From: Vladimir Sementsov-Ogievskiy
Rename the type to be reused. Old name is "what is it for". To be
natively reused for other needs, let's name it exactly "what is it".
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
block/monitor/bitmap-qmp-cmds.c| 6 +++---
include/block/block_int-g
From: Vladimir Sementsov-Ogievskiy
Hi all!
Here is small improvement for bitmap exporting interface.
v2: Sorry for the noise, me trying to find a email service, that don't
consider sending patch series by git-send-email as as spam :/
Vladimir Sementsov-Ogievskiy (3):
qapi: rename BlockDirtyB
From: Vladimir Sementsov-Ogievskiy
Hi all! Current logic of relying on search through backing chain is not
safe neither convenient.
Sometimes it leads to necessity of extra bitmap copying. Also, we are
going to add "snapshot-access" driver, to access some snapshot state
through NBD. And this dri
From: Vladimir Sementsov-Ogievskiy
Add simple test that new interface introduced in previous commit works.
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
tests/qemu-iotests/223 | 16 +
tests/qemu-iotests/223.out | 47 --
2 files changed, 61
Markus Armbruster writes:
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type errors.
>
> diff --git a/semihosting/conf
15.03.2022 00:32, Vladimir Sementsov-Ogievskiy wrote:
From: Vladimir Sementsov-Ogievskiy
Hi all!
Here is small improvement for bitmap exporting interface.
v2: Sorry for the noise, me trying to find a email service, that don't
consider sending patch series by git-send-email as as spam :/
Aha
According to the NBD spec, a server that advertises
NBD_FLAG_CAN_MULTI_CONN promises that multiple client connections will
not see any cache inconsistencies: when properly separated by a single
flush, actions performed by one client will be visible to another
client, regardless of which client did
The next patch wants to adjust whether the NBD server code advertises
MULTI_CONN based on whether it is known if the server limits to
exactly one client. For a server started by QMP, this information is
obtained through nbd_server_start (which can support more than one
export); but for qemu-nbd (w
Prefer the :option:`--name` form when cross-referencing other options
from the qemu-nbd documentation.
Signed-off-by: Eric Blake
---
docs/tools/qemu-nbd.rst | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/docs/tools/qemu-nbd.rst b/docs/tools/qemu-nbd.rst
index 2b
v2 was here:
https://lists.gnu.org/archive/html/qemu-devel/2022-02/msg03314.html
Since then:
- expose the knob through 'qemu-nbd -m on|off|auto'
- reuse QAPI OnOffAuto type instead of rolling my own
- rewrite the iotest from bash into python (thanks: Nir, Vladimir)
- split out easy preliminary fix
On Montag, 14. März 2022 20:48:47 CET Alex Bennée wrote:
> Markus Armbruster writes:
> > g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> > for two reasons. One, it catches multiplication overflowing size_t.
> > Two, it returns T * rather than void *, which lets the compile
Markus Armbruster writes:
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type errors.
>
> This commit only touches alloc
Hi all!
Here is small improvement for bitmap exporting interface.
Vladimir Sementsov-Ogievskiy (3):
qapi: rename BlockDirtyBitmapMergeSource to BlockDirtyBitmapOrStr
qapi: nbd-export: allow select bitmaps by node/name pair
iotests/223: check new possibility of exporting bitmaps by node/name
From: Vladimir Sementsov-Ogievskiy
Rename the type to be reused. Old name is "what is it for". To be
natively reused for other needs, let's name it exactly "what is it".
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
block/monitor/bitmap-qmp-cmds.c| 6 +++---
include/block/block_int-g
On 3/14/22 09:21, Paolo Bonzini wrote:
But perhaps someone will try to use templates to replace repeated inclusion (which is
common in hw/display) and others will follow suit.
The code in fpu/ desperately calls out for overloading and templates. At present it is a
tangle of _Generic and multi
Il lun 14 mar 2022, 15:37 Stefan Hajnoczi ha scritto:
> On Mon, Mar 14, 2022 at 10:32:01AM +0100, Paolo Bonzini wrote:
> > +//
> > +
> > +// CoroutineFn does not even need anything more than what
> > +// BaseCoroutine provides, so it's just a type alias. The magic
> > +/
Markus Armbruster writes:
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type errors.
