Commit Graph

466 Commits

Author SHA1 Message Date
Peter Maydell
7fa124b273 Error reporting patches for 2016-06-20
-----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQIbBAABAgAGBQJXaAQPAAoJEDhwtADrkYZTbPMP+OmXlf1IwcUONhaHqD0SXAdC
 Q5NNHheNHbIdzrsOmgj2GAxCeGG08REPrdYLb8N9lI3Fev8ggQ0v3d9qiStL6Nvz
 2B7DeTxXba9Qmc2gY86RR4f7T6VK+8k1Kj9f3YcAoXHucdjDLRmtUozwES3Kk8mn
 oeBqd07yuTAO93xQhF6BDv3mfagPZV+kp15Se04ufWPpT9eG3Cp9Pl4/Yad/wJUr
 2B4pIvYnhISVlZIOvpuzdydBaCb/U0YefjWBDkawLwRBtM/kpNKoDdqr2UA0yChQ
 Xulgpt+UQLZ4aOYGLkzqBT7v3kDeEnYUFDdE03xZBLbHddOU05NEhLgzRVUEAdUS
 EAKWvWu692USDh9a2r0AYLqZW6J0+Fyi4fZ34gSuP69oi74CIvjJHDAEwsSdt6bm
 WHCchvfIk9kinzGiFA3hkAbVWAbI0EpB0J7k3mgCVVVelBpF9b9U4M2ydx9ZPufX
 t4ULXxQO9Qyxr7r0uiznugQAjIRwLMDOPPd1F5Vi+LX0wF3Pcrtc/F+3ehoLBjBD
 6zZA5lZVlUEISdzzytfjeX3ch5vRFwlujrTd2anxRuduuBV1/ztvE1v0Nsiii6Sw
 15BaHQJvBnMnAIO5LfhnAaVpIBk0jpQpwVyRcNUpeZ1VU2qlqVS4zxYZ0FDQgyer
 IZa1qYJ0Sz+oMkq9RXU=
 =TptL
 -----END PGP SIGNATURE-----

Merge remote-tracking branch 'remotes/armbru/tags/pull-error-2016-06-20' into staging

Error reporting patches for 2016-06-20

# gpg: Signature made Mon 20 Jun 2016 15:56:15 BST
# gpg:                using RSA key 0x3870B400EB918653
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>"
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>"
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* remotes/armbru/tags/pull-error-2016-06-20:
  log: Fix qemu_set_log_filename() error handling
  log: Fix qemu_set_dfilter_ranges() error reporting
  log: Plug memory leak on multiple -dfilter
  coccinelle: Remove unnecessary variables for function return value
  error: Remove unnecessary local_err variables
  error: Remove NULL checks on error_propagate() calls
  vl: Error messages need to go to stderr, fix some

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2016-06-20 16:19:18 +01:00
Eduardo Habkost
621ff94d50 error: Remove NULL checks on error_propagate() calls
error_propagate() already ignores local_err==NULL, so there's no
need to check it before calling.

Coccinelle patch used to perform the changes added to
scripts/coccinelle/error_propagate_null.cocci.

Reviewed-by: Eric Blake <eblake@redhat.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <1465855078-19435-2-git-send-email-ehabkost@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-06-20 16:38:13 +02:00
Stefan Hajnoczi
17bd51f936 blockjob: move iostatus reset out of block_job_enter()
The QMP block-job-resume command and cancellation may want to reset the
job's iostatus.  The next patches add a user who does not want to reset
iostatus so move it up to block_job_enter() callers.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 1466096189-6477-2-git-send-email-stefanha@redhat.com
2016-06-20 11:44:12 +01:00
Max Reitz
274fccee2b block/mirror: Fix target backing BDS
Currently, we are trying to move the backing BDS from the source to the
target in bdrv_replace_in_backing_chain() which is called from
mirror_exit(). However, mirror_complete() already tries to open the
target's backing chain with a call to bdrv_open_backing_file().

First, we should only set the target's backing BDS once. Second, the
mirroring block job has a better idea of what to set it to than the
generic code in bdrv_replace_in_backing_chain() (in fact, the latter's
conditions on when to move the backing BDS from source to target are not
really correct).

Therefore, remove that code from bdrv_replace_in_backing_chain() and
leave it to mirror_complete().

