- Linux-stable-mirror - lists.linaro.org

[PATCH 1/3] block: Remove queue freezing from several sysfs store callbacks

by Bart Van Assche

Freezing the request queue from inside sysfs store callbacks may cause a deadlock in combination with the dm-multipath driver and the queue_if_no_path option. Additionally, freezing the request queue slows down system boot on systems where sysfs attributes are set synchronously. Fix this by removing the blk_mq_freeze_queue() / blk_mq_unfreeze_queue() calls from the store callbacks that do not strictly need these callbacks. This patch may cause a small delay in applying the new settings. This patch affects the following sysfs attributes: * io_poll_delay * io_timeout * nomerges * read_ahead_kb * rq_affinity Here is an example of a deadlock triggered by running test srp/002: task:multipathd Call Trace: <TASK> __schedule+0x8c1/0x1bf0 schedule+0xdd/0x270 schedule_preempt_disabled+0x1c/0x30 __mutex_lock+0xb89/0x1650 mutex_lock_nested+0x1f/0x30 dm_table_set_restrictions+0x823/0xdf0 __bind+0x166/0x590 dm_swap_table+0x2a7/0x490 do_resume+0x1b1/0x610 dev_suspend+0x55/0x1a0 ctl_ioctl+0x3a5/0x7e0 dm_ctl_ioctl+0x12/0x20 __x64_sys_ioctl+0x127/0x1a0 x64_sys_call+0xe2b/0x17d0 do_syscall_64+0x96/0x3a0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 </TASK> task:(udev-worker) Call Trace: <TASK> __schedule+0x8c1/0x1bf0 schedule+0xdd/0x270 blk_mq_freeze_queue_wait+0xf2/0x140 blk_mq_freeze_queue_nomemsave+0x23/0x30 queue_ra_store+0x14e/0x290 queue_attr_store+0x23e/0x2c0 sysfs_kf_write+0xde/0x140 kernfs_fop_write_iter+0x3b2/0x630 vfs_write+0x4fd/0x1390 ksys_write+0xfd/0x230 __x64_sys_write+0x76/0xc0 x64_sys_call+0x276/0x17d0 do_syscall_64+0x96/0x3a0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 </TASK> Cc: Christoph Hellwig <hch(a)lst.de> Cc: Nilay Shroff <nilay(a)linux.ibm.com> Cc: stable(a)vger.kernel.org Fixes: af2814149883 ("block: freeze the queue in queue_attr_store") Signed-off-by: Bart Van Assche <bvanassche(a)acm.org> --- block/blk-sysfs.c | 16 +--------------- 1 file changed, 1 insertion(+), 15 deletions(-) diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index b2b9b89d6967..ab34fe62f4da 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -105,7 +105,6 @@ queue_ra_store(struct gendisk *disk, const char *page, size_t count) { unsigned long ra_kb; ssize_t ret; - unsigned int memflags; struct request_queue *q = disk->queue; ret = queue_var_store(&ra_kb, page, count); @@ -116,10 +115,8 @@ queue_ra_store(struct gendisk *disk, const char *page, size_t count) * calculated from the queue limits by queue_limits_commit_update. */ mutex_lock(&q->limits_lock); - memflags = blk_mq_freeze_queue(q); disk->bdi->ra_pages = ra_kb >> (PAGE_SHIFT - 10); mutex_unlock(&q->limits_lock); - blk_mq_unfreeze_queue(q, memflags); return ret; } @@ -317,21 +314,18 @@ static ssize_t queue_nomerges_store(struct gendisk *disk, const char *page, size_t count) { unsigned long nm; - unsigned int memflags; struct request_queue *q = disk->queue; ssize_t ret = queue_var_store(&nm, page, count); if (ret < 0) return ret; - memflags = blk_mq_freeze_queue(q); blk_queue_flag_clear(QUEUE_FLAG_NOMERGES, q); blk_queue_flag_clear(QUEUE_FLAG_NOXMERGES, q); if (nm == 2) blk_queue_flag_set(QUEUE_FLAG_NOMERGES, q); else if (nm) blk_queue_flag_set(QUEUE_FLAG_NOXMERGES, q); - blk_mq_unfreeze_queue(q, memflags); return ret; } @@ -351,7 +345,6 @@ queue_rq_affinity_store(struct gendisk *disk, const char *page, size_t count) #ifdef CONFIG_SMP struct request_queue *q = disk->queue; unsigned long val; - unsigned int memflags; ret = queue_var_store(&val, page, count); if (ret < 0) @@ -363,7 +356,6 @@ queue_rq_affinity_store(struct gendisk *disk, const char *page, size_t count) * are accessed individually using atomic test_bit operation. So we * don't grab any lock while updating these flags. */ - memflags = blk_mq_freeze_queue(q); if (val == 2) { blk_queue_flag_set(QUEUE_FLAG_SAME_COMP, q); blk_queue_flag_set(QUEUE_FLAG_SAME_FORCE, q); @@ -374,7 +366,6 @@ queue_rq_affinity_store(struct gendisk *disk, const char *page, size_t count) blk_queue_flag_clear(QUEUE_FLAG_SAME_COMP, q); blk_queue_flag_clear(QUEUE_FLAG_SAME_FORCE, q); } - blk_mq_unfreeze_queue(q, memflags); #endif return ret; } @@ -388,11 +379,9 @@ static ssize_t queue_poll_delay_store(struct gendisk *disk, const char *page, static ssize_t queue_poll_store(struct gendisk *disk, const char *page, size_t count) { - unsigned int memflags; ssize_t ret = count; struct request_queue *q = disk->queue; - memflags = blk_mq_freeze_queue(q); if (!(q->limits.features & BLK_FEAT_POLL)) { ret = -EINVAL; goto out; @@ -401,7 +390,6 @@ static ssize_t queue_poll_store(struct gendisk *disk, const char *page, pr_info_ratelimited("writes to the poll attribute are ignored.\n"); pr_info_ratelimited("please use driver specific parameters instead.\n"); out: - blk_mq_unfreeze_queue(q, memflags); return ret; } @@ -414,7 +402,7 @@ static ssize_t queue_io_timeout_show(struct gendisk *disk, char *page) static ssize_t queue_io_timeout_store(struct gendisk *disk, const char *page, size_t count) { - unsigned int val, memflags; + unsigned int val; int err; struct request_queue *q = disk->queue; @@ -422,9 +410,7 @@ static ssize_t queue_io_timeout_store(struct gendisk *disk, const char *page, if (err || val == 0) return -EINVAL; - memflags = blk_mq_freeze_queue(q); blk_queue_rq_timeout(q, msecs_to_jiffies(val)); - blk_mq_unfreeze_queue(q, memflags); return count; }

