This is a new protocol driver that exclusively opens a host NVMe
controller through VFIO. It achieves better latency than linux-aio by
completely bypassing host kernel vfs/block layer.
$rw-$bs-$iodepth linux-aio nvme://
----------------------------------------
randread-4k-1 10.5k 21.6k
randread-512k-1 745 1591
randwrite-4k-1 30.7k 37.0k
randwrite-512k-1 1945 1980
(unit: IOPS)
The driver also integrates with the polling mechanism of iothread.
This patch is co-authored by Paolo and me.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <20180116060901.17413-4-famz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Now that CoQueues can use a QemuMutex for thread-safety, there is no
need for curl to roll its own coroutine queue. Coroutines can be
placed directly on the queue instead of using a list of CURLAIOCBs.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20180203153935.8056-6-pbonzini@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Since mirror job supports efficient zero out target mechanism (see
in mirror_dirty_init()), implement bdrv_get_info to make it work
over NBD. Such improvement will allow using the largest chunk possible
and will decrease the number of NBD_CMD_WRITE_ZEROES requests on the wire.
Signed-off-by: Edgar Kaziakhmedov <edgar.kaziakhmedov@virtuozzo.com>
Message-Id: <20180118115158.17219-1-edgar.kaziakhmedov@virtuozzo.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
Block layer patches
# gpg: Signature made Tue 23 Jan 2018 12:38:36 GMT
# gpg: using RSA key 0x7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74 56FE 7F09 B272 C88F 2FD6
* remotes/kevin/tags/for-upstream: (29 commits)
iotests: Disable some tests for compat=0.10
iotests: Split 177 into two parts for compat=0.10
iotests: Make 059 pass on machines with little RAM
iotests: Filter compat-dependent info in 198
iotests: Make 191 work with qcow2 options
iotests: Make 184 image-less
iotests: Make 089 compatible with compat=0.10
iotests: Fix 067 for compat=0.10
iotests: Fix 059's reference output
iotests: Fix 051 for compat=0.10
iotests: Fix 020 for vmdk
iotests: Skip 103 for refcount_bits=1
iotests: Forbid 020 for non-file protocols
iotests: Drop format-specific in _filter_img_info
iotests: Fix _img_info for backslashes
block/vmdk: Add blkdebug events
block/qcow: Add blkdebug events
qcow2: No persistent dirty bitmaps for compat=0.10
block/vmdk: Fix , instead of ; at end of line
qemu-iotests: Fix locking issue in 102
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Persistent dirty bitmaps require a properly functioning
autoclear_features field, or we cannot track when an unsupporting
program might overwrite them. Therefore, we cannot support them for
compat=0.10 images.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20171123020832.8165-3-mreitz@redhat.com
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
The find_desc_by_name() from util/qemu-option.c relies on the .name not being
NULL to call strcmp(). This check becomes unsafe when the list is not
NULL-terminated, which is the case of nbd_runtime_opts in block/nbd.c, and can
result in segmentation fault when strcmp() tries to access an invalid memory:
#0 0x00007fff8c75f7d4 in __strcmp_power9 () from /lib64/libc.so.6
#1 0x00000000102d3ec8 in find_desc_by_name (desc=0x1036d6f0, name=0x28e46670 "server.path") at util/qemu-option.c:166
#2 0x00000000102d93e0 in qemu_opts_absorb_qdict (opts=0x28e47a80, qdict=0x28e469a0, errp=0x7fffec247c98) at util/qemu-option.c:1026
#3 0x000000001012a2e4 in nbd_open (bs=0x28e42290, options=0x28e469a0, flags=24578, errp=0x7fffec247d80) at block/nbd.c:406
#4 0x00000000100144e8 in bdrv_open_driver (bs=0x28e42290, drv=0x1036e070 <bdrv_nbd_unix>, node_name=0x0, options=0x28e469a0, open_flags=24578, errp=0x7fffec247f50) at block.c:1135
#5 0x0000000010015b04 in bdrv_open_common (bs=0x28e42290, file=0x0, options=0x28e469a0, errp=0x7fffec247f50) at block.c:1395
>From gdb, the desc[i].name was not NULL and resulted in strcmp() accessing an
invalid memory:
>>> p desc[5]
$8 = {
name = 0x1037f098 "R27A",
type = 1561964883,
help = 0xc0bbb23e <error: Cannot access memory at address 0xc0bbb23e>,
def_value_str = 0x2 <error: Cannot access memory at address 0x2>
}
>>> p desc[6]
$9 = {
name = 0x103dac78 <__gcov0.do_qemu_init_bdrv_nbd_init> "\001",
type = 272101528,
help = 0x29ec0b754403e31f <error: Cannot access memory at address 0x29ec0b754403e31f>,
def_value_str = 0x81f343b9 <error: Cannot access memory at address 0x81f343b9>
}
This patch fixes the segmentation fault in strcmp() by adding a NULL element at
the end of nbd_runtime_opts.desc list, which is the common practice to most of
other structs like runtime_opts in block/null.c. Thus, the desc[i].name != NULL
check becomes safe because it will not evaluate to true when .desc list reached
its end.
