This patch changes where the throttling state is stored (used to be the
BlockDriverState, now it is the BlockBackend), but it doesn't actually
make it a BB level feature yet. For example, throttling is still
disabled when the BDS is detached from the BB.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
As a first step towards moving I/O throttling to the BlockBackend level,
this patch changes all pointers in struct ThrottleGroup from referencing
a BlockDriverState to referencing a BlockBackend.
This change is valid because we made sure that throttling can only be
enabled on BDSes which have a BB attached.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
In some cases, we want to take a quorum child offline, and take
another child online.
Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Changlong Xie <xiecl.fnst@cn.fujitsu.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Message-id: 1462865799-19402-2-git-send-email-xiecl.fnst@cn.fujitsu.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
The block layer has a couple of cases where it can lose
Force Unit Access semantics when writing a large block of
zeroes, such that the request returns before the zeroes
have been guaranteed to land on underlying media.
SCSI does not support FUA during WRITESAME(10/16); FUA is only
supported if it falls back to WRITE(10/16). But where the
underlying device is new enough to not need a fallback, it
means that any upper layer request with FUA semantics was
silently ignoring BDRV_REQ_FUA.
Conversely, NBD has situations where it can support FUA but not
ZERO_WRITE; when that happens, the generic block layer fallback
to bdrv_driver_pwritev() (or the older bdrv_co_writev() in qemu
2.6) was losing the FUA flag.
The problem of losing flags unrelated to ZERO_WRITE has been
latent in bdrv_co_do_write_zeroes() since commit aa7bfbff, but
back then, it did not matter because there was no FUA flag. It
became observable when commit 93f5e6d8 paved the way for flags
that can impact correctness, when we should have been using
bdrv_co_writev_flags() with modified flags. Compare to commit
9eeb6dd, which got flag manipulation right in
bdrv_co_do_zero_pwritev().
Symptoms: I tested with qemu-io with default writethrough cache
(which is supposed to use FUA semantics on every write), and
targetted an NBD client connected to a server that intentionally
did not advertise NBD_FLAG_SEND_FUA. When doing 'write 0 512',
the NBD client sent two operations (NBD_CMD_WRITE then
NBD_CMD_FLUSH) to get the fallback FUA semantics; but when doing
'write -z 0 512', the NBD client sent only NBD_CMD_WRITE.
The fix is do to a cleanup bdrv_co_flush() at the end of the
operation if any step in the middle relied on a BDS that does
not natively support FUA for that step (note that we don't
need to flush after every operation, if the operation is broken
into chunks based on bounce-buffer sizing). Each BDS gains a
new flag .supported_zero_flags, which parallels the use of
.supported_write_flags but only when accessing a zero write
operation (the flags MUST be different, because of SCSI having
different semantics based on WRITE vs. WRITESAME; and also
because BDRV_REQ_MAY_UNMAP only makes sense on zero writes).
Also fix some documentation to describe -ENOTSUP semantics,
particularly since iscsi depends on those semantics.
Down the road, we may want to add a driver where its
.bdrv_co_pwritev() honors all three of BDRV_REQ_FUA,
BDRV_REQ_ZERO_WRITE, and BDRV_REQ_MAY_UNMAP, and advertise
this via bs->supported_write_flags for blocks opened by that
driver; such a driver should NOT supply .bdrv_co_write_zeroes
nor .supported_zero_flags. But none of the drivers touched
in this patch want to do that (the act of writing zeroes is
different enough from normal writes to deserve a second
callback).
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Pre-patch, .supported_write_flags lives at the driver level, which
means we are blindly declaring that all block devices using a
given driver will either equally support FUA, or that we need a
fallback at the block layer. But there are drivers where FUA
support is a per-block decision: the NBD block driver is dependent
on the remote server advertising NBD_FLAG_SEND_FUA (and has
fallback code to duplicate the flush that the block layer would do
if NBD had not set .supported_write_flags); and the iscsi block
driver is dependent on the mode sense bits advertised by the
underlying device (and is currently silently ignoring FUA requests
if the underlying device does not support FUA).
The fix is to make supported flags as a per-BDS option, set during
.bdrv_open(). This patch moves the variable and fixes NBD and iscsi
to set it only conditionally; later patches will then further
simplify the NBD driver to quit duplicating work done at the block
layer, as well as tackle the fact that SCSI does not support FUA
semantics on WRITESAME(10/16) but only on WRITE(10/16).
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Vmdk images have metadata to indicate the vmware virtual
hardware version image was created/tested to run with.
