May 2023 - Devel - libvirt List Archives

[PATCH V3 0/3] migration: add qemu parallel migration options
by Jiang Jiacheng 18 May '23

18 May '23

Add compress method zlib and zstd for parallel migration and new migration options to set qemu's parameter related with parallel migration(multifd-compression, multifd-zlib-level and multifd-zstd-level). These parameters has been supported by QEMU since 5.0. v3 of: https://listman.redhat.com/archives/libvir-list/2023-February/237604.html diff to v2: * merge the processing of new method into 'qemuMigrationParamsSetCompression' * improve descriptions for the new options. Jiang Jiacheng (3): Add public API for parallel compression method virsh: Add migrate options to set parallel compress level qemu: support set parallel migration compression method docs/manpages/virsh.rst | 29 ++++++++---- include/libvirt/libvirt-domain.h | 30 ++++++++++-- src/qemu/qemu_migration.h | 2 + src/qemu/qemu_migration_params.c | 80 +++++++++++++++++++++++++++++++- src/qemu/qemu_migration_params.h | 3 ++ tools/virsh-domain.c | 26 +++++++++++ 6 files changed, 156 insertions(+), 14 deletions(-) -- 2.33.0

2 12

[PATCH 0/7] docs: Cleanups of generated pages and generators
by Peter Krempa 17 May '23

17 May '23

Browse the affected content at: https://pipo.sk.gitlab.io/-/libvirt/-/jobs/4302989355/artifacts/website/hvs… https://pipo.sk.gitlab.io/-/libvirt/-/jobs/4302989355/artifacts/website/htm… https://pipo.sk.gitlab.io/-/libvirt/-/jobs/4302989355/artifacts/website/htm… Peter Krempa (7): docs: Remove XSLT table of contents generator scripts: hvsupport: Properly register virConnectOpenAuth/virConnectOpenReadOnly APIs hvsupport: Split out common APIs from hypervisor API section css: Remove override of width for 'hvsupport' page docs: html: Add a manually written index page docs: newapi.xsl: Remove support for generating index page docs: xsl: Simplify templating XSL docs/css/libvirt.css | 2 - docs/html/index.rst | 65 +++++++++++++++++++++++ docs/html/meson.build | 46 +++++++++++++++- docs/newapi.xsl | 119 ++++++++++++++++-------------------------- docs/page.xsl | 66 ----------------------- scripts/hvsupport.py | 71 +++++++++++++++++++------ 6 files changed, 208 insertions(+), 161 deletions(-) create mode 100644 docs/html/index.rst -- 2.40.1

2 8

[PATCH v3] vfio/pci: Propagate ACPI notifications to user-space via eventfd
by Grzegorz Jaszczyk 17 May '23

