As out-of-tree builds are preferred for qemu, running the qemu-iotests
in that out-of-tree build should be supported as well. To do so, a
symbolic link has to be created pointing to the check script in the
source directory. That script will check whether it has been run through
a symlink, and if so, will assume it is run in the build tree. All
output and temporary operations performed by iotests are then redirected
here and, unless specified otherwise by the user, QEMU_PROG etc. will be
set to paths appropriate for the build tree.

Also, drop making every test case executable if it is not yet, as this
would modify the source tree which is not desired for out-of-tree runs
and should be fixed in the repository anyway.
Signed-off-by: NMax Reitz <mreitz@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

When failure occurs, 'ret' need be set, or may return 0 to indicate
success. Previously, an error was set in errp, but 0 was returned
anyway. So let bdrv_append_temp_snapshot() return an error code and
use that for the bdrv_open() return value.

Also, error_propagate() need be called only one time within a function.
Signed-off-by: NChen Gang <gang.chen.5i5j@gmail.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

The to-replace-node-name is designed to allow repairing a broken Quorum file.
This patch introduces a new class TestRepairQuorum testing that the feature
works.
Some further work will be done on QEMU to improve the robustness of the tests.
Signed-off-by: NBenoit Canet <benoit@irqsave.net>
Reviewed-by: NMax Reitz <mreitz@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

drive-mirror will bdrv_swap the new BDS named node-name with the one
pointed by replaces when the mirroring is finished.
Signed-off-by: NBenoit Canet <benoit@irqsave.net>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

Commit 5a2d2cbd screwed up the the value of members device and action,
breaking tests/qemu-iotests/041.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Tested-By: NBenoit Canet <benoit@irqsave.net>
Reviewed-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NLuiz Capitulino <lcapitulino@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

Commit bcada37b dropped the (up to now undocumented) members type, len,
offset, speed, breaking tests/qemu-iotests/040 and 041.

Restore and document them.  This fixes 040, and partially fixes 041.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Tested-By: NBenoit Canet <benoit@irqsave.net>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

The old name is misleading in its new usage, so rename it.
Signed-off-by: NFam Zheng <famz@redhat.com>
Tested-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

This drops request handling code from dataplane, and uses code from
hw/block/virtio-blk.c.

It starts to use multiwrite as non-dataplane does.

Dataplane sets VirtIOBlock.complete_request to vring version, and calls
into non-dataplane's process handling. In complete_request_early,
qiov.size is added to vring push length, because it's also called in rw
completion now.
Signed-off-by: NFam Zheng <famz@redhat.com>
Tested-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

The BH must be called in the AioContext of bs. Currently it is only the
main loop, but with coming changes, it could also be a dataplane
IOThread.
Signed-off-by: NFam Zheng <famz@redhat.com>
Tested-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

So that dataplane can use virtio_blk_handle_request and
virtio_submit_multiwrite.
Signed-off-by: NFam Zheng <famz@redhat.com>
Tested-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

virtio_blk_req_complete will call VirtIOBlock.complete_request() to push
data and notify guest. No functional change.

Later, this will allow dataplane to provide it's own (vring_) version.
Signed-off-by: NFam Zheng <famz@redhat.com>
Tested-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

Make query-blockstats safe for dataplane by acquiring the
BlockDriverState's AioContext.  This ensures that the dataplane IOThread
and the main loop's monitor code do not race.

Note the assumption that acquiring the drive's BDS AioContext also
protects ->file and ->backing_hd.  This assumption is made by other
aio_context_acquire() callers too.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NFam Zheng <famz@redhat.com>
Tested-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

This function is only called from block/qapi.c.  There is no need to
keep it public.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NFam Zheng <famz@redhat.com>
Tested-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

out_sg is checked by iov_to_buf below, so it can be dropped.

Add assert and iov_discard_back around in_sg, as the in_sg is handled in
dataplane code.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

VirtIOBlockReq is allocated in process_request, and freed in command
functions.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

The virtio code currently assumes that the outhdr is in its own iovec.
This is not guaranteed by the spec, so we should relax this assumption.

Convert the VirtIOBlockReq.out field to structrue so that we can use
iov_to_buf and then discard the header from the beginning of iovec.
Suggested-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

In current virtio spec, inhdr is a single byte, and is unlikely to
change for both functionality and compatibility considerations.
Non-dataplane uses .in, and we are on the way to converge them. So
let's unify it to get cleaner code.

Remove .inhdr and use .in.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

Field "inhdr" is added temporarily for a more mechanical change, and
will be dropped in the next commit.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

Since it's set but not used.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

The block layer will handle the unaligned request.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

This will make converging with dataplane code easier.

Add virtio_blk_free_request to handle the freeing of request internal
fields.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

For later reusing by dataplane code.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

This new argument can be used to specify the node-name of the new mirrored BDS.
Signed-off-by: NBenoit Canet <benoit@irqsave.net>
Reviewed-by: NMax Reitz <mreitz@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

On read operations when this parameter is set and some replicas are corrupted
while quorum can be reached quorum will proceed to rewrite the correct version
of the data to fix the corrupted replicas.

This will shine with SSD where the FTL will remap the same block at another
place on rewrite.
Signed-off-by: NBenoit Canet <benoit@irqsave.net>
Reviewed-by: NMax Reitz <mreitz@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

If 'base' is smaller than the overlay image being committed into it,
then the base image will be grown in commit_run via bdrv_truncate().

This tests to make sure that this works, and the bdrv_truncate() is
not blocked when it shouldn't be.
Signed-off-by: NJeff Cody <jcody@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

If we check for the RESIZE blocker in bdrv_truncate(), that means a
commit will fail if the overlay layer is larger than the base, due to
the backing blocker.