Depending on what kind of mirroring is performed, we furthermore want to
use different strategies to open the target's backing chain:

- If blockdev-mirror is used, we can assume the user made sure that the
  target already has the correct backing chain. In particular, we should
  not try to open a backing file if the target does not have any yet.

- If drive-mirror with mode=absolute-paths is used, we can and should
  reuse the already existing chain of nodes that the source BDS is in.
  In case of sync=full, no backing BDS is required; with sync=top, we
  just link the source's backing BDS to the target, and with sync=none,
  we use the source BDS as the target's backing BDS.
  We should not try to open these backing files anew because this would
  lead to two BDSs existing per physical file in the backing chain, and
  we would like to avoid such concurrent access.

- If drive-mirror with mode=existing is used, we have to use the
  information provided in the physical image file which means opening
  the target's backing chain completely anew, just as it has been done
  already.
  If the target's backing chain shares images with the source, this may
  lead to multiple BDSs per physical image file. But since we cannot
  reliably ascertain this case, there is nothing we can do about it.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20160610185750.30956-3-mreitz@redhat.com
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
2016-06-16 15:20:37 +02:00
Alberto Garcia
f0f55deda2 block: use the block job list in qmp_query_block_jobs()
qmp_query_block_jobs() uses bdrv_next() to look for block jobs, but
this function can only find those in top-level BlockDriverStates.

This patch uses block_job_next() instead.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Message-id: a8b7e5497b7c1fa67c12fcceae1630d01c3b1f96.1464346103.git.berto@igalia.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
2016-06-16 15:20:37 +02:00
Colin Lord
38a53d506b blockdev: clarify error on attempt to open locked tray
When opening a device with a locked tray, gives an error explaining the
device tray is locked and that the user should wait and try again. This
is less confusing than the previous error, which simply stated that the
tray was locked.

Signed-off-by: Colin Lord <clord@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-06-16 15:19:55 +02:00
Colin Lord
bf18bee547 blockdev: clean up error handling in do_open_tray
Returns negative error codes and accompanying error messages in cases where
the device has no tray or the tray is locked and isn't forced open. This
extra information should result in better flexibility in functions that
call do_open_tray.

Suggested by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Colin Lord <clord@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-06-08 10:21:09 +02:00
Fam Zheng
efd7556708 blockdev-backup: Don't move target AioContext if it's attached
If the BDS is attached, it will want to stay on the AioContext where its
BlockBackend is. Don't call bdrv_set_aio_context in this case.

Signed-off-by: Fam Zheng <famz@redhat.com>
Message-id: 1463969978-24970-3-git-send-email-famz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2016-06-07 14:40:50 +01:00
Fam Zheng
0d97891312 blockdev-backup: Use bdrv_lookup_bs on target
This allows backing up to a BDS that has not been attached to any BB.

Signed-off-by: Fam Zheng <famz@redhat.com>
Message-id: 1463969978-24970-2-git-send-email-famz@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2016-06-07 14:40:50 +01:00
Kevin Wolf
5c438bc68c backup: Use BlockBackend for I/O
This changes the backup block job to use the job's BlockBackend for
performing its I/O. job->bs isn't used by the backup code any more
afterwards.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-05-25 19:04:21 +02:00
Kevin Wolf
e253f4b897 mirror: Use BlockBackend for I/O
This changes the mirror block job to use the job's BlockBackend for
performing its I/O. job->bs isn't used by the mirroring code any more
afterwards.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-05-25 19:04:21 +02:00
Kevin Wolf
b880481579 mirror: Allow target that already has a BlockBackend
We had to forbid mirroring to a target BDS that already had a BB
attached because the node swapping at job completion would add a second
BB and we didn't support multiple BBs on a single BDS at the time. Now
we do, so we can lift the restriction.

As we allow additional BlockBackends for the target, we must expect
other users to be sending requests. There may no requests be in flight
during the graph modification, so we have to drain those users now.

The core part of this patch is a revert of commit 40365552.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-05-25 19:04:21 +02:00
Max Reitz
109525ad6a block: Drop errp parameter from blk_new()
blk_new() cannot fail so its Error ** parameter has become superfluous.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-05-25 19:04:10 +02:00
Max Reitz
5b3639371c block: Make bdrv_open() return a BDS
There are no callers to bdrv_open() or bdrv_open_inherit() left that
pass a pointer to a non-NULL BDS pointer as the first argument of these
functions, so we can finally drop that parameter and just make them
return the new BDS.