4 months

3
2
0 0

[PATCH v2 0/4] HID: avoid setting up battery timer when not needed in hid-apple and magicmouse

by Aditya Garg

Both hid-apple and hid-magicmouse require set up a battery timer for certain devices in order to fetch battery status. However, the timer is being set unconditionally for all devices. This patch series introduces checks to ensure that the battery timer is only set up for devices that actually require it. v2: - Address the cases of out_err and err_stop_hw left in v1 - Create a function to check if the device is a USB Magic Mouse 2 or Magic Trackpad 2 to reduce code duplication. - Add 2 new patches that convert the battery timeout to use secs_to_jiffies() instead of msecs_to_jiffies(). Aditya Garg (4): HID: apple: avoid setting up battery timer for devices without battery HID: magicmouse: avoid setting up battery timer when not needed HID: apple: use secs_to_jiffies() for battery timeout HID: magicmouse: use secs_to_jiffies() for battery timeout drivers/hid/hid-apple.c | 21 +++++++----- drivers/hid/hid-magicmouse.c | 66 ++++++++++++++++++++++-------------- 2 files changed, 54 insertions(+), 33 deletions(-) Range-diff against v1: 1: 05b6ac964 ! 1: 41c49e1d6 HID: apple: avoid setting up battery timer for devices without battery @@ drivers/hid/hid-apple.c: static int apple_probe(struct hid_device *hdev, if (quirks & APPLE_BACKLIGHT_CTL) apple_backlight_init(hdev); +@@ drivers/hid/hid-apple.c: static int apple_probe(struct hid_device *hdev, + return 0; + + out_err: +- timer_delete_sync(&asc->battery_timer); ++ if (quirks & APPLE_RDESC_BATTERY) ++ timer_delete_sync(&asc->battery_timer); ++ + hid_hw_stop(hdev); + return ret; + } @@ drivers/hid/hid-apple.c: static void apple_remove(struct hid_device *hdev) { struct apple_sc *asc = hid_get_drvdata(hdev); 2: 25b52facf ! 2: a1617042f HID: magicmouse: avoid setting up battery timer when not needed @@ Commit message Signed-off-by: Aditya Garg <gargaditya08(a)live.com> ## drivers/hid/hid-magicmouse.c ## +@@ drivers/hid/hid-magicmouse.c: static void magicmouse_enable_mt_work(struct work_struct *work) + hid_err(msc->hdev, "unable to request touch data (%d)\n", ret); + } + ++static bool is_usb_magicmouse2(__u32 vendor, __u32 product) ++{ ++ if (vendor != USB_VENDOR_ID_APPLE) ++ return false; ++ return product == USB_DEVICE_ID_APPLE_MAGICMOUSE2 || ++ product == USB_DEVICE_ID_APPLE_MAGICMOUSE2_USBC; ++} ++ ++static bool is_usb_magictrackpad2(__u32 vendor, __u32 product) ++{ ++ if (vendor != USB_VENDOR_ID_APPLE) ++ return false; ++ return product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2 || ++ product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2_USBC; ++} ++ + static int magicmouse_fetch_battery(struct hid_device *hdev) + { + #ifdef CONFIG_HID_BATTERY_STRENGTH + struct hid_report_enum *report_enum; + struct hid_report *report; + +- if (!hdev->battery || hdev->vendor != USB_VENDOR_ID_APPLE || +- (hdev->product != USB_DEVICE_ID_APPLE_MAGICMOUSE2 && +- hdev->product != USB_DEVICE_ID_APPLE_MAGICMOUSE2_USBC && +- hdev->product != USB_DEVICE_ID_APPLE_MAGICTRACKPAD2 && +- hdev->product != USB_DEVICE_ID_APPLE_MAGICTRACKPAD2_USBC)) ++ if (!hdev->battery || ++ (!is_usb_magicmouse2(hdev->vendor, hdev->product) && ++ !is_usb_magictrackpad2(hdev->vendor, hdev->product))) + return -1; + + report_enum = &hdev->report_enum[hdev->battery_report_type]; @@ drivers/hid/hid-magicmouse.c: static int magicmouse_probe(struct hid_device *hdev, return ret; } @@ drivers/hid/hid-magicmouse.c: static int magicmouse_probe(struct hid_device *hde - ((id->product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2 || - id->product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2_USBC) && - hdev->type != HID_TYPE_USBMOUSE))) -- return 0; -+ if (id->vendor == USB_VENDOR_ID_APPLE) { -+ bool is_mouse2 = (id->product == USB_DEVICE_ID_APPLE_MAGICMOUSE2 || -+ id->product == USB_DEVICE_ID_APPLE_MAGICMOUSE2_USBC); -+ bool is_trackpad2 = (id->product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2 || -+ id->product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2_USBC); -+ -+ if (is_mouse2 || is_trackpad2) { -+ timer_setup(&msc->battery_timer, magicmouse_battery_timer_tick, 0); -+ mod_timer(&msc->battery_timer, -+ jiffies + msecs_to_jiffies(USB_BATTERY_TIMEOUT_MS)); -+ magicmouse_fetch_battery(hdev); -+ } -+ -+ if (is_mouse2 || (is_trackpad2 && hdev->type != HID_TYPE_USBMOUSE)) -+ return 0; ++ if (is_usb_magicmouse2(id->vendor, id->product) || ++ is_usb_magictrackpad2(id->vendor, id->product)) { ++ timer_setup(&msc->battery_timer, magicmouse_battery_timer_tick, 0); ++ mod_timer(&msc->battery_timer, ++ jiffies + msecs_to_jiffies(USB_BATTERY_TIMEOUT_MS)); ++ magicmouse_fetch_battery(hdev); + } ++ ++ if (is_usb_magicmouse2(id->vendor, id->product) || ++ (is_usb_magictrackpad2(id->vendor, id->product) && ++ hdev->type != HID_TYPE_USBMOUSE)) + return 0; if (!msc->input) { - hid_err(hdev, "magicmouse input not registered\n"); +@@ drivers/hid/hid-magicmouse.c: static int magicmouse_probe(struct hid_device *hdev, + + return 0; + err_stop_hw: +- timer_delete_sync(&msc->battery_timer); ++ if (is_usb_magicmouse2(id->vendor, id->product) || ++ is_usb_magictrackpad2(id->vendor, id->product)) ++ timer_delete_sync(&msc->battery_timer); ++ + hid_hw_stop(hdev); + return ret; + } @@ drivers/hid/hid-magicmouse.c: static void magicmouse_remove(struct hid_device *hdev) if (msc) { cancel_delayed_work_sync(&msc->work); - timer_delete_sync(&msc->battery_timer); -+ if (hdev->vendor == USB_VENDOR_ID_APPLE && -+ (hdev->product == USB_DEVICE_ID_APPLE_MAGICMOUSE2 || -+ hdev->product == USB_DEVICE_ID_APPLE_MAGICMOUSE2_USBC || -+ hdev->product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2 || -+ hdev->product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2_USBC)) -+ ++ if (is_usb_magicmouse2(hdev->vendor, hdev->product) || ++ is_usb_magictrackpad2(hdev->vendor, hdev->product)) + timer_delete_sync(&msc->battery_timer); } hid_hw_stop(hdev); +@@ drivers/hid/hid-magicmouse.c: static const __u8 *magicmouse_report_fixup(struct hid_device *hdev, __u8 *rdesc, + * 0x05, 0x01, // Usage Page (Generic Desktop) 0 + * 0x09, 0x02, // Usage (Mouse) 2 + */ +- if (hdev->vendor == USB_VENDOR_ID_APPLE && +- (hdev->product == USB_DEVICE_ID_APPLE_MAGICMOUSE2 || +- hdev->product == USB_DEVICE_ID_APPLE_MAGICMOUSE2_USBC || +- hdev->product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2 || +- hdev->product == USB_DEVICE_ID_APPLE_MAGICTRACKPAD2_USBC) && ++ if ((is_usb_magicmouse2(hdev->vendor, hdev->product) || ++ is_usb_magictrackpad2(hdev->vendor, hdev->product)) && + *rsize == 83 && rdesc[46] == 0x84 && rdesc[58] == 0x85) { + hid_info(hdev, + "fixing up magicmouse battery report descriptor\n"); -: --------- > 3: 3ebe25998 HID: apple: use secs_to_jiffies() for battery timeout -: --------- > 4: 1d3400714 HID: magicmouse: use secs_to_jiffies() for battery timeout -- 2.49.0