This is a regression in behavior from 2.0; currently, commit will try to
grow the size of the base image to match the overlay size, if the
overlay size is larger.

By moving this into the QMP command qmp_block_resize(), it allows
usage of bdrv_truncate() within block jobs.
Signed-off-by: NJeff Cody <jcody@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

When running a libvirt test suite I've noticed the qemu-img is
crashing occasionally. Tracing the problem down led me to the
following valgrind output:

qemu.git $ valgrind -q ./qemu-img create -f qed -obacking_file=/dev/null,backing_fmt=raw qed
==14881== Invalid write of size 8
==14881==    at 0x1D263F: qemu_opts_create (qemu-option.c:692)
==14881==    by 0x130782: bdrv_img_create (block.c:5531)
==14881==    by 0x118DE0: img_create (qemu-img.c:462)
==14881==    by 0x11E7E4: main (qemu-img.c:2830)
==14881==  Address 0x11fedd38 is 24 bytes inside a block of size 232 free'd
==14881==    at 0x4C2CA5E: realloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==14881==    by 0x592D35E: g_realloc (in /usr/lib64/libglib-2.0.so.0.3800.2)
==14881==    by 0x1D38D8: qemu_opts_append (qemu-option.c:1129)
==14881==    by 0x13075E: bdrv_img_create (block.c:5528)
==14881==    by 0x118DE0: img_create (qemu-img.c:462)
==14881==    by 0x11E7E4: main (qemu-img.c:2830)
==14881==
Formatting 'qed', fmt=qed size=0 backing_file='/dev/null' backing_fmt='raw' cluster_size=65536
==14881== Invalid write of size 8
==14881==    at 0x1D28BE: qemu_opts_del (qemu-option.c:750)
==14881==    by 0x130BF3: bdrv_img_create (block.c:5638)
==14881==    by 0x118DE0: img_create (qemu-img.c:462)
==14881==    by 0x11E7E4: main (qemu-img.c:2830)
==14881==  Address 0x11fedd38 is 24 bytes inside a block of size 232 free'd
==14881==    at 0x4C2CA5E: realloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==14881==    by 0x592D35E: g_realloc (in /usr/lib64/libglib-2.0.so.0.3800.2)
==14881==    by 0x1D38D8: qemu_opts_append (qemu-option.c:1129)
==14881==    by 0x13075E: bdrv_img_create (block.c:5528)
==14881==    by 0x118DE0: img_create (qemu-img.c:462)
==14881==    by 0x11E7E4: main (qemu-img.c:2830)
==14881==

The problem is apparently in the qemu_opts_append(). Well, if it
gets called twice or more. On the first call, when @dst is NULL
some initialization is done during which @dst->head list gets
initialized. The list is initialized in a way, so that the list
tail points at the list head. However, the next time
qemu_opts_append() is called for new options to be added,
g_realloc() may move @dst to a new address making the old list tail
point at an invalid address. If that's the case, we must update the
list pointers.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

The only semantic change is that bs->open_flags gets BDRV_O_PROTOCOL set
now. This isn't useful, but it doesn't hurt either. The code that was
previously skipped by 'goto done' is automatically disabled because
protocol drivers don't support backing files (and if they did, this
would probably be a fix) and can't have snapshot_flags set.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

Since we parse backing.* options to add a backing file from the command
line when the driver didn't assign one, it has been possible to have a
backing file for e.g. raw images (it just was never accessed).

This is obvious nonsense and should be rejected.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

This recursion was introduced in commit 505d7583 in order to allow
nesting image formats. It only ever takes effect when the user
explicitly specifies a driver name and that driver isn't suitable for
the protocol level.

We can check this earlier in bdrv_open() and if the explicitly
requested driver is a format driver, clear BDRV_O_PROTOCOL so that
another bs->file layer is opened.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

It doesn't do much any more, we can move the code to bdrv_open() now.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NBenoit Canet <benoit@irqsave.net>
Reviewed-by: NEric Blake <eblake@redhat.com>

This moves the bdrv_open_file() call a bit down so that it can use the
bdrv_open() code that selects the right block driver.

The code between the old and the new call site is either common code
(the error message for an unknown driver has been unified now) or
doesn't run with cleared BDRV_O_PROTOCOL (added an if block in one
place, whereas the right path was already asserted in another place)
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NBenoit Canet <benoit@irqsave.net>

The "driver" entry in the options QDict is now only missing if we're
opening an image with format probing.

We also catch cases now where both the drv argument and a "driver"
option is specified, e.g. by specifying -drive format=qcow2,driver=raw
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

bs->options now contains the modified version of the options.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

The idea of bdrv_fill_options() is to convert every parameter for
opening images, in particular the filename and flags, to entries in the
options QDict.

This patch starts with moving the filename parsing and driver probing
part from bdrv_file_open() to the new function.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

upcoming libnfs will feature internal readahead support.
Add a knob to pass the optional readahead value as a URL
parameter.
Signed-off-by: NPeter Lieven <pl@kamp.de>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

this patch fixes the incorrect usage of strncmp and
adds simple error checking by means of parse_uint_full
instead of atoi for the supplied URL parameters.
Signed-off-by: NPeter Lieven <pl@kamp.de>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

All behavior and invariant should hold for images with 0 length, so
add a class to repeat all the tests in TestSingleDrive.

Hide two unapplicable test methods that would fail with 0 image length
because it's also used as cluster size.
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

There should be a BLOCK_JOB_READY event with active commit, regardless
of image length. Let's test the 0 length image case, and make sure it
goes through the ready->complete process.
Signed-off-by: NFam Zheng <famz@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>