linux/drivers/virtio, branch v6.16

virtio_ring: Fix error reporting in virtqueue_resize

2025-07-03T09:40:02Z

The virtqueue_resize() function was not correctly propagating error codes from its internal resize helper functions, specifically virtqueue_resize_packet() and virtqueue_resize_split(). If these helpers returned an error, but the subsequent call to virtqueue_enable_after_reset() succeeded, the original error from the resize operation would be masked. Consequently, virtqueue_resize() could incorrectly report success to its caller despite an underlying resize failure. This change restores the original code behavior: if (vdev->config->enable_vq_after_reset(_vq)) return -EBUSY; return err; Fix: commit ad48d53b5b3f ("virtio_ring: separate the logic of reset/enable from virtqueue_resize") Cc: xuanzhuo@linux.alibaba.com Signed-off-by: Laurent Vivier Acked-by: Jason Wang Link: https://patch.msgid.link/20250521092236.661410-2-lvivier@redhat.com Tested-by: Lei Yang Acked-by: Michael S. Tsirkin Signed-off-by: Paolo Abeni

virtio_rtc: Add RTC class driver

2025-05-27T14:27:54Z

Expose the virtio-rtc UTC-like clock as an RTC clock to userspace - if it is present, and if it does not step on leap seconds. The RTC class enables the virtio-rtc device to resume the system from sleep states on RTC alarm. Support RTC alarm if the virtio-rtc alarm feature is present. The virtio-rtc device signals an alarm by marking an alarmq buffer as used. Peculiarities ------------- A virtio-rtc clock is a bit special for an RTC clock in that - the clock may step (also backwards) autonomously at any time and - the device, and its notification mechanism, will be reset during boot or resume from sleep. The virtio-rtc device avoids that the driver might miss an alarm. The device signals an alarm whenever the clock has reached or passed the alarm time, and also when the device is reset (on boot or resume from sleep), if the alarm time is in the past. Open Issue ---------- The CLOCK_BOOTTIME_ALARM will use the RTC clock to wake up from sleep, and implicitly assumes that no RTC clock steps will occur during sleep. The RTC class driver does not know whether the current alarm is a real-time alarm or a boot-time alarm. Perhaps this might be handled by the driver also setting a virtio-rtc monotonic alarm (which uses a clock similar to CLOCK_BOOTTIME_ALARM). The virtio-rtc monotonic alarm would just be used to wake up in case it was a CLOCK_BOOTTIME_ALARM alarm. Otherwise, the behavior should not differ from other RTC class drivers. Signed-off-by: Peter Hilber Acked-by: Alexandre Belloni Message-Id: <20250509160734.1772-5-quic_philber@quicinc.com> Signed-off-by: Michael S. Tsirkin

virtio_rtc: Add Arm Generic Timer cross-timestamping

2025-05-27T14:27:54Z

For platforms using the Arm Generic Timer, add precise cross-timestamping support to virtio_rtc. Always report the CP15 virtual counter as the HW counter in use by arm_arch_timer, since the Linux kernel's usage of the Arm Generic Timer should always be compatible with this. Signed-off-by: Peter Hilber Message-Id: <20250509160734.1772-4-quic_philber@quicinc.com> Signed-off-by: Michael S. Tsirkin