>
> This commit only touches alloc
Markus Armbruster writes:
> This is the semantic patch from commit b45c03f585 "arm: Use g_new() &
> friends where that makes obvious sense".
>
> Signed-off-by: Markus Armbruster
Reviewed-by: Alex Bennée
--
Alex Bennée
On Mon, Mar 14, 2022 at 1:23 PM Stefan Hajnoczi wrote:
>
> On Tue, Feb 15, 2022 at 06:59:38PM +0800, Xie Yongji wrote:
> > This supports passing NULL ops to blk_set_dev_ops()
> > so that we can remove stale ops in some cases.
> >
> > Signed-off-by: Xie Yongji
> > ---
> > block/block-backend.c |
Hi Christian,
On 14/3/22 17:42, Christian Schoenebeck wrote:
On Montag, 14. März 2022 17:01:07 CET Markus Armbruster wrote:
g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
for two reasons. One, it catches multiplication overflowing size_t.
Two, it returns T * rather than
14.03.2022 19:40, Vladimir Sementsov-Ogievskiy wrote:
14.03.2022 18:16, Vladimir Sementsov-Ogievskiy wrote:
14.03.2022 17:47, Eric Blake wrote:
On Mon, Mar 14, 2022 at 02:04:15PM +0300, Vladimir Sementsov-Ogievskiy wrote:
Old vsement...@virtuozzo.com is not accessible anymore.
Signed-off-by:
On Mon, Mar 14, 2022 at 05:52:32PM +0100, Markus Armbruster wrote:
> Peter Maydell writes:
>
> > On Mon, 14 Mar 2022 at 16:01, Markus Armbruster wrote:
> >>
> >> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> >> for two reasons. One, it catches multiplication overflowin
On Wed, Feb 16, 2022 at 07:14:58PM +0200, Nir Soffer wrote:
> > > I'm not the best at writing python iotests; I welcome a language
> > > translation of this aspect.
> >
> >
> >
> > Let me try:)
>
> Thanks! This is much nicer and will be easier to maintain.
>
> >
> >
> > #!/usr/bin/env python3
> >
On Tue, Feb 15, 2022 at 06:59:38PM +0800, Xie Yongji wrote:
> This supports passing NULL ops to blk_set_dev_ops()
> so that we can remove stale ops in some cases.
>
> Signed-off-by: Xie Yongji
> ---
> block/block-backend.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a
On Mon, 2022-03-14 at 13:35 +, Stefan Hajnoczi wrote:
> On Fri, Mar 11, 2022 at 11:40:30AM +0100, Nicolas Saenz Julienne wrote:
> > On Thu, 2022-03-10 at 10:45 +, Stefan Hajnoczi wrote:
> > > On Thu, Mar 03, 2022 at 04:13:07PM +0100, Nicolas Saenz Julienne wrote:
> > > > @@ -537,10 +546,19
On 3/14/22 17:01, Markus Armbruster wrote:
g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
for two reasons. One, it catches multiplication overflowing size_t.
Two, it returns T * rather than void *, which lets the compiler catch
more type errors.
This commit only touches a
On Mon, Mar 14, 2022 at 10:31:47AM +0100, Paolo Bonzini wrote:
> This was compiled with GCC 11 only. Coroutine support was added in
> GCC 10, released in 2020, which IIRC is much newer than the most recent
> release we support.
Currrently we target 7.4:
commit 2a85a08c998e418a46a308095893f2236
Peter Maydell writes:
> On Mon, 14 Mar 2022 at 16:01, Markus Armbruster wrote:
>>
>> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
>> for two reasons. One, it catches multiplication overflowing size_t.
>> Two, it returns T * rather than void *, which lets the compiler ca
On Montag, 14. März 2022 17:01:07 CET Markus Armbruster wrote:
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type errors.
>
On 14/3/22 17:01, Markus Armbruster wrote:
g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
for two reasons. One, it catches multiplication overflowing size_t.
Two, it returns T * rather than void *, which lets the compiler catch
more type errors.
This series only touches a
14.03.2022 18:16, Vladimir Sementsov-Ogievskiy wrote:
14.03.2022 17:47, Eric Blake wrote:
On Mon, Mar 14, 2022 at 02:04:15PM +0300, Vladimir Sementsov-Ogievskiy wrote:
Old vsement...@virtuozzo.com is not accessible anymore.
