On 7/12/22 04:13, Sam Li wrote:
Signed-off-by: Sam Li
---
block/file-posix.c | 60
include/block/block-common.h | 4 +--
2 files changed, 62 insertions(+), 2 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 3161d39ea4..4
Hannes Reinecke 于2022年7月12日周二 14:10写道：
>
> On 7/12/22 04:13, Sam Li wrote:
> > By adding zone management operations in BlockDriver, storage
> > controller emulation can use the new block layer APIs including
> > zone_report and zone_mgmt(open, close, finish, reset).
> >
> > Signed-off-by: Sam Li
On 7/12/22 04:13, Sam Li wrote:
Add zoned storage commands of the device: zone_open(zo), zone_close(zc),
zone_reset(zs), zone_report(zp), zone_finish(zf).
For example, it can be called by:
./build/qemu-io --image-opts driver=zoned_host_device, filename=/dev/nullb0
-c "zone_report 0 0 1"
Signed-
On 7/12/22 04:13, Sam Li wrote:
Use sysfs attribute files to get the zoned device information in case
that ioctl() commands of zone management interface won't work.
Signed-off-by: Sam Li
---
block/file-posix.c | 38 +++---
1 file changed, 27 insertions(+), 11
On 7/12/22 04:13, Sam Li wrote:
By adding zone management operations in BlockDriver, storage
controller emulation can use the new block layer APIs including
zone_report and zone_mgmt(open, close, finish, reset).
Signed-off-by: Sam Li
---
block/block-backend.c| 41 ++
block/c
Markus Armbruster 于2022年7月12日周二 13:47写道：
>
> Sam Li writes:
>
> > This patch series adds support for zoned device to virtio-blk emulation.
> > Zoned
> > Storage can support sequential writes, which reduces write amplification in
> > SSD,
> > leading to higher write throughput and increased capa
Sam Li writes:
> This patch series adds support for zoned device to virtio-blk emulation. Zoned
> Storage can support sequential writes, which reduces write amplification in
> SSD,
> leading to higher write throughput and increased capacity.
Forgive me if this has already been discussed, or is
On 7/8/22 21:05, John Snow wrote:
The fedora container has since been split apart, so there's no suitable
nearby target that would support "test-mingw" as it requires both x32
and x64 support -- so either fedora-cross-win32 nor fedora-cross-win64
would be truly suitable.
Just remove this test as
On 7/8/22 21:04, John Snow wrote:
If you try to run a 16 or 32 threaded test, you're going to run out of
memory very quickly with qom-test and a few others. Bump the memory
limit to try to scale with larger-core machines.
Granted, this means that a 16 core processor is going to ask for 16GB,
but
On 7/8/22 21:04, John Snow wrote:
This is listed twice by accident; we require genisoimage to run the
test, so remove the unconditional entry.
Signed-off-by: John Snow
Reviewed-by: Thomas Huth
Reviewed-by: Daniel P. Berrangé
---
tests/vm/Makefile.include | 2 +-
1 file changed, 1 insertion(+)
On 7/8/22 21:04, John Snow wrote:
The old CentOS image didn't work anymore because it was already EOL at
the beginning of 2022.
Signed-off-by: John Snow
Reviewed-by: Thomas Huth
Reviewed-by: Daniel P. Berrangé
---
tests/vm/centos | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
R
On 7/8/22 21:04, John Snow wrote:
Ubuntu 18.04 is out of our support window, and Ubuntu 20.04 does not
support i386 anymore. The debian project does, but they do not provide
any cloud images for it, a new expect-style script would have to be
written.
Since we have i386 cross-compiler tests hoste
On 7/8/22 21:04, John Snow wrote:
18.04 has fallen out of our support window, so move ubuntu.aarch64
forward to ubuntu 20.04, which is now our oldest supported Ubuntu
release.
Notes:
This checksum changes periodically; use a fixed point image with a known
checksum so that the image isn't re-dow
On 7/8/22 21:04, John Snow wrote:
Switch this test over to using a cloud image like the base CentOS8 VM
test, which helps make this script a bit simpler too.
Note: At time of writing, this test seems pretty flaky when run without
KVM support for aarch64. Certain unit tests like migration-test,
v
On 7/8/22 21:04, John Snow wrote:
If the initial setup fails, you've permanently altered the state of the
downloaded image in an unknowable way. Use 'cp' like our other test
setup scripts do.
Signed-off-by: John Snow
Reviewed-by: Thomas Huth
Reviewed-by: Daniel P. Berrangé
---
tests/vm/centos
---
block/file-posix.c | 8 +++-
qapi/block-core.json | 7 +--
2 files changed, 12 insertions(+), 3 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index e9ad1d8e1e..4e0aa02acf 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -3737,6 +3737,12 @@ static void
Putting zoned/non-zoned BlockDrivers on top of each other is not
allowed.