Allow users to specify that version via new 'hwversion'
option.
[ kwolf: Adjust qemu-iotests common.filter ]
Signed-off-by: Janne Karhunen <Janne.Karhunen@gmail.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
There are no block drivers left that implement the old .bdrv_read/write
interface, so it can be removed now. This gets us rid of the
corresponding emulation functions, too.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Many parts of the block layer are already byte granularity. The block
driver interface, however, was still missing an interface that allows
making use of this. This patch introduces a new BlockDriver interface,
which is based on coroutines, vectored, has flags and uses a byte
granularity. This is now the preferred interface for new drivers.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
It used to be an internal helper function just for implementing
bdrv_co_do_readv/writev(), but now that it's a public interface, it
deserves a name without "do" in it.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Extract the handling of io_plug "depth" from linux-aio.c and let the
main bdrv_drain loop do nothing but wait on I/O.
Like the two newly introduced functions, bdrv_io_plug and bdrv_io_unplug
now operate on all children. The visit order is now symmetrical between
plug and unplug, making it possible for formats to implement plug/unplug.
Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Extract the handling of throttling from bdrv_flush_io_queue. These
new functions will soon become BdrvChildRole callbacks, as they can
be generalized to "beginning of drain" and "end of drain".
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We want to remove throttled_reqs from block/io.c. This is the easy
part---hide the handling of throttled_reqs during disable/enable of
throttling within throttle-groups.c.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Using the nested aio_poll() in coroutine is a bad idea. This patch
replaces the aio_poll loop in bdrv_drain with a BH, if called in
coroutine.
For example, the bdrv_drain() in mirror.c can hang when a guest issued
request is pending on it in qemu_co_mutex_lock().
Mirror coroutine in this case has just finished a request, and the block
job is about to complete. It calls bdrv_drain() which waits for the
other coroutine to complete. The other coroutine is a scsi-disk request.
The deadlock happens when the latter is in turn pending on the former to
yield/terminate, in qemu_co_mutex_lock(). The state flow is as below
(assuming a qcow2 image):
mirror coroutine scsi-disk coroutine
-------------------------------------------------------------
do last write
qcow2:qemu_co_mutex_lock()
...
scsi disk read
tracked request begin
qcow2:qemu_co_mutex_lock.enter
qcow2:qemu_co_mutex_unlock()
bdrv_drain
while (has tracked request)
aio_poll()
In the scsi-disk coroutine, the qemu_co_mutex_lock() will never return
because the mirror coroutine is blocked in the aio_poll(blocking=true).
With this patch, the added qemu_coroutine_yield() allows the scsi-disk
coroutine to make progress as expected:
mirror coroutine scsi-disk coroutine
-------------------------------------------------------------
do last write
qcow2:qemu_co_mutex_lock()
...
scsi disk read
tracked request begin
qcow2:qemu_co_mutex_lock.enter
qcow2:qemu_co_mutex_unlock()
bdrv_drain.enter
> schedule BH
> qemu_coroutine_yield()
> qcow2:qemu_co_mutex_lock.return
> ...
tracked request end
...
(resumed from BH callback)
bdrv_drain.return
...
Reported-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-id: 1459855253-5378-2-git-send-email-famz@redhat.com
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
The only remaining users were block jobs (mirror and backup) which
unconditionally enabled WCE on the BlockBackend of the target image. As
these block jobs don't go through BlockBackend for their I/O requests,
they aren't affected by this setting anyway but always get a writeback
mode, so that call can be removed.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
The previous patches have successively made blk->enable_write_cache the
true source for the information whether a writethrough mode must be
implemented. The corresponding BDRV_O_CACHE_WB is only useless baggage
we're carrying around, so now's the time to remove it.
At the same time, we remove the 'cache.writeback' option parsing on the
BDS level as the only effect was setting the BDRV_O_CACHE_WB flag.
This change requires test cases that explicitly enabled the option to
drop it. Other than that and the change of the error message when
writethrough is enabled on the BDS level (from "Can't set writethrough
mode" to "doesn't support the option"), there should be no change in
behaviour.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
This function will allow drivers to implement BDRV_REQ_FUA natively
instead of sending a separate flush after the write.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Now that WCE is handled on the BlockBackend level, the flag is
meaningless for BDSes. As the schema requires us to fill the field,
we return an enabled write cache for them.
Note that this means that querying the BlockBackend name may return
writethrough as the cache information, whereas querying the node-name of
the root of that same BlockBackend will return writeback.
