Re: [PATCH 0/3] Audio test cleanups

2023-05-09 Thread Ján Tomko
On a Tuesday in 2023, Peter Krempa wrote: Few things I've cleaned up before realizing that: https://gitlab.com/libvirt/libvirt/-/issues/473 can't really be solved without qemu assistance. Peter Krempa (3): qemuxml2xmlout: Replace symlinks of all 'audio-' tests by real files

[PATCH 3/3] qemuxml2xmltest: Modernize all 'audio-' cases

2023-05-09 Thread Peter Krempa
Use DO_TEST_CAPS_LATEST to run with the latest capapbilities. Signed-off-by: Peter Krempa --- xml => audio-alsa-best.x86_64-latest.xml} | 5 +- xml => audio-alsa-full.x86_64-latest.xml} | 5 +- ...l => audio-alsa-minimal.x86_64-latest.xml} | 5 +- ...=>

[PATCH 1/3] qemuxml2xmlout: Replace symlinks of all 'audio-' tests by real files

2023-05-09 Thread Peter Krempa
Symlinks are hard to maintain and especially un-cool when attempting to test against real capapbilities. Replace symlinks by real files first so that we can switch to real caps and see the difference. Signed-off-by: Peter Krempa --- tests/qemuxml2xmloutdata/audio-alsa-best.xml | 44

[PATCH 0/3] Audio test cleanups

2023-05-09 Thread Peter Krempa
Few things I've cleaned up before realizing that: https://gitlab.com/libvirt/libvirt/-/issues/473 can't really be solved without qemu assistance. Peter Krempa (3): qemuxml2xmlout: Replace symlinks of all 'audio-' tests by real files qemuxml2argvtest: Use real caps instead of fake caps for

[PATCH 2/3] qemuxml2argvtest: Use real caps instead of fake caps for 'audio-default-*' cases

2023-05-09 Thread Peter Krempa
Convert all of the 'audio-default-*' cases to use capabilities from qemu-4.2 instead of the fake caps. Signed-off-by: Peter Krempa --- ...default-sdl.args => audio-default-sdl.x86_64-4.2.0.args} | 6 -- ...ult-spice.args => audio-default-spice.x86_64-4.2.0.args} | 6 --

[PATCH 2/2] qemu: Drop @forceVFIO argument of qemuDomainGetMemLockLimitBytes() and qemuDomainAdjustMaxMemLock()

2023-05-09 Thread Michal Privoznik
After previous cleanup, there's not a single caller that would call either qemuDomainGetMemLockLimitBytes() or qemuDomainAdjustMaxMemLock() with @forceVFIO set. All callers pass false. Drop the unneeded argument from both functions. Signed-off-by: Michal Privoznik --- src/qemu/qemu_domain.c |

[PATCH 0/2] qemu_domin: Account for NVMe disks when calculating memlock limit on hotplug

2023-05-09 Thread Michal Privoznik
*** BLURB HERE *** Michal Prívozník (2): qemu_domin: Account for NVMe disks when calculating memlock limit on hotplug qemu: Drop @forceVFIO argument of qemuDomainGetMemLockLimitBytes() and qemuDomainAdjustMaxMemLock() src/qemu/qemu_domain.c | 75

[PATCH 1/2] qemu_domin: Account for NVMe disks when calculating memlock limit on hotplug

2023-05-09 Thread Michal Privoznik
During hotplug of a NVMe disk we need to adjust the memlock limit. The computation of the limit is handled by qemuDomainGetMemLockLimitBytes() which looks at given domain definition and accounts for various device types (as different types require different amounts). But during disk hotplug the

Re: [PATCH 0/5] qemu: Clean up testing of sound devices and implement 'multichannel' support for 'usb-audio'

2023-05-09 Thread Ján Tomko
On a Tuesday in 2023, Peter Krempa wrote: Peter Krempa (5): docs: formatdomain: Use code blocks to emphasize various sound device options qemuxml2(argv|xml)test: Remove 'sound' case qemuxml2(argv|xml)test: Modernize 'sound-device' case conf: Register autoptr cleanup for