virtio_rtc: Add PTP clocks

2025-05-27T14:27:54Z

Expose the virtio_rtc clocks as PTP clocks to userspace, similar to ptp_kvm. virtio_rtc can expose multiple clocks, e.g. a UTC clock and a monotonic clock. Userspace should distinguish different clocks through the name assigned by the driver. In particular, UTC-like clocks can also be distinguished by if and how leap seconds are smeared. udev rules such as the following can be used to get different symlinks for different clock types: SUBSYSTEM=="ptp", ATTR{clock_name}=="Virtio PTP type 0/variant 0", SYMLINK += "ptp_virtio" SUBSYSTEM=="ptp", ATTR{clock_name}=="Virtio PTP type 1/variant 0", SYMLINK += "ptp_virtio_tai" SUBSYSTEM=="ptp", ATTR{clock_name}=="Virtio PTP type 2/variant 0", SYMLINK += "ptp_virtio_monotonic" SUBSYSTEM=="ptp", ATTR{clock_name}=="Virtio PTP type 3/variant 0", SYMLINK += "ptp_virtio_smear_unspecified" SUBSYSTEM=="ptp", ATTR{clock_name}=="Virtio PTP type 3/variant 1", SYMLINK += "ptp_virtio_smear_noon_linear" SUBSYSTEM=="ptp", ATTR{clock_name}=="Virtio PTP type 3/variant 2", SYMLINK += "ptp_virtio_smear_sls" SUBSYSTEM=="ptp", ATTR{clock_name}=="Virtio PTP type 4/variant 0", SYMLINK += "ptp_virtio_maybe_smeared" The preferred PTP clock reading method is ioctl PTP_SYS_OFFSET_PRECISE2, through the ptp_clock_info.getcrosststamp() op. For now, PTP_SYS_OFFSET_PRECISE2 will return -EOPNOTSUPP through a weak function. PTP_SYS_OFFSET_PRECISE2 requires cross-timestamping support for specific clocksources, which will be added in the following. If the clocksource specific code is enabled, check that the Virtio RTC device supports the respective HW counter before obtaining an actual cross-timestamp from the Virtio device. The Virtio RTC device response time may be higher than the timekeeper seqcount increment interval. Therefore, obtain the cross-timestamp before calling get_device_system_crosststamp(). As a fallback, support the ioctl PTP_SYS_OFFSET_EXTENDED2 for all platforms. Assume that concurrency issues during PTP clock removal are avoided by the posix_clock framework. Kconfig recursive dependencies prevent virtio_rtc from implicitly enabling PTP_1588_CLOCK, therefore just warn the user if PTP_1588_CLOCK is not available. Since virtio_rtc should in the future also expose clocks as RTC class devices, do not depend VIRTIO_RTC on PTP_1588_CLOCK. Signed-off-by: Peter Hilber Message-Id: <20250509160734.1772-3-quic_philber@quicinc.com> Signed-off-by: Michael S. Tsirkin

virtio_rtc: Add module and driver core

2025-05-27T14:27:54Z

Add the virtio_rtc module and driver core. The virtio_rtc module implements a driver compatible with the proposed Virtio RTC device specification. The Virtio RTC (Real Time Clock) device provides information about current time. The device can provide different clocks, e.g. for the UTC or TAI time standards, or for physical time elapsed since some past epoch. The driver can read the clocks with simple or more accurate methods. Implement the core, which interacts with the Virtio RTC device. Apart from this, the core does not expose functionality outside of the virtio_rtc module. Follow-up patches will expose PTP clocks and an RTC Class device. Provide synchronous messaging, which is enough for the expected time synchronization use cases through PTP clocks (similar to ptp_kvm) or RTC Class device. Signed-off-by: Peter Hilber Message-Id: <20250509160734.1772-2-quic_philber@quicinc.com> Signed-off-by: Michael S. Tsirkin

virtio-pci: Fix result size returned for the admin command completion

2025-05-27T14:27:53Z

The result size returned by virtio_pci_admin_dev_parts_get() is 8 bytes larger than the actual result data size. This occurs because the result_sg_size field of the command is filled with the result length from virtqueue_get_buf(), which includes both the data size and an additional 8 bytes of status. This oversized result size causes two issues: 1. The state transferred to the destination includes 8 bytes of extra data at the end. 2. The allocated buffer in the kernel may be smaller than the returned size, leading to failures when reading beyond the allocated size. The commit fixes this by subtracting the status size from the result of virtqueue_get_buf(). This fix has been tested through live migrations with virtio-net, virtio-net-transitional, and virtio-blk devices. Fixes: 704806ca400e ("virtio: Extend the admin command to include the result size") Signed-off-by: Israel Rukshin Reviewed-by: Parav Pandit Reviewed-by: Max Gurtovoy Message-Id: <1745318025-23103-1-git-send-email-israelr@nvidia.com> Signed-off-by: Michael S. Tsirkin