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
Hi all!
That's my n
VMDK disk data is stored in extents, which may or may not be separate
from bs->file. VmdkExtent.file points to where they are stored. Each
that is stored in bs->file will simply reuse the exact pointer value of
bs->file.
(That is why vmdk_free_extents() will unref VmdkExtent.file (e->file)
only
This is the semantic patch from commit b45c03f585 "arm: Use g_new() &
friends where that makes obvious sense".
Signed-off-by: Markus Armbruster
---
scripts/coccinelle/use-g_new-etc.cocci | 75 ++
1 file changed, 75 insertions(+)
create mode 100644 scripts/coccinelle/use-
On Mon, 14 Mar 2022 at 16:01, Markus Armbruster wrote:
>
> g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
> for two reasons. One, it catches multiplication overflowing size_t.
> Two, it returns T * rather than void *, which lets the compiler catch
> more type errors.
>
> Th
This should work for all format drivers that support reopening, so test
it.
(This serves as a regression test for HEAD^: This test used to fail for
VMDK before HEAD^.)
Signed-off-by: Hanna Reitz
---
tests/qemu-iotests/tests/reopen-file | 88
tests/qemu-iotests/tests
Hi,
A couple of months ago I noticed that changing a vmdk node’s file child
through blockdev-reopen would crash qemu. I started writing a fix at
some point, got distracted, but now here it is.
Hanna Reitz (2):
block/vmdk: Fix reopening bs->file
iotests/reopen-file: Test reopening file child
On Wed, Mar 09, 2022 at 04:21:01PM +0300, Vladimir Sementsov-Ogievskiy wrote:
> 09.03.2022 10:48, Rao Lei wrote:
> > During the IO stress test, the IO request coroutine has a probability that
> > is
> > can't be awakened when the NBD server is killed.
> >
> > The GDB stack is as follows:
> > Whe
On 3/14/22 15:07, Stefan Hajnoczi wrote:
If we can reach a consensus about C++ language usage in QEMU then I'm in
favor of using C++ coroutines. It's probably not realistic to think we
can limit C++ language usage to just coroutines forever. Someone finds
another C++ feature they absolutely need
g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
for two reasons. One, it catches multiplication overflowing size_t.
Two, it returns T * rather than void *, which lets the compiler catch
more type errors.
This commit only touches allocations with size arguments of the form
si
g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
for two reasons. One, it catches multiplication overflowing size_t.
Two, it returns T * rather than void *, which lets the compiler catch
more type errors.
This commit only touches allocations with size arguments of the form
si
g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
for two reasons. One, it catches multiplication overflowing size_t.
Two, it returns T * rather than void *, which lets the compiler catch
more type errors.
This series only touches allocations with size arguments of the form
si
On Wed, Feb 16, 2022 at 11:08:06AM +0300, Vladimir Sementsov-Ogievskiy wrote:
> 16.02.2022 02:24, Eric Blake wrote:
> > > > +++ b/tests/qemu-iotests/tests/nbd-multiconn
> > > > @@ -0,0 +1,188 @@
> > > > +#!/usr/bin/env bash
> > > > +# group: rw auto quick
> > > > +#
> > > > +# Test that qemu-nbd M
14.03.2022 17:47, Eric Blake wrote:
On Mon, Mar 14, 2022 at 02:04:15PM +0300, Vladimir Sementsov-Ogievskiy wrote:
Old vsement...@virtuozzo.com is not accessible anymore.
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
Hi all!
That's my new address: v.sementsov...@ya.ru , the old one is not
a
My apologies for not notifying earlier but I have been looking into these
errors.
Hopefully should be able to get a fix sent out within the week.
Thanks
Jonah
On 3/7/22 17:46, Michael S. Tsirkin wrote:
On Mon, Mar 07, 2022 at 08:08:33AM -0500, Jonah Palmer wrote:
This series introduces new Q
On Mon, Mar 14, 2022 at 02:04:15PM +0300, Vladimir Sementsov-Ogievskiy wrote:
> Old vsement...@virtuozzo.com is not accessible anymore.
>
> Signed-off-by: Vladimir Sementsov-Ogievskiy
> ---
>
> Hi all!
