and kept up to date when changing
AioContexts. Actually applying the stored AioContext to the root node
is saved for another commit.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Block jobs require that all of the nodes the job is using are in the
same AioContext. Therefore all BdrvChild objects of the job propagate
.(can_)set_aio_context to all other job nodes, so that the switch is
checked and performed consistently even if both nodes are in different
subtrees.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Filter drivers that support .bdrv_co_pwrite_zeroes can safely advertise
BDRV_REQ_NO_FALLBACK because they just forward the request flags to
their child node.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Acked-by: Eric Blake <eblake@redhat.com>
While child_job_drained_begin() calls to job_pause(), the job doesn't
actually transition between states until it runs again and reaches a
pause point. This means bdrv_drained_begin() may return with some jobs
using the node still having 'busy == true'.
As a consequence, block_job_detach_aio_context() may get into a
deadlock, waiting for the job to be actually paused, while the coroutine
servicing the job is yielding and doesn't get the opportunity to get
scheduled again. This situation can be reproduced by issuing a
'block-commit' immediately followed by a 'device_del'.
To ensure bdrv_drained_begin() only returns when the jobs have been
paused, we change mirror_drained_poll() to only confirm it's quiesced
when job->paused == true and there aren't any in-flight requests, except
if we reached that point by a drained section initiated by the
mirror/commit job itself.
The other block jobs shouldn't need any changes, as the default
drained_poll() behavior is to only confirm it's quiesced if the job is
not busy or completed.
Signed-off-by: Sergio Lopez <slp@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Currently, BlockDriver.bdrv_refresh_filename() is supposed to both
refresh the filename (BDS.exact_filename) and set BDS.full_open_options.
Now that we have generic code in the central bdrv_refresh_filename() for
creating BDS.full_open_options, we can drop the latter part from all
BlockDriver.bdrv_refresh_filename() implementations.
This also means that we can drop all of the existing default code for
this from the global bdrv_refresh_filename() itself.
Furthermore, we now have to call BlockDriver.bdrv_refresh_filename()
after having set BDS.full_open_options, because the block driver's
implementation should now be allowed to depend on BDS.full_open_options
being set correctly.
Finally, with this patch we can drop the @options parameter from
BlockDriver.bdrv_refresh_filename(); also, add a comment on this
function's purpose in block/block_int.h while touching its interface.
This completely obsoletes blklogwrite's implementation of
.bdrv_refresh_filename().
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190201192935.18394-25-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
bdrv_refresh_filename() should invoke itself recursively on all
children, not just on file.
With that change, we can remove the manual invocations in blkverify,
quorum, commit, mirror, and blklogwrites.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Message-id: 20190201192935.18394-3-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
The mirror_start_job() function used for the commit-active job blocks
the source, target and all intermediate nodes for the duration of the
job.
target <- intermediate <- source
Since 4ef85a9c23 this function creates a dummy mirror_top_bs that
goes on top of the source node, and it is this dummy node that gets
blocked instead. The source node is never blocked or added to the
job's list of nodes.
target <- intermediate <- source <- mirror_top
At the moment I don't think it is possible to exploit this problem
because any additional job on 'source' would either be forbidden for
other reasons or it would need to involve an additional node that is
blocked, causing an error.
This can be seen in the error messages, however, because they never
refer to the source node being blocked:
$ qemu-img create -f qcow2 hd0.qcow2 1M
$ qemu-img create -f qcow2 -b hd0.qcow2 hd1.qcow2
$ qemu-io -c 'write 0 1M' hd0.qcow2
$ $QEMU -drive if=none,file=hd1.qcow2,node-name=hd1
{ "execute": "qmp_capabilities" }
{ "execute": "block-commit", "arguments": {"device": "hd1", "speed": 256}}
{ "execute": "block-stream", "arguments": {"device": "hd1"}}
{ "error": {"class": "GenericError",
"desc": "Node 'hd0' is busy: block device is in use by block job: commit"}}
After this patch the error message refers to 'hd1', as it should.
The expected output of iotest 141 also needs to be updated for the
same reason.
Signed-off-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
At the moment I don't see how to make this function fail after the
dirty bitmap has been created, but if that was possible then we would
hit the assert(QLIST_EMPTY(&bs->dirty_bitmaps)) in bdrv_close().
Signed-off-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Use bdrv_dirty_bitmap_next_dirty_area() instead of
bdrv_dirty_iter_next_area(), because of the following problems of
bdrv_dirty_iter_next_area():
1. Using HBitmap iterators we should carefully handle unaligned offset,
as first call to hbitmap_iter_next() may return a value less than
original offset (actually, it will be original offset rounded down to
bitmap granularity). This handling is not done in
do_sync_target_write().