This may appear odd at first, but it actually makes sense because it
correctly repesents the layer that implements the WCE handling. This
becomes more apparent when you consider nodes that are the root node of
multiple BlockBackends, where each BB can have its own WCE setting.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Whether a write cache is used or not is a decision that concerns the
user (e.g. the guest device) rather than the backend. It was already
logically part of the BB level as bdrv_move_feature_fields() always kept
it on top of the BDS tree; with this patch, the core of it (the actual
flag and the additional flushes) is also implemented there.
Direct callers of bdrv_open() must pass BDRV_O_CACHE_WB now if bs
doesn't have a BlockBackend attached.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
It's like bdrv_parse_cache_flags(), except that writethrough mode isn't
included in the flags, but returned as a separate bool.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
This patch adds callback for flush request. This callback is responsible
for flushing whole block devices stack. bdrv_flush function does not
proceed to underlying devices. It should be performed by this callback
function, if needed.
Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
For a couple of releases we have been warning
Encrypted images are deprecated
Support for them will be removed in a future release.
You can use 'qemu-img convert' to convert your image to an unencrypted one.
This warning was issued by system emulators, qemu-img, qemu-nbd
and qemu-io. Such a broad warning was issued because the original
intention was to rip out all the code for dealing with encryption
inside the QEMU block layer APIs.
The new block encryption framework used for the LUKS driver does
not rely on the unloved block layer API for encryption keys,
instead using the QOM 'secret' object type. It is thus no longer
appropriate to warn about encryption unconditionally.
When the qcow/qcow2 drivers are converted to use the new encryption
framework too, it will be practical to keep AES-CBC support present
for use in qemu-img, qemu-io & qemu-nbd to allow for interoperability
with older QEMU versions and liberation of data from existing encrypted
qcow2 files.
This change moves the warning out of the generic block code and
into the qcow/qcow2 drivers. Further, the warning is set to only
appear when running the system emulators, since qemu-img, qemu-io,
qemu-nbd are expected to support qcow2 encryption long term now that
the maint burden has been eliminated.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
When opening an image it is useful to know whether the caller
intends to perform I/O on the image or not. In the case of
encrypted images this will allow the block driver to avoid
having to prompt for decryption keys when we merely want to
query header metadata about the image. eg qemu-img info
This flag is enforced at the top level only, since even if
we don't want todo I/O on the 'qcow2' file payload, the
underlying 'file' driver will still need todo I/O to read
the qcow2 header, for example.
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The function is unused since commit f21d96d0 ('block: Use BdrvChild in
BlockBackend').
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
The call in hmp_drive_del() is dead code because blk_remove_bs() is
called a few lines above. The only other remaining user is
bdrv_delete(), which only abuses bdrv_make_anon() to remove it from the
named nodes list. This path inlines the list entry removal into
bdrv_delete() and removes bdrv_make_anon().
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
qemu-common.h should only be included by .c files. Its file comment
explains why: "No header file should depend on qemu-common.h, as this
would easily lead to circular header dependencies."
qemu/iov.h includes qemu-common.h for QEMUIOVector stuff. Move all
that to qemu/iov.h and drop the ill-advised include. Include
qemu/iov.h where the QEMUIOVector stuff is now missing.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Re-run scripts/clean-includes to apply the previous commit's
corrections and updates. Besides redundant qemu/typedefs.h, this only
finds a redundant config-host.h include in ui/egl-helpers.c. No idea
how that escaped the previous runs.
Some manual whitespace trimming around dropped includes squashed in.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch introduces blk_co_preadv() as a central function on the
BlockBackend level that is supposed to handle all read requests from the
BB to its root BDS eventually.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Add a function for iterating over all monitor-owned BlockDriverStates so
the generic block layer can do so.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Move bdrv_commit_all() and bdrv_flush_all() to the BlockBackend level.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The only code change is making bdrv_dirty_bitmap_truncate public. It is
used in block.c.
Also two long lines (bdrv_get_dirty) are wrapped.
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Message-id: 1457412306-18940-5-git-send-email-famz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
Following patches to refactor and move block dirty bitmap code could use
this.
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Message-id: 1457412306-18940-4-git-send-email-famz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
This patch adds an option to the drive_add HMP command to create only a
BlockDriverState without a BlockBackend on top.
The motivation for this is that libvirt needs to specify options to a
migration target (specifically, detect-zeroes). drive-mirror doesn't
allow specifying options, and the proper way to do this is to create the
target BDS separately with blockdev-add (where you can specify options)
and then use blockdev-mirror to that BDS.