>
> That's my new address: v.sementsov...@ya.ru , the old one is not
> available anymore.
I
On Mon, Mar 14, 2022 at 10:32:01AM +0100, Paolo Bonzini wrote:
> +//
> +
> +// CoroutineFn does not even need anything more than what
> +// BaseCoroutine provides, so it's just a type alias. The magic
> +// is all in ValuePromise.
> +//
> +// Suspended CoroutineFns are cha
On Sun, Mar 13, 2022 at 04:18:40PM +0100, Paolo Bonzini wrote:
> On 3/11/22 13:17, Daniel P. Berrangé wrote:
> The question is what is easier to maintain, stack switching code that is
> becoming less and less portable (status quo with SafeStack, CET and the TLS
> issues that Stefan has worked on),
With the *nop* job_lock/unlock placed, rename the job functions
of the job API that are always under job_mutex, adding "_locked"
suffix.
List of functions that get this suffix:
job_txn_unref job_txn_add_job
job_ref job_unref
job_enter_cond job_finish_sync
job_is_
Once job lock is used and aiocontext is removed, mirror has
to perform job operations under the same critical section,
using the helpers prepared in previous commit.
Note: at this stage, job_{lock/unlock} and job lock guard macros
are *nop*.
Signed-off-by: Emanuele Giuseppe Esposito
Reviewed-by:
In preparation to the job_lock/unlock usage, create _locked
duplicates of some functions, since they will be sometimes called with
job_mutex held (mostly within job.c),
and sometimes without (mostly from JobDrivers using the job API).
Therefore create a _locked version of such function, so that it
On Mon, Mar 14, 2022 at 10:31:47AM +0100, Paolo Bonzini wrote:
> However, there are no ramifications to actual coroutine code, except
> for the template syntax "CoroutineFn" for the function and
> the mandatory co_await/co_return keywords... both of which are an
> improvement, really: the fact tha
Old vsement...@virtuozzo.com is not accessible anymore.
Signed-off-by: Vladimir Sementsov-Ogievskiy
---
Hi all!
That's my new address: v.sementsov...@ya.ru , the old one is not
available anymore.
I've also subscribed this new address for qemu-devel mailing list, but
not yet get any message fro
On 14/3/22 10:31, Paolo Bonzini wrote:
This is invalid in C++.
Signed-off-by: Paolo Bonzini
---
util/trace-events | 16
1 file changed, 8 insertions(+), 8 deletions(-)
Reviewed-by: Philippe Mathieu-Daudé
In order to make it thread safe, implement a "fake rwlock",
where we allow reads under BQL *or* job_mutex held, but
writes only under BQL *and* job_mutex.
The only write we have is in child_job_set_aio_ctx, which always
happens under drain (so the job is paused).
For this reason, introduce job_set
All these functions assume that the lock is not held, and acquire
it internally.
These functions will be useful when job_lock is globally applied,
as they will allow callers to access the job struct fields
without worrying about the job lock.
Update also the comments in blockjob.c (and move them
On 11/3/22 23:16, Murilo Opsfelder Araujo wrote:
Building QEMU on Fedora 37 (Rawhide Prerelease) ppc64le failed with the
following error:
$ ../configure --prefix=/usr/local/qemu-disabletcg
--target-list=ppc-softmmu,ppc64-softmmu --disable-tcg --disable-linux-user
...
$ make -j$(n
On 3/14/22 14:33, Philippe Mathieu-Daudé wrote:
On 14/3/22 10:31, Paolo Bonzini wrote:
Signed-off-by: Paolo Bonzini
---
scripts/tracetool/format/h.py | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/scripts/tracetool/format/h.py
b/scripts/tracetool/format/h.py
ind
drv->drained_poll() is only implemented in mirror, and allows
it to drain from within the coroutine. The mirror implementation uses
in_drain flag to recognize when it is draining from coroutine,
and consequently avoid deadlocking (wait the poll condition in
child_job_drained_poll to wait for itself
With the *nop* job_lock/unlock placed, rename the static
functions that are always under job_mutex, adding "_locked" suffix.
List of functions that get this suffix:
job_txn_refjob_txn_del_job
job_txn_apply job_state_transition
job_should_pause job_event_cance
Same as AIO_WAIT_WHILE macro, but if we are in the Main loop
do not release and then acquire ctx_ 's aiocontext.
