October 2025 - Linux-stable-mirror

[PATCH 5.10/5.15] cpufreq: davinci: Fix clk use after free

by Vasiliy Kovalev

From: Uwe Kleine-König <u.kleine-koenig(a)pengutronix.de> [ Upstream commit 5d8f384a9b4fc50f6a18405f1c08e5a87a77b5b3 ] The remove function first frees the clks and only then calls cpufreq_unregister_driver(). If one of the cpufreq callbacks is called just before cpufreq_unregister_driver() is run, the freed clks might be used. Fixes: 6601b8030de3 ("davinci: add generic CPUFreq driver for DaVinci") Signed-off-by: Uwe Kleine-König <u.kleine-koenig(a)pengutronix.de> Acked-by: Viresh Kumar <viresh.kumar(a)linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> (cherry picked from commit a5f024d0e6f91e05c816ad4ee8837173369dd5cb) [ kovalev: bp to fix CVE-2023-53544 ] Signed-off-by: Vasiliy Kovalev <kovalev(a)altlinux.org> --- drivers/cpufreq/davinci-cpufreq.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/cpufreq/davinci-cpufreq.c b/drivers/cpufreq/davinci-cpufreq.c index 91f477a6cbc4..62380f05aec8 100644 --- a/drivers/cpufreq/davinci-cpufreq.c +++ b/drivers/cpufreq/davinci-cpufreq.c @@ -133,12 +133,14 @@ static int __init davinci_cpufreq_probe(struct platform_device *pdev) static int __exit davinci_cpufreq_remove(struct platform_device *pdev) { + cpufreq_unregister_driver(&davinci_driver); + clk_put(cpufreq.armclk); if (cpufreq.asyncclk) clk_put(cpufreq.asyncclk); - return cpufreq_unregister_driver(&davinci_driver); + return 0; } static struct platform_driver davinci_cpufreq_driver = { -- 2.50.1

2 weeks, 3 days

1
0
0 0

[PATCH v2] ACPI: video: Fix use-after-free in acpi_video_switch_brightness()

by Yuhao Jiang

The switch_brightness_work delayed work accesses device->brightness and device->backlight, which are freed by acpi_video_dev_unregister_backlight() during device removal. If the work executes after acpi_video_bus_unregister_backlight() frees these resources, it causes a use-after-free when acpi_video_switch_brightness() dereferences device->brightness or device->backlight. Fix this by calling cancel_delayed_work_sync() for each device's switch_brightness_work before unregistering its backlight resources. This ensures the work completes before the memory is freed. Fixes: 8ab58e8e7e097 ("ACPI / video: Fix backlight taking 2 steps on a brightness up/down keypress") Cc: stable(a)vger.kernel.org Signed-off-by: Yuhao Jiang <danisjiang(a)gmail.com> --- Changes in v2: - Move cancel_delayed_work_sync() to acpi_video_bus_unregister_backlight() instead of acpi_video_bus_put_devices() for better logic clarity and to prevent potential UAF of device->brightness - Correct Fixes tag to point to 8ab58e8e7e097 which introduced the delayed work - Link to v1: https://lore.kernel.org/all/20251022040859.2102914-1-danisjiang@gmail.com --- drivers/acpi/acpi_video.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/acpi/acpi_video.c b/drivers/acpi/acpi_video.c index 103f29661576..64709658bdc4 100644 --- a/drivers/acpi/acpi_video.c +++ b/drivers/acpi/acpi_video.c @@ -1869,8 +1869,10 @@ static int acpi_video_bus_unregister_backlight(struct acpi_video_bus *video) error = unregister_pm_notifier(&video->pm_nb); mutex_lock(&video->device_list_lock); - list_for_each_entry(dev, &video->video_device_list, entry) + list_for_each_entry(dev, &video->video_device_list, entry) { + cancel_delayed_work_sync(&dev->switch_brightness_work); acpi_video_dev_unregister_backlight(dev); + } mutex_unlock(&video->device_list_lock); video->backlight_registered = false; -- 2.34.1