virtgpu: don't reset on shutdown

2025-04-18T14:05:49Z

It looks like GPUs are used after shutdown is invoked. Thus, breaking virtio gpu in the shutdown callback is not a good idea - guest hangs attempting to finish console drawing, with these warnings: [ 20.504464] WARNING: CPU: 0 PID: 568 at drivers/gpu/drm/virtio/virtgpu_vq.c:358 virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.505685] Modules linked in: nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 rfkill ip_set nf_tables nfnetlink vfat fat intel_rapl_msr intel_rapl_common intel_uncore_frequency_common nfit libnvdimm kvm_intel kvm rapl iTCO_wdt iTCO_vendor_support virtio_gpu virtio_dma_buf pcspkr drm_shmem_helper i2c_i801 drm_kms_helper lpc_ich i2c_smbus virtio_balloon joydev drm fuse xfs libcrc32c ahci libahci crct10dif_pclmul crc32_pclmul crc32c_intel libata virtio_net ghash_clmulni_intel net_failover virtio_blk failover serio_raw dm_mirror dm_region_hash dm_log dm_mod [ 20.511847] CPU: 0 PID: 568 Comm: kworker/0:3 Kdump: loaded Tainted: G W ------- --- 5.14.0-578.6675_1757216455.el9.x86_64 #1 [ 20.513157] Hardware name: Red Hat KVM/RHEL, BIOS edk2-20241117-3.el9 11/17/2024 [ 20.513918] Workqueue: events drm_fb_helper_damage_work [drm_kms_helper] [ 20.514626] RIP: 0010:virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.515332] Code: 00 00 48 85 c0 74 0c 48 8b 78 08 48 89 ee e8 51 50 00 00 65 ff 0d 42 e3 74 3f 0f 85 69 ff ff ff 0f 1f 44 00 00 e9 5f ff ff ff <0f> 0b e9 3f ff ff ff 48 83 3c 24 00 74 0e 49 8b 7f 40 48 85 ff 74 [ 20.517272] RSP: 0018:ff34f0a8c0787ad8 EFLAGS: 00010282 [ 20.517820] RAX: 00000000fffffffb RBX: 0000000000000000 RCX: 0000000000000820 [ 20.518565] RDX: 0000000000000000 RSI: ff34f0a8c0787be0 RDI: ff218bef03a26300 [ 20.519308] RBP: ff218bef03a26300 R08: 0000000000000001 R09: ff218bef07224360 [ 20.520059] R10: 0000000000008dc0 R11: 0000000000000002 R12: ff218bef02630028 [ 20.520806] R13: ff218bef0263fb48 R14: ff218bef00cb8000 R15: ff218bef07224360 [ 20.521555] FS: 0000000000000000(0000) GS:ff218bef7ba00000(0000) knlGS:0000000000000000 [ 20.522397] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 20.522996] CR2: 000055ac4f7871c0 CR3: 000000010b9f2002 CR4: 0000000000771ef0 [ 20.523740] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 20.524477] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400 [ 20.525223] PKRU: 55555554 [ 20.525515] Call Trace: [ 20.525777] [ 20.526003] ? show_trace_log_lvl+0x1c4/0x2df [ 20.526464] ? show_trace_log_lvl+0x1c4/0x2df [ 20.526925] ? virtio_gpu_queue_fenced_ctrl_buffer+0x82/0x2c0 [virtio_gpu] [ 20.527643] ? virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.528282] ? __warn+0x7e/0xd0 [ 20.528621] ? virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.529256] ? report_bug+0x100/0x140 [ 20.529643] ? handle_bug+0x3c/0x70 [ 20.530010] ? exc_invalid_op+0x14/0x70 [ 20.530421] ? asm_exc_invalid_op+0x16/0x20 [ 20.530862] ? virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.531506] ? virtio_gpu_queue_ctrl_sgs+0x174/0x290 [virtio_gpu] [ 20.532148] virtio_gpu_queue_fenced_ctrl_buffer+0x82/0x2c0 [virtio_gpu] [ 20.532843] virtio_gpu_primary_plane_update+0x3e2/0x460 [virtio_gpu] [ 20.533520] drm_atomic_helper_commit_planes+0x108/0x320 [drm_kms_helper] [ 20.534233] drm_atomic_helper_commit_tail+0x45/0x80 [drm_kms_helper] [ 20.534914] commit_tail+0xd2/0x130 [drm_kms_helper] [ 20.535446] drm_atomic_helper_commit+0x11b/0x140 [drm_kms_helper] [ 20.536097] drm_atomic_commit+0xa4/0xe0 [drm] [ 20.536588] ? __pfx___drm_printfn_info+0x10/0x10 [drm] [ 20.537162] drm_atomic_helper_dirtyfb+0x192/0x270 [drm_kms_helper] [ 20.537823] drm_fbdev_shmem_helper_fb_dirty+0x43/0xa0 [drm_shmem_helper] [ 20.538536] drm_fb_helper_damage_work+0x87/0x160 [drm_kms_helper] [ 20.539188] process_one_work+0x194/0x380 [ 20.539612] worker_thread+0x2fe/0x410 [ 20.540007] ? __pfx_worker_thread+0x10/0x10 [ 20.540456] kthread+0xdd/0x100 [ 20.540791] ? __pfx_kthread+0x10/0x10 [ 20.541190] ret_from_fork+0x29/0x50 [ 20.541566] [ 20.541802] ---[ end trace 0000000000000000 ]--- It looks like the shutdown is called in the middle of console drawing, so we should either wait for it to finish, or let drm handle the shutdown. This patch implements this second option: Add an option for drivers to bypass the common break+reset handling. As DRM is careful to flush/synchronize outstanding buffers, it looks like GPU can just have a NOP there. Reviewed-by: Eric Auger Tested-by: Eric Auger Fixes: 8bd2fa086a04 ("virtio: break and reset virtio devices on device_shutdown()") Cc: Eric Auger Cc: Jocelyn Falempe Signed-off-by: Michael S. Tsirkin Message-Id: <8490dbeb6f79ed039e6c11d121002618972538a3.1744293540.git.mst@redhat.com>

