[PATCH] qemuProcessSetupDisksTransientHotplug: Add checking set-action capability

From: Masayoshi Mizuma The VM is terminated abnormally when is set to the disk option and the qemu doesn't have set-action capability. # virsh start guest01 error: Failed to start domain 'guest01' error: internal error: qemu unexpectedly closed the monitor # Add checking the capabilit

Re: [RFC PATCH 05/10] qemu_capabilities: Added capability for qemu's "ebpf_rss_fds".

Hi, No problem, those patches are RFC. If there would be issues with the libvirt interface - qemu patches should not be applied either. On Fri, Aug 20, 2021 at 3:59 PM Michal Prívozník wrote: > On 8/20/21 2:57 PM, Michal Prívozník wrote: > > On 7/28/21 10:17 AM, Andrew Melnychenko wrote: > >> Ad

Re: [RFC PATCH 01/10] domain_conf: Added configs for RSS and Hash report.

Hi, I'll squash it and add tests. On Fri, Aug 20, 2021 at 3:57 PM Michal Prívozník wrote: > On 7/28/21 10:17 AM, Andrew Melnychenko wrote: > > Added "rss" and "rss_hash_report" configuration that should be used with > > qemu virtio RSS. > > Both options are triswitches. Used as "driver" options

Re: [RFC PATCH 02/10] qemu_capabilities: Added capabilites for qemu's "rss" and "hash".

Hi, Ok, I'll update in RFC v2. On Fri, Aug 20, 2021 at 3:57 PM Michal Prívozník wrote: > On 7/28/21 10:17 AM, Andrew Melnychenko wrote: > > Added qemu's property check for virtio-net. > > And added capability QEMU_CAPS_VIRTIO_RSS. > > With "rss" and "rss_hash_report" from domain config, > > qemu

Re: [RFC PATCH 03/10] qemu_command: Added "rss" and "hash" properties.

Hi, Yea, it is only RFC - later I'll add all tests for review in v2. On Fri, Aug 20, 2021 at 3:57 PM Michal Prívozník wrote: > On 7/28/21 10:17 AM, Andrew Melnychenko wrote: > > If domain config contains "rss" and/or "rss_hash_report" options for > driver. > > Also if the qemu has device capabil

Re: [RFC PATCH 04/10] virsocket: Added receive for multiple fds.

Hi, > virSocketRecvFD() > { > int fds[1]; > > virSocketRecvMultipleFDs(sock, fds, 1, fdflags); > return fds[0]; > } > Yea, it's a good idea. On Fri, Aug 20, 2021 at 3:57 PM Michal Prívozník wrote: > On 7/28/21 10:17 AM, Andrew Melnychenko wrote: > > Similar to virSocketRecvFD() added virS

Re: [RFC PATCH 06/10] qemu_capabilities: Added capability for ebpf helper path.

Hi, Well, qemu-ebpf-rss-helper is kinda unique for each qemu emulator, that's why it's stored with qemucaps. On Fri, Aug 20, 2021 at 3:57 PM Michal Prívozník wrote: > On 7/28/21 10:17 AM, Andrew Melnychenko wrote: > > Added qmp call to receive helper path. > > Also added monitor functions. > >

Re: [libvirt PATCH 01/11] glibcompat: Add wrapper for g_auto*

On a Wednesday in 2021, Tim Wiederhake wrote: See comment for rationale. Missing commit message ;) Signed-off-by: Tim Wiederhake --- src/util/glibcompat.h | 20 1 file changed, 20 insertions(+) diff --git a/src/util/glibcompat.h b/src/util/glibcompat.h index 697687b967.

Re: [libvirt PATCH 06/11] virthread: Introduce WITH_VIR_MUTEX_LOCK_GUARD

On a Wednesday in 2021, Tim Wiederhake wrote: Modeled after "WITH_QEMU_LOCK_GUARD" (see qemu's include/qemu/lockable.h). Uses "__LINE__" instead of "__COUNTER__", as the latter is a GNU extension. We mandate gnu99 in meson.build. But it seems both would work. See comment for typical usage.