Once all Aiocontext locks go away, this macro will replace
AIO_WAIT_WHILE.
Reviewed-by: Stefan Hajnoczi
Signed-off-by: Emanuele Giuseppe Esposito
---
include/block/aio-wait.h | 17 +
Graph initialization functions like blk_new(), bdrv_new() and so on
should not run in a coroutine. In fact, they might invoke a drain
(for example blk_insert_bs eventually calls bdrv_replace_child_noperm)
that in turn can invoke callbacks like bdrv_do_drained_begin_quiesce(),
that asserts exactly t
Not sure what the atomic here was supposed to do, since job.busy
is protected by the job lock. Since the whole function
is called under job_mutex, just remove the atomic.
Reviewed-by: Stefan Hajnoczi
Signed-off-by: Emanuele Giuseppe Esposito
---
blockjob.c | 2 +-
1 file changed, 1 insertion(+)
From: Paolo Bonzini
We want to make sure access of job->aio_context is always done
under either BQL or job_mutex. The problem is that using
aio_co_enter(job->aiocontext, job->co) in job_start and job_enter_cond
makes the coroutine immediately resume, so we can't hold the job lock.
And caching it
Introduce the job locking mechanism through the whole job API,
following the comments in job.h and requirements of job-monitor
(like the functions in job-qmp.c, assume lock is held) and
job-driver (like in mirror.c and all other JobDriver, lock is not held).
Use the _locked helpers introduced bef
job mutex will be used to protect the job struct elements and list,
replacing AioContext locks.
Right now use a shared lock for all jobs, in order to keep things
simple. Once the AioContext lock is gone, we can introduce per-job
locks.
To simplify the switch from aiocontext to job lock, introduce
Change the job_{lock/unlock} and macros to use job_mutex.
Now that they are not nop anymore, remove the aiocontext
to avoid deadlocks.
Therefore:
- when possible, remove completely the aiocontext lock/unlock pair
- if it is used by some other function too, reduce the locking
section as much as po
job_event_* functions can all be static, as they are not used
outside job.c.
Same applies for job_txn_add_job().
Signed-off-by: Emanuele Giuseppe Esposito
---
include/qemu/job.h | 18 --
job.c | 22 +++---
2 files changed, 19 insertions(+), 21 deleti
On Fri, Mar 11, 2022 at 11:40:30AM +0100, Nicolas Saenz Julienne wrote:
> On Thu, 2022-03-10 at 10:45 +, Stefan Hajnoczi wrote:
> > On Thu, Mar 03, 2022 at 04:13:07PM +0100, Nicolas Saenz Julienne wrote:
> > > @@ -537,10 +546,19 @@
> > > # 0 means that the engine will use its d
Just as for the job API, rename block_job functions that are
always called under job lock.
No functional change intended.
Signed-off-by: Emanuele Giuseppe Esposito
---
block.c | 3 ++-
block/backup.c | 4 ++--
blockdev.c | 12 +++-
blockjob.c
We are always using the given bs AioContext, so there is no need
to take the job ones (which is identical anyways).
This also reduces the point we need to check when protecting
job.aio_context field.
Reviewed-by: Stefan Hajnoczi
Signed-off-by: Emanuele Giuseppe Esposito
---
block/commit.c | 4 +
On 14/3/22 10:31, Paolo Bonzini wrote:
Signed-off-by: Paolo Bonzini
---
scripts/tracetool/format/h.py | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/scripts/tracetool/format/h.py b/scripts/tracetool/format/h.py
index e94f0be7da..2d92fa8bd2 100644
--- a/scripts/trac
Add missing job synchronization in the unit tests, with
explicit locks.
Note: at this stage, job_{lock/unlock} and job lock guard macros
are *nop*.
Signed-off-by: Emanuele Giuseppe Esposito
---
tests/unit/test-bdrv-drain.c | 76 -
tests/unit/test-block-iothread.c |
Both blockdev.c and job-qmp.c have TOC/TOU conditions, because
they first search for the job and then perform an action on it.
Therefore, we need to do the search + action under the same
job mutex critical section.
Note: at this stage, job_{lock/unlock} and job lock guard macros
are *nop*.