17 May '23

From: Grzegorz Jaszczyk <jaz(a)semihalf.com> To allow pass-through devices receiving ACPI notifications, permit to register ACPI notify handler (via introduced new ioctl) for a given device. The handler role is to receive and propagate such ACPI notifications to the user-space through the user provided eventfd. This allows VMM to receive and propagate them further to the VM, where the actual driver for pass-through device resides and can react to device specific notifications accordingly. The eventfd usage ensures VMM and device isolation: it allows to use a dedicated channel associated with the device for such events, such that the VMM has direct access. Since the eventfd counter is used as ACPI notification value placeholder, the eventfd signaling needs to be serialized in order to not end up with notification values being coalesced. Therefore ACPI notification values are buffered and signalized one by one, when the previous notification value has been consumed. Signed-off-by: Grzegorz Jaszczyk <jaz(a)semihalf.com> --- Changelog v2..v3: - Fix compilation warnings when building with "W=1" Changelog v1..v2: - The v2 implementation is actually completely different then v1: instead of using acpi netlink events for propagating ACPI notifications to the user space take advantage of eventfd, which can provide better VMM and device isolation: it allows to use a dedicated channel associated with the device for such events, such that the VMM has direct access. - Using eventfd counter as notification value placeholder was suggested in v1 and requires additional serialization logic introduced in v2. - Since the vfio-pci supports non-ACPI platforms address !CONFIG_ACPI case. - v1 discussion: https://patchwork.kernel.org/project/kvm/patch/20230307220553.631069-1-jaz@… --- drivers/vfio/pci/vfio_pci_core.c | 214 +++++++++++++++++++++++++++++++ include/linux/vfio_pci_core.h | 11 ++ include/uapi/linux/vfio.h | 15 +++ 3 files changed, 240 insertions(+) diff --git a/drivers/vfio/pci/vfio_pci_core.c b/drivers/vfio/pci/vfio_pci_core.c index a5ab416cf476..2d6101e89fde 100644 --- a/drivers/vfio/pci/vfio_pci_core.c +++ b/drivers/vfio/pci/vfio_pci_core.c @@ -10,6 +10,7 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +#include <linux/acpi.h> #include <linux/aperture.h> #include <linux/device.h> #include <linux/eventfd.h> @@ -679,6 +680,70 @@ void vfio_pci_core_disable(struct vfio_pci_core_device *vdev) } EXPORT_SYMBOL_GPL(vfio_pci_core_disable); +struct notification_queue { + int notification_val; + struct list_head notify_val_next; +}; + +#if IS_ENABLED(CONFIG_ACPI) +static void vfio_pci_core_acpi_notify(acpi_handle handle, u32 event, void *data) +{ + struct vfio_pci_core_device *vdev = (struct vfio_pci_core_device *)data; + struct vfio_acpi_notification *acpi_notify = vdev->acpi_notification; + struct notification_queue *entry; + + entry = kmalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) + return; + + entry->notification_val = event; + INIT_LIST_HEAD(&entry->notify_val_next); + + mutex_lock(&acpi_notify->notification_list_lock); + list_add_tail(&entry->notify_val_next, &acpi_notify->notification_list); + mutex_unlock(&acpi_notify->notification_list_lock); + + schedule_work(&acpi_notify->acpi_notification_work); +} + +static void vfio_pci_acpi_notify_close_device(struct vfio_pci_core_device *vdev) +{ + struct vfio_acpi_notification *acpi_notify = vdev->acpi_notification; + struct pci_dev *pdev = vdev->pdev; + struct acpi_device *adev = ACPI_COMPANION(&pdev->dev); + struct notification_queue *entry, *entry_tmp; + u64 cnt; + + if (!acpi_notify || !acpi_notify->acpi_notify_trigger) + return; + + acpi_remove_notify_handler(adev->handle, ACPI_DEVICE_NOTIFY, + vfio_pci_core_acpi_notify); + + eventfd_ctx_remove_wait_queue(acpi_notify->acpi_notify_trigger, + &acpi_notify->wait, &cnt); + + flush_work(&acpi_notify->acpi_notification_work); + + mutex_lock(&acpi_notify->notification_list_lock); + list_for_each_entry_safe(entry, entry_tmp, + &acpi_notify->notification_list, + notify_val_next) { + list_del(&entry->notify_val_next); + kfree(entry); + } + mutex_unlock(&acpi_notify->notification_list_lock); + + eventfd_ctx_put(acpi_notify->acpi_notify_trigger); + + kfree(acpi_notify); + + vdev->acpi_notification = NULL; +} +#else +static void vfio_pci_acpi_notify_close_device(struct vfio_pci_core_device *vdev) {} +#endif /* CONFIG_ACPI */ + void vfio_pci_core_close_device(struct vfio_device *core_vdev) { struct vfio_pci_core_device *vdev = @@ -705,6 +770,8 @@ void vfio_pci_core_close_device(struct vfio_device *core_vdev) vdev->req_trigger = NULL; } mutex_unlock(&vdev->igate); + + vfio_pci_acpi_notify_close_device(vdev); } EXPORT_SYMBOL_GPL(vfio_pci_core_close_device); @@ -882,6 +949,151 @@ int vfio_pci_core_register_dev_region(struct vfio_pci_core_device *vdev, } EXPORT_SYMBOL_GPL(vfio_pci_core_register_dev_region); +#if IS_ENABLED(CONFIG_ACPI) +static int vfio_pci_eventfd_wakeup(wait_queue_entry_t *wait, unsigned int mode, + int sync, void *key) +{ + struct vfio_acpi_notification *acpi_notify = + container_of(wait, struct vfio_acpi_notification, wait); + __poll_t flags = key_to_poll(key); + + /* + * eventfd_read signalize EPOLLOUT at the end of its function - this + * means previous eventfd with its notification value was consumed so + * the next notification can be signalized now if pending - schedule + * proper work. + */ + if (flags & EPOLLOUT) { + mutex_unlock(&acpi_notify->notification_lock); + schedule_work(&acpi_notify->acpi_notification_work); + } + + return 0; +} + +static void vfio_pci_ptable_queue_proc(struct file *file, + wait_queue_head_t *wqh, poll_table *pt) +{ + struct vfio_acpi_notification *acpi_notify = + container_of(pt, struct vfio_acpi_notification, pt); + + add_wait_queue(wqh, &acpi_notify->wait); +} + +static void acpi_notification_work_fn(struct work_struct *work) +{ + struct vfio_acpi_notification *acpi_notify; + struct notification_queue *entry; + + acpi_notify = container_of(work, struct vfio_acpi_notification, + acpi_notification_work); + + mutex_lock(&acpi_notify->notification_list_lock); + if (list_empty(&acpi_notify->notification_list) || !acpi_notify->acpi_notify_trigger) + goto out; + + /* + * If the previous eventfd was not yet consumed by user-space lets hold + * on and exit. The notification function will be rescheduled when + * signaling eventfd will be possible (when the EPOLLOUT will be + * signalized and unlocks notify_events). + */ + if (!mutex_trylock(&acpi_notify->notification_lock)) + goto out; + + entry = list_first_entry(&acpi_notify->notification_list, + struct notification_queue, notify_val_next); + + list_del(&entry->notify_val_next); + mutex_unlock(&acpi_notify->notification_list_lock); + + eventfd_signal(acpi_notify->acpi_notify_trigger, entry->notification_val); + + kfree(entry); + + return; +out: + mutex_unlock(&acpi_notify->notification_list_lock); +} + +static int vfio_pci_ioctl_acpi_notify_eventfd(struct vfio_pci_core_device *vdev, struct + vfio_irq_info __user *arg) +{ + struct file *acpi_notify_trigger_file; + struct vfio_acpi_notification *acpi_notify; + struct pci_dev *pdev = vdev->pdev; + struct acpi_device *adev = ACPI_COMPANION(&pdev->dev); + struct vfio_acpi_notify_eventfd entry; + struct eventfd_ctx *efdctx; + acpi_status status; + + if (!adev) + return -ENODEV; + + if (copy_from_user(&entry, arg, sizeof(entry))) + return -EFAULT; + + if (entry.notify_eventfd < 0) + return -EINVAL; + + efdctx = eventfd_ctx_fdget(entry.notify_eventfd); + if (IS_ERR(efdctx)) + return PTR_ERR(efdctx); + + vdev->acpi_notification = kzalloc(sizeof(*acpi_notify), GFP_KERNEL); + if (!vdev->acpi_notification) + return -ENOMEM; + + acpi_notify = vdev->acpi_notification; + + INIT_WORK(&acpi_notify->acpi_notification_work, acpi_notification_work_fn); + INIT_LIST_HEAD(&acpi_notify->notification_list); + + acpi_notify->acpi_notify_trigger = efdctx; + + mutex_init(&acpi_notify->notification_lock); + + /* + * Install custom wake-up handler to be notified whenever underlying + * eventfd is consumed by the user-space. + */ + init_waitqueue_func_entry(&acpi_notify->wait, vfio_pci_eventfd_wakeup); + init_poll_funcptr(&acpi_notify->pt, vfio_pci_ptable_queue_proc); + + acpi_notify_trigger_file = eventfd_fget(entry.notify_eventfd); + vfs_poll(acpi_notify_trigger_file, &acpi_notify->pt); + + status = acpi_install_notify_handler(adev->handle, ACPI_DEVICE_NOTIFY, + vfio_pci_core_acpi_notify, (void *)vdev); + + if (ACPI_FAILURE(status)) { + u64 cnt; + + pci_err(pdev, "Failed to install notify handler: %s", + acpi_format_exception(status)); + + eventfd_ctx_remove_wait_queue(acpi_notify->acpi_notify_trigger, + &acpi_notify->wait, &cnt); + + flush_work(&acpi_notify->acpi_notification_work); + + eventfd_ctx_put(acpi_notify->acpi_notify_trigger); + + kfree(acpi_notify); + + return -ENODEV; + } + + return 0; +} +#else +static int vfio_pci_ioctl_acpi_notify_eventfd(struct vfio_pci_core_device *vdev, struct + vfio_irq_info __user *arg) +{ + return -ENODEV; +} +#endif /* CONFIG_ACPI */ + static int vfio_pci_ioctl_get_info(struct vfio_pci_core_device *vdev, struct vfio_device_info __user *arg) { @@ -1398,6 +1610,8 @@ long vfio_pci_core_ioctl(struct vfio_device *core_vdev, unsigned int cmd, return vfio_pci_ioctl_reset(vdev, uarg); case VFIO_DEVICE_SET_IRQS: return vfio_pci_ioctl_set_irqs(vdev, uarg); + case VFIO_ACPI_NOTIFY_EVENTFD: + return vfio_pci_ioctl_acpi_notify_eventfd(vdev, uarg); default: return -ENOTTY; } diff --git a/include/linux/vfio_pci_core.h b/include/linux/vfio_pci_core.h index 367fd79226a3..3711e8a1c6f0 100644 --- a/include/linux/vfio_pci_core.h +++ b/include/linux/vfio_pci_core.h @@ -49,6 +49,16 @@ struct vfio_pci_region { u32 flags; }; +struct vfio_acpi_notification { + struct eventfd_ctx *acpi_notify_trigger; + struct work_struct acpi_notification_work; + struct list_head notification_list; + struct mutex notification_list_lock; + struct mutex notification_lock; + poll_table pt; + wait_queue_entry_t wait; +}; + struct vfio_pci_core_device { struct vfio_device vdev; struct pci_dev *pdev; @@ -96,6 +106,7 @@ struct vfio_pci_core_device { struct mutex vma_lock; struct list_head vma_list; struct rw_semaphore memory_lock; + struct vfio_acpi_notification *acpi_notification; }; /* Will be exported for vfio pci drivers usage */ diff --git a/include/uapi/linux/vfio.h b/include/uapi/linux/vfio.h index 0552e8dcf0cb..d4b602d8f4b2 100644 --- a/include/uapi/linux/vfio.h +++ b/include/uapi/linux/vfio.h @@ -1622,6 +1622,21 @@ struct vfio_iommu_spapr_tce_remove { }; #define VFIO_IOMMU_SPAPR_TCE_REMOVE _IO(VFIO_TYPE, VFIO_BASE + 20) +/** + * VFIO_ACPI_NOTIFY_EVENTFD - _IOW(VFIO_TYPE, VFIO_BASE + 21, struct vfio_acpi_notify_eventfd) + * + * Register ACPI notify handler for a given device which will allow to receive + * and propagate ACPI notifications to the user-space through the user provided + * eventfd. + * + * Return: 0 on success, -errno on failure. + */ +struct vfio_acpi_notify_eventfd { + __s32 notify_eventfd; + __u32 reserved; +}; +#define VFIO_ACPI_NOTIFY_EVENTFD _IO(VFIO_TYPE, VFIO_BASE + 21) + /* ***************************************************************** */ #endif /* _UAPIVFIO_H */ -- 2.40.1.495.gc816e09b53d-goog