However, libvirt can't use blockdev-add as long as it is still
experimental, and we're expecting that it will still take some time, so
we need to resort to drive_add.
The problem with drive_add is that so far it always created a BB, and
BDSes with a BB can't be used as a mirroring target as long as we don't
support multiple BBs per BDS - and while we're working towards that
goal, it's another thing that will still take some time.
So to achieve the goal, the simplest solution to provide the
functionality now without adding one-off options to the mirror QMP
commands is to extend drive_add to create nodes without BBs.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Since commit 91a097e, we end up with a somewhat weird cache mode
configuration with snapshot=on: The commit broke the cache mode
inheritance for the snapshot overlay so that it is opened as
writethrough instead of unsafe now. The following bdrv_append() call to
put it on top of the tree swaps the WCE flag with the snapshot's backing
file (i.e. the originally given file), so what we eventually get is
cache=writeback on the temporary overlay and
cache=writethrough,cache.no-flush=on on the real image file.
This patch changes things so that the temporary overlay gets
cache=unsafe again like it used to, and the real images get whatever the
user specified. This means that cache.direct is now respected even with
snapshot=on, and in the case of committing changes, the final flush is
no longer ignored except explicitly requested by the user.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Clean up includes so that osdep.h is included first and headers
which it implies are not included manually.
This commit was created with scripts/clean-includes.
NB: If this commit breaks compilation for your out-of-tree
patchseries or fork, then you need to make sure you add
#include "qemu/osdep.h" to any new .c files that you have.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Eric Blake <eblake@redhat.com>
This extends the NBD protocol handling code so that it is capable
of negotiating TLS support during the connection setup. This involves
requesting the STARTTLS protocol option before any other NBD options.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Message-Id: <1455129674-17255-14-git-send-email-berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Now that all callers are converted to use I/O channels for
initial connection setup, it is possible to switch the core
NBD protocol handling core over to use QIOChannel APIs for
actual sockets I/O.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Message-Id: <1455129674-17255-7-git-send-email-berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
With a mirror job running on a virtio-blk dataplane disk, sending "q" to
HMP will cause a dead loop in block_job_finish_sync.
This is because the aio_poll() only processes the AIO context of bs
which has no more work to do, while the main loop BH that is scheduled
for setting the job->completed flag is never processed.
Fix this by adding a flag in BlockJob structure, to track which context
to poll for the block job to make progress. Its value is set to true
when block_job_coroutine_complete() is called, and is checked in
block_job_finish_sync to determine which context to poll.
Suggested-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-id: 1454379144-29807-1-git-send-email-famz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
This fixes a regression introduced with commit 3f09bfbc7. Multiple
bugs arise in conjunction with live snapshots and mirroring operations
(which include active layer commit).
After a live snapshot occurs, the active layer and the base layer both
have a non-NULL tqe_prev field in the device_list, although the base
node's tqe_prev field points to a NULL entry. This non-NULL tqe_prev
field occurs after the bdrv_append() in the external snapshot calls
change_parent_backing_link().
In change_parent_backing_link(), when the previous active layer is
removed from device_list, the device_list.tqe_prev pointer is not
set to NULL.
The operating scheme in the block layer is to indicate that a BDS belongs
in the bdrv_states device_list iff the device_list.tqe_prev pointer
is non-NULL.
This patch does two things:
1.) Introduces a new block layer helper bdrv_device_remove() to remove a
BDS from the device_list, and
2.) uses that new API, which also fixes the regression once used in
change_parent_backing_link().
Signed-off-by: Jeff Cody <jcody@redhat.com>
Message-id: 0cd51e11c0666c04ddb7c05293fe94afeb551e89.1454376655.git.jcody@redhat.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
The added parameter can be used to return the BDS pointer which the
valid offset is referring to. Its value should be ignored unless
BDRV_BLOCK_OFFSET_VALID in ret is set.
Until block drivers fill in the right value, let's clear it explicitly
right before calling .bdrv_get_block_status.
The "bs->file" condition in bdrv_co_get_block_status is kept now to keep iotest
case 102 passing, and will be fixed once all drivers return the right file
pointer.
Signed-off-by: Fam Zheng <famz@redhat.com>
Message-id: 1453780743-16806-2-git-send-email-famz@redhat.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
As a side effect, we can now make x-blockdev-del's check whether a BDS
is actually owned by the monitor explicit.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