Signed
Categorize the fields in struct Job to understand which ones
need to be protected by the job mutex and which don't.
Signed-off-by: Emanuele Giuseppe Esposito
---
include/qemu/job.h | 59 ++
1 file changed, 34 insertions(+), 25 deletions(-)
diff --git
In this series, we want to remove the AioContext lock and instead
use the already existent job_mutex to protect the job structures
and list. This is part of the work to get rid of AioContext lock
usage in favour of smaller granularity locks.
In order to simplify reviewer's job, job lock/unlock fun
On Fri, Mar 11, 2022 at 11:17:22AM +0100, Nicolas Saenz Julienne wrote:
> On Thu, 2022-03-10 at 10:25 +, Stefan Hajnoczi wrote:
> > On Thu, Mar 03, 2022 at 03:58:20PM +0100, Nicolas Saenz Julienne wrote:
> > > @@ -2935,13 +2947,6 @@ qemu_syms = custom_target('qemu.syms', output:
> > > 'qemu.sy
On 14/3/22 10:31, Paolo Bonzini wrote:
qemu_co_rwlock_wrlock stores the current coroutine in a loc variable,
use it instead of calling qemu_coroutine_self() again.
Signed-off-by: Paolo Bonzini
---
util/qemu-coroutine-lock.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Reviewed-by:
On Mon, 2022-03-14 at 13:33 +, Stefan Hajnoczi wrote:
> On Fri, Mar 11, 2022 at 11:17:22AM +0100, Nicolas Saenz Julienne wrote:
> > On Thu, 2022-03-10 at 10:25 +, Stefan Hajnoczi wrote:
> > > On Thu, Mar 03, 2022 at 03:58:20PM +0100, Nicolas Saenz Julienne wrote:
> > > > @@ -2935,13 +2947,6
There will be a problem in this test when we will add
subtree drains in bdrv_replace_child_noperm:
test_detach_indirect is only interested in observing the first
call to .drained_begin. In the original test, there was only a single
subtree drain; however, with additional drains introduced in
bdrv_
This test uses a callback of an I/O function (blk_aio_preadv)
to modify the graph, using bdrv_attach_child.
This is simply not allowed anymore. I/O cannot change the graph.
Before "block/io.c: make bdrv_do_drained_begin_quiesce static
and introduce bdrv_drained_begin_no_poll", the test would simpl
bdrv_parent_drained_begin_single() is also called by
bdrv_replace_child_noperm(). The latter is often called
from coroutines, for example in bdrv_co_create_opts() callbacks.
This can potentially create deadlocks, because if the drain_saldo
in bdrv_replace_child_noperm is > 0, the coroutine will st
Using bdrv_do_drained_begin_quiesce() in bdrv_child_cb_drained_begin()
is not a good idea: the callback might be called when running
a drain in a coroutine, and bdrv_drained_begin_poll() does not
handle that case, resulting in assertion failure.
Instead, bdrv_do_drained_begin with no recursion and
Doing the opposite can make ->detach() (more precisely
bdrv_unapply_subtree_drain() in bdrv_child_cb_detach) undo the subtree_drain
just performed to protect the removal of the child from the graph,
thus making the fully-enabled assert_bdrv_graph_writable fail.
Note that assert_bdrv_graph_writable
Doing the opposite can make adding the child node to a non-drained node,
as apply_subtree_drain is only done in ->attach() and thus make
assert_bdrv_graph_writable fail.
This can happen for example during a transaction rollback (test 245,
test_io_with_graph_changes):
1. a node is removed from the
First, use run_job() instead of the current logic to run the
stream job. Then, use auto_finalize=False to be sure that
the job is not automatically deleted once it is done.
In this way, if the job finishes before we want, it is not
finalized yet so the other commands can still execute
without fail
We need to prevent coroutines from calling BDRV_POLL_WHILE, because
it can create deadlocks. This is done by firstly creating a bottom half
and then yielding. The bh is then scheduled in the main loop, performs
the drain and polling, and then resumes the coroutine.
The problem is that currently we
This serie aims to remove and clean up some bugs that came up
when trying to replace the AioContext lock and still protect
BlockDriverState fields.