2 2

[PATCH 00/12] docs: Finish conversion of non-generated docs to rST and fix and cleanup CSS
by Peter Krempa 17 May '23

17 May '23

Browse the results at: https://pipo.sk.gitlab.io/-/libvirt/-/jobs/4294950526/artifacts/website/ind… Peter Krempa (12): css: Drop styles for '#projects' id css: Drop style for '#changelog' id css: Drop style for 'p.image' selector css: Drop styles for '.mail' class css: Drop styles for '.gitmirror' class css: mobile: Replace tabs with spaces css: mobile: Fix responsive design of 'docs' and 'knowledgebase' pages css: mobile: Fix hiding of big logo in mobile layout css: mobile: Make colums in "3 column" mobile layout wider css: Fix styling of the "3 panel" pages docs: acl: Convert to 'rst' docs: index: Convert to 'rst' docs/acl.html.in | 101 ------------------------------ docs/acl.rst | 76 +++++++++++++++++++++++ docs/aclpolkit.rst | 2 +- docs/css/libvirt.css | 143 ++++++++----------------------------------- docs/css/mobile.css | 136 ++++++++++++++++++++-------------------- docs/index.html.in | 87 -------------------------- docs/index.rst | 70 +++++++++++++++++++++ docs/meson.build | 15 +---- 8 files changed, 245 insertions(+), 385 deletions(-) delete mode 100644 docs/acl.html.in create mode 100644 docs/acl.rst delete mode 100644 docs/index.html.in create mode 100644 docs/index.rst -- 2.40.1

2 15

[PATCH 0/3] qemu: Introduce pipewire audio backend
by Michal Privoznik 16 May '23

