On Mon, Mar 18, 2024 at 10:02 AM Michael S. Tsirkin wrote:
>
> On Mon, Mar 18, 2024 at 12:31:26PM +0800, Jason Wang wrote:
> > On Fri, Mar 15, 2024 at 11:59 PM Kevin Wolf wrote:
> > >
> > > VDUSE requires that virtqueues are first enabled before the DRIVER_OK
> > > status flag is set; with the
On Tue, Mar 12, 2024 at 11:12:04AM -0400, Stefan Hajnoczi wrote:
> It is possible to hit the sysctl vm.max_map_count limit when the
> coroutine pool size becomes large. Each coroutine requires two mappings
> (one for the stack and one for the guard page). QEMU can crash with
> "failed to set up
> Subject: [EXTERNAL] [PATCH v3 for 9.1 5/6] vhost/vhost-user: Add
> VIRTIO_F_NOTIFICATION_DATA to vhost feature bits
>
> Prioritize security for external emails: Confirm sender and content safety
> before clicking links or opening attachments
>
>
Hello,
pretty much the $Subject.
We had reports of very bad performance with the default options when running:
$ time qemu-img convert -c -p -O qcow2 file.qcow2 file.disk
real7m13.220s
user6m56.682s
sys 0m13.038s
I studied this and noticed that for every coroutine a thread was
On Thu, Mar 14, 2024 at 05:58:23PM +0100, Kevin Wolf wrote:
> Kevin Wolf (2):
> nbd/server: Fix race in draining the export
> iotests: Add test for reset/AioContext switches with NBD exports
>
> nbd/server.c | 15 ++---
>
From: Thomas Huth
Using "-drive ...,backing.file.filename=..." only works with the
file protocol, but not with URIs, so mark this test accordingly.
Signed-off-by: Thomas Huth
Message-ID: <2024031508.153201-5-th...@redhat.com>
Reviewed-by: Kevin Wolf
Signed-off-by: Kevin Wolf
---
From: Thomas Huth
When running iotest 033 with the ssh protocol, it fails with:
033 fail [14:48:31] [14:48:41] 10.2soutput mismatch
--- /.../tests/qemu-iotests/033.out
+++ /.../tests/qemu-iotests/scratch/qcow2-ssh-033/033.out.bad
@@ -174,6 +174,7 @@
512 bytes, X
When draining an NBD export, nbd_drained_begin() first sets
client->quiescing so that nbd_client_receive_next_request() won't start
any new request coroutines. Then nbd_drained_poll() tries to makes sure
that we wait for any existing request coroutines by checking that
client->nb_requests has
From: Thomas Huth
These tests 188, 189 and 198 use qemu-io with --image-opts with additional
hard-coded parameters for the file protocol, so they cannot work for other
protocols. Thus we have to limit these tests to the file protocol only.
Signed-off-by: Thomas Huth
Message-ID:
From: Thomas Huth
Commit b25b387fa592 updated the iotests 134 and 158 to use the --image-opts
parameter for qemu-io with file protocol related options, but forgot to
update the _supported_proto line accordingly. So let's do that now.
Fixes: b25b387fa5 ("qcow2: convert QCow2 to use QCryptoBlock
From: Thomas Huth
The test fails completely when you try to use it with a different
protocol, e.g. with "./check -ssh -qcow2 156".
The test uses some hand-crafted JSON statements which cannot work with other
protocols, thus let's change this test to only support the 'file' protocol.
From: Thomas Huth
Tests 263, 284 and detect-zeroes-registered-buf use qemu-io
with --image-opts so we have to enforce IMGOPTSSYNTAX=true here
to get $TEST_IMG in shape for other protocols than "file".
Signed-off-by: Thomas Huth
Message-ID: <2024031508.153201-9-th...@redhat.com>
From: Markus Armbruster
When external_snapshot_abort() rejects a BlockDriverState without a
medium, it creates an error like this:
error_setg(errp, "Device '%s' has no medium", device);
Trouble is @device can be null. My system formats null as "(null)",
but other systems might crash.
Calling job_pause_point() while holding the graph reader lock
potentially results in a deadlock: bdrv_graph_wrlock() first drains
everything, including the mirror job, which pauses it. The job is only
unpaused at the end of the drain section, which is when the graph writer
lock has been
From: Thomas Huth
iotest 114 uses "truncate" and the qcow2.py script on the destination file,
which both cannot deal with URIs. Thus this test needs the "file" protocol,
otherwise it fails with an error message like this:
truncate: cannot open
From: Thomas Huth
Tests that use "--blockdev" with the "file" driver cannot work with
other protocols, so we should mark them accordingly.
Signed-off-by: Thomas Huth
Message-ID: <2024031508.153201-10-th...@redhat.com>
Reviewed-by: Kevin Wolf
Signed-off-by: Kevin Wolf
---
This replicates the scenario in which the bug was reported.
Unfortunately this relies on actually executing a guest (so that the
firmware initialises the virtio-blk device and moves it to its
configured iothread), so this can't make use of the qtest accelerator
like most other test cases. I tried
From: Fiona Ebner
Failure was noticed when running the tests for the qcow2 image format.