virtio_ring: Fix data race by tagging event_triggered as racy for KCSAN

2025-04-14T13:51:31Z

syzbot reports a data-race when accessing the event_triggered, here is the simplified stack when the issue occurred: ================================================================== BUG: KCSAN: data-race in virtqueue_disable_cb / virtqueue_enable_cb_delayed write to 0xffff8881025bc452 of 1 bytes by task 3288 on cpu 0: virtqueue_enable_cb_delayed+0x42/0x3c0 drivers/virtio/virtio_ring.c:2653 start_xmit+0x230/0x1310 drivers/net/virtio_net.c:3264 __netdev_start_xmit include/linux/netdevice.h:5151 [inline] netdev_start_xmit include/linux/netdevice.h:5160 [inline] xmit_one net/core/dev.c:3800 [inline] read to 0xffff8881025bc452 of 1 bytes by interrupt on cpu 1: virtqueue_disable_cb_split drivers/virtio/virtio_ring.c:880 [inline] virtqueue_disable_cb+0x92/0x180 drivers/virtio/virtio_ring.c:2566 skb_xmit_done+0x5f/0x140 drivers/net/virtio_net.c:777 vring_interrupt+0x161/0x190 drivers/virtio/virtio_ring.c:2715 __handle_irq_event_percpu+0x95/0x490 kernel/irq/handle.c:158 handle_irq_event_percpu kernel/irq/handle.c:193 [inline] value changed: 0x01 -> 0x00 ================================================================== When the data race occurs, the function virtqueue_enable_cb_delayed() sets event_triggered to false, and virtqueue_disable_cb_split/packed() reads it as false due to the race condition. Since event_triggered is an unreliable hint used for optimization, this should only cause the driver temporarily suggest that the device not send an interrupt notification when the event index is used. Fix this KCSAN reported data-race issue by explicitly tagging the access as data_racy. Reported-by: syzbot+efe683d57990864b8c8e@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/67c7761a.050a0220.15b4b9.0018.GAE@google.com/ Signed-off-by: Zhongqiu Han Message-Id: <20250312130412.3516307-1-quic_zhonhan@quicinc.com> Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang

virtio_pci: Use self group type for cap commands

2025-04-14T13:51:31Z

Section 2.12.1.2 of v1.4 of the VirtIO spec states: The device and driver capabilities commands are currently defined for self group type. 1. VIRTIO_ADMIN_CMD_CAP_ID_LIST_QUERY 2. VIRTIO_ADMIN_CMD_DEVICE_CAP_GET 3. VIRTIO_ADMIN_CMD_DRIVER_CAP_SET Fixes: bfcad518605d ("virtio: Manage device and driver capabilities via the admin commands") Signed-off-by: Daniel Jurgens Reviewed-by: Parav Pandit Message-Id: <20250304161442.90700-1-danielj@nvidia.com> Signed-off-by: Michael S. Tsirkin

Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost

2025-04-02T01:52:54Z

Pull virtio updates from Michael Tsirkin: "A small number of improvements all over the place: - shutdown has been reworked to reset devices - virtio fs is now allowed in vduse - vhost-scsi memory use has been reduced - cleanups, fixes all over the place A couple more fixes are being tested and will be merged after rc1" * tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost: vhost-scsi: Reduce response iov mem use vhost-scsi: Allocate iov_iter used for unaligned copies when needed vhost-scsi: Stop duplicating se_cmd fields vhost-scsi: Dynamically allocate scatterlists vhost-scsi: Return queue full for page alloc failures during copy vhost-scsi: Add better resource allocation failure handling vhost-scsi: Allocate T10 PI structs only when enabled vhost-scsi: Reduce mem use by moving upages to per queue vduse: add virtio_fs to allowed dev id sound/virtio: Fix cancel_sync warnings on uninitialized work_structs vdpa/mlx5: Fix oversized null mkey longer than 32bit vdpa/mlx5: Fix mlx5_vdpa_get_config() endianness on big-endian machines vhost-scsi: Fix handling of multiple calls to vhost_scsi_set_endpoint tools: virtio/linux/module.h add MODULE_DESCRIPTION() define. tools: virtio/linux/compiler.h: Add data_race() define. tools/virtio: Add DMA_MAPPING_ERROR and sg_dma_len api define for virtio test virtio: break and reset virtio devices on device_shutdown()