Generally, the following pattern is applied:

    bs = NULL;
    ret = bdrv_open(&bs, ..., &local_err);
    if (ret < 0) {
        error_propagate(errp, local_err);
        ...
    }

by

    bs = bdrv_open(..., errp);
    if (!bs) {
        ret = -EINVAL;
        ...
    }

Of course, there are only a few instances where the pattern is really
pure.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-05-25 19:04:10 +02:00
Kevin Wolf
88be7b4be4 block: Fix bdrv_next() memory leak
The bdrv_next() users all leaked the BdrvNextIterator after completing
the iteration. Simply changing bdrv_next() to free the iterator before
returning NULL at the end of list doesn't work because some callers exit
the loop before looking at all BDSes.

This patch moves the BdrvNextIterator from the heap to the stack of
the caller and switches to a bdrv_first()/bdrv_next() interface for
initialising the iterator.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
2016-05-25 19:04:10 +02:00
John Snow
3a3086b72a block: clarify error message for qmp-eject
If you use HMP's eject but the CDROM tray is locked, you may get a
confusing error message informing you that the "tray isn't open."

As this is the point of eject, we can do a little better and help
clarify that the tray was locked and that it (might) open up later,
so try again.

It's not ideal, but it makes the semantics of the (legacy) eject
command more understandable to end users when they try to use it.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-05-19 16:45:31 +02:00
Kevin Wolf
1f0c461b82 block: Remove BlockDriverState.blk
This patch removes the remaining users of bs->blk, which will allow us
to have multiple BBs on top of a single BDS. In the meantime, all checks
that are currently in place to prevent the user from creating such
setups can be switched to bdrv_has_blk() instead of accessing BDS.blk.

Future patches can allow them and e.g. enable users to mirror to a block
device that already has a BlockBackend on it.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-05-19 16:45:31 +02:00
Kevin Wolf
7c8eece45b block: Avoid bs->blk in bdrv_next()
We need to introduce a separate BdrvNextIterator struct that can keep
more state than just the current BDS in order to avoid using the bs->blk
pointer.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-05-19 16:45:31 +02:00
Kevin Wolf
b26ded9a7d Revert "block: Forbid I/O throttling on nodes with multiple parents for 2.6"
This reverts commit 76b223200e.

Now that I/O throttling is fully done on the BlockBackend level, there
is no reason any more to block I/O throttling for nodes with multiple
parents as the parents don't influence each other any more.

Conflicts:
	block.c

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
2016-05-19 16:45:31 +02:00
Kevin Wolf
7ca7f0f6db block: Decouple throttling from BlockDriverState
This moves the throttling related part of the BDS life cycle management
to BlockBackend. The throttling group reference is now kept even when no
medium is inserted.

With this commit, throttling isn't disabled and then re-enabled any more
during graph reconfiguration. This fixes the temporary breakage of I/O
throttling when used with live snapshots or block jobs that manipulate
the graph.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
2016-05-19 16:45:30 +02:00
Kevin Wolf
97148076e8 block: Move I/O throttling configuration functions to BlockBackend
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
2016-05-19 16:45:30 +02:00
Kevin Wolf
27ccdd5259 block: Move throttling fields from BDS to BB
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>
2016-05-19 16:45:30 +02:00
Kevin Wolf
a5614993d7 block: Make sure throttled BDSes always have a BB
It was already true in principle that a throttled BDS always has a BB
attached, except that the order of operations while attaching or
detaching a BDS to/from a BB wasn't careful enough.

This commit breaks graph manipulations while I/O throttling is enabled.
It would have been possible to keep things working with some temporary
hacks, but quite cumbersome, so it's not worth the hassle. We'll fix
things again in a minute.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
2016-05-19 16:45:29 +02:00
Wen Congyang
7f82159769 qmp: add monitor command to add/remove a child
The new QMP command name is x-blockdev-change. It's just for adding/removing
quorum's child now, and doesn't support all kinds of children, all kinds of
operations, nor all block drivers. So it is experimental now.