Fixes: 0bd779e27e ("crypto: Introduce 'detached-header' field in
QCryptoBlockInfoLUKS")
Signed-off-by: Fiona Ebner
Message-ID: <20240216101415.293769-1-f.eb...@proxmox.com>
Reviewed-by: Daniel P. Berrangé
The following changes since commit ba49d760eb04630e7b15f423ebecf6c871b8f77b:
Merge tag 'pull-maintainer-final-130324-1' of https://gitlab.com/stsquad/qemu
into staging (2024-03-13 15:12:14 +)
are available in the Git repository at:
https://repo.or.cz/qemu/kevin.git tags/for-upstream
From: Abhiram Tilak
When running the command `qemu-img snapshot -l SNAPSHOT` the output of
VM_CLOCK (measures the offset between host and VM clock) cannot to
accommodate values in the order of thousands (4-digit).
This line [1] hints on the problem. Additionally, the column width for
the
From: Thomas Huth
The hand-crafted json statement in this test only works if the test
is run with the "file" protocol, so mark this test accordingly.
Signed-off-by: Thomas Huth
Message-ID: <2024031508.153201-3-th...@redhat.com>
Reviewed-by: Kevin Wolf
Signed-off-by: Kevin Wolf
---
On Mon, Mar 18, 2024 at 12:37:19PM +0100, Kevin Wolf wrote:
> Am 14.03.2024 um 15:29 hat Stefan Hajnoczi geschrieben:
> > On Wed, Mar 13, 2024 at 04:30:00PM +0100, Kevin Wolf wrote:
> > > Calling job_pause_point() while holding the graph reader lock
> > > potentially results in a deadlock:
Am 16.02.2024 um 11:14 hat Fiona Ebner geschrieben:
> Failure was noticed when running the tests for the qcow2 image format.
>
> Fixes: 0bd779e27e ("crypto: Introduce 'detached-header' field in
> QCryptoBlockInfoLUKS")
> Signed-off-by: Fiona Ebner
Thanks, applied to the block branch.
Kevin
Am 15.03.2024 um 12:10 hat Thomas Huth geschrieben:
> I recently wanted to check for some changes that I did to the URI handling
> in the block layer code, but I had to discover that a lot of iotests only
> work with the raw file format when using a protocol that is not "file",
> i.e. "./check
Mon, Mar 18, 2024 at 12:22:02PM CET, jonah.pal...@oracle.com wrote:
>
>
>On 3/16/24 11:45 AM, Jiri Pirko wrote:
>> Fri, Mar 15, 2024 at 05:55:51PM CET, jonah.pal...@oracle.com wrote:
>> > The goal of these patches are to add support to a variety of virtio and
>> > vhost devices for the
Am 06.03.2024 um 15:28 hat Markus Armbruster geschrieben:
> When external_snapshot_abort() rejects a BlockDriverState without a
> medium, it creates an error like this:
>
> error_setg(errp, "Device '%s' has no medium", device);
>
> Trouble is @device can be null. My system formats null
Am 14.03.2024 um 15:29 hat Stefan Hajnoczi geschrieben:
> On Wed, Mar 13, 2024 at 04:30:00PM +0100, Kevin Wolf wrote:
> > Calling job_pause_point() while holding the graph reader lock
> > potentially results in a deadlock: bdrv_graph_wrlock() first drains
> > everything, including the mirror job,
On 3/16/24 11:45 AM, Jiri Pirko wrote:
Fri, Mar 15, 2024 at 05:55:51PM CET, jonah.pal...@oracle.com wrote:
The goal of these patches are to add support to a variety of virtio and
vhost devices for the VIRTIO_F_NOTIFICATION_DATA transport feature. This
feature indicates that a driver will
Markus Armbruster writes:
> Philippe posted "[PATCH v2 00/22] qapi: Kill 'qapi/qmp/qerror.h' for
> good" a couple of months ago. I cherry-picked just its simplest parts
> for now.
>
> Markus Armbruster (1):
> error: Drop superfluous #include "qapi/qmp/qerror.h"
Queued for 9.1 with PATCH 08's
On 18/3/24 09:58, Markus Armbruster wrote:
Philippe Mathieu-Daudé writes:
On 12/3/24 16:26, Zhao Liu wrote:
On Tue, Mar 12, 2024 at 03:13:41PM +0100, Markus Armbruster wrote:
Date: Tue, 12 Mar 2024 15:13:41 +0100
From: Markus Armbruster
Subject: [PATCH 08/10] qapi: Correct error message
On Fri, Mar 15, 2024 at 04:59:49PM +0100, Kevin Wolf wrote:
> VDUSE requires that virtqueues are first enabled before the DRIVER_OK
> status flag is set; with the current API of the kernel module, it is
> impossible to enable the opposite order in our block export code because
> userspace is not
On Mon, Mar 18, 2024 at 12:31:26PM +0800, Jason Wang wrote:
> On Fri, Mar 15, 2024 at 11:59 PM Kevin Wolf wrote:
> >
> > VDUSE requires that virtqueues are first enabled before the DRIVER_OK
> > status flag is set; with the current API of the kernel module, it is
> > impossible to enable the
Philippe Mathieu-Daudé writes:
> On 12/3/24 16:26, Zhao Liu wrote:
>> On Tue, Mar 12, 2024 at 03:13:41PM +0100, Markus Armbruster wrote:
>>> Date: Tue, 12 Mar 2024 15:13:41 +0100
>>> From: Markus Armbruster
>>> Subject: [PATCH 08/10] qapi: Correct error message for 'vcpu_dirty_limit'
>>>
33 matches
Mail list logo