4 months

3
7
0 0

Re: [PATCH v2 1/3] drm/amdgpu: Dirty cleared blocks on free

by Christian König

On 02.07.25 13:58, Arunpravin Paneer Selvam wrote: > Hi Christian, > > On 7/2/2025 1:27 PM, Christian König wrote: >> On 01.07.25 21:08, Arunpravin Paneer Selvam wrote: >>> Set the dirty bit when the memory resource is not cleared >>> during BO release. >>> >>> v2(Christian): >>> - Drop the cleared flag set to false. >>> - Improve the amdgpu_vram_mgr_set_clear_state() function. >>> >>> Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam(a)amd.com> >>> Suggested-by: Christian König <christian.koenig(a)amd.com> >>> Cc: stable(a)vger.kernel.org >>> Fixes: a68c7eaa7a8f ("drm/amdgpu: Enable clear page functionality") >>> --- >>> drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 1 - >>> drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h | 5 ++++- >>> 2 files changed, 4 insertions(+), 2 deletions(-) >>> >>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c >>> index 9c5df35f05b7..86eb6d47dcc5 100644 >>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c >>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c >>> @@ -409,7 +409,6 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, >>> if (r) { >>> goto error; >>> } else if (wipe_fence) { >>> - amdgpu_vram_mgr_set_cleared(bo->resource); >> Mhm, that looks incorrect to me. >> >> Why don't we consider the resource cleared after it go wiped during eviction? > > Modifying the resource flag here doesn't go into effect until we call the drm_buddy_free_list() in amdgpu_vram_mgr_del(). This BO will be cleared once again after executing amdgpu_bo_release_notify(). With the new implementation, there's a chance that changing the resource flag the second time would cause the WARN_ON to occur. Hence I removed the resource cleared function call in amdgpu_move_blit. Thanks, Arun. Something fishy is going on that we don't fully understand. What happens here is that we move from VRAM to GTT, clear the VRAM BO after the move and set the flag. When the BO is destroyed the it is backed by GTT and not VRAM any more, so no clear operation and no flag setting is performed. It looks more like we forget to clear the flag in some cases. Regards, Christian. > >> Regards, >> Christian. >> >>> dma_fence_put(fence); >>> fence = wipe_fence; >>> } >>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h >>> index b256cbc2bc27..2c88d5fd87da 100644 >>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h >>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h >>> @@ -66,7 +66,10 @@ to_amdgpu_vram_mgr_resource(struct ttm_resource *res) >>> >>> static inline void amdgpu_vram_mgr_set_cleared(struct ttm_resource *res) >>> { >>> - to_amdgpu_vram_mgr_resource(res)->flags |= DRM_BUDDY_CLEARED; >>> + struct amdgpu_vram_mgr_resource *ares = to_amdgpu_vram_mgr_resource(res); >>> + >>> + WARN_ON(ares->flags & DRM_BUDDY_CLEARED); >>> + ares->flags |= DRM_BUDDY_CLEARED; >>> } >>> >>> #endif