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: 1462865799-19402-4-git-send-email-xiecl.fnst@cn.fujitsu.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2016-05-12 15:33:23 +02:00
Wei Jiangang
547cb1574e block: Fix typo in comment
s/imlement/implement/

Signed-off-by: Wei Jiangang <weijg.fnst@cn.fujitsu.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-05-12 15:22:08 +02:00
Kevin Wolf
76b223200e block: Forbid I/O throttling on nodes with multiple parents for 2.6
As the patches to move I/O throttling to BlockBackend didn't make it in
time for the 2.6 release, but the release adds new ways of configuring
VMs whose behaviour would change once the move is done, we need to
outlaw such configurations temporarily.

The problem exists whenever a BDS has more users than just its BB, for
example it is used as a backing file for another node. (This wasn't
possible in 2.5 yet as we introduced node references to specify a
backing file only recently.) In these cases, the throttling would
apply to these other users now, but after moving throttling to the
BlockBackend the other users wouldn't be throttled any more.

This patch prevents making new references to a throttled node as well as
using monitor commands to throttle a node with multiple parents.

Compared to 2.5 this changes behaviour in some corner cases where
references were allowed before, like bs->file or Quorum children. It
seems reasonable to assume that users didn't use I/O throttling on such
low level nodes. With the upcoming move of throttling into BlockBackend,
such configurations won't be possible anyway.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
2016-04-05 09:22:28 +02:00
Paolo Bonzini
5cf87fd68e block: forbid x-blockdev-del from acting on DriveInfo
Failing on -drive/drive_add created BlockBackends was a
requirement for x-blockdev-del, but it sneaked through
the patch review.  Let's fix it now.

Example:

$ x86_64-softmmu/qemu-system-x86_64 -drive if=none,file=null-co://,id=null -qmp stdio
>> {'execute':'qmp_capabilities'}
<< {"return": {}}
>> {'execute':'x-blockdev-del','arguments':{'id':'null'}}
<< {"error": {"class": "GenericError", "desc": "Deleting block backend added with drive-add is not supported"}}

And without a DriveInfo:

>> { "execute": "blockdev-add", "arguments": { "options": { "driver":"null-co", "id":"null2"}}}
<< {"return": {}}
>> {'execute':'x-blockdev-del','arguments':{'id':'null2'}}
<< {"return": {}}

Suggested-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-04-05 09:22:28 +02:00
Kevin Wolf
61de4c6808 block: Remove BDRV_O_CACHE_WB
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>
2016-03-30 12:16:03 +02:00
Kevin Wolf
04feb4a507 block: Use bdrv_parse_cache_mode() in drive_init()
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-03-30 12:16:02 +02:00
Kevin Wolf
72e775c7d9 block: Always set writeback mode in blk_new_open()
All callers of blk_new_open() either don't rely on the WCE bit set after
blk_new_open() because they explicitly set it anyway, or they pass
BDRV_O_CACHE_WB unconditionally.

This patch changes blk_new_open() so that it always enables writeback
mode and asserts that BDRV_O_CACHE_WB is clear. For those callers that
used to pass BDRV_O_CACHE_WB unconditionally, the flag is removed now.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-03-30 12:16:01 +02:00
Kevin Wolf
e4b24b497e block: blockdev_init(): Call blk_set_enable_write_cache() explicitly
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-03-30 12:16:01 +02:00
Kevin Wolf
73ac451f34 block: Reject writethrough mode except at the root
Writethrough mode is going to become a BlockBackend feature rather than
a BDS one, so forbid it in places where we won't be able to support it
when the code finally matches the envisioned design.

We only allowed setting the cache mode of non-root nodes after the 2.5
release, so we're still free to make this change.

The target of block jobs is now always opened in a writeback mode
because it doesn't have a BlockBackend attached. This makes more sense
anyway because block jobs know when to flush. If the graph is modified
on job completion, the original cache mode moves to the new root, so
for the guest device writethough always stays enabled if it was
configured this way.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
2016-03-30 11:59:32 +02:00
Kevin Wolf
4c8449832c block: Remove copy-on-read from bdrv_move_feature_fields()
Ever since we first introduced bdrv_append() in commit 8802d1fd ('qapi:
Introduce blockdev-group-snapshot-sync command'), the copy-on-read flag
was moved to the new top layer when taking a snapshot. The only problem
is that it doesn't make a whole lot of sense.