2 weeks, 3 days

2
2
0 0

[PATCH 5.10.y] ionic: catch failure from devlink_alloc

by Vasiliy Kovalev

From: Shannon Nelson <shannon.nelson(a)amd.com> commit 4a54903ff68ddb33b6463c94b4eb37fc584ef760 upstream. Add a check for NULL on the alloc return. If devlink_alloc() fails and we try to use devlink_priv() on the NULL return, the kernel gets very unhappy and panics. With this fix, the driver load will still fail, but at least it won't panic the kernel. Fixes: df69ba43217d ("ionic: Add basic framework for IONIC Network device driver") Signed-off-by: Shannon Nelson <shannon.nelson(a)amd.com> Reviewed-by: Simon Horman <simon.horman(a)corigine.com> Reviewed-by: Jiri Pirko <jiri(a)nvidia.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> [ kovalev: bp to fix CVE-2023-53470 ] Signed-off-by: Vasiliy Kovalev <kovalev(a)altlinux.org> --- drivers/net/ethernet/pensando/ionic/ionic_devlink.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_devlink.c b/drivers/net/ethernet/pensando/ionic/ionic_devlink.c index 3d94064c685d..52ebecf2521e 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_devlink.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_devlink.c @@ -65,6 +65,8 @@ struct ionic *ionic_devlink_alloc(struct device *dev) struct devlink *dl; dl = devlink_alloc(&ionic_dl_ops, sizeof(struct ionic)); + if (!dl) + return NULL; return devlink_priv(dl); } -- 2.50.1

2 weeks, 3 days

1
0
0 0

[PATCH v7 1/8] iommu: Disable SVA when CONFIG_X86 is set

by Lu Baolu

In the IOMMU Shared Virtual Addressing (SVA) context, the IOMMU hardware shares and walks the CPU's page tables. The x86 architecture maps the kernel's virtual address space into the upper portion of every process's page table. Consequently, in an SVA context, the IOMMU hardware can walk and cache kernel page table entries. The Linux kernel currently lacks a notification mechanism for kernel page table changes, specifically when page table pages are freed and reused. The IOMMU driver is only notified of changes to user virtual address mappings. This can cause the IOMMU's internal caches to retain stale entries for kernel VA. Use-After-Free (UAF) and Write-After-Free (WAF) conditions arise when kernel page table pages are freed and later reallocated. The IOMMU could misinterpret the new data as valid page table entries. The IOMMU might then walk into attacker-controlled memory, leading to arbitrary physical memory DMA access or privilege escalation. This is also a Write-After-Free issue, as the IOMMU will potentially continue to write Accessed and Dirty bits to the freed memory while attempting to walk the stale page tables. Currently, SVA contexts are unprivileged and cannot access kernel mappings. However, the IOMMU will still walk kernel-only page tables all the way down to the leaf entries, where it realizes the mapping is for the kernel and errors out. This means the IOMMU still caches these intermediate page table entries, making the described vulnerability a real concern. Disable SVA on x86 architecture until the IOMMU can receive notification to flush the paging cache before freeing the CPU kernel page table pages. Fixes: 26b25a2b98e4 ("iommu: Bind process address spaces to devices") Cc: stable(a)vger.kernel.org Suggested-by: Jason Gunthorpe <jgg(a)nvidia.com> Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> --- drivers/iommu/iommu-sva.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/iommu/iommu-sva.c b/drivers/iommu/iommu-sva.c index 1a51cfd82808..a0442faad952 100644 --- a/drivers/iommu/iommu-sva.c +++ b/drivers/iommu/iommu-sva.c @@ -77,6 +77,9 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm if (!group) return ERR_PTR(-ENODEV); + if (IS_ENABLED(CONFIG_X86)) + return ERR_PTR(-EOPNOTSUPP); + mutex_lock(&iommu_sva_lock); /* Allocate mm->pasid if necessary. */ -- 2.43.0

2 weeks, 3 days

2
1
0 0

[PATCH 5.10.y] firewire: net: fix use after free in fwnet_finish_incoming_packet()

by Vasiliy Kovalev

From: Zhang Shurong <zhang_shurong(a)foxmail.com> commit 3ff256751a2853e1ffaa36958ff933ccc98c6cb5 upstream. The netif_rx() function frees the skb so we can't dereference it to save the skb->len. Signed-off-by: Zhang Shurong <zhang_shurong(a)foxmail.com> Link: https://lore.kernel.org/r/tencent_3B3D24B66ED66A6BB73CC0E63C6A14E45109@qq.c… Signed-off-by: Takashi Sakamoto <o-takashi(a)sakamocchi.jp> [ kovalev: bp to fix CVE-2023-53432 ] Signed-off-by: Vasiliy Kovalev <kovalev(a)altlinux.org> --- drivers/firewire/net.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c index 715e491dfbc3..8504e95ede90 100644 --- a/drivers/firewire/net.c +++ b/drivers/firewire/net.c @@ -489,7 +489,7 @@ static int fwnet_finish_incoming_packet(struct net_device *net, bool is_broadcast, u16 ether_type) { struct fwnet_device *dev; - int status; + int status, len; __be64 guid; switch (ether_type) { @@ -546,13 +546,15 @@ static int fwnet_finish_incoming_packet(struct net_device *net, } skb->protocol = protocol; } + + len = skb->len; status = netif_rx(skb); if (status == NET_RX_DROP) { net->stats.rx_errors++; net->stats.rx_dropped++; } else { net->stats.rx_packets++; - net->stats.rx_bytes += skb->len; + net->stats.rx_bytes += len; } return 0; -- 2.50.1