Re: [libvirt PATCH 05/11] virthread: Introduce virLockGuard

On a Wednesday in 2021, Tim Wiederhake wrote: Locks a virMutex on creation and unlocks it in its destructor. Typical usage: void function(virMutex *m) { vir_g_autoptr(virLockGuard) lock = virLockGuardNew(m); Since the developer does not need to refer to the 'lock' variable, wrapp

Re: [libvirt PATCH 01/11] glibcompat: Add wrapper for g_auto*

On Wed, Aug 25, 2021 at 02:43:20PM +0200, Tim Wiederhake wrote: > See comment for rationale. > > Signed-off-by: Tim Wiederhake > --- > src/util/glibcompat.h | 20 > 1 file changed, 20 insertions(+) > > diff --git a/src/util/glibcompat.h b/src/util/glibcompat.h > index 69768

[RFC 3/7] virCgroupSetFCAppid: properly handle when appid is NULL

With introduction of live changes of appid we should also support removal of the appid from VM. This is done by writing empty appid part to appid_store file. Signed-off-by: Pavel Hrdina --- src/util/vircgroup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/util/vircgrou

[RFC 6/7] qemu: implement virDomainSetFibreChannelAppid API

Signed-off-by: Pavel Hrdina --- src/qemu/qemu_driver.c | 75 ++ 1 file changed, 75 insertions(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index f1f961c51c..e8c819bfb9 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -

[RFC 5/7] remote: add RPC support for the virDomainSetFibreChannelAppid API

Signed-off-by: Pavel Hrdina --- src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 14 +- src/remote_protocol-structs | 6 ++ 3 files changed, 20 insertions(+), 1 deletion(-) diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index b64a86af63.

[RFC 7/7] tools: introduce virsh setappid command

Signed-off-by: Pavel Hrdina --- docs/manpages/virsh.rst | 14 + tools/virsh-domain.c| 65 + 2 files changed, 79 insertions(+) diff --git a/docs/manpages/virsh.rst b/docs/manpages/virsh.rst index 2204bed3bb..5875b523a4 100644 --- a/docs/manpages

[RFC 2/7] cgroup: extract setting fibre channel appid into virCgroupSetFCAppid

Signed-off-by: Pavel Hrdina --- src/libvirt_private.syms | 1 + src/qemu/qemu_cgroup.c | 17 + src/util/vircgroup.c | 24 src/util/vircgroup.h | 3 +++ 4 files changed, 29 insertions(+), 16 deletions(-) diff --git a/src/libvirt_private.syms b/

[RFC 4/7] src: introduce virDomainSetFibreChannelAppid API

Signed-off-by: Pavel Hrdina --- include/libvirt/libvirt-domain.h | 4 +++ src/driver-hypervisor.h | 6 + src/libvirt-domain.c | 44 src/libvirt_public.syms | 1 + 4 files changed, 55 insertions(+) diff --git a/include/libvirt/

[RFC 1/7] conf: extract appid validation to virDomainDefResourceAppidValidate

This will be needed by future patches adding appid API to allow changing it for running VMs. Signed-off-by: Pavel Hrdina --- src/conf/domain_validate.c | 42 +++--- src/conf/domain_validate.h | 2 ++ src/libvirt_private.syms | 1 + 3 files changed, 28 insertio

[RFC 0/7] introduce support for live appid updates

I'm posting this as an RFC mainly because I'm not sure how to model the new API. This patches introduce a new naive API that will change only the APPID and nothing else. Currently there are no other known features related to Fibre Channel resources so this non-extendable API will be sufficient, ho