The use case for manually enabled CoR is to avoid reading data twice
from a slow remote image, so we want to save it to a local overlay, say
an ISO image accessed via HTTP to a local qcow2 overlay. When taking a
snapshot, we end up with a backing chain like this:

    http <- local.qcow2 <- snap_overlay.qcow2

There is no point in doing CoR from local.qcow2 into snap_overlay.qcow2,
we just want to keep copying data from the remote source into
local.qcow2.

The other use case of CoR is in the context of streaming, which isn't
very interesting for bdrv_move_feature_fields() because op blockers
prevent this combination.

This patch makes the copy-on-read flag stay on the image for which it
was originally set and prevents it from being propagated to the new
overlay. It is no longer intended to move CoR to the BlockBackend level.
In order for this to make sense, we also need to keep the respective
image read-write.

As a side effect of these changes, creating a live snapshot image (as
opposed to using an existing externally created one) on top of a COR
block device works now. It used to fail because it tried to open its
backing file both read-only and with COR.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
2016-03-30 11:59:32 +02:00
Kevin Wolf
63eaaae08c block: Remove bdrv_make_anon()
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>
2016-03-30 11:59:32 +02:00
Veronia Bahaa
f348b6d1a5 util: move declarations out of qemu-common.h
Move declarations out of qemu-common.h for functions declared in
utils/ files: e.g. include/qemu/path.h for utils/path.c.
Move inline functions out of qemu-common.h and into new files (e.g.
include/qemu/bcd.h)

Signed-off-by: Veronia Bahaa <veroniabahaa@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2016-03-22 22:20:17 +01:00
Eric Blake
32bafa8fdd qapi: Don't special-case simple union wrappers
Simple unions were carrying a special case that hid their 'data'
QMP member from the resulting C struct, via the hack method
QAPISchemaObjectTypeVariant.simple_union_type().  But by using
the work we started by unboxing flat union and alternate
branches, coupled with the ability to visit the members of an
implicit type, we can now expose the simple union's implicit
type in qapi-types.h:

| struct q_obj_ImageInfoSpecificQCow2_wrapper {
|     ImageInfoSpecificQCow2 *data;
| };
|
| struct q_obj_ImageInfoSpecificVmdk_wrapper {
|     ImageInfoSpecificVmdk *data;
| };
...
| struct ImageInfoSpecific {
|     ImageInfoSpecificKind type;
|     union { /* union tag is @type */
|         void *data;
|-        ImageInfoSpecificQCow2 *qcow2;
|-        ImageInfoSpecificVmdk *vmdk;
|+        q_obj_ImageInfoSpecificQCow2_wrapper qcow2;
|+        q_obj_ImageInfoSpecificVmdk_wrapper vmdk;
|     } u;
| };

Doing this removes asymmetry between QAPI's QMP side and its
C side (both sides now expose 'data'), and means that the
treatment of a simple union as sugar for a flat union is now
equivalent in both languages (previously the two approaches used
a different layer of dereferencing, where the simple union could
be converted to a flat union with equivalent C layout but
different {} on the wire, or to an equivalent QMP wire form
but with different C representation).  Using the implicit type
also lets us get rid of the simple_union_type() hack.

Of course, now all clients of simple unions have to adjust from
using su->u.member to using su->u.member.data; while this touches
a number of files in the tree, some earlier cleanup patches
helped minimize the change to the initialization of a temporary
variable rather than every single member access.  The generated
qapi-visit.c code is also affected by the layout change:

|@@ -7393,10 +7393,10 @@ void visit_type_ImageInfoSpecific_member
|     }
|     switch (obj->type) {
|     case IMAGE_INFO_SPECIFIC_KIND_QCOW2:
|-        visit_type_ImageInfoSpecificQCow2(v, "data", &obj->u.qcow2, &err);
|+        visit_type_q_obj_ImageInfoSpecificQCow2_wrapper_members(v, &obj->u.qcow2, &err);
|         break;
|     case IMAGE_INFO_SPECIFIC_KIND_VMDK:
|-        visit_type_ImageInfoSpecificVmdk(v, "data", &obj->u.vmdk, &err);
|+        visit_type_q_obj_ImageInfoSpecificVmdk_wrapper_members(v, &obj->u.vmdk, &err);
|         break;
|     default:
|         abort();

Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1458254921-17042-13-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-03-18 10:29:26 +01:00
Max Reitz
9aaf28c61d block: Remove bdrv_states list
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-03-17 15:47:57 +01:00
Max Reitz
262b4e8f74 block: Add bdrv_next_monitor_owned()
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>
2016-03-17 15:47:56 +01:00
Max Reitz
7c735873d9 blockdev: Remove blk_hide_on_behalf_of_hmp_drive_del()
We can basically inline it in hmp_drive_del(); monitor_remove_blk() is
called already, so we just need to call bdrv_make_anon(), too.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-03-17 15:47:56 +01:00
Max Reitz
efaa7c4eeb blockdev: Split monitor reference from BB creation
Before this patch, blk_new() automatically assigned a name to the new
BlockBackend and considered it referenced by the monitor. This patch
removes the implicit monitor_add_blk() call from blk_new() (and
consequently the monitor_remove_blk() call from blk_delete(), too) and
thus blk_new() (and related functions) no longer take a BB name
argument.

In fact, there is only a single point where blk_new()/blk_new_open() is
called and the new BB is monitor-owned, and that is in blockdev_init().
Besides thus relieving us from having to invent names for all of the BBs
we use in qemu-img, this fixes a bug where qemu cannot create a new
image if there already is a monitor-owned BB named "image".

If a BB and its BDS tree are created in a single operation, as of this
patch the BDS tree will be created before the BB is given a name
(whereas it was the other way around before). This results in minor
change to the output of iotest 087, whose reference output is amended
accordingly.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-03-17 15:47:56 +01:00
Max Reitz
da31d594cf block: Use blk_{commit,flush}_all() consistently
Replace bdrv_commmit_all() and bdrv_flush_all() by their BlockBackend
equivalents.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2016-03-17 15:47:56 +01:00
Kevin Wolf
f8746fb804 block: Fix memory leak in hmp_drive_add_node()
hmp_drive_add_node() leaked qdict in the error path when no node-name is
specified.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
2016-03-17 15:47:56 +01:00
Kevin Wolf
23f7fcb295 block: Fix qemu_root_bds_opts.head initialisation
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-03-17 15:47:56 +01:00
Kevin Wolf
2073d410ce hmp: Extend drive_del to delete nodes without BB
Now that we can use drive_add to create new nodes without a BB, we also
want to be able to delete such nodes again.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
2016-03-14 16:46:43 +01:00
Kevin Wolf
abb21ac3e6 hmp: 'drive_add -n' for creating a node without BB
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>
2016-03-14 16:46:43 +01:00
Kevin Wolf
a81d616437 block: Fix cache mode defaults in bds_tree_init()
Without setting explicit defaults in the options, blockdev-add without
an ID ended up defaulting to writethrough. It should be writeback as
documented.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
2016-03-14 16:46:43 +01:00
Kevin Wolf
73176bee99 block: Fix snapshot=on cache modes
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>
2016-03-14 16:46:43 +01:00
Kevin Wolf
f86b8b584b blockdev: Snapshotting must not open second instance of old top
Calling bdrv_img_create() with a size of -1 means that it determines the
size automatically by opening the backing file. However, in the case of
live snapshots, the backing file is already opened and we must avoid
opening the same image twice at the same time. Apart from that, just
getting the size from the already existing BDS is a lot less overhead
than opening a new instance.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Jeff Cody <jcody@redhat.com>
2016-03-14 16:46:43 +01:00
Eric Blake
10f759079e qapi: Avoid use of 'data' member of QAPI unions
QAPI code generators currently create a 'void *data' member as
part of the anonymous union embedded in the C struct corresponding
to a QAPI union.  However, directly assigning to this member of
the union feels a bit fishy, when we can assign to another member
of the struct instead.

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Message-Id: <1457021813-10704-9-git-send-email-eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-03-05 10:41:58 +01:00
Alyssa Milburn
156abc2f90 blockdev: unset inappropriate flags when changing medium
Most importantly, this removes BDRV_O_TEMPORARY, to avoid unlink()ing an
image which replaces a snapshotted one.

Signed-off-by: Alyssa Milburn <fuzzie@fuzzie.org>
Message-id: 20160206133618.GA16635@li141-249.members.linode.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2016-02-22 16:54:14 +01:00