They were part of the serie "Removal of Aiocontext lock
through drains: protect bdrv_replace_child_noperm", but since
that serie is still a work in pr
Paolo Bonzini writes:
> Casting to/from void* must be explicit in C++. g_new0 takes care of that.
>
> Signed-off-by: Paolo Bonzini
> ---
> include/qemu/timer.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/include/qemu/timer.h b/include/qemu/timer.h
> index 88ef11468
Murilo Opsfelder Araujo writes:
> Building QEMU on Fedora 37 (Rawhide Prerelease) ppc64le failed with the
> following error:
>
> $ ../configure --prefix=/usr/local/qemu-disabletcg
> --target-list=ppc-softmmu,ppc64-softmmu --disable-tcg --disable-linux-user
> ...
> $ make -j$(nproc)
>
Convert "T coroutine_fn" annotations to the new type CoroutineFn,
and add co_await as needed.
_Generic is replaced by an overloaded constructor. C++ also does
not like & on a temporary, so that is replaced by a function
qemu_make_co_lockable_nonnull that hides it from the compiler.
Signed-off-by
Signed-off-by: Paolo Bonzini
---
configure | 44 +-
include/qemu/coroutine.h | 445 +-
include/qemu/coroutine_int.h | 8 +
util/coroutine-stackless.cc | 145 ++
util/meson.build
Signed-off-by: Paolo Bonzini
---
util/trace-events | 28 ++--
1 file changed, 14 insertions(+), 14 deletions(-)
diff --git a/util/trace-events b/util/trace-events
index c8f53d7d9f..5bc718eff7 100644
--- a/util/trace-events
+++ b/util/trace-events
@@ -1,10 +1,10 @@
# See
Disable a lot of code that I can't be bothered to convert right now.
Signed-off-by: Paolo Bonzini
---
include/qemu/coroutine.h| 2 ++
qapi/qmp-dispatch.c | 2 ++
tests/unit/meson.build | 2 +-
tests/unit/test-coroutine.c | 6 ++
util/async.c| 2 ++
util/
Signed-off-by: Paolo Bonzini
---
tests/unit/meson.build| 6 +-
.../{test-coroutine.c => test-coroutine.cc} | 140 +-
2 files changed, 76 insertions(+), 70 deletions(-)
rename tests/unit/{test-coroutine.c => test-coroutine.cc} (81%)
diff --git a/tests/
Signed-off-by: Paolo Bonzini
---
include/block/aio.h | 5 +
include/fpu/softfloat-types.h | 4
include/qemu/bitops.h | 3 +++
include/qemu/bswap.h | 10 +++---
include/qemu/coroutine.h | 4
include/qemu/host-utils.h | 4
include/qemu
Signed-off-by: Paolo Bonzini
---
configure | 4 ++--
meson.build | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/configure b/configure
index 886000346a..091710ec03 100755
--- a/configure
+++ b/configure
@@ -157,8 +157,8 @@ update_cxxflags() {
# Set QEMU_CXXFLAGS from
This is invalid in C++.
Signed-off-by: Paolo Bonzini
---
util/trace-events | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/util/trace-events b/util/trace-events
index 5bc718eff7..9e23c11d11 100644
--- a/util/trace-events
+++ b/util/trace-events
@@ -3,8 +3,8 @
Since it operates on a given coroutine, qemu_coroutine_get_aio_context
can be called from outside coroutine context.
This is for example how qio_channel_restart_read uses it.
Signed-off-by: Paolo Bonzini
---
include/qemu/coroutine.h | 2 +-
util/qemu-coroutine.c| 2 +-
2 files changed, 2 in
Signed-off-by: Paolo Bonzini
---
scripts/tracetool/format/h.py | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/scripts/tracetool/format/h.py b/scripts/tracetool/format/h.py
index e94f0be7da..2d92fa8bd2 100644
--- a/scripts/tracetool/format/h.py
+++ b/scripts/tracetool/f
Signed-off-by: Paolo Bonzini
---
include/qemu/compiler.h | 4
1 file changed, 4 insertions(+)
diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
index 3baa5e3790..18848f0d49 100644
--- a/include/qemu/compiler.h
+++ b/include/qemu/compiler.h
@@ -72,6 +72,10 @@
int:(x) ?
1 - 100 of 106 matches
Mail list logo