Signed-off-by: Sam Li
---
block.c | 7 +++
block/file-posix.c | 2 ++
include/block/block_int-common.h | 5 +
3 files changed, 14 insertions(+)
diff --git a/block.c b/block.c
We have added new block layer APIs of zoned block devices. Test it with:
(1) Create a null_blk device, run each zone operation on it and see
whether reporting right zone information.
Signed-off-by: Sam Li
---
tests/qemu-iotests/tests/zoned.sh | 69 +++
1 file changed,
Signed-off-by: Sam Li
---
block/file-posix.c | 60
include/block/block-common.h | 4 +--
2 files changed, 62 insertions(+), 2 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 3161d39ea4..42708012ff 100644
--- a/block/file-pos
Use sysfs attribute files to get the zoned device information in case
that ioctl() commands of zone management interface won't work.
Signed-off-by: Sam Li
---
block/file-posix.c | 38 +++---
1 file changed, 27 insertions(+), 11 deletions(-)
diff --git a/block/fil
This is the virtio_blk.h header file from Dmitry's "virtio-blk: add
support for zoned block devices" patch. It introduces
virtio_blk_zoned_characteristics struct from Dmitry's virtio-blk zoned
storage spec.
Signed-off-by: Dmitry Fomichev
Signed-off-by: Sam Li
---
include/standard-headers/linux/
This patch series adds support for zoned device to virtio-blk emulation. Zoned
Storage can support sequential writes, which reduces write amplification in SSD,
leading to higher write throughput and increased capacity.
v4:
- add block layer APIs (revision)
- add configurations for zoned block devi
Signed-off-by: Sam Li
---
block/raw-format.c | 13 +
1 file changed, 13 insertions(+)
diff --git a/block/raw-format.c b/block/raw-format.c
index 69fd650eaf..96bdb6c1e2 100644
--- a/block/raw-format.c
+++ b/block/raw-format.c
@@ -314,6 +314,17 @@ static int coroutine_fn raw_co_pdiscar
Add zoned storage commands of the device: zone_open(zo), zone_close(zc),
zone_reset(zs), zone_report(zp), zone_finish(zf).
For example, it can be called by:
./build/qemu-io --image-opts driver=zoned_host_device, filename=/dev/nullb0
-c "zone_report 0 0 1"
Signed-off-by: Sam Li
---
block/io.c
By adding zone management operations in BlockDriver, storage
controller emulation can use the new block layer APIs including
zone_report and zone_mgmt(open, close, finish, reset).
Signed-off-by: Sam Li
---
block/block-backend.c| 41 ++
block/coroutines.h | 5 +
b
This patch adds the 'check-venv' target as a requisite of all meson
driven check-* targets. As of this commit, it will only install the
"qemu" namespace package from the source tree, and nothing else.
In the future, the "qemu" namespace package in qemu.git will begin to
require an external qemu.qm
When iotests are invoked manually from
e.g. $build/tests/qemu-iotests/check, it is not necessarily guaranteed
that we'll have run 'check-venv' yet.
With this patch, teach testenv.py how to create its own environment.
Note: this self-bootstrapping is fairly rudimentary and will miss
certain trigge
Essentially, the changes to testenv.py here mimic the changes that occur when
you "source venv/bin/activate.fish" or similar.
(1) update iotest's internal notion of which python binary to use,
(2) export the VIRTUAL_ENV variable,
(3) front-load the venv/bin directory to PATH.
If the venv director
NetBSD needs "py37-pip" in order to create virtual environments.
Signed-off-by: John Snow
---
tests/vm/netbsd | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/vm/netbsd b/tests/vm/netbsd
index 45aa9a7fda7..53fe508e487 100755
--- a/tests/vm/netbsd
+++ b/tests/vm/netbsd
@@ -31,6 +31,7 @@
Hi, here's another RFC for bringing external Python dependencies to the
QEMU test suite.
This is mostly a refresh of a version I sent out before, but mixes in my
VM improvement test as an optional pre-requisite to improve VM test
stability to ensure that the BSDs all pass with the new
infrastructu
The version of pip ("20.0.2") that ships with Ubuntu 20.04 has a bug
where it will try to attempt building a wheel even if the "wheel" python
package that enables it to do so is not installed. Even though pip
continues gracefully from source, The result is a lot of irrelevant
failure output.
Upstr
This patch uses a dummy package and setup.cfg/setup.py files to manage
optional dependency groups for the test venv specification. Now, there's
a core set of dependencies which for now includes just "qemu" (but soon,
qemu.qmp) and a separate, optional 'avocado' group that includes
avocado-framework
This splits the venv creation logic out of the Makefile and into a
Python script.