Re: [PATCH] qemu: Report domain name in unexpectedly closed monitor message

2023-05-09 Thread Peter Krempa
On Tue, May 09, 2023 at 14:51:08 +0300, Michal Privoznik wrote: > When QEMU closes the monitor suddenly, the following error > message is reported: > > internal error: qemu unexpectedly closed the monitor: ... > > And this works. But other error messages produced in the same > function include

[PATCH] qemu: Report domain name in unexpectedly closed monitor message

2023-05-09 Thread Michal Privoznik
When QEMU closes the monitor suddenly, the following error message is reported: internal error: qemu unexpectedly closed the monitor: ... And this works. But other error messages produced in the same function include domain name too. Do that for the unexpectedly closed monitor message too.

[PATCH 3/5] qemuxml2(argv|xml)test: Modernize 'sound-device' case

2023-05-09 Thread Peter Krempa
Drop the unnecessary disk definition and use x86_64 emulator. For 'qemuxml2argvtest' replace the fake-caps invocation by a 4.2.0 version-locked invocation and add a '_CAPS_LATEST' invocation. For 'qemuxml2xmltest' convert to use '_CAPS_LATEST' only. There are no sound-device relevant changes

[PATCH 5/5] conf: qemu: Add support for multi-channel mode for 'usb' sound cards

2023-05-09 Thread Peter Krempa
Allow users controlling the multi-channel mode by adding a 'multichannel' property parsed for USB audio devices and wire up the support in the qemu driver. Closes: https://gitlab.com/libvirt/libvirt/-/issues/472 Signed-off-by: Peter Krempa --- docs/formatdomain.rst | 5

[PATCH 0/5] qemu: Clean up testing of sound devices and implement 'multichannel' support for 'usb-audio'

2023-05-09 Thread Peter Krempa
Peter Krempa (5): docs: formatdomain: Use code blocks to emphasize various sound device options qemuxml2(argv|xml)test: Remove 'sound' case qemuxml2(argv|xml)test: Modernize 'sound-device' case conf: Register autoptr cleanup for 'virDomainSoundDef' and refactor

[PATCH 1/5] docs: formatdomain: Use code blocks to emphasize various sound device options

2023-05-09 Thread Peter Krempa
Emphasize the various sound card models and other config options by using ``...``. Signed-off-by: Peter Krempa --- docs/formatdomain.rst | 50 +-- 1 file changed, 25 insertions(+), 25 deletions(-) diff --git a/docs/formatdomain.rst

[PATCH 2/5] qemuxml2(argv|xml)test: Remove 'sound' case

2023-05-09 Thread Peter Krempa
The test case is a subset of what the 'sound-device' case tests. Signed-off-by: Peter Krempa --- tests/qemuxml2argvdata/sound.args | 37 - tests/qemuxml2argvdata/sound.xml | 34 --- tests/qemuxml2argvtest.c | 1 -

[PATCH 4/5] conf: Register autoptr cleanup for 'virDomainSoundDef' and refactor virDomainSoundDefParseXML

2023-05-09 Thread Peter Krempa
Use our modern cleanup path pattern. Signed-off-by: Peter Krempa --- src/conf/domain_conf.c | 19 +++ src/conf/domain_conf.h | 1 + 2 files changed, 8 insertions(+), 12 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index ac7165bad4..204b6a85e1

Re: [PATCH] logging: Provide empty VIRTLOGD_ARGS in the unit file

2023-05-09 Thread Ján Tomko
On a Tuesday in 2023, Michal Privoznik wrote: For all our daemons, we provide VIRXXXD_ARGS env var in the unit file. The variable can then be overridden in corresponding file: EnvironmentFile=-@initconfdir@/virtxxxd The daemon is then executed as: ExecStart=@sbindir@/virtxxxd $VIRTXXXD_ARGS

[PATCH] logging: Provide empty VIRTLOGD_ARGS in the unit file

2023-05-09 Thread Michal Privoznik
For all our daemons, we provide VIRXXXD_ARGS env var in the unit file. The variable can then be overridden in corresponding file: EnvironmentFile=-@initconfdir@/virtxxxd The daemon is then executed as: ExecStart=@sbindir@/virtxxxd $VIRTXXXD_ARGS But virlogd is exception, for no good