4 months

2
1
0 0

Please cherry-pick this commit to v5.15.y, 5.10.y and 5.4.y: 23e118a48acf ("PCI: hv: Do not set PCI_COMMAND_MEMORY to reduce VM boot time")

by Dexuan Cui

Hi, The commit has been in v6.1.y for 3+ years (but not in the 5.x stable branches): 23e118a48acf ("PCI: hv: Do not set PCI_COMMAND_MEMORY to reduce VM boot time") The commit can be cherry-picked cleanly to the latest 5.x stable branches, i.e. 5.15.185, 5.10.238 and 5.4.294. Some Linux distro kernels are still based on the 5.x stable branches, and they have been carrying the commit explicitly. It would be great to have the commit in the upstream 5.x branches, so they don't have to carry the commit explicitly. There is a Debian 10.3 user who wants the commit, but the commit is absent from Debian 10.3's kernel, whose full version string is: Linux version 5.10.0-0.deb10.30-amd64 (debian-kernel(a)lists.debian.org) (gcc-8 (Debian 8.3.0-6) 8.3.0, GNU ld (GNU Binutils for Debian) 2.31.1) #1 SMP Debian 5.10.218-1~deb10u1 (2024-06-12) If the commit is cherry-picked into the upstream 5.10.y, the next Debian 10.3 kernel update would have the commit automatically. Thanks! -- Dexuan

4 months

2
4
0 0

[PATCH v3 1/2] platform/x86: int3472: add hpd pin support

by Dongcheng Yan

Typically HDMI to MIPI CSI-2 bridges have a pin to signal image data is being received. On the host side this is wired to a GPIO for polling or interrupts. This includes the Lontium HDMI to MIPI CSI-2 bridges lt6911uxe and lt6911uxc. The GPIO "hpd" is used already by other HDMI to CSI-2 bridges, use it here as well. Signed-off-by: Dongcheng Yan <dongcheng.yan(a)intel.com> --- drivers/platform/x86/intel/int3472/common.h | 1 + drivers/platform/x86/intel/int3472/discrete.c | 6 ++++++ 2 files changed, 7 insertions(+) diff --git a/drivers/platform/x86/intel/int3472/common.h b/drivers/platform/x86/intel/int3472/common.h index 51b818e62a25..4593d567caf4 100644 --- a/drivers/platform/x86/intel/int3472/common.h +++ b/drivers/platform/x86/intel/int3472/common.h @@ -23,6 +23,7 @@ #define INT3472_GPIO_TYPE_CLK_ENABLE 0x0c #define INT3472_GPIO_TYPE_PRIVACY_LED 0x0d #define INT3472_GPIO_TYPE_HANDSHAKE 0x12 +#define INT3472_GPIO_TYPE_HOTPLUG_DETECT 0x13 #define INT3472_PDEV_MAX_NAME_LEN 23 #define INT3472_MAX_SENSOR_GPIOS 3 diff --git a/drivers/platform/x86/intel/int3472/discrete.c b/drivers/platform/x86/intel/int3472/discrete.c index 394975f55d64..efa3bc7af193 100644 --- a/drivers/platform/x86/intel/int3472/discrete.c +++ b/drivers/platform/x86/intel/int3472/discrete.c @@ -191,6 +191,10 @@ static void int3472_get_con_id_and_polarity(struct int3472_discrete_device *int3 *con_id = "privacy-led"; *gpio_flags = GPIO_ACTIVE_HIGH; break; + case INT3472_GPIO_TYPE_HOTPLUG_DETECT: + *con_id = "hpd"; + *gpio_flags = GPIO_ACTIVE_HIGH; + break; case INT3472_GPIO_TYPE_POWER_ENABLE: *con_id = "avdd"; *gpio_flags = GPIO_ACTIVE_HIGH; @@ -221,6 +225,7 @@ static void int3472_get_con_id_and_polarity(struct int3472_discrete_device *int3 * 0x0b Power enable * 0x0c Clock enable * 0x0d Privacy LED + * 0x13 Hotplug detect * * There are some known platform specific quirks where that does not quite * hold up; for example where a pin with type 0x01 (Power down) is mapped to @@ -290,6 +295,7 @@ static int skl_int3472_handle_gpio_resources(struct acpi_resource *ares, switch (type) { case INT3472_GPIO_TYPE_RESET: case INT3472_GPIO_TYPE_POWERDOWN: + case INT3472_GPIO_TYPE_HOTPLUG_DETECT: ret = skl_int3472_map_gpio_to_sensor(int3472, agpio, con_id, gpio_flags); if (ret) err_msg = "Failed to map GPIO pin to sensor\n"; base-commit: 4d1e8c8f11c611db5828e4bae7292bc295eea8ef -- 2.34.1