2. bdrv_dirty_iter_next_area() handles unaligned max_offset
incorrectly:
look at the code:
if (max_offset == iter->bitmap->size) {
/* If max_offset points to the image end, round it up by the
* bitmap granularity */
gran_max_offset = ROUND_UP(max_offset, granularity);
} else {
gran_max_offset = max_offset;
}
ret = hbitmap_iter_next(&iter->hbi, false);
if (ret < 0 || ret + granularity > gran_max_offset) {
return false;
}
and assume that max_offset != iter->bitmap->size but still unaligned.
if 0 < ret < max_offset we found dirty area, but the function can
return false in this case (if ret + granularity > max_offset).
3. bdrv_dirty_iter_next_area() uses inefficient loop to find the end of
the dirty area. Let's use more efficient hbitmap_next_zero instead
(bdrv_dirty_bitmap_next_dirty_area() do so)
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Most list head structs need not be given a name. In most cases the
name is given just in case one is going to use QTAILQ_LAST, QTAILQ_PREV
or reverse iteration, but this does not apply to lists of other kinds,
and even for QTAILQ in practice this is only rarely needed. In addition,
we will soon reimplement those macros completely so that they do not
need a name for the head struct. So clean up everything, not giving a
name except in the rare case where it is necessary.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Marking a function coroutine_fn currently has no effect on the compiler,
but it documents that this function must be called from coroutine
context and it may yield. This is important information for the
programmer.
Also, if we ever transition to a stackless coroutine implementation,
then it's likely that the annotation will become mandatory so the
compiler can use the correct calling convention for coroutine functions.
Cc: Max Reitz <mreitz@redhat.com>
Cc: John Snow <jsnow@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The 'block-commit' QMP command is implemented internally using two
different drivers. If the source image is the active layer then the
mirror driver is used (commit_active_start()), otherwise the commit
driver is used (commit_start()).
In both cases the destination image must be put temporarily in
read-write mode. This is done correctly in the latter case, but what
commit_active_start() does is copy all flags instead.
This patch replaces the bdrv_reopen() calls in that function with
bdrv_reopen_set_read_only() so that only the read-only status is
changed.
A similar change is made in mirror_exit(), which is also used by the
'drive-mirror' and 'blockdev-mirror' commands.
Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Let start from the beginning:
Commit b9e413dd37 (in 2.9)
"block: explicitly acquire aiocontext in aio callbacks that need it"
added pairs of aio_context_acquire/release to mirror_write_complete and
mirror_read_complete, when they were aio callbacks for blk_aio_* calls.
Then, commit 2e1990b26e (in 3.0) "block/mirror: Convert to coroutines"
dropped these blk_aio_* calls, than mirror_write_complete and
mirror_read_complete are not callbacks more, and don't need additional
aiocontext acquiring. Furthermore, mirror_read_complete calls
blk_co_pwritev inside these pair of aio_context_acquire/release, which
leads to the following dead-lock with mirror:
(gdb) info thr
Id Target Id Frame
3 Thread (LWP 145412) "qemu-system-x86" syscall ()
2 Thread (LWP 145416) "qemu-system-x86" __lll_lock_wait ()
* 1 Thread (LWP 145411) "qemu-system-x86" __lll_lock_wait ()
(gdb) bt
#0 __lll_lock_wait ()
#1 _L_lock_812 ()
#2 __GI___pthread_mutex_lock
#3 qemu_mutex_lock_impl (mutex=0x561032dce420 <qemu_global_mutex>,
file=0x5610327d8654 "util/main-loop.c", line=236) at
util/qemu-thread-posix.c:66
#4 qemu_mutex_lock_iothread_impl
#5 os_host_main_loop_wait (timeout=480116000) at util/main-loop.c:236
#6 main_loop_wait (nonblocking=0) at util/main-loop.c:497
#7 main_loop () at vl.c:1892
#8 main
Printing contents of qemu_global_mutex, I see that "__owner = 145416",
so, thr1 is main loop, and now it wants BQL, which is owned by thr2.
(gdb) thr 2
(gdb) bt
#0 __lll_lock_wait ()
#1 _L_lock_870 ()
#2 __GI___pthread_mutex_lock
#3 qemu_mutex_lock_impl (mutex=0x561034d25dc0, ...