Re: [libvirt PATCH 03/11] virxml: Simplify VIR_XPATH_NODE_AUTORESTORE

On a Wednesday in 2021, Tim Wiederhake wrote: Signed-off-by: Tim Wiederhake --- src/util/virxml.h | 6 ++ 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/src/util/virxml.h b/src/util/virxml.h index 06fb7aebd8..b4716630b9 100644 --- a/src/util/virxml.h +++ b/src/util/virxml.h @@

Re: [libvirt PATCH 02/11] VIR_XPATH_NODE_AUTORESTORE: Add semicolon

On a Wednesday in 2021, Tim Wiederhake wrote: This is preparation for the next patch which will make the use of a semicolon after VIR_XPATH_NODE_AUTORESTORE mandatory. Signed-off-by: Tim Wiederhake --- src/conf/backup_conf.c | 2 +- src/conf/checkpoint_conf.c | 2 +- src/c

[libvirt PATCH] src: fix generation of default resource partition

Now that resource structure can have appid as well we need to adapt code that creates default resource partition if not provided by user. Otherwise starting a VM with appid defined would fail with following error: error: unsupported configuration: Resource partition '(null)' must start with '

Re: [libvirt PATCH 02/11] scripts: group-qemu-caps: store paths in helper variables

On 8/25/21 2:19 PM, Pavel Hrdina wrote: > On Tue, Aug 24, 2021 at 04:25:05PM +0200, Ján Tomko wrote: >> Avoid repetition and specifying the path to the header file twice. >> >> Signed-off-by: Ján Tomko >> --- >> scripts/group-qemu-caps.py | 9 ++--- >> 1 file changed, 6 insertions(+), 3 delet

Re: [libvirt PATCH] tests: Fix typos

On a Wednesday in 2021, Tim Wiederhake wrote: Signed-off-by: Tim Wiederhake --- tests/qemuxml2argvtest.c | 2 +- tests/testutilsqemu.c| 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) Reviewed-by: Ján Tomko Jano signature.asc Description: PGP signature

Re: [PATCH 3/4] virscsi: Drop @tmp from virSCSIDeviceListDel

On Wed, Aug 25, 2021 at 14:54:58 +0200, Michal Privoznik wrote: > Clang on Rawhide started to complain that @tmp variable in > virSCSIDeviceListDel() is set but not used. This is obviously a > false positive because the variable is used to free device stolen > from the list. Anyway, we can do witho

Re: [PATCH 2/4] test: Drop unused @cfg from qemu*test

On Wed, Aug 25, 2021 at 14:54:57 +0200, Michal Privoznik wrote: > In qemumigrationcookiexmltest and qemustatusxml2xmltest there is > @cfg variable that is unused. It's set via virQEMUDriverGetConfig() > but then never used. Drop it. > > Signed-off-by: Michal Privoznik > --- > tests/qemumigration

Re: [PATCH 1/4] virpci: Avoid Clang false positive

Please pick a better summary. Something along: virpci: Clarify lifetime of temporary object On Wed, Aug 25, 2021 at 14:54:56 +0200, Michal Privoznik wrote: > The virPCIDeviceIsBehindSwitchLackingACS() function checks > whether given PCI device is not behind a switch that lacks ACS. > It does so

Re: [PATCH] qemu: snapshot: Translate 'volume' disks before attempting offline snapshot manipulation

On a Wednesday in 2021, Peter Krempa wrote: When the VM is inactive the 'virStorageSource' struct doesn't have the necessary data pointing to the actual storage. This is a problem for inactive snapshot operations on VMs which use disk type='volume'. Add the translation steps for reversion and de

[PATCH] qemu: snapshot: Translate 'volume' disks before attempting offline snapshot manipulation

When the VM is inactive the 'virStorageSource' struct doesn't have the necessary data pointing to the actual storage. This is a problem for inactive snapshot operations on VMs which use disk type='volume'. Add the translation steps for reversion and deletion of snapshots. Resolves: https://bugzil