4 months

8
12
0 0

[REGRESSION][BISECTED] Performance Regression in IOMMU/VT-d Since Kernel 6.10

by Ioanna Alifieraki

#regzbot introduced: 129dab6e1286 Hello everyone, We've identified a performance regression that starts with linux kernel 6.10 and persists through 6.16(tested at commit e540341508ce). Bisection pointed to commit: 129dab6e1286 ("iommu/vt-d: Use cache_tag_flush_range_np() in iotlb_sync_map"). The issue occurs when running fio against two NVMe devices located under the same PCIe bridge (dual-port NVMe configuration). Performance drops compared to configurations where the devices are on different bridges. Observed Performance: - Before the commit: ~6150 MiB/s, regardless of NVMe device placement. - After the commit: -- Same PCIe bridge: ~4985 MiB/s -- Different PCIe bridges: ~6150 MiB/s Currently we can only reproduce the issue on a Z3 metal instance on gcp. I suspect the issue can be reproducible if you have a dual port nvme on any machine. At [1] there's a more detailed description of the issue and details on the reproducer. Could you please advise on the appropriate path forward to mitigate or address this regression? Thanks, Jo [1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2115738

4 months

2
4
0 0

[PATCH net] net: libwx: fix double put of page to page_pool

by Jiawen Wu

wx_dma_sync_frag() incorrectly attempted to return the page to page_pool, which is already handled via buffer reuse or wx_build_skb() paths. Under high MTU and high throughput, this causes list corruption inside page_pool due to double free. [ 876.949950] Call Trace: [ 876.949951] <IRQ> [ 876.949952] __rmqueue_pcplist+0x53/0x2c0 [ 876.949955] alloc_pages_bulk_noprof+0x2e0/0x660 [ 876.949958] __page_pool_alloc_pages_slow+0xa9/0x400 [ 876.949961] page_pool_alloc_pages+0xa/0x20 [ 876.949963] wx_alloc_rx_buffers+0xd7/0x110 [libwx] [ 876.949967] wx_clean_rx_irq+0x262/0x430 [libwx] [ 876.949971] wx_poll+0x92/0x130 [libwx] [ 876.949975] __napi_poll+0x28/0x190 [ 876.949977] net_rx_action+0x301/0x3f0 [ 876.949980] ? srso_alias_return_thunk+0x5/0xfbef5 [ 876.949981] ? profile_tick+0x30/0x70 [ 876.949983] ? srso_alias_return_thunk+0x5/0xfbef5 [ 876.949984] ? srso_alias_return_thunk+0x5/0xfbef5 [ 876.949986] ? timerqueue_add+0xa3/0xc0 [ 876.949988] ? srso_alias_return_thunk+0x5/0xfbef5 [ 876.949989] ? __raise_softirq_irqoff+0x16/0x70 [ 876.949991] ? srso_alias_return_thunk+0x5/0xfbef5 [ 876.949993] ? srso_alias_return_thunk+0x5/0xfbef5 [ 876.949994] ? wx_msix_clean_rings+0x41/0x50 [libwx] [ 876.949998] handle_softirqs+0xf9/0x2c0 Fixes: 3c47e8ae113a ("net: libwx: Support to receive packets in NAPI") Cc: stable(a)vger.kernel.org Signed-off-by: Jiawen Wu <jiawenwu(a)trustnetic.com> --- drivers/net/ethernet/wangxun/libwx/wx_lib.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c index c57cc4f27249..246b18a0285e 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c @@ -174,10 +174,6 @@ static void wx_dma_sync_frag(struct wx_ring *rx_ring, skb_frag_off(frag), skb_frag_size(frag), DMA_FROM_DEVICE); - - /* If the page was released, just unmap it. */ - if (unlikely(WX_CB(skb)->page_released)) - page_pool_put_full_page(rx_ring->page_pool, rx_buffer->page, false); } static struct wx_rx_buffer *wx_get_rx_buffer(struct wx_ring *rx_ring, -- 2.48.1