2 weeks, 3 days

1
0
0 0

+ iommu-disable-sva-when-config_x86-is-set.patch added to mm-new branch

by Andrew Morton

The patch titled Subject: iommu: disable SVA when CONFIG_X86 is set has been added to the -mm mm-new branch. Its filename is iommu-disable-sva-when-config_x86-is-set.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patche… This patch will later appear in the mm-new branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Note, mm-new is a provisional staging ground for work-in-progress patches, and acceptance into mm-new is a notification for others take notice and to finish up reviews. Please do not hesitate to respond to review feedback and post updated versions to replace or incrementally fixup patches in mm-new. Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via the mm-everything branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there every 2-3 working days ------------------------------------------------------ From: Lu Baolu <baolu.lu(a)linux.intel.com> Subject: iommu: disable SVA when CONFIG_X86 is set Date: Wed, 22 Oct 2025 16:26:27 +0800 Patch series "Fix stale IOTLB entries for kernel address space", v7. This proposes a fix for a security vulnerability related to IOMMU Shared Virtual Addressing (SVA). In an SVA context, an IOMMU can cache kernel page table entries. When a kernel page table page is freed and reallocated for another purpose, the IOMMU might still hold stale, incorrect entries. This can be exploited to cause a use-after-free or write-after-free condition, potentially leading to privilege escalation or data corruption. This solution introduces a deferred freeing mechanism for kernel page table pages, which provides a safe window to notify the IOMMU to invalidate its caches before the page is reused. This patch (of 8): In the IOMMU Shared Virtual Addressing (SVA) context, the IOMMU hardware shares and walks the CPU's page tables. The x86 architecture maps the kernel's virtual address space into the upper portion of every process's page table. Consequently, in an SVA context, the IOMMU hardware can walk and cache kernel page table entries. The Linux kernel currently lacks a notification mechanism for kernel page table changes, specifically when page table pages are freed and reused. The IOMMU driver is only notified of changes to user virtual address mappings. This can cause the IOMMU's internal caches to retain stale entries for kernel VA. Use-After-Free (UAF) and Write-After-Free (WAF) conditions arise when kernel page table pages are freed and later reallocated. The IOMMU could misinterpret the new data as valid page table entries. The IOMMU might then walk into attacker-controlled memory, leading to arbitrary physical memory DMA access or privilege escalation. This is also a Write-After-Free issue, as the IOMMU will potentially continue to write Accessed and Dirty bits to the freed memory while attempting to walk the stale page tables. Currently, SVA contexts are unprivileged and cannot access kernel mappings. However, the IOMMU will still walk kernel-only page tables all the way down to the leaf entries, where it realizes the mapping is for the kernel and errors out. This means the IOMMU still caches these intermediate page table entries, making the described vulnerability a real concern. Disable SVA on x86 architecture until the IOMMU can receive notification to flush the paging cache before freeing the CPU kernel page table pages. Link: https://lkml.kernel.org/r/20251022082635.2462433-1-baolu.lu@linux.intel.com Link: https://lkml.kernel.org/r/20251022082635.2462433-2-baolu.lu@linux.intel.com Fixes: 26b25a2b98e4 ("iommu: Bind process address spaces to devices") Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> Suggested-by: Jason Gunthorpe <jgg(a)nvidia.com> Cc: Alistair Popple <apopple(a)nvidia.com> Cc: Andy Lutomirski <luto(a)kernel.org> Cc: Borislav Betkov <bp(a)alien8.de> Cc: Dave Hansen <dave.hansen(a)intel.com> Cc: David Hildenbrand <david(a)redhat.com> Cc: Ingo Molnar <mingo(a)redhat.com> Cc: Jann Horn <jannh(a)google.com> Cc: Jean-Philippe Brucker <jean-philippe(a)linaro.org> Cc: Joerg Roedel <joro(a)8bytes.org> Cc: Kevin Tian <kevin.tian(a)intel.com> Cc: Liam Howlett <liam.howlett(a)oracle.com> Cc: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Cc: Matthew Wilcox (Oracle) <willy(a)infradead.org> Cc: Michal Hocko <mhocko(a)kernel.org> Cc: Mike Rapoport <rppt(a)kernel.org> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Robin Murohy <robin.murphy(a)arm.com> Cc: Thomas Gleinxer <tglx(a)linutronix.de> Cc: "Uladzislau Rezki (Sony)" <urezki(a)gmail.com> Cc: Vasant Hegde <vasant.hegde(a)amd.com> Cc: Vinicius Costa Gomes <vinicius.gomes(a)intel.com> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: Will Deacon <will(a)kernel.org> Cc: Yi Lai <yi1.lai(a)intel.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- drivers/iommu/iommu-sva.c | 3 +++ 1 file changed, 3 insertions(+) --- a/drivers/iommu/iommu-sva.c~iommu-disable-sva-when-config_x86-is-set +++ a/drivers/iommu/iommu-sva.c @@ -77,6 +77,9 @@ struct iommu_sva *iommu_sva_bind_device( if (!group) return ERR_PTR(-ENODEV); + if (IS_ENABLED(CONFIG_X86)) + return ERR_PTR(-EOPNOTSUPP); + mutex_lock(&iommu_sva_lock); /* Allocate mm->pasid if necessary. */ _ Patches currently in -mm which might be from baolu.lu(a)linux.intel.com are iommu-disable-sva-when-config_x86-is-set.patch x86-mm-use-pagetable_free.patch iommu-sva-invalidate-stale-iotlb-entries-for-kernel-address-space.patch