[PATCH 3/4] virscsi: Drop @tmp from virSCSIDeviceListDel

Clang on Rawhide started to complain that @tmp variable in virSCSIDeviceListDel() is set but not used. This is obviously a false positive because the variable is used to free device stolen from the list. Anyway, we can do without the variable so in this specific case let's fix our code to appease C

[PATCH 4/4] DO NOT MERGE

This patch is intentionally wrong (it's leaking memory). I just include it to illustrate Clang bug. Signed-off-by: Michal Privoznik --- src/rpc/virnetclient.c | 9 ++--- src/util/viridentity.c | 3 ++- 2 files changed, 8 insertions(+), 4 deletions(-) diff --git a/src/rpc/virnetclient.c b/sr

[PATCH 2/4] test: Drop unused @cfg from qemu*test

In qemumigrationcookiexmltest and qemustatusxml2xmltest there is @cfg variable that is unused. It's set via virQEMUDriverGetConfig() but then never used. Drop it. Signed-off-by: Michal Privoznik --- tests/qemumigrationcookiexmltest.c | 2 -- tests/qemustatusxml2xmltest.c | 2 -- 2 files cha

[PATCH 0/4] Appease Clang

With its recent update Clang-13.0.1-rc1 started to report spurious warnings. Basically it is concerned about our use of g_auto*: g_autofree char *var = NULL; var = func(); It fails to see that @var is there to automatically free retval of func(). I've reported the bug here: https://bugs.l

[PATCH 1/4] virpci: Avoid Clang false positive

The virPCIDeviceIsBehindSwitchLackingACS() function checks whether given PCI device is not behind a switch that lacks ACS. It does so by starting at given device and traversing up, one parent at time towards the root. The parent device is obtained via virPCIDeviceGetParent() which allocates new vir

[libvirt PATCH] tests: Fix typos

Signed-off-by: Tim Wiederhake --- tests/qemuxml2argvtest.c | 2 +- tests/testutilsqemu.c| 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 0cc4e8a73b..aa2afd878b 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qem

Re: [PATCH 00/22] qemu: Fix mess with lifecycle actions and unbreak transient disks

On 8/24/21 4:44 PM, Peter Krempa wrote: > While investigating how to fix issue with transient disks breaking when > -no-shutdown is not used I've ended up figuring out that lifecycle > action handling in the qemu driver is very broken. > > Unbreak the handling by rejecting some actions which were

[libvirt PATCH 11/11] networkBridgeNameValidate: Cleanup

Signed-off-by: Tim Wiederhake --- src/network/bridge_driver.c | 7 ++- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 5ccc3817d4..7c46a65c72 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.

[libvirt PATCH 10/11] networkBridgeNameValidate: Use virLockGuard

Signed-off-by: Tim Wiederhake --- src/network/bridge_driver.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index e8b9ffa1fc..5ccc3817d4 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@

[libvirt PATCH 07/11] virChrdevFDStreamCloseCb: Use virLockGuard

Signed-off-by: Tim Wiederhake --- src/conf/virchrdev.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/src/conf/virchrdev.c b/src/conf/virchrdev.c index 5d6de68427..8ca1c1062c 100644 --- a/src/conf/virchrdev.c +++ b/src/conf/virchrdev.c @@ -237,12 +237,10 @@ static void vir

[libvirt PATCH 08/11] virChrdevFree: Use virLockGuard

Signed-off-by: Tim Wiederhake --- src/conf/virchrdev.c | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/conf/virchrdev.c b/src/conf/virchrdev.c index 8ca1c1062c..a95b8976ac 100644 --- a/src/conf/virchrdev.c +++ b/src/conf/virchrdev.c @@ -291,10 +291,10 @@ void virCh

[libvirt PATCH 09/11] virChrdevOpen: Use virLockGuard

Signed-off-by: Tim Wiederhake --- src/conf/virchrdev.c | 14 -- 1 file changed, 4 insertions(+), 10 deletions(-) diff --git a/src/conf/virchrdev.c b/src/conf/virchrdev.c index a95b8976ac..7acbfd7f6a 100644 --- a/src/conf/virchrdev.c +++ b/src/conf/virchrdev.c @@ -326,6 +326,7 @@ int