Reported-by: R. Nageswara Sastry <nasastry@in.ibm.com>
Buglink: https://bugs.launchpad.net/qemu/+bug/1727259
Signed-off-by: Murilo Opsfelder Araujo <muriloo@linux.vnet.ibm.com>
Message-Id: <20180105133241.14141-2-muriloo@linux.vnet.ibm.com>
CC: qemu-stable@nongnu.org
Fixes: 7ccc44fd7d
Signed-off-by: Eric Blake <eblake@redhat.com>
The bdrv_reopen*() implementation doesn't like it if the graph is
changed between queuing nodes for reopen and actually reopening them
(one of the reasons is that queuing can be recursive).
So instead of draining the device only in bdrv_reopen_multiple(),
require that callers already drained all affected nodes, and assert this
in bdrv_reopen_queue().
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Since commit bde70715, base is the only node that is reopened in
commit_start(). This means that the code, which still involves an
explicit BlockReopenQueue, can now be simplified by using bdrv_reopen().
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
We need to remember how many of the drain sections in which a node is
were recursive (i.e. subtree drain rather than node drain), so that they
can be correctly applied when children are added or removed during the
drained section.
With this change, it is safe to modify the graph even inside a
bdrv_subtree_drained_begin/end() section.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
bdrv_drained_begin() waits for the completion of requests in the whole
subtree, but it only actually keeps its immediate bs parameter quiesced
until bdrv_drained_end().
Add a version that keeps the whole subtree drained. As of this commit,
graph changes cannot be allowed during a subtree drained section, but
this will be fixed soon.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
This is in preparation for subtree drains, i.e. drained sections that
affect not only a single node, but recursively all child nodes, too.
Calling the parent callbacks for drain is pointless when we just came
from that parent node recursively and leads to multiple increases of
bs->quiesce_counter in a single drain call. Don't do it.
In order for this to work correctly, the parent callback must be called
for every bdrv_drain_begin/end() call, not only for the outermost one:
If we have a node N with two parents A and B, recursive draining of A
should cause the quiesce_counter of B to increase because its child N is
drained independently of B. If now B is recursively drained, too, A must
increase its quiesce_counter because N is drained independently of A
only now, even if N is going from quiesce_counter 1 to 2.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
bdrv_do_drained_begin() restricts the call of parent callbacks and
aio_disable_external() to the outermost drain section, but the block
driver callbacks are always called. bdrv_do_drained_end() must match
this behaviour, otherwise nodes stay drained even if begin/end calls
were balanced.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Block jobs are already paused using the BdrvChildRole drain callbacks,
so we don't need an additional block_job_pause_all() call.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
bdrv_drained_begin() doesn't increase bs->quiesce_counter recursively
and also doesn't notify other parent nodes of children, which both means
that the child nodes are not actually drained, and bdrv_drained_begin()
is providing useful functionality only on a single node.
To keep things consistent, we also shouldn't call the block driver
callbacks recursively.
A proper recursive drain version that provides an actually working
drained section for child nodes will be introduced later.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Since bdrv_co_preadv does all neccessary checks including
reading after the end of the backing file, avoid duplication
of verification before bdrv_co_preadv call.
Signed-off-by: Edgar Kaziakhmedov <edgar.kaziakhmedov@virtuozzo.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Drain requests are propagated to child nodes, parent nodes and directly
to the AioContext. The order in which this happened was different
between all combinations of drain/drain_all and begin/end.
The correct order is to keep children only drained when their parents
are also drained. This means that at the start of a drained section, the
AioContext needs to be drained first, the parents second and only then
the children. The correct order for the end of a drained section is the
opposite.
This patch changes the three other functions to follow the example of
bdrv_drained_begin(), which is the only one that got it right.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
The device is drained, so there is no point in waiting for requests at
the end of the drained section. Remove the bdrv_drain_recurse() calls
there.
The bdrv_drain_recurse() calls were introduced in commit 481cad48e5
in order to call the .bdrv_co_drain_end() driver callback. This is now
done by a separate bdrv_drain_invoke() call.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Now that the bdrv_drain_invoke() calls are pulled up to the callers of
bdrv_drain_recurse(), the 'begin' parameter isn't needed any more.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
bdrv_drain_all_begin() used to call the .bdrv_co_drain_begin() driver
callback inside its polling loop. This means that how many times it got
called for each node depended on long it had to poll the event loop.
This is obviously not right and results in nodes that stay drained even
after bdrv_drain_all_end(), which calls .bdrv_co_drain_begin() once per
node.
Fix bdrv_drain_all_begin() to call the callback only once, too.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
This change separates bdrv_drain_invoke(), which calls the BlockDriver
drain callbacks, from bdrv_drain_recurse(). Instead, the function
performs its own recursion now.
One reason for this is that bdrv_drain_recurse() can be called multiple
times by bdrv_drain_all_begin(), but the callbacks may only be called
once. The separation is necessary to fix this bug.
The other reason is that we intend to go to a model where we call all
driver callbacks first, and only then start polling. This is not fully
achieved yet with this patch, as bdrv_drain_invoke() contains a
BDRV_POLL_WHILE() loop for the block driver callbacks, which can still
call callbacks for any unrelated event. It's a step in this direction
anyway.
Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
we currently report an "iSCSI Failure" in iscsi_co_generic_cb if the task
hasn't completed with SCSI_STATUS_GOOD. However, we expect a failure in
some cases and handle it gracefully. This is the case for misaligned UNMAPs
and WRITESAME10/16 calls without UNMAP. In this case a failure in the
logs can be quite misleading.
While we are at it improve the logging to reveal which operation failed
at what LBA.
Signed-off-by: Peter Lieven <pl@kamp.de>
Message-Id: <1512733868-9009-3-git-send-email-pl@kamp.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The AioContext pointer argument to co_aio_sleep_ns() is only used for
the sleep timer. It does not affect where the caller coroutine is
resumed.
Due to changes to coroutine and AIO APIs it is now possible to drop the
AioContext pointer argument. This is safe to do since no caller has
specific requirements for which AioContext the timer must run in.
This patch drops the AioContext pointer argument and renames the
function to simplify the API.
Reported-by: Paolo Bonzini <pbonzini@redhat.com>
Reported-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 20171109102652.6360-1-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
If curl_global_init() fails, per the documentation no other curl
functions may be called, so make sure to check the return value.
Also, some minor changes to the initialization latch variable 'inited':
- Make it static in the file, for clarity
- Change the name for clarity
- Make it a bool
Signed-off-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Signed-off-by: Jeff Cody <jcody@redhat.com>
'tag' is already checked in the lines immediately preceding this check,
and set to non-NULL if NULL. No need to check again, it hasn't changed.
Signed-off-by: Jeff Cody <jcody@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Signed-off-by: Jeff Cody <jcody@redhat.com>
Set fake progress for non-dirty clusters in copy_bitmap initialization,
to. It simplifies code and allows further refactoring.
This patch changes user's view of backup progress, but formally it
doesn't changed: progress hops are just moved to the beginning.
Actually it's just a point of view: when do we actually skip clusters?
We can say in the very beginning, that we skip these clusters and do
not think about them later.
Of course, if go through disk sequentially, it's logical to say, that
we skip clusters between copied portions to the left and to the right
of them. But even now copying progress is not sequential because of
write notifiers. Future patches will introduce new backup architecture
which will do copying in several coroutines in parallel, so it will
make no sense to publish fake progress by parts in parallel with
other copying requests.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
Message-id: 20171012135313.227864-5-vsementsov@virtuozzo.com
Signed-off-by: Jeff Cody <jcody@redhat.com>