16 May '23

You'd expect to see src/qemu/qemu_capabilities.c changed, wouldn't you. Well, there's no way to query for supported audio backends, see: https://gitlab.com/libvirt/libvirt/-/issues/473 Michal Prívozník (3): conf: Introduce pipewire audio backend qemu: Generate cmd line for pipewire audio backend NEWS: Document pipewire audio backend NEWS.rst | 10 +++ docs/formatdomain.rst | 35 +++++++++- src/conf/domain_conf.c | 70 +++++++++++++++++++ src/conf/domain_conf.h | 12 ++++ src/conf/schemas/domaincommon.rng | 37 ++++++++++ src/qemu/qemu_command.c | 63 +++++++++++++++++ src/qemu/qemu_validate.c | 1 + .../audio-many-backends.x86_64-latest.args | 2 + .../qemuxml2argvdata/audio-many-backends.xml | 1 + .../audio-pipewire-best.x86_64-latest.args | 36 ++++++++++ .../qemuxml2argvdata/audio-pipewire-best.xml | 43 ++++++++++++ .../audio-pipewire-full.x86_64-latest.args | 36 ++++++++++ .../qemuxml2argvdata/audio-pipewire-full.xml | 43 ++++++++++++ .../audio-pipewire-minimal.x86_64-latest.args | 36 ++++++++++ .../audio-pipewire-minimal.xml | 36 ++++++++++ tests/qemuxml2argvtest.c | 12 ++++ .../audio-pipewire-best.xml | 43 ++++++++++++ .../audio-pipewire-full.xml | 43 ++++++++++++ .../audio-pipewire-minimal.xml | 36 ++++++++++ tests/qemuxml2xmltest.c | 3 + 20 files changed, 596 insertions(+), 2 deletions(-) create mode 100644 tests/qemuxml2argvdata/audio-pipewire-best.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/audio-pipewire-best.xml create mode 100644 tests/qemuxml2argvdata/audio-pipewire-full.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/audio-pipewire-full.xml create mode 100644 tests/qemuxml2argvdata/audio-pipewire-minimal.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/audio-pipewire-minimal.xml create mode 100644 tests/qemuxml2xmloutdata/audio-pipewire-best.xml create mode 100644 tests/qemuxml2xmloutdata/audio-pipewire-full.xml create mode 100644 tests/qemuxml2xmloutdata/audio-pipewire-minimal.xml -- 2.39.3

4 7

[PATCH 0/2] qemu_domin: Account for NVMe disks when calculating memlock limit on hotplug
by Michal Privoznik 16 May '23

16 May '23

*** 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 +++++++++++++++++++++++++++-------------- src/qemu/qemu_domain.h | 9 ++--- src/qemu/qemu_hotplug.c | 16 ++++----- src/qemu/qemu_process.c | 2 +- tests/qemumemlocktest.c | 2 +- 5 files changed, 64 insertions(+), 40 deletions(-) -- 2.39.3

2 4

[PATCH 0/2] conf: Fix formatting of 'associativity' and 'policy' cache attributes
by Peter Krempa 16 May '23

16 May '23

Peter Krempa (2): virDomainNumaDefNodeCacheParseXML: Refactor parsing of cache XML conf: numa: Allow formatting 'none' values for 'associativity' and 'policy' of cache src/conf/numa_conf.c | 73 +++++-------------- .../numatune-hmat-none.x86_64-latest.args | 49 +++++++++++++ tests/qemuxml2argvdata/numatune-hmat-none.xml | 54 ++++++++++++++ tests/qemuxml2argvtest.c | 1 + .../numatune-hmat-none.x86_64-latest.xml | 55 ++++++++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 178 insertions(+), 55 deletions(-) create mode 100644 tests/qemuxml2argvdata/numatune-hmat-none.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/numatune-hmat-none.xml create mode 100644 tests/qemuxml2xmloutdata/numatune-hmat-none.x86_64-latest.xml -- 2.40.1

2 3

[libvirt PATCH] docs: Fix broken link to Windows MSI downloads
by Kashyap Chamarthy 16 May '23

16 May '23

Right now the URL is 404 due to missing .html. Reported by: Sebastien Wains <swains(a)redhat.com> Signed-off-by: Kashyap Chamarthy <kchamart(a)redhat.com> --- docs/windows.html.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/windows.html.in b/docs/windows.html.in index 0f2d9061c4..561c08cfa1 100644 --- a/docs/windows.html.in +++ b/docs/windows.html.in @@ -17,7 +17,7 @@ <p> Users who need pre-built Windows DLLs of libvirt are advised to use the <a href="https://virt-manager.org">Virt Viewer</a> - pre-compiled <a href="https://virt-manager.org/download/">Windows MSI packages</a> + pre-compiled <a href="https://virt-manager.org/download.html">Windows MSI packages</a> </p> <p> -- 2.40.0