#4 aio_context_acquire (ctx=0x561034d25d60)
#5 dma_blk_cb
#6 dma_blk_io
#7 dma_blk_read
#8 ide_dma_cb
#9 bmdma_cmd_writeb
#10 bmdma_write
#11 memory_region_write_accessor
#12 access_with_adjusted_size
#15 flatview_write
#16 address_space_write
#17 address_space_rw
#18 kvm_handle_io
#19 kvm_cpu_exec
#20 qemu_kvm_cpu_thread_fn
#21 qemu_thread_start
#22 start_thread
#23 clone ()
Printing mutex in fr 2, I see "__owner = 145411", so thr2 wants aio
context mutex, which is owned by thr1. Classic dead-lock.
Then, let's check that aio context is hold by mirror coroutine: just
print coroutine stack of first tracked request in mirror job target:
(gdb) [...]
(gdb) qemu coroutine 0x561035dd0860
#0 qemu_coroutine_switch
#1 qemu_coroutine_yield
#2 qemu_co_mutex_lock_slowpath
#3 qemu_co_mutex_lock
#4 qcow2_co_pwritev
#5 bdrv_driver_pwritev
#6 bdrv_aligned_pwritev
#7 bdrv_co_pwritev
#8 blk_co_pwritev
#9 mirror_read_complete () at block/mirror.c:232
#10 mirror_co_read () at block/mirror.c:370
#11 coroutine_trampoline
#12 __start_context
Yes it is mirror_read_complete calling blk_co_pwritev after acquiring
aio context.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
For purposes of minimum code movement, refactor the mirror_exit
callback to use the post-finalization callbacks in a trivial way.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20180906130225.5118-7-jsnow@redhat.com
Reviewed-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
[mreitz: Added comment for the mirror_exit() function]
Signed-off-by: Max Reitz <mreitz@redhat.com>
In cases where we abort the block/mirror job, there's no point in
installing the new backing chain before we finish aborting.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20180906130225.5118-6-jsnow@redhat.com
Reviewed-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Add support for taking and passing forward job creation flags.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Message-id: 20180906130225.5118-3-jsnow@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
Change the manual deferment to mirror_exit into the implicit
callback to job_exit and the mirror_exit callback.
This does change the order of some bdrv_unref calls and job_completed,
but thanks to the new context in which we call .exit, this is safe to
defer the possible flushing of any nodes to the job_finalize_single
cleanup stage.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20180830015734.19765-6-jsnow@redhat.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Jobs presently use both an Error object in the case of the create job,
and char strings in the case of generic errors elsewhere.
Unify the two paths as just j->err, and remove the extra argument from
job_completed. The integer error code for job_completed is kept for now,
to be removed shortly in a separate patch.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20180830015734.19765-3-jsnow@redhat.com
[mreitz: Dropped a superfluous g_strdup()]
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Presently we codify the entry point for a job as the "start" callback,
but a more apt name would be "run" to clarify the idea that when this
function returns we consider the job to have "finished," except for
any cleanup which occurs in separate callbacks later.
As part of this clarification, change the signature to include an error
object and a return code. The error ptr is not yet used, and the return
code while captured, will be overwritten by actions in the job_completed
function.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20180830015734.19765-2-jsnow@redhat.com
Reviewed-by: Jeff Cody <jcody@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
.bdrv_close handler is optional after previous commit, no needs to keep
empty functions more.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
blockdev-mirror with the same node for source and target segfaults
today: A node is in its own backing chain, so mirror_start_job() decides
that this is an active commit. When adding the intermediate nodes with
block_job_add_bdrv(), it starts the iteration through the subchain with
the backing file of source, though, so it never reaches target and
instead runs into NULL at the base.
While we could fix that by starting with source itself, there is no
point in allowing mirroring a node into itself and I wouldn't be
surprised if this caused more problems later.
So just check for this scenario and error out.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Other I/O functions are already using a BdrvChild pointer in the API, so
make discard do the same. It makes it possible to initiate the same
permission checks before doing I/O, and much easier to share the
helper functions for this, which will be added and used by write,
truncate and copy range paths.
Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
This patch allows the user to specify whether to use active or only
background mode for mirror block jobs. Currently, this setting will
remain constant for the duration of the entire block job.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Message-id: 20180613181823.13618-14-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
This patch implements active synchronous mirroring. In active mode, the
passive mechanism will still be in place and is used to copy all
initially dirty clusters off the source disk; but every write request
will write data both to the source and the target disk, so the source
cannot be dirtied faster than data is mirrored to the target. Also,
once the block job has converged (BLOCK_JOB_READY sent), source and
target are guaranteed to stay in sync (unless an error occurs).
Active mode is completely optional and currently disabled at runtime. A
later patch will add a way for users to enable it.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Message-id: 20180613181823.13618-13-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