2 weeks, 3 days

1
0
0 0

[PATCH 5.10.y] RDMA/mlx5: Fix mlx5_ib_get_hw_stats when used for device

by Vasiliy Kovalev

From: Shay Drory <shayd(a)nvidia.com> commit 38b50aa44495d5eb4218f0b82fc2da76505cec53 upstream. Currently, when mlx5_ib_get_hw_stats() is used for device (port_num = 0), there is a special handling in order to use the correct counters, but, port_num is being passed down the stack without any change. Also, some functions assume that port_num >=1. As a result, the following oops can occur. BUG: unable to handle page fault for address: ffff89510294f1a8 #PF: supervisor write access in kernel mode #PF: error_code(0x0002) - not-present page PGD 0 P4D 0 Oops: 0002 [#1] SMP CPU: 8 PID: 1382 Comm: devlink Tainted: G W 6.1.0-rc4_for_upstream_base_2022_11_10_16_12 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 RIP: 0010:_raw_spin_lock+0xc/0x20 Call Trace: <TASK> mlx5_ib_get_native_port_mdev+0x73/0xe0 [mlx5_ib] do_get_hw_stats.constprop.0+0x109/0x160 [mlx5_ib] mlx5_ib_get_hw_stats+0xad/0x180 [mlx5_ib] ib_setup_device_attrs+0xf0/0x290 [ib_core] ib_register_device+0x3bb/0x510 [ib_core] ? atomic_notifier_chain_register+0x67/0x80 __mlx5_ib_add+0x2b/0x80 [mlx5_ib] mlx5r_probe+0xb8/0x150 [mlx5_ib] ? auxiliary_match_id+0x6a/0x90 auxiliary_bus_probe+0x3c/0x70 ? driver_sysfs_add+0x6b/0x90 really_probe+0xcd/0x380 __driver_probe_device+0x80/0x170 driver_probe_device+0x1e/0x90 __device_attach_driver+0x7d/0x100 ? driver_allows_async_probing+0x60/0x60 ? driver_allows_async_probing+0x60/0x60 bus_for_each_drv+0x7b/0xc0 __device_attach+0xbc/0x200 bus_probe_device+0x87/0xa0 device_add+0x404/0x940 ? dev_set_name+0x53/0x70 __auxiliary_device_add+0x43/0x60 add_adev+0x99/0xe0 [mlx5_core] mlx5_attach_device+0xc8/0x120 [mlx5_core] mlx5_load_one_devl_locked+0xb2/0xe0 [mlx5_core] devlink_reload+0x133/0x250 devlink_nl_cmd_reload+0x480/0x570 ? devlink_nl_pre_doit+0x44/0x2b0 genl_family_rcv_msg_doit.isra.0+0xc2/0x110 genl_rcv_msg+0x180/0x2b0 ? devlink_nl_cmd_region_read_dumpit+0x540/0x540 ? devlink_reload+0x250/0x250 ? devlink_put+0x50/0x50 ? genl_family_rcv_msg_doit.isra.0+0x110/0x110 netlink_rcv_skb+0x54/0x100 genl_rcv+0x24/0x40 netlink_unicast+0x1f6/0x2c0 netlink_sendmsg+0x237/0x490 sock_sendmsg+0x33/0x40 __sys_sendto+0x103/0x160 ? handle_mm_fault+0x10e/0x290 ? do_user_addr_fault+0x1c0/0x5f0 __x64_sys_sendto+0x25/0x30 do_syscall_64+0x3d/0x90 entry_SYSCALL_64_after_hwframe+0x46/0xb0 Fix it by setting port_num to 1 in order to get device status and remove unused variable. Fixes: aac4492ef23a ("IB/mlx5: Update counter implementation for dual port RoCE") Link: https://lore.kernel.org/r/98b82994c3cd3fa593b8a75ed3f3901e208beb0f.16722317… Signed-off-by: Shay Drory <shayd(a)nvidia.com> Reviewed-by: Patrisious Haddad <phaddad(a)nvidia.com> Signed-off-by: Leon Romanovsky <leon(a)kernel.org> [ kovalev: bp to fix CVE-2023-53393 ] Signed-off-by: Vasiliy Kovalev <kovalev(a)altlinux.org> --- drivers/infiniband/hw/mlx5/counters.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/counters.c b/drivers/infiniband/hw/mlx5/counters.c index 33636268d43d..29b38cf055be 100644 --- a/drivers/infiniband/hw/mlx5/counters.c +++ b/drivers/infiniband/hw/mlx5/counters.c @@ -249,7 +249,6 @@ static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, const struct mlx5_ib_counters *cnts = get_counters(dev, port_num - 1); struct mlx5_core_dev *mdev; int ret, num_counters; - u8 mdev_port_num; if (!stats) return -EINVAL; @@ -270,8 +269,9 @@ static int mlx5_ib_get_hw_stats(struct ib_device *ibdev, } if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) { - mdev = mlx5_ib_get_native_port_mdev(dev, port_num, - &mdev_port_num); + if (!port_num) + port_num = 1; + mdev = mlx5_ib_get_native_port_mdev(dev, port_num, NULL); if (!mdev) { /* If port is not affiliated yet, its in down state * which doesn't have any counters yet, so it would be -- 2.50.1