[libvirt PATCH 06/11] virthread: Introduce WITH_VIR_MUTEX_LOCK_GUARD

Modeled after "WITH_QEMU_LOCK_GUARD" (see qemu's include/qemu/lockable.h). Uses "__LINE__" instead of "__COUNTER__", as the latter is a GNU extension. See comment for typical usage. Signed-off-by: Tim Wiederhake --- src/util/virthread.h | 20 1 file changed, 20 insertions(+

[libvirt PATCH 05/11] virthread: Introduce virLockGuard

Locks a virMutex on creation and unlocks it in its destructor. Typical usage: void function(virMutex *m) { vir_g_autoptr(virLockGuard) lock = virLockGuardNew(m); /* `m` is locked, and released automatically on scope exit */ ... while (expression) {

[libvirt PATCH 01/11] glibcompat: Add wrapper for g_auto*

See comment for rationale. Signed-off-by: Tim Wiederhake --- src/util/glibcompat.h | 20 1 file changed, 20 insertions(+) diff --git a/src/util/glibcompat.h b/src/util/glibcompat.h index 697687b967..fe495c5c32 100644 --- a/src/util/glibcompat.h +++ b/src/util/glibcompat.h @

[libvirt PATCH 04/11] internal: Add CONCAT macro

Using the two-step idiom to force resolution of other macros, e.g.: #define bar BAR CONCAT_(foo, bar) // foobar CONCAT(foo, bar) // fooBAR Signed-off-by: Tim Wiederhake --- src/internal.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/internal.h b/src/internal.h index

[libvirt PATCH 02/11] VIR_XPATH_NODE_AUTORESTORE: Add semicolon

This is preparation for the next patch which will make the use of a semicolon after VIR_XPATH_NODE_AUTORESTORE mandatory. Signed-off-by: Tim Wiederhake --- src/conf/backup_conf.c | 2 +- src/conf/checkpoint_conf.c | 2 +- src/conf/cpu_conf.c| 2 +- src/c

[libvirt PATCH 03/11] virxml: Simplify VIR_XPATH_NODE_AUTORESTORE

Signed-off-by: Tim Wiederhake --- src/util/virxml.h | 6 ++ 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/src/util/virxml.h b/src/util/virxml.h index 06fb7aebd8..b4716630b9 100644 --- a/src/util/virxml.h +++ b/src/util/virxml.h @@ -358,10 +358,8 @@ G_DEFINE_AUTO_CLEANUP_CLEAR_

[libvirt PATCH 00/11] Automatic mutex management

libvirt currently manages mutexes manually. Manual mutex management suffers from the same drawbacks as manual memory management, and at least one lock-unlock mismatch exists currently in the code base, see https://listman.redhat.com/archives/libvir-list/2021-August/msg00125.html. This series lays

Re: [libvirt PATCH 02/11] scripts: group-qemu-caps: store paths in helper variables

On Tue, Aug 24, 2021 at 04:25:05PM +0200, Ján Tomko wrote: > Avoid repetition and specifying the path to the header file twice. > > Signed-off-by: Ján Tomko > --- > scripts/group-qemu-caps.py | 9 ++--- > 1 file changed, 6 insertions(+), 3 deletions(-) > > diff --git a/scripts/group-qemu-ca

Re: [libvirt PATCH 01/11] scripts: group-qemu-caps: read file separately in load_caps_flags

On Tue, Aug 24, 2021 at 04:25:04PM +0200, Ján Tomko wrote: > Split the file by lines and store it in a list. > Switch the rest of the function to operate on this list, > to prepare for splitting out the fire reading logic. > > Signed-off-by: Ján Tomko > --- > scripts/group-qemu-caps.py | 30

[PATCH 1/2] NEWS: Mention bug with incorrect format recorded for qcow2+luks overlays

Signed-off-by: Peter Krempa --- NEWS.rst | 8 1 file changed, 8 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index 21df0e1602..a868c575c7 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -21,6 +21,14 @@ v7.7.0 (unreleased) * **Bug fixes** + * qemu: Record proper ``backing`` format for over

[PATCH 0/2] NEWS updates for the upcoming release

Note that the changes described in 2/2 are not pushed yet. Peter Krempa (2): NEWS: Mention bug with incorrect format recorded for qcow2+luks overlays NEWS: Update with description of changes to lifecycle action handling NEWS.rst | 38 ++ 1 file changed

[PATCH 2/2] NEWS: Update with description of changes to lifecycle action handling

Signed-off-by: Peter Krempa --- NEWS.rst | 30 ++ 1 file changed, 30 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index a868c575c7..cdbd6e48a7 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -19,6 +19,36 @@ v7.7.0 (unreleased) * **Improvements** + * qemu: Lifecycle acti

[PATCH 5/5] Add get domaincaps unit test

Signed-off-by: Haibin Huang --- tests/domaincapsdata/bhyve_basic.x86_64.xml | 1 + tests/domaincapsdata/bhyve_fbuf.x86_64.xml| 1 + tests/domaincapsdata/bhyve_uefi.x86_64.xml| 1 + tests/domaincapsdata/empty.xml| 1 + tests/domaincapsdata/libxl-xenfv.xml

[PATCH 2/5] qemu: Add command-line to generate SGX EPC memory backend

From: Lin Yang According to the result parsing from xml, add the argument of SGX EPC memory backend into QEMU command line: #qemu-system-x86_64 \ .. \ -object memory-backend-epc,id=mem1,size=64M,prealloc=on \ -object memory-backend-epc,id=mem2,size=28M \ -

[PATCH 4/5] Support to query SGX capability

1.Add SGX feature in domain capabilities 2.Get sgx capabilities by query-sgx-capabilities 3.Transfer the B to KB for epc_size Signed-off-by: Haibin Huang --- docs/schemas/domaincaps.rng| 19 - src/conf/domain_capabilities.c | 29 +++ src/conf/domain_capabilities.h | 13 +++ src

[PATCH 3/5] Add unit tests for guest VM creation command with SGX EPC

From: Lin Yang --- .../sgx-epc.x86_64-5.1.0.args | 37 + tests/qemuxml2argvdata/sgx-epc.xml| 40 +++ tests/qemuxml2argvtest.c | 1 + 3 files changed, 78 insertions(+) create mode 100644 tests/qemuxml2argvdata/sgx-

[PATCH 1/5] conf: Introduce SGX EPC element into device memory xml

From: Lin Yang ... 512 ... --- docs/schemas/domaincommon.rng| 1 + src/conf/domain_conf.c | 5 + src/conf/domain_conf.h | 1 + src/conf/domain_validate.c | 1 + src/qemu/qemu_alias.c| 3 +++ src/qemu/qemu_command.c

Re: [PATCH 0/4] storage pool define: add support for validation against schema

On 8/24/21 4:51 PM, Kristina Hanicova wrote: > > Kristina Hanicova (4): > api: add virStoragePoolDefineFlags > storage_conf: add validation against schema in pool define > storage_driver & test_driver: allow VIR_STORAGE_POOL_DEFINE_VALIDATE > flag > virsh: add support for '--validate'

Re: [PATCH] lxcxml2xmltest: Substitute 'inactive' variable with 'active'

On 7/21/21 10:25 AM, Kristina Hanicova wrote: > I removed negation from the name of a variable to make the code > more readable. > > Signed-off-by: Kristina Hanicova > --- > tests/lxcxml2xmltest.c | 12 ++-- > 1 file changed, 6 insertions(+), 6 deletions(-) Reviewed-by: Michal Privoznik