4 months

2
2
0 0

[PATCH] drm/v3d: Disable interrupts before resetting the GPU

by Maíra Canal

Currently, an interrupt can be triggered during a GPU reset, which can lead to GPU hangs and NULL pointer dereference in an interrupt context as shown in the following trace: [ 314.035040] Unable to handle kernel NULL pointer dereference at virtual address 00000000000000c0 [ 314.043822] Mem abort info: [ 314.046606] ESR = 0x0000000096000005 [ 314.050347] EC = 0x25: DABT (current EL), IL = 32 bits [ 314.055651] SET = 0, FnV = 0 [ 314.058695] EA = 0, S1PTW = 0 [ 314.061826] FSC = 0x05: level 1 translation fault [ 314.066694] Data abort info: [ 314.069564] ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000 [ 314.075039] CM = 0, WnR = 0, TnD = 0, TagAccess = 0 [ 314.080080] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0 [ 314.085382] user pgtable: 4k pages, 39-bit VAs, pgdp=0000000102728000 [ 314.091814] [00000000000000c0] pgd=0000000000000000, p4d=0000000000000000, pud=0000000000000000 [ 314.100511] Internal error: Oops: 0000000096000005 [#1] PREEMPT SMP [ 314.106770] Modules linked in: v3d i2c_brcmstb vc4 snd_soc_hdmi_codec gpu_sched drm_shmem_helper drm_display_helper cec drm_dma_helper drm_kms_helper drm drm_panel_orientation_quirks snd_soc_core snd_compress snd_pcm_dmaengine snd_pcm snd_timer snd backlight [ 314.129654] CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 6.12.25+rpt-rpi-v8 #1 Debian 1:6.12.25-1+rpt1 [ 314.139388] Hardware name: Raspberry Pi 4 Model B Rev 1.4 (DT) [ 314.145211] pstate: 600000c5 (nZCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 314.152165] pc : v3d_irq+0xec/0x2e0 [v3d] [ 314.156187] lr : v3d_irq+0xe0/0x2e0 [v3d] [ 314.160198] sp : ffffffc080003ea0 [ 314.163502] x29: ffffffc080003ea0 x28: ffffffec1f184980 x27: 021202b000000000 [ 314.170633] x26: ffffffec1f17f630 x25: ffffff8101372000 x24: ffffffec1f17d9f0 [ 314.177764] x23: 000000000000002a x22: 000000000000002a x21: ffffff8103252000 [ 314.184895] x20: 0000000000000001 x19: 00000000deadbeef x18: 0000000000000000 [ 314.192026] x17: ffffff94e51d2000 x16: ffffffec1dac3cb0 x15: c306000000000000 [ 314.199156] x14: 0000000000000000 x13: b2fc982e03cc5168 x12: 0000000000000001 [ 314.206286] x11: ffffff8103f8bcc0 x10: ffffffec1f196868 x9 : ffffffec1dac3874 [ 314.213416] x8 : 0000000000000000 x7 : 0000000000042a3a x6 : ffffff810017a180 [ 314.220547] x5 : ffffffec1ebad400 x4 : ffffffec1ebad320 x3 : 00000000000bebeb [ 314.227677] x2 : 0000000000000000 x1 : 0000000000000000 x0 : 0000000000000000 [ 314.234807] Call trace: [ 314.237243] v3d_irq+0xec/0x2e0 [v3d] [ 314.240906] __handle_irq_event_percpu+0x58/0x218 [ 314.245609] handle_irq_event+0x54/0xb8 [ 314.249439] handle_fasteoi_irq+0xac/0x240 [ 314.253527] handle_irq_desc+0x48/0x68 [ 314.257269] generic_handle_domain_irq+0x24/0x38 [ 314.261879] gic_handle_irq+0x48/0xd8 [ 314.265533] call_on_irq_stack+0x24/0x58 [ 314.269448] do_interrupt_handler+0x88/0x98 [ 314.273624] el1_interrupt+0x34/0x68 [ 314.277193] el1h_64_irq_handler+0x18/0x28 [ 314.281281] el1h_64_irq+0x64/0x68 [ 314.284673] default_idle_call+0x3c/0x168 [ 314.288675] do_idle+0x1fc/0x230 [ 314.291895] cpu_startup_entry+0x3c/0x50 [ 314.295810] rest_init+0xe4/0xf0 [ 314.299030] start_kernel+0x5e8/0x790 [ 314.302684] __primary_switched+0x80/0x90 [ 314.306691] Code: 940029eb 360ffc13 f9442ea0 52800001 (f9406017) [ 314.312775] ---[ end trace 0000000000000000 ]--- [ 314.317384] Kernel panic - not syncing: Oops: Fatal exception in interrupt [ 314.324249] SMP: stopping secondary CPUs [ 314.328167] Kernel Offset: 0x2b9da00000 from 0xffffffc080000000 [ 314.334076] PHYS_OFFSET: 0x0 [ 314.336946] CPU features: 0x08,00002013,c0200000,0200421b [ 314.342337] Memory Limit: none [ 314.345382] ---[ end Kernel panic - not syncing: Oops: Fatal exception in interrupt ]--- Before resetting the GPU, it's necessary to disable all interrupts and deal with any interrupt handler still in-flight. Otherwise, the GPU might reset with jobs still running, or yet, an interrupt could be handled during the reset. Cc: stable(a)vger.kernel.org Fixes: 57692c94dcbe ("drm/v3d: Introduce a new DRM driver for Broadcom V3D V3.x+") Signed-off-by: Maíra Canal <mcanal(a)igalia.com> --- drivers/gpu/drm/v3d/v3d_drv.h | 8 ++++++++ drivers/gpu/drm/v3d/v3d_gem.c | 2 ++ drivers/gpu/drm/v3d/v3d_irq.c | 37 +++++++++++++++++++++++++---------- 3 files changed, 37 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h index b51f0b648a08..411e47702f8a 100644 --- a/drivers/gpu/drm/v3d/v3d_drv.h +++ b/drivers/gpu/drm/v3d/v3d_drv.h @@ -101,6 +101,12 @@ enum v3d_gen { V3D_GEN_71 = 71, }; +enum v3d_irq { + V3D_CORE_IRQ, + V3D_HUB_IRQ, + V3D_MAX_IRQS, +}; + struct v3d_dev { struct drm_device drm; @@ -112,6 +118,8 @@ struct v3d_dev { bool single_irq_line; + int irq[V3D_MAX_IRQS]; + struct v3d_perfmon_info perfmon_info; void __iomem *hub_regs; diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c index d7d16da78db3..37bf5eecdd2c 100644 --- a/drivers/gpu/drm/v3d/v3d_gem.c +++ b/drivers/gpu/drm/v3d/v3d_gem.c @@ -134,6 +134,8 @@ v3d_reset(struct v3d_dev *v3d) if (false) v3d_idle_axi(v3d, 0); + v3d_irq_disable(v3d); + v3d_idle_gca(v3d); v3d_reset_sms(v3d); v3d_reset_v3d(v3d); diff --git a/drivers/gpu/drm/v3d/v3d_irq.c b/drivers/gpu/drm/v3d/v3d_irq.c index 2cca5d3a26a2..a515a301e480 100644 --- a/drivers/gpu/drm/v3d/v3d_irq.c +++ b/drivers/gpu/drm/v3d/v3d_irq.c @@ -260,7 +260,7 @@ v3d_hub_irq(int irq, void *arg) int v3d_irq_init(struct v3d_dev *v3d) { - int irq1, ret, core; + int irq, ret, core; INIT_WORK(&v3d->overflow_mem_work, v3d_overflow_mem_work); @@ -271,17 +271,24 @@ v3d_irq_init(struct v3d_dev *v3d) V3D_CORE_WRITE(core, V3D_CTL_INT_CLR, V3D_CORE_IRQS(v3d->ver)); V3D_WRITE(V3D_HUB_INT_CLR, V3D_HUB_IRQS(v3d->ver)); - irq1 = platform_get_irq_optional(v3d_to_pdev(v3d), 1); - if (irq1 == -EPROBE_DEFER) - return irq1; - if (irq1 > 0) { - ret = devm_request_irq(v3d->drm.dev, irq1, + irq = platform_get_irq_optional(v3d_to_pdev(v3d), 1); + if (irq == -EPROBE_DEFER) + return irq; + if (irq > 0) { + v3d->irq[V3D_CORE_IRQ] = irq; + + ret = devm_request_irq(v3d->drm.dev, v3d->irq[V3D_CORE_IRQ], v3d_irq, IRQF_SHARED, "v3d_core0", v3d); if (ret) goto fail; - ret = devm_request_irq(v3d->drm.dev, - platform_get_irq(v3d_to_pdev(v3d), 0), + + irq = platform_get_irq(v3d_to_pdev(v3d), 0); + if (irq < 0) + return irq; + v3d->irq[V3D_HUB_IRQ] = irq; + + ret = devm_request_irq(v3d->drm.dev, v3d->irq[V3D_HUB_IRQ], v3d_hub_irq, IRQF_SHARED, "v3d_hub", v3d); if (ret) @@ -289,8 +296,12 @@ v3d_irq_init(struct v3d_dev *v3d) } else { v3d->single_irq_line = true; - ret = devm_request_irq(v3d->drm.dev, - platform_get_irq(v3d_to_pdev(v3d), 0), + irq = platform_get_irq(v3d_to_pdev(v3d), 0); + if (irq < 0) + return irq; + v3d->irq[V3D_CORE_IRQ] = irq; + + ret = devm_request_irq(v3d->drm.dev, v3d->irq[V3D_CORE_IRQ], v3d_irq, IRQF_SHARED, "v3d", v3d); if (ret) @@ -331,6 +342,12 @@ v3d_irq_disable(struct v3d_dev *v3d) V3D_CORE_WRITE(core, V3D_CTL_INT_MSK_SET, ~0); V3D_WRITE(V3D_HUB_INT_MSK_SET, ~0); + /* Finish any interrupt handler still in flight. */ + for (int i = 0; i < V3D_MAX_IRQS; i++) { + if (v3d->irq[i]) + synchronize_irq(v3d->irq[i]); + } + /* Clear any pending interrupts we might have left. */ for (core = 0; core < v3d->cores; core++) V3D_CORE_WRITE(core, V3D_CTL_INT_CLR, V3D_CORE_IRQS(v3d->ver)); -- 2.50.0