2 weeks, 3 days

1
0
0 0

[PATCH 2/2] io_uring/sqpoll: be smarter on when to update the stime usage

by Jens Axboe

The current approach is a bit naive, and hence calls the time querying way too often. Only start the "doing work" timer when there's actual work to do, and then use that information to terminate (and account) the work time once done. This greatly reduces the frequency of these calls, when they cannot have changed anyway. Running a basic random reader that is setup to use SQPOLL, a profile before this change shows these as the top cycle consumers: + 32.60% iou-sqp-1074 [kernel.kallsyms] [k] thread_group_cputime_adjusted + 19.97% iou-sqp-1074 [kernel.kallsyms] [k] thread_group_cputime + 12.20% io_uring io_uring [.] submitter_uring_fn + 4.13% iou-sqp-1074 [kernel.kallsyms] [k] getrusage + 2.45% iou-sqp-1074 [kernel.kallsyms] [k] io_submit_sqes + 2.18% iou-sqp-1074 [kernel.kallsyms] [k] __pi_memset_generic + 2.09% iou-sqp-1074 [kernel.kallsyms] [k] cputime_adjust and after this change, top of profile looks as follows: + 36.23% io_uring io_uring [.] submitter_uring_fn + 23.26% iou-sqp-819 [kernel.kallsyms] [k] io_sq_thread + 10.14% iou-sqp-819 [kernel.kallsyms] [k] io_sq_tw + 6.52% iou-sqp-819 [kernel.kallsyms] [k] tctx_task_work_run + 4.82% iou-sqp-819 [kernel.kallsyms] [k] nvme_submit_cmds.part.0 + 2.91% iou-sqp-819 [kernel.kallsyms] [k] io_submit_sqes [...] 0.02% iou-sqp-819 [kernel.kallsyms] [k] cputime_adjust where it's spending the cycles on things that actually matter. Reported-by: Fengnan Chang <changfengnan(a)bytedance.com> Cc: stable(a)vger.kernel.org Fixes: 3fcb9d17206e ("io_uring/sqpoll: statistics of the true utilization of sq threads") Signed-off-by: Jens Axboe <axboe(a)kernel.dk> --- io_uring/sqpoll.c | 43 ++++++++++++++++++++++++++++++++----------- 1 file changed, 32 insertions(+), 11 deletions(-) diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c index 2b816fdb9866..e22f072c7d5f 100644 --- a/io_uring/sqpoll.c +++ b/io_uring/sqpoll.c @@ -170,6 +170,11 @@ static inline bool io_sqd_events_pending(struct io_sq_data *sqd) return READ_ONCE(sqd->state); } +struct io_sq_time { + bool started; + u64 usec; +}; + u64 io_sq_cpu_usec(struct task_struct *tsk) { u64 utime, stime; @@ -179,12 +184,24 @@ u64 io_sq_cpu_usec(struct task_struct *tsk) return stime; } -static void io_sq_update_worktime(struct io_sq_data *sqd, u64 usec) +static void io_sq_update_worktime(struct io_sq_data *sqd, struct io_sq_time *ist) +{ + if (!ist->started) + return; + ist->started = false; + sqd->work_time += io_sq_cpu_usec(current) - ist->usec; +} + +static void io_sq_start_worktime(struct io_sq_time *ist) { - sqd->work_time += io_sq_cpu_usec(current) - usec; + if (ist->started) + return; + ist->started = true; + ist->usec = io_sq_cpu_usec(current); } -static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries) +static int __io_sq_thread(struct io_ring_ctx *ctx, struct io_sq_data *sqd, + bool cap_entries, struct io_sq_time *ist) { unsigned int to_submit; int ret = 0; @@ -197,6 +214,8 @@ static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries) if (to_submit || !wq_list_empty(&ctx->iopoll_list)) { const struct cred *creds = NULL; + io_sq_start_worktime(ist); + if (ctx->sq_creds != current_cred()) creds = override_creds(ctx->sq_creds); @@ -278,7 +297,6 @@ static int io_sq_thread(void *data) unsigned long timeout = 0; char buf[TASK_COMM_LEN] = {}; DEFINE_WAIT(wait); - u64 start; /* offload context creation failed, just exit */ if (!current->io_uring) { @@ -313,6 +331,7 @@ static int io_sq_thread(void *data) mutex_lock(&sqd->lock); while (1) { bool cap_entries, sqt_spin = false; + struct io_sq_time ist = { }; if (io_sqd_events_pending(sqd) || signal_pending(current)) { if (io_sqd_handle_event(sqd)) @@ -321,9 +340,8 @@ static int io_sq_thread(void *data) } cap_entries = !list_is_singular(&sqd->ctx_list); - start = io_sq_cpu_usec(current); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { - int ret = __io_sq_thread(ctx, cap_entries); + int ret = __io_sq_thread(ctx, sqd, cap_entries, &ist); if (!sqt_spin && (ret > 0 || !wq_list_empty(&ctx->iopoll_list))) sqt_spin = true; @@ -331,15 +349,18 @@ static int io_sq_thread(void *data) if (io_sq_tw(&retry_list, IORING_TW_CAP_ENTRIES_VALUE)) sqt_spin = true; - list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) - if (io_napi(ctx)) + list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { + if (io_napi(ctx)) { + io_sq_start_worktime(&ist); io_napi_sqpoll_busy_poll(ctx); + } + } + + io_sq_update_worktime(sqd, &ist); if (sqt_spin || !time_after(jiffies, timeout)) { - if (sqt_spin) { - io_sq_update_worktime(sqd, start); + if (sqt_spin) timeout = jiffies + sqd->sq_thread_idle; - } if (unlikely(need_resched())) { mutex_unlock(&sqd->lock); cond_resched(); -- 2.51.0

2 weeks, 3 days

1
0
0 0

[PATCH 1/2] io_uring/sqpoll: switch away from getrusage() for CPU accounting

by Jens Axboe

getrusage() does a lot more than what the SQPOLL accounting needs, the latter only cares about (and uses) the stime. Rather than do a full RUSAGE_SELF summation, just query the used stime instead. Cc: stable(a)vger.kernel.org Fixes: 3fcb9d17206e ("io_uring/sqpoll: statistics of the true utilization of sq threads") Reviewed-by: Gabriel Krisman Bertazi <krisman(a)suse.de> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> --- io_uring/fdinfo.c | 8 ++++---- io_uring/sqpoll.c | 32 ++++++++++++++++++-------------- io_uring/sqpoll.h | 1 + 3 files changed, 23 insertions(+), 18 deletions(-) diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c index ff3364531c77..294c75a8a3bd 100644 --- a/io_uring/fdinfo.c +++ b/io_uring/fdinfo.c @@ -59,7 +59,6 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) { struct io_overflow_cqe *ocqe; struct io_rings *r = ctx->rings; - struct rusage sq_usage; unsigned int sq_mask = ctx->sq_entries - 1, cq_mask = ctx->cq_entries - 1; unsigned int sq_head = READ_ONCE(r->sq.head); unsigned int sq_tail = READ_ONCE(r->sq.tail); @@ -152,14 +151,15 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) * thread termination. */ if (tsk) { + u64 usec; + get_task_struct(tsk); rcu_read_unlock(); - getrusage(tsk, RUSAGE_SELF, &sq_usage); + usec = io_sq_cpu_usec(tsk); put_task_struct(tsk); sq_pid = sq->task_pid; sq_cpu = sq->sq_cpu; - sq_total_time = (sq_usage.ru_stime.tv_sec * 1000000 - + sq_usage.ru_stime.tv_usec); + sq_total_time = usec; sq_work_time = sq->work_time; } else { rcu_read_unlock(); diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c index a3f11349ce06..2b816fdb9866 100644 --- a/io_uring/sqpoll.c +++ b/io_uring/sqpoll.c @@ -11,6 +11,7 @@ #include <linux/audit.h> #include <linux/security.h> #include <linux/cpuset.h> +#include <linux/sched/cputime.h> #include <linux/io_uring.h> #include <uapi/linux/io_uring.h> @@ -169,6 +170,20 @@ static inline bool io_sqd_events_pending(struct io_sq_data *sqd) return READ_ONCE(sqd->state); } +u64 io_sq_cpu_usec(struct task_struct *tsk) +{ + u64 utime, stime; + + task_cputime_adjusted(tsk, &utime, &stime); + do_div(stime, 1000); + return stime; +} + +static void io_sq_update_worktime(struct io_sq_data *sqd, u64 usec) +{ + sqd->work_time += io_sq_cpu_usec(current) - usec; +} + static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries) { unsigned int to_submit; @@ -255,26 +270,15 @@ static bool io_sq_tw_pending(struct llist_node *retry_list) return retry_list || !llist_empty(&tctx->task_list); } -static void io_sq_update_worktime(struct io_sq_data *sqd, struct rusage *start) -{ - struct rusage end; - - getrusage(current, RUSAGE_SELF, &end); - end.ru_stime.tv_sec -= start->ru_stime.tv_sec; - end.ru_stime.tv_usec -= start->ru_stime.tv_usec; - - sqd->work_time += end.ru_stime.tv_usec + end.ru_stime.tv_sec * 1000000; -} - static int io_sq_thread(void *data) { struct llist_node *retry_list = NULL; struct io_sq_data *sqd = data; struct io_ring_ctx *ctx; - struct rusage start; unsigned long timeout = 0; char buf[TASK_COMM_LEN] = {}; DEFINE_WAIT(wait); + u64 start; /* offload context creation failed, just exit */ if (!current->io_uring) { @@ -317,7 +321,7 @@ static int io_sq_thread(void *data) } cap_entries = !list_is_singular(&sqd->ctx_list); - getrusage(current, RUSAGE_SELF, &start); + start = io_sq_cpu_usec(current); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { int ret = __io_sq_thread(ctx, cap_entries); @@ -333,7 +337,7 @@ static int io_sq_thread(void *data) if (sqt_spin || !time_after(jiffies, timeout)) { if (sqt_spin) { - io_sq_update_worktime(sqd, &start); + io_sq_update_worktime(sqd, start); timeout = jiffies + sqd->sq_thread_idle; } if (unlikely(need_resched())) { diff --git a/io_uring/sqpoll.h b/io_uring/sqpoll.h index b83dcdec9765..fd2f6f29b516 100644 --- a/io_uring/sqpoll.h +++ b/io_uring/sqpoll.h @@ -29,6 +29,7 @@ void io_sq_thread_unpark(struct io_sq_data *sqd); void io_put_sq_data(struct io_sq_data *sqd); void io_sqpoll_wait_sq(struct io_ring_ctx *ctx); int io_sqpoll_wq_cpu_affinity(struct io_ring_ctx *ctx, cpumask_var_t mask); +u64 io_sq_cpu_usec(struct task_struct *tsk); static inline struct task_struct *sqpoll_task_locked(struct io_sq_data *sqd) { -- 2.51.0

2 weeks, 3 days

1
0
0 0

[PATCH 1/2] io_uring/sqpoll: switch away from getrusage() for CPU accounting

by Jens Axboe

getrusage() does a lot more than what the SQPOLL accounting needs, the latter only cares about (and uses) the stime. Rather than do a full RUSAGE_SELF summation, just query the used stime instead. Cc: stable(a)vger.kernel.org Fixes: 3fcb9d17206e ("io_uring/sqpoll: statistics of the true utilization of sq threads") Signed-off-by: Jens Axboe <axboe(a)kernel.dk> --- io_uring/fdinfo.c | 9 +++++---- io_uring/sqpoll.c | 34 ++++++++++++++++++++-------------- io_uring/sqpoll.h | 1 + 3 files changed, 26 insertions(+), 18 deletions(-) diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c index ff3364531c77..966e06b078f6 100644 --- a/io_uring/fdinfo.c +++ b/io_uring/fdinfo.c @@ -59,7 +59,6 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) { struct io_overflow_cqe *ocqe; struct io_rings *r = ctx->rings; - struct rusage sq_usage; unsigned int sq_mask = ctx->sq_entries - 1, cq_mask = ctx->cq_entries - 1; unsigned int sq_head = READ_ONCE(r->sq.head); unsigned int sq_tail = READ_ONCE(r->sq.tail); @@ -152,14 +151,16 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) * thread termination. */ if (tsk) { + struct timespec64 ts; + get_task_struct(tsk); rcu_read_unlock(); - getrusage(tsk, RUSAGE_SELF, &sq_usage); + ts = io_sq_cpu_time(tsk); put_task_struct(tsk); sq_pid = sq->task_pid; sq_cpu = sq->sq_cpu; - sq_total_time = (sq_usage.ru_stime.tv_sec * 1000000 - + sq_usage.ru_stime.tv_usec); + sq_total_time = (ts.tv_sec * 1000000 + + ts.tv_nsec / 1000); sq_work_time = sq->work_time; } else { rcu_read_unlock(); diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c index a3f11349ce06..8705b0aa82e0 100644 --- a/io_uring/sqpoll.c +++ b/io_uring/sqpoll.c @@ -11,6 +11,7 @@ #include <linux/audit.h> #include <linux/security.h> #include <linux/cpuset.h> +#include <linux/sched/cputime.h> #include <linux/io_uring.h> #include <uapi/linux/io_uring.h> @@ -169,6 +170,22 @@ static inline bool io_sqd_events_pending(struct io_sq_data *sqd) return READ_ONCE(sqd->state); } +struct timespec64 io_sq_cpu_time(struct task_struct *tsk) +{ + u64 utime, stime; + + task_cputime_adjusted(tsk, &utime, &stime); + return ns_to_timespec64(stime); +} + +static void io_sq_update_worktime(struct io_sq_data *sqd, struct timespec64 start) +{ + struct timespec64 ts; + + ts = timespec64_sub(io_sq_cpu_time(current), start); + sqd->work_time += ts.tv_sec * 1000000 + ts.tv_nsec / 1000; +} + static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries) { unsigned int to_submit; @@ -255,23 +272,12 @@ static bool io_sq_tw_pending(struct llist_node *retry_list) return retry_list || !llist_empty(&tctx->task_list); } -static void io_sq_update_worktime(struct io_sq_data *sqd, struct rusage *start) -{ - struct rusage end; - - getrusage(current, RUSAGE_SELF, &end); - end.ru_stime.tv_sec -= start->ru_stime.tv_sec; - end.ru_stime.tv_usec -= start->ru_stime.tv_usec; - - sqd->work_time += end.ru_stime.tv_usec + end.ru_stime.tv_sec * 1000000; -} - static int io_sq_thread(void *data) { struct llist_node *retry_list = NULL; struct io_sq_data *sqd = data; struct io_ring_ctx *ctx; - struct rusage start; + struct timespec64 start; unsigned long timeout = 0; char buf[TASK_COMM_LEN] = {}; DEFINE_WAIT(wait); @@ -317,7 +323,7 @@ static int io_sq_thread(void *data) } cap_entries = !list_is_singular(&sqd->ctx_list); - getrusage(current, RUSAGE_SELF, &start); + start = io_sq_cpu_time(current); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { int ret = __io_sq_thread(ctx, cap_entries); @@ -333,7 +339,7 @@ static int io_sq_thread(void *data) if (sqt_spin || !time_after(jiffies, timeout)) { if (sqt_spin) { - io_sq_update_worktime(sqd, &start); + io_sq_update_worktime(sqd, start); timeout = jiffies + sqd->sq_thread_idle; } if (unlikely(need_resched())) { diff --git a/io_uring/sqpoll.h b/io_uring/sqpoll.h index b83dcdec9765..84ed2b312e88 100644 --- a/io_uring/sqpoll.h +++ b/io_uring/sqpoll.h @@ -29,6 +29,7 @@ void io_sq_thread_unpark(struct io_sq_data *sqd); void io_put_sq_data(struct io_sq_data *sqd); void io_sqpoll_wait_sq(struct io_ring_ctx *ctx); int io_sqpoll_wq_cpu_affinity(struct io_ring_ctx *ctx, cpumask_var_t mask); +struct timespec64 io_sq_cpu_time(struct task_struct *tsk); static inline struct task_struct *sqpoll_task_locked(struct io_sq_data *sqd) { -- 2.51.0

2 weeks, 3 days

2
2
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror October 2025