2 3

[PATCH] virmockstathelpers: Adapt to musl-1.2.4
by Michal Privoznik 15 May '23

15 May '23

With musl-1.2.3: I get the following macros defined (from $builddir/meson-config.h): #define WITH_LSTAT 1 #define WITH_LSTAT64 1 #define WITH_LSTAT_DECL 1 #define WITH_STAT 1 #define WITH_STAT64 1 #define WITH_STAT_DECL 1 #define WITH___LXSTAT 1 #define WITH___LXSTAT64 1 #define WITH___XSTAT 1 #define WITH___XSTAT64 1 which in turn means the virmockstathelpers.c ends up defining: MOCK_STAT64 MOCK_LSTAT64 But with musl-1.2.4 everything changes and the set of defined macros gets simplified to: #define WITH_LSTAT 1 #define WITH_LSTAT_DECL 1 #define WITH_STAT 1 #define WITH_STAT_DECL 1 #define WITH___LXSTAT 1 #define WITH___XSTAT 1 which results in no MOCK_* macros defined in virmockstathelpers.c, i.e. no stat() mocking, nada. The reason for this simplification are thess musl commits [1][2] which removed all 64 bit aliases. And that's not what our logic for deciding what flavor of stat() to mock counted with. Nevertheless, we do build with Alpine Linux in our CI, so how come we don't see this problem there? Well, simply because Alpine Linux maintainers decided to revert the commits [3][4]. But on distributions that use vanilla musl, this problem can be seen easily. 1: https://git.musl-libc.org/cgit/musl/commit/?id=246f1c811448f37a44b41cd8df8d… 2: https://git.musl-libc.org/cgit/musl/commit/?id=25e6fee27f4a293728dd15b65917… 3: https://git.alpinelinux.org/aports/commit/main/musl?id=6a5563fbb45b3d9d6067… 4: https://git.alpinelinux.org/aports/commit/main/musl?id=a089bd852f8983623fa8… Resolves: https://bugs.gentoo.org/906167 Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tests/virmockstathelpers.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/tests/virmockstathelpers.c b/tests/virmockstathelpers.c index 5b1f3b08a7..a02ee04503 100644 --- a/tests/virmockstathelpers.c +++ b/tests/virmockstathelpers.c @@ -64,6 +64,10 @@ * building on Apple Silicon (aarch64) those functions are missing * from the header altogether and should not be mocked. * + * Starting from 1.2.4, the musl library dropped 64-bit aliases, i.e. there's + * no stat64() only stat(). BTW: while musl implements __xstat(), it's never + * declared. And even the implementation is but a direct call to stat(). + * * With all this in mind the list of functions we have to mock will depend * on several factors * @@ -82,7 +86,9 @@ #if !defined(__APPLE__) # if !defined(WITH___XSTAT_DECL) # if defined(WITH_STAT) -# if !defined(WITH___XSTAT) && !defined(WITH_STAT64) +# if (!defined(WITH___XSTAT) && !defined(WITH_STAT64)) || \ + (defined(WITH_STAT) && defined(WITH_STAT_DECL) && !defined(WITH_STAT64) && \ + !defined(WITH___XSTAT64)) /* glibc || musl */ # define MOCK_STAT # endif # endif @@ -99,7 +105,9 @@ # endif /* WITH___XSTAT_DECL */ # if !defined(WITH___LXSTAT_DECL) # if defined(WITH_LSTAT) -# if !defined(WITH___LXSTAT) && !defined(WITH_LSTAT64) +# if (!defined(WITH___LXSTAT) && !defined(WITH_LSTAT64)) || \ + (defined(WITH_LSTAT) && defined(WITH_LSTAT_DECL) && !defined(WITH_LSTAT64) \ + && !defined(WITH___LXSTAT64)) /* glibc || musl */ # define MOCK_LSTAT # endif # endif -- 2.39.3

3 2

[PATCH 0/3] backup: Fix schema and default filename filling logic
by Peter Krempa 15 May '23

15 May '23

Reported by 'philipp' on OFTC. Peter Krempa (3): virDomainBackupDefAssignStore: Restructure control flow conf: backup: Fix logic for generating default backup filenames schemas: backup: Allow missing 'type' attribute for backup disk src/conf/backup_conf.c | 17 +++++++++++------ src/conf/schemas/domainbackup.rng | 4 ++++ tests/domainbackupxml2xmlin/backup-push.xml | 3 +++ tests/domainbackupxml2xmlout/backup-push.xml | 9 +++++++++ 4 files changed, 27 insertions(+), 6 deletions(-) -- 2.40.1

2 4