One reason for doing this is to be able to call the venv bootstrapper
from places outside of the Makefile, e.g. configure and iotests. Another
reason is to be able to add "offline" logic to modify the behavior of
the
On Fri, May 27, 2022, 12:29 PM Kevin Wolf wrote:
> Am 26.05.2022 um 16:21 hat John Snow geschrieben:
> > On Thu, May 26, 2022, 3:54 AM Daniel P. Berrangé
> > wrote:
> >
> > > On Wed, May 25, 2022 at 08:25:12PM -0400, John Snow wrote:
> > > > If you invoke the check script from outside of the tes
On Mon, Jul 11, 2022 at 5:16 PM John Snow wrote:
>
> On Fri, Jun 24, 2022 at 3:53 PM Vladimir Sementsov-Ogievskiy
> wrote:
> >
> > I've spent much time trying to debug hanging pipeline in gitlab. I
> > started from and idea that I have problem in code in my series (which
> > has some timeouts). F
On Fri, Jun 24, 2022 at 3:53 PM Vladimir Sementsov-Ogievskiy
wrote:
>
> I've spent much time trying to debug hanging pipeline in gitlab. I
> started from and idea that I have problem in code in my series (which
> has some timeouts). Finally I found that the problem is that I've used
> QEMUMachine
That seems a lot closer!
Now I'm going to vocation from tomorrow up to the end of week, so I'll return
next Monday.
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
In this series, we want to remove the AioContext lock and instead
use the already existent job_mutex to protect the job struct
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
The same job lock is being used also to protect some of blockjob fields.
Categorize them just as done in job.h.
Thanks!
Signed-off-by: Emanuele Giuseppe Esposito
---
include/block/blockjob.h | 17 ++---
1 file changed, 14 ins
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
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 r
On Sun, Jul 10, 2022 at 08:08:49PM +0800, Chao Gao wrote:
> When we measure FIO read performance (cache=writethrough, bs=4k,
> iodepth=64) in VMs, ~80K/s notifications (e.g., EPT_MISCONFIG) are observed
> from guest to qemu.
>
> It turns out those frequent notificatons are caused by interference f
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
iostatus is the only field (together with .job) that needs
protection using the job mutex.
It is set in the main loop (GLOBAL_STATE functions) but read
in I/O code (block_job_error_action).
In order to protect it, change block_job_iostatus_set_
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
They all are called with job_lock held, in job_event_*_locked()
Signed-off-by: Emanuele Giuseppe Esposito
Reviewed-by: Stefan Hajnoczi
Reviewed-by: Vladimir Sementsov-Ogievskiy
--
Best regards,
Vladimir
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
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
R
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
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 f
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
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
also in job_create of course, but it
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
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: E
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
Now that the API offers also _locked() functions, take advantage
of it and give also the caller control to take the lock and call
_locked functions.
This makes sense especially when we have for loops, because it
makes no sense to have:
for(job
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
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*.
Sig
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
Add missing job synchronization in the unit tests, with
explicit locks.
We are deliberately using _locked functions wrapped by a guard
instead of a normal call because the normal call will be removed
in future, as the only usage is limited to th
Am 11/07/2022 um 14:04 schrieb Vladimir Sementsov-Ogievskiy:
> On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
>> Just as done with job.h, create _locked() functions in blockjob.h
>>
>> These functions will be later useful when caller has already taken
>> the lock. All blockjob _locked functi
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
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
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
Just as done with job.h, create _locked() functions in blockjob.h
These functions will be later useful when caller has already taken
the lock. All blockjob _locked functions call job _locked functions.
Note: at this stage, job_{lock/unlock} and
Commit 5f76a7aac156ca75680dad5df4a385fd0b58f6b1 is looking harmless from
the first glance, but it has changed things a lot. 'libvirt' uses it to
detect that it should follow new initialization way and this changes
things considerably. With this procedure followed, blockdev_init() is
not called anym
On 7/11/22 10:30, Emanuele Giuseppe Esposito wrote:
Am 08/07/2022 um 21:25 schrieb Vladimir Sementsov-Ogievskiy:
static bool job_started(Job *job)
So we can call it both with mutex locked and without. Hope it never race
with job_start.
Where exactly do you see it called with mutex not h
On 7/6/22 23:15, Emanuele Giuseppe Esposito wrote:
This comment applies more on job, it was left in blockjob as in the past
the whole job logic was implemented there.
Note: at this stage, job_{lock/unlock} and job lock guard macros
are*nop*.
No functional change intended.
Signed-off-by: Emanue
Am 08/07/2022 um 21:25 schrieb Vladimir Sementsov-Ogievskiy:
>> static bool job_started(Job *job)
>
> So we can call it both with mutex locked and without. Hope it never race
> with job_start.
Where exactly do you see it called with mutex not held?
I don't see it anywhere, and if you agree
56 matches
Mail list logo