4 months

3
3
0 0

[PATCH net v2] net: libwx: fix the incorrect display of the queue number

by Jiawen Wu

When setting "ethtool -L eth0 combined 1", the number of RX/TX queue is changed to be 1. RSS is disabled at this moment, and the indices of FDIR have not be changed in wx_set_rss_queues(). So the combined count still shows the previous value. This issue was introduced when supporting FDIR. Fix it for those devices that support FDIR. Fixes: 34744a7749b3 ("net: txgbe: add FDIR info to ethtool ops") Cc: stable(a)vger.kernel.org Signed-off-by: Jiawen Wu <jiawenwu(a)trustnetic.com> Reviewed-by: Simon Horman <horms(a)kernel.org> --- v1 -> v2: - fix it in wx_set_rss_queues() --- drivers/net/ethernet/wangxun/libwx/wx_lib.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c index 85c606805e27..55e252789db3 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c @@ -1705,6 +1705,7 @@ static void wx_set_rss_queues(struct wx *wx) clear_bit(WX_FLAG_FDIR_HASH, wx->flags); + wx->ring_feature[RING_F_FDIR].indices = 1; /* Use Flow Director in addition to RSS to ensure the best * distribution of flows across cores, even when an FDIR flow * isn't matched. -- 2.48.1

4 months

3
2
0 0

FAILED: patch "[PATCH] staging: rtl8723bs: Avoid memset() in aes_cipher() and" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y git checkout FETCH_HEAD git cherry-pick -x a55bc4ffc06d8c965a7d6f0a01ed0ed41380df28 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025063055-overfed-dispute-71ba@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From a55bc4ffc06d8c965a7d6f0a01ed0ed41380df28 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor <nathan(a)kernel.org> Date: Mon, 9 Jun 2025 14:13:14 -0700 Subject: [PATCH] staging: rtl8723bs: Avoid memset() in aes_cipher() and aes_decipher() After commit 6f110a5e4f99 ("Disable SLUB_TINY for build testing"), which causes CONFIG_KASAN to be enabled in allmodconfig again, arm64 allmodconfig builds with older versions of clang (15 through 17) show an instance of -Wframe-larger-than (which breaks the build with CONFIG_WERROR=y): drivers/staging/rtl8723bs/core/rtw_security.c:1287:5: error: stack frame size (2208) exceeds limit (2048) in 'rtw_aes_decrypt' [-Werror,-Wframe-larger-than] 1287 | u32 rtw_aes_decrypt(struct adapter *padapter, u8 *precvframe) | ^ This comes from aes_decipher() being inlined in rtw_aes_decrypt(). Running the same build with CONFIG_FRAME_WARN=128 shows aes_cipher() also uses a decent amount of stack, just under the limit of 2048: drivers/staging/rtl8723bs/core/rtw_security.c:864:19: warning: stack frame size (1952) exceeds limit (128) in 'aes_cipher' [-Wframe-larger-than] 864 | static signed int aes_cipher(u8 *key, uint hdrlen, | ^ -Rpass-analysis=stack-frame-layout only shows one large structure on the stack, which is the ctx variable inlined from aes128k128d(). A good number of the other variables come from the additional checks of fortified string routines, which are present in memset(), which both aes_cipher() and aes_decipher() use to initialize some temporary buffers. In this case, since the size is known at compile time, these additional checks should not result in any code generation changes but allmodconfig has several sanitizers enabled, which may make it harder for the compiler to eliminate the compile time checks and the variables that come about from them. The memset() calls are just initializing these buffers to zero, so use '= {}' instead, which is used all over the kernel and does the exact same thing as memset() without the fortify checks, which drops the stack usage of these functions by a few hundred kilobytes. drivers/staging/rtl8723bs/core/rtw_security.c:864:19: warning: stack frame size (1584) exceeds limit (128) in 'aes_cipher' [-Wframe-larger-than] 864 | static signed int aes_cipher(u8 *key, uint hdrlen, | ^ drivers/staging/rtl8723bs/core/rtw_security.c:1271:5: warning: stack frame size (1456) exceeds limit (128) in 'rtw_aes_decrypt' [-Wframe-larger-than] 1271 | u32 rtw_aes_decrypt(struct adapter *padapter, u8 *precvframe) | ^ Cc: stable(a)vger.kernel.org Fixes: 554c0a3abf21 ("staging: Add rtl8723bs sdio wifi driver") Signed-off-by: Nathan Chancellor <nathan(a)kernel.org> Reviewed-by: Dan Carpenter <dan.carpenter(a)linaro.org> Link: https://lore.kernel.org/r/20250609-rtl8723bs-fix-clang-arm64-wflt-v1-1-e2ac… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> diff --git a/drivers/staging/rtl8723bs/core/rtw_security.c b/drivers/staging/rtl8723bs/core/rtw_security.c index 1e9eff01b1aa..e9f382c280d9 100644 --- a/drivers/staging/rtl8723bs/core/rtw_security.c +++ b/drivers/staging/rtl8723bs/core/rtw_security.c @@ -868,29 +868,21 @@ static signed int aes_cipher(u8 *key, uint hdrlen, num_blocks, payload_index; u8 pn_vector[6]; - u8 mic_iv[16]; - u8 mic_header1[16]; - u8 mic_header2[16]; - u8 ctr_preload[16]; + u8 mic_iv[16] = {}; + u8 mic_header1[16] = {}; + u8 mic_header2[16] = {}; + u8 ctr_preload[16] = {}; /* Intermediate Buffers */ - u8 chain_buffer[16]; - u8 aes_out[16]; - u8 padded_buffer[16]; + u8 chain_buffer[16] = {}; + u8 aes_out[16] = {}; + u8 padded_buffer[16] = {}; u8 mic[8]; uint frtype = GetFrameType(pframe); uint frsubtype = GetFrameSubType(pframe); frsubtype = frsubtype>>4; - memset((void *)mic_iv, 0, 16); - memset((void *)mic_header1, 0, 16); - memset((void *)mic_header2, 0, 16); - memset((void *)ctr_preload, 0, 16); - memset((void *)chain_buffer, 0, 16); - memset((void *)aes_out, 0, 16); - memset((void *)padded_buffer, 0, 16); - if ((hdrlen == WLAN_HDR_A3_LEN) || (hdrlen == WLAN_HDR_A3_QOS_LEN)) a4_exists = 0; else @@ -1080,15 +1072,15 @@ static signed int aes_decipher(u8 *key, uint hdrlen, num_blocks, payload_index; signed int res = _SUCCESS; u8 pn_vector[6]; - u8 mic_iv[16]; - u8 mic_header1[16]; - u8 mic_header2[16]; - u8 ctr_preload[16]; + u8 mic_iv[16] = {}; + u8 mic_header1[16] = {}; + u8 mic_header2[16] = {}; + u8 ctr_preload[16] = {}; /* Intermediate Buffers */ - u8 chain_buffer[16]; - u8 aes_out[16]; - u8 padded_buffer[16]; + u8 chain_buffer[16] = {}; + u8 aes_out[16] = {}; + u8 padded_buffer[16] = {}; u8 mic[8]; uint frtype = GetFrameType(pframe); @@ -1096,14 +1088,6 @@ static signed int aes_decipher(u8 *key, uint hdrlen, frsubtype = frsubtype>>4; - memset((void *)mic_iv, 0, 16); - memset((void *)mic_header1, 0, 16); - memset((void *)mic_header2, 0, 16); - memset((void *)ctr_preload, 0, 16); - memset((void *)chain_buffer, 0, 16); - memset((void *)aes_out, 0, 16); - memset((void *)padded_buffer, 0, 16); - /* start to decrypt the payload */ num_blocks = (plen-8) / 16; /* plen including LLC, payload_length and mic) */

4 months

4
3
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror