September 2025 - Linux-stable-mirror

FAILED: patch "[PATCH] iommufd: Fix race during abort for file descriptors" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-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-6.12.y git checkout FETCH_HEAD git cherry-pick -x 4e034bf045b12852a24d5d33f2451850818ba0c1 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025092933-conform-unclaimed-d167@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 4e034bf045b12852a24d5d33f2451850818ba0c1 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe <jgg(a)ziepe.ca> Date: Tue, 16 Sep 2025 20:47:10 -0300 Subject: [PATCH] iommufd: Fix race during abort for file descriptors fput() doesn't actually call file_operations release() synchronously, it puts the file on a work queue and it will be released eventually. This is normally fine, except for iommufd the file and the iommufd_object are tied to gether. The file has the object as it's private_data and holds a users refcount, while the object is expected to remain alive as long as the file is. When the allocation of a new object aborts before installing the file it will fput() the file and then go on to immediately kfree() the obj. This causes a UAF once the workqueue completes the fput() and tries to decrement the users refcount. Fix this by putting the core code in charge of the file lifetime, and call __fput_sync() during abort to ensure that release() is called before kfree. __fput_sync() is a bit too tricky to open code in all the object implementations. Instead the objects tell the core code where the file pointer is and the core will take care of the life cycle. If the object is successfully allocated then the file will hold a users refcount and the iommufd_object cannot be destroyed. It is worth noting that close(); ioctl(IOMMU_DESTROY); doesn't have an issue because close() is already using a synchronous version of fput(). The UAF looks like this: BUG: KASAN: slab-use-after-free in iommufd_eventq_fops_release+0x45/0xc0 drivers/iommu/iommufd/eventq.c:376 Write of size 4 at addr ffff888059c97804 by task syz.0.46/6164 CPU: 0 UID: 0 PID: 6164 Comm: syz.0.46 Not tainted syzkaller #0 PREEMPT(full) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/18/2025 Call Trace: <TASK> __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:378 [inline] print_report+0xcd/0x630 mm/kasan/report.c:482 kasan_report+0xe0/0x110 mm/kasan/report.c:595 check_region_inline mm/kasan/generic.c:183 [inline] kasan_check_range+0x100/0x1b0 mm/kasan/generic.c:189 instrument_atomic_read_write include/linux/instrumented.h:96 [inline] atomic_fetch_sub_release include/linux/atomic/atomic-instrumented.h:400 [inline] __refcount_dec include/linux/refcount.h:455 [inline] refcount_dec include/linux/refcount.h:476 [inline] iommufd_eventq_fops_release+0x45/0xc0 drivers/iommu/iommufd/eventq.c:376 __fput+0x402/0xb70 fs/file_table.c:468 task_work_run+0x14d/0x240 kernel/task_work.c:227 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] exit_to_user_mode_loop+0xeb/0x110 kernel/entry/common.c:43 exit_to_user_mode_prepare include/linux/irq-entry-common.h:225 [inline] syscall_exit_to_user_mode_work include/linux/entry-common.h:175 [inline] syscall_exit_to_user_mode include/linux/entry-common.h:210 [inline] do_syscall_64+0x41c/0x4c0 arch/x86/entry/syscall_64.c:100 entry_SYSCALL_64_after_hwframe+0x77/0x7f Link: https://patch.msgid.link/r/1-v1-02cd136829df+31-iommufd_syz_fput_jgg@nvidia… Cc: stable(a)vger.kernel.org Fixes: 07838f7fd529 ("iommufd: Add iommufd fault object") Reviewed-by: Nicolin Chen <nicolinc(a)nvidia.com> Reviewed-by: Nirmoy Das <nirmoyd(a)nvidia.com> Reviewed-by: Kevin Tian <kevin.tian(a)intel.com> Tested-by: Nicolin Chen <nicolinc(a)nvidia.com> Reported-by: syzbot+80620e2d0d0a33b09f93(a)syzkaller.appspotmail.com Closes: https://lore.kernel.org/r/68c8583d.050a0220.2ff435.03a2.GAE@google.com Signed-off-by: Jason Gunthorpe <jgg(a)nvidia.com> diff --git a/drivers/iommu/iommufd/eventq.c b/drivers/iommu/iommufd/eventq.c index fc4de63b0bce..e23d9ee4fe38 100644 --- a/drivers/iommu/iommufd/eventq.c +++ b/drivers/iommu/iommufd/eventq.c @@ -393,12 +393,12 @@ static int iommufd_eventq_init(struct iommufd_eventq *eventq, char *name, const struct file_operations *fops) { struct file *filep; - int fdno; spin_lock_init(&eventq->lock); INIT_LIST_HEAD(&eventq->deliver); init_waitqueue_head(&eventq->wait_queue); + /* The filep is fput() by the core code during failure */ filep = anon_inode_getfile(name, fops, eventq, O_RDWR); if (IS_ERR(filep)) return PTR_ERR(filep); @@ -408,10 +408,7 @@ static int iommufd_eventq_init(struct iommufd_eventq *eventq, char *name, eventq->filep = filep; refcount_inc(&eventq->obj.users); - fdno = get_unused_fd_flags(O_CLOEXEC); - if (fdno < 0) - fput(filep); - return fdno; + return get_unused_fd_flags(O_CLOEXEC); } static const struct file_operations iommufd_fault_fops = @@ -452,7 +449,6 @@ int iommufd_fault_alloc(struct iommufd_ucmd *ucmd) return 0; out_put_fdno: put_unused_fd(fdno); - fput(fault->common.filep); return rc; } @@ -536,7 +532,6 @@ int iommufd_veventq_alloc(struct iommufd_ucmd *ucmd) out_put_fdno: put_unused_fd(fdno); - fput(veventq->common.filep); out_abort: iommufd_object_abort_and_destroy(ucmd->ictx, &veventq->common.obj); out_unlock_veventqs: diff --git a/drivers/iommu/iommufd/main.c b/drivers/iommu/iommufd/main.c index a9d4decc8ba1..88be2e157245 100644 --- a/drivers/iommu/iommufd/main.c +++ b/drivers/iommu/iommufd/main.c @@ -23,6 +23,7 @@ #include "iommufd_test.h" struct iommufd_object_ops { + size_t file_offset; void (*pre_destroy)(struct iommufd_object *obj); void (*destroy)(struct iommufd_object *obj); void (*abort)(struct iommufd_object *obj); @@ -131,10 +132,30 @@ void iommufd_object_abort(struct iommufd_ctx *ictx, struct iommufd_object *obj) void iommufd_object_abort_and_destroy(struct iommufd_ctx *ictx, struct iommufd_object *obj) { - if (iommufd_object_ops[obj->type].abort) - iommufd_object_ops[obj->type].abort(obj); + const struct iommufd_object_ops *ops = &iommufd_object_ops[obj->type]; + + if (ops->file_offset) { + struct file **filep = ((void *)obj) + ops->file_offset; + + /* + * A file should hold a users refcount while the file is open + * and put it back in its release. The file should hold a + * pointer to obj in their private data. Normal fput() is + * deferred to a workqueue and can get out of order with the + * following kfree(obj). Using the sync version ensures the + * release happens immediately. During abort we require the file + * refcount is one at this point - meaning the object alloc + * function cannot do anything to allow another thread to take a + * refcount prior to a guaranteed success. + */ + if (*filep) + __fput_sync(*filep); + } + + if (ops->abort) + ops->abort(obj); else - iommufd_object_ops[obj->type].destroy(obj); + ops->destroy(obj); iommufd_object_abort(ictx, obj); } @@ -659,6 +680,12 @@ void iommufd_ctx_put(struct iommufd_ctx *ictx) } EXPORT_SYMBOL_NS_GPL(iommufd_ctx_put, "IOMMUFD"); +#define IOMMUFD_FILE_OFFSET(_struct, _filep, _obj) \ + .file_offset = (offsetof(_struct, _filep) + \ + BUILD_BUG_ON_ZERO(!__same_type( \ + struct file *, ((_struct *)NULL)->_filep)) + \ + BUILD_BUG_ON_ZERO(offsetof(_struct, _obj))) + static const struct iommufd_object_ops iommufd_object_ops[] = { [IOMMUFD_OBJ_ACCESS] = { .destroy = iommufd_access_destroy_object, @@ -669,6 +696,7 @@ static const struct iommufd_object_ops iommufd_object_ops[] = { }, [IOMMUFD_OBJ_FAULT] = { .destroy = iommufd_fault_destroy, + IOMMUFD_FILE_OFFSET(struct iommufd_fault, common.filep, common.obj), }, [IOMMUFD_OBJ_HW_QUEUE] = { .destroy = iommufd_hw_queue_destroy, @@ -691,6 +719,7 @@ static const struct iommufd_object_ops iommufd_object_ops[] = { [IOMMUFD_OBJ_VEVENTQ] = { .destroy = iommufd_veventq_destroy, .abort = iommufd_veventq_abort, + IOMMUFD_FILE_OFFSET(struct iommufd_veventq, common.filep, common.obj), }, [IOMMUFD_OBJ_VIOMMU] = { .destroy = iommufd_viommu_destroy,

2 months, 1 week

2
1
0 0

FAILED: patch "[PATCH] i40e: add mask to apply valid bits for itr_idx" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-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-5.4.y git checkout FETCH_HEAD git cherry-pick -x eac04428abe9f9cb203ffae4600791ea1d24eb18 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025092927-captivate-suspense-fdf7@gregkh' --subject-prefix 'PATCH 5.4.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From eac04428abe9f9cb203ffae4600791ea1d24eb18 Mon Sep 17 00:00:00 2001 From: Lukasz Czapnik <lukasz.czapnik(a)intel.com> Date: Wed, 13 Aug 2025 12:45:17 +0200 Subject: [PATCH] i40e: add mask to apply valid bits for itr_idx The ITR index (itr_idx) is only 2 bits wide. When constructing the register value for QINT_RQCTL, all fields are ORed together. Without masking, higher bits from itr_idx may overwrite adjacent fields in the register. Apply I40E_QINT_RQCTL_ITR_INDX_MASK to ensure only the intended bits are set. Fixes: 5c3c48ac6bf5 ("i40e: implement virtual device interface") Cc: stable(a)vger.kernel.org Signed-off-by: Lukasz Czapnik <lukasz.czapnik(a)intel.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov(a)intel.com> Signed-off-by: Przemek Kitszel <przemyslaw.kitszel(a)intel.com> Reviewed-by: Simon Horman <horms(a)kernel.org> Tested-by: Rafal Romanowski <rafal.romanowski(a)intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen(a)intel.com> diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index f29941c00342..f9b2197f0942 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -448,7 +448,7 @@ static void i40e_config_irq_link_list(struct i40e_vf *vf, u16 vsi_id, (qtype << I40E_QINT_RQCTL_NEXTQ_TYPE_SHIFT) | (pf_queue_id << I40E_QINT_RQCTL_NEXTQ_INDX_SHIFT) | BIT(I40E_QINT_RQCTL_CAUSE_ENA_SHIFT) | - (itr_idx << I40E_QINT_RQCTL_ITR_INDX_SHIFT); + FIELD_PREP(I40E_QINT_RQCTL_ITR_INDX_MASK, itr_idx); wr32(hw, reg_idx, reg); }

2 months, 1 week

2
1
0 0

[PATCH] mtd: nand: raw: gpmi: fix clocks when CONFIG_PM=N

by Maarten Zanders

Commit f04ced6d545e ("mtd: nand: raw: gpmi: improve power management handling") moved all clock handling into PM callbacks. With CONFIG_PM disabled, those callbacks are missing, leaving the driver unusable. Add clock init/teardown for !CONFIG_PM builds to restore basic operation. Keeping the driver working without requiring CONFIG_PM is preferred over adding a Kconfig dependency. Fixes: f04ced6d545e ("mtd: nand: raw: gpmi: improve power management handling") Signed-off-by: Maarten Zanders <maarten(a)zanders.be> Cc: stable(a)vger.kernel.org --- drivers/mtd/nand/raw/gpmi-nand/gpmi-nand.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/mtd/nand/raw/gpmi-nand/gpmi-nand.c b/drivers/mtd/nand/raw/gpmi-nand/gpmi-nand.c index f4e68008ea03..a750f5839e34 100644 --- a/drivers/mtd/nand/raw/gpmi-nand/gpmi-nand.c +++ b/drivers/mtd/nand/raw/gpmi-nand/gpmi-nand.c @@ -145,6 +145,9 @@ static int __gpmi_enable_clk(struct gpmi_nand_data *this, bool v) return ret; } +#define gpmi_enable_clk(x) __gpmi_enable_clk(x, true) +#define gpmi_disable_clk(x) __gpmi_enable_clk(x, false) + static int gpmi_init(struct gpmi_nand_data *this) { struct resources *r = &this->resources; @@ -2765,6 +2768,11 @@ static int gpmi_nand_probe(struct platform_device *pdev) pm_runtime_enable(&pdev->dev); pm_runtime_set_autosuspend_delay(&pdev->dev, 500); pm_runtime_use_autosuspend(&pdev->dev); +#ifndef CONFIG_PM + ret = gpmi_enable_clk(this); + if (ret) + goto exit_acquire_resources; +#endif ret = gpmi_init(this); if (ret) @@ -2800,6 +2808,9 @@ static void gpmi_nand_remove(struct platform_device *pdev) release_resources(this); pm_runtime_dont_use_autosuspend(&pdev->dev); pm_runtime_disable(&pdev->dev); +#ifndef CONFIG_PM + gpmi_disable_clk(this); +#endif } static int gpmi_pm_suspend(struct device *dev) @@ -2846,9 +2857,6 @@ static int gpmi_pm_resume(struct device *dev) return 0; } -#define gpmi_enable_clk(x) __gpmi_enable_clk(x, true) -#define gpmi_disable_clk(x) __gpmi_enable_clk(x, false) - static int gpmi_runtime_suspend(struct device *dev) { struct gpmi_nand_data *this = dev_get_drvdata(dev); -- 2.51.0

2 months, 1 week

2
3
0 0

FAILED: patch "[PATCH] drm/ast: Use msleep instead of mdelay for edid read" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.1-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-6.1.y git checkout FETCH_HEAD git cherry-pick -x c7c31f8dc54aa3c9b2c994b5f1ff7e740a654e97 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025092921-consensus-mystified-6396@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From c7c31f8dc54aa3c9b2c994b5f1ff7e740a654e97 Mon Sep 17 00:00:00 2001 From: Nirmoy Das <nirmoyd(a)nvidia.com> Date: Wed, 17 Sep 2025 12:43:46 -0700 Subject: [PATCH] drm/ast: Use msleep instead of mdelay for edid read The busy-waiting in `mdelay()` can cause CPU stalls and kernel timeouts during boot. Signed-off-by: Nirmoy Das <nirmoyd(a)nvidia.com> Reviewed-by: Thomas Zimmermann <tzimmermann(a)suse.de> Tested-by: Carol L Soto csoto(a)nvidia.com<mailto:csoto@nvidia.com> Fixes: 594e9c04b586 ("drm/ast: Create the driver for ASPEED proprietory Display-Port") Cc: KuoHsiang Chou <kuohsiang_chou(a)aspeedtech.com> Cc: Thomas Zimmermann <tzimmermann(a)suse.de> Cc: Dave Airlie <airlied(a)redhat.com> Cc: Jocelyn Falempe <jfalempe(a)redhat.com> Cc: dri-devel(a)lists.freedesktop.org Cc: <stable(a)vger.kernel.org> # v5.19+ Signed-off-by: Thomas Zimmermann <tzimmermann(a)suse.de> Link: https://lore.kernel.org/r/20250917194346.2905522-1-nirmoyd@nvidia.com diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index 19c04687b0fe..8e650a02c528 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -134,7 +134,7 @@ static int ast_astdp_read_edid_block(void *data, u8 *buf, unsigned int block, si * 3. The Delays are often longer a lot when system resume from S3/S4. */ if (j) - mdelay(j + 1); + msleep(j + 1); /* Wait for EDID offset to show up in mirror register */ vgacrd7 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xd7);

2 months, 1 week

2
1
0 0

FAILED: patch "[PATCH] drm/ast: Use msleep instead of mdelay for edid read" failed to apply to 6.6-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.6-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-6.6.y git checkout FETCH_HEAD git cherry-pick -x c7c31f8dc54aa3c9b2c994b5f1ff7e740a654e97 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025092920-backspin-glade-7b6c@gregkh' --subject-prefix 'PATCH 6.6.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From c7c31f8dc54aa3c9b2c994b5f1ff7e740a654e97 Mon Sep 17 00:00:00 2001 From: Nirmoy Das <nirmoyd(a)nvidia.com> Date: Wed, 17 Sep 2025 12:43:46 -0700 Subject: [PATCH] drm/ast: Use msleep instead of mdelay for edid read The busy-waiting in `mdelay()` can cause CPU stalls and kernel timeouts during boot. Signed-off-by: Nirmoy Das <nirmoyd(a)nvidia.com> Reviewed-by: Thomas Zimmermann <tzimmermann(a)suse.de> Tested-by: Carol L Soto csoto(a)nvidia.com<mailto:csoto@nvidia.com> Fixes: 594e9c04b586 ("drm/ast: Create the driver for ASPEED proprietory Display-Port") Cc: KuoHsiang Chou <kuohsiang_chou(a)aspeedtech.com> Cc: Thomas Zimmermann <tzimmermann(a)suse.de> Cc: Dave Airlie <airlied(a)redhat.com> Cc: Jocelyn Falempe <jfalempe(a)redhat.com> Cc: dri-devel(a)lists.freedesktop.org Cc: <stable(a)vger.kernel.org> # v5.19+ Signed-off-by: Thomas Zimmermann <tzimmermann(a)suse.de> Link: https://lore.kernel.org/r/20250917194346.2905522-1-nirmoyd@nvidia.com diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index 19c04687b0fe..8e650a02c528 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -134,7 +134,7 @@ static int ast_astdp_read_edid_block(void *data, u8 *buf, unsigned int block, si * 3. The Delays are often longer a lot when system resume from S3/S4. */ if (j) - mdelay(j + 1); + msleep(j + 1); /* Wait for EDID offset to show up in mirror register */ vgacrd7 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xd7);

2 months, 1 week

2
1
0 0

FAILED: patch "[PATCH] iommufd: Fix race during abort for file descriptors" failed to apply to 6.16-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.16-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-6.16.y git checkout FETCH_HEAD git cherry-pick -x 4e034bf045b12852a24d5d33f2451850818ba0c1 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025092931-uneven-never-0b1f@gregkh' --subject-prefix 'PATCH 6.16.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 4e034bf045b12852a24d5d33f2451850818ba0c1 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe <jgg(a)ziepe.ca> Date: Tue, 16 Sep 2025 20:47:10 -0300 Subject: [PATCH] iommufd: Fix race during abort for file descriptors fput() doesn't actually call file_operations release() synchronously, it puts the file on a work queue and it will be released eventually. This is normally fine, except for iommufd the file and the iommufd_object are tied to gether. The file has the object as it's private_data and holds a users refcount, while the object is expected to remain alive as long as the file is. When the allocation of a new object aborts before installing the file it will fput() the file and then go on to immediately kfree() the obj. This causes a UAF once the workqueue completes the fput() and tries to decrement the users refcount. Fix this by putting the core code in charge of the file lifetime, and call __fput_sync() during abort to ensure that release() is called before kfree. __fput_sync() is a bit too tricky to open code in all the object implementations. Instead the objects tell the core code where the file pointer is and the core will take care of the life cycle. If the object is successfully allocated then the file will hold a users refcount and the iommufd_object cannot be destroyed. It is worth noting that close(); ioctl(IOMMU_DESTROY); doesn't have an issue because close() is already using a synchronous version of fput(). The UAF looks like this: BUG: KASAN: slab-use-after-free in iommufd_eventq_fops_release+0x45/0xc0 drivers/iommu/iommufd/eventq.c:376 Write of size 4 at addr ffff888059c97804 by task syz.0.46/6164 CPU: 0 UID: 0 PID: 6164 Comm: syz.0.46 Not tainted syzkaller #0 PREEMPT(full) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/18/2025 Call Trace: <TASK> __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:378 [inline] print_report+0xcd/0x630 mm/kasan/report.c:482 kasan_report+0xe0/0x110 mm/kasan/report.c:595 check_region_inline mm/kasan/generic.c:183 [inline] kasan_check_range+0x100/0x1b0 mm/kasan/generic.c:189 instrument_atomic_read_write include/linux/instrumented.h:96 [inline] atomic_fetch_sub_release include/linux/atomic/atomic-instrumented.h:400 [inline] __refcount_dec include/linux/refcount.h:455 [inline] refcount_dec include/linux/refcount.h:476 [inline] iommufd_eventq_fops_release+0x45/0xc0 drivers/iommu/iommufd/eventq.c:376 __fput+0x402/0xb70 fs/file_table.c:468 task_work_run+0x14d/0x240 kernel/task_work.c:227 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] exit_to_user_mode_loop+0xeb/0x110 kernel/entry/common.c:43 exit_to_user_mode_prepare include/linux/irq-entry-common.h:225 [inline] syscall_exit_to_user_mode_work include/linux/entry-common.h:175 [inline] syscall_exit_to_user_mode include/linux/entry-common.h:210 [inline] do_syscall_64+0x41c/0x4c0 arch/x86/entry/syscall_64.c:100 entry_SYSCALL_64_after_hwframe+0x77/0x7f Link: https://patch.msgid.link/r/1-v1-02cd136829df+31-iommufd_syz_fput_jgg@nvidia… Cc: stable(a)vger.kernel.org Fixes: 07838f7fd529 ("iommufd: Add iommufd fault object") Reviewed-by: Nicolin Chen <nicolinc(a)nvidia.com> Reviewed-by: Nirmoy Das <nirmoyd(a)nvidia.com> Reviewed-by: Kevin Tian <kevin.tian(a)intel.com> Tested-by: Nicolin Chen <nicolinc(a)nvidia.com> Reported-by: syzbot+80620e2d0d0a33b09f93(a)syzkaller.appspotmail.com Closes: https://lore.kernel.org/r/68c8583d.050a0220.2ff435.03a2.GAE@google.com Signed-off-by: Jason Gunthorpe <jgg(a)nvidia.com> diff --git a/drivers/iommu/iommufd/eventq.c b/drivers/iommu/iommufd/eventq.c index fc4de63b0bce..e23d9ee4fe38 100644 --- a/drivers/iommu/iommufd/eventq.c +++ b/drivers/iommu/iommufd/eventq.c @@ -393,12 +393,12 @@ static int iommufd_eventq_init(struct iommufd_eventq *eventq, char *name, const struct file_operations *fops) { struct file *filep; - int fdno; spin_lock_init(&eventq->lock); INIT_LIST_HEAD(&eventq->deliver); init_waitqueue_head(&eventq->wait_queue); + /* The filep is fput() by the core code during failure */ filep = anon_inode_getfile(name, fops, eventq, O_RDWR); if (IS_ERR(filep)) return PTR_ERR(filep); @@ -408,10 +408,7 @@ static int iommufd_eventq_init(struct iommufd_eventq *eventq, char *name, eventq->filep = filep; refcount_inc(&eventq->obj.users); - fdno = get_unused_fd_flags(O_CLOEXEC); - if (fdno < 0) - fput(filep); - return fdno; + return get_unused_fd_flags(O_CLOEXEC); } static const struct file_operations iommufd_fault_fops = @@ -452,7 +449,6 @@ int iommufd_fault_alloc(struct iommufd_ucmd *ucmd) return 0; out_put_fdno: put_unused_fd(fdno); - fput(fault->common.filep); return rc; } @@ -536,7 +532,6 @@ int iommufd_veventq_alloc(struct iommufd_ucmd *ucmd) out_put_fdno: put_unused_fd(fdno); - fput(veventq->common.filep); out_abort: iommufd_object_abort_and_destroy(ucmd->ictx, &veventq->common.obj); out_unlock_veventqs: diff --git a/drivers/iommu/iommufd/main.c b/drivers/iommu/iommufd/main.c index a9d4decc8ba1..88be2e157245 100644 --- a/drivers/iommu/iommufd/main.c +++ b/drivers/iommu/iommufd/main.c @@ -23,6 +23,7 @@ #include "iommufd_test.h" struct iommufd_object_ops { + size_t file_offset; void (*pre_destroy)(struct iommufd_object *obj); void (*destroy)(struct iommufd_object *obj); void (*abort)(struct iommufd_object *obj); @@ -131,10 +132,30 @@ void iommufd_object_abort(struct iommufd_ctx *ictx, struct iommufd_object *obj) void iommufd_object_abort_and_destroy(struct iommufd_ctx *ictx, struct iommufd_object *obj) { - if (iommufd_object_ops[obj->type].abort) - iommufd_object_ops[obj->type].abort(obj); + const struct iommufd_object_ops *ops = &iommufd_object_ops[obj->type]; + + if (ops->file_offset) { + struct file **filep = ((void *)obj) + ops->file_offset; + + /* + * A file should hold a users refcount while the file is open + * and put it back in its release. The file should hold a + * pointer to obj in their private data. Normal fput() is + * deferred to a workqueue and can get out of order with the + * following kfree(obj). Using the sync version ensures the + * release happens immediately. During abort we require the file + * refcount is one at this point - meaning the object alloc + * function cannot do anything to allow another thread to take a + * refcount prior to a guaranteed success. + */ + if (*filep) + __fput_sync(*filep); + } + + if (ops->abort) + ops->abort(obj); else - iommufd_object_ops[obj->type].destroy(obj); + ops->destroy(obj); iommufd_object_abort(ictx, obj); } @@ -659,6 +680,12 @@ void iommufd_ctx_put(struct iommufd_ctx *ictx) } EXPORT_SYMBOL_NS_GPL(iommufd_ctx_put, "IOMMUFD"); +#define IOMMUFD_FILE_OFFSET(_struct, _filep, _obj) \ + .file_offset = (offsetof(_struct, _filep) + \ + BUILD_BUG_ON_ZERO(!__same_type( \ + struct file *, ((_struct *)NULL)->_filep)) + \ + BUILD_BUG_ON_ZERO(offsetof(_struct, _obj))) + static const struct iommufd_object_ops iommufd_object_ops[] = { [IOMMUFD_OBJ_ACCESS] = { .destroy = iommufd_access_destroy_object, @@ -669,6 +696,7 @@ static const struct iommufd_object_ops iommufd_object_ops[] = { }, [IOMMUFD_OBJ_FAULT] = { .destroy = iommufd_fault_destroy, + IOMMUFD_FILE_OFFSET(struct iommufd_fault, common.filep, common.obj), }, [IOMMUFD_OBJ_HW_QUEUE] = { .destroy = iommufd_hw_queue_destroy, @@ -691,6 +719,7 @@ static const struct iommufd_object_ops iommufd_object_ops[] = { [IOMMUFD_OBJ_VEVENTQ] = { .destroy = iommufd_veventq_destroy, .abort = iommufd_veventq_abort, + IOMMUFD_FILE_OFFSET(struct iommufd_veventq, common.filep, common.obj), }, [IOMMUFD_OBJ_VIOMMU] = { .destroy = iommufd_viommu_destroy,

2 months, 1 week

2
1
0 0

[PATCH RESEND] of: unittest: Fix device reference count leak in of_unittest_pci_node_verify

by Ma Ke

In of_unittest_pci_node_verify(), when the add parameter is false, device_find_any_child() obtains a reference to a child device. This function implicitly calls get_device() to increment the device's reference count before returning the pointer. However, the caller fails to properly release this reference by calling put_device(), leading to a device reference count leak. As the comment of device_find_any_child states: "NOTE: you will need to drop the reference with put_device() after use". Cc: stable(a)vger.kernel.org Fixes: 26409dd04589 ("of: unittest: Add pci_dt_testdrv pci driver") Signed-off-by: Ma Ke <make24(a)iscas.ac.cn> --- drivers/of/unittest.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/of/unittest.c b/drivers/of/unittest.c index e3503ec20f6c..d225e73781fe 100644 --- a/drivers/of/unittest.c +++ b/drivers/of/unittest.c @@ -4271,7 +4271,7 @@ static struct platform_driver unittest_pci_driver = { static int of_unittest_pci_node_verify(struct pci_dev *pdev, bool add) { struct device_node *pnp, *np = NULL; - struct device *child_dev; + struct device *child_dev = NULL; char *path = NULL; const __be32 *reg; int rc = 0; @@ -4306,6 +4306,8 @@ static int of_unittest_pci_node_verify(struct pci_dev *pdev, bool add) kfree(path); if (np) of_node_put(np); + if (child_dev) + put_device(child_dev); return rc; } -- 2.17.1

2 months, 1 week

2
1
0 0

FAILED: patch "[PATCH] i40e: fix validation of VF state in get resources" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-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-5.4.y git checkout FETCH_HEAD git cherry-pick -x 877b7e6ffc23766448236e8732254534c518ba42 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025092913-monotype-pouch-f2e0@gregkh' --subject-prefix 'PATCH 5.4.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 877b7e6ffc23766448236e8732254534c518ba42 Mon Sep 17 00:00:00 2001 From: Lukasz Czapnik <lukasz.czapnik(a)intel.com> Date: Wed, 13 Aug 2025 12:45:15 +0200 Subject: [PATCH] i40e: fix validation of VF state in get resources VF state I40E_VF_STATE_ACTIVE is not the only state in which VF is actually active so it should not be used to determine if a VF is allowed to obtain resources. Use I40E_VF_STATE_RESOURCES_LOADED that is set only in i40e_vc_get_vf_resources_msg() and cleared during reset. Fixes: 61125b8be85d ("i40e: Fix failed opcode appearing if handling messages from VF") Cc: stable(a)vger.kernel.org Signed-off-by: Lukasz Czapnik <lukasz.czapnik(a)intel.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov(a)intel.com> Signed-off-by: Przemek Kitszel <przemyslaw.kitszel(a)intel.com> Reviewed-by: Simon Horman <horms(a)kernel.org> Tested-by: Rafal Romanowski <rafal.romanowski(a)intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen(a)intel.com> diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index c85715f75435..5ef3dc43a8a0 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -1464,6 +1464,7 @@ static void i40e_trigger_vf_reset(struct i40e_vf *vf, bool flr) * functions that may still be running at this point. */ clear_bit(I40E_VF_STATE_INIT, &vf->vf_states); + clear_bit(I40E_VF_STATE_RESOURCES_LOADED, &vf->vf_states); /* In the case of a VFLR, the HW has already reset the VF and we * just need to clean up, so don't hit the VFRTRIG register. @@ -2130,7 +2131,10 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg) size_t len = 0; int ret; - if (!i40e_sync_vf_state(vf, I40E_VF_STATE_INIT)) { + i40e_sync_vf_state(vf, I40E_VF_STATE_INIT); + + if (!test_bit(I40E_VF_STATE_INIT, &vf->vf_states) || + test_bit(I40E_VF_STATE_RESOURCES_LOADED, &vf->vf_states)) { aq_ret = -EINVAL; goto err; } @@ -2233,6 +2237,7 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg) vf->default_lan_addr.addr); } set_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states); + set_bit(I40E_VF_STATE_RESOURCES_LOADED, &vf->vf_states); err: /* send the response back to the VF */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h index 5cf74f16f433..f558b45725c8 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h @@ -41,7 +41,8 @@ enum i40e_vf_states { I40E_VF_STATE_MC_PROMISC, I40E_VF_STATE_UC_PROMISC, I40E_VF_STATE_PRE_ENABLE, - I40E_VF_STATE_RESETTING + I40E_VF_STATE_RESETTING, + I40E_VF_STATE_RESOURCES_LOADED, }; /* VF capabilities */

2 months, 1 week

2
1
0 0

FAILED: patch "[PATCH] i40e: fix idx validation in config queues msg" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-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-5.4.y git checkout FETCH_HEAD git cherry-pick -x f1ad24c5abe1eaef69158bac1405a74b3c365115 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025092937-trash-dainty-32e4@gregkh' --subject-prefix 'PATCH 5.4.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From f1ad24c5abe1eaef69158bac1405a74b3c365115 Mon Sep 17 00:00:00 2001 From: Lukasz Czapnik <lukasz.czapnik(a)intel.com> Date: Wed, 13 Aug 2025 12:45:13 +0200 Subject: [PATCH] i40e: fix idx validation in config queues msg Ensure idx is within range of active/initialized TCs when iterating over vf->ch[idx] in i40e_vc_config_queues_msg(). Fixes: c27eac48160d ("i40e: Enable ADq and create queue channel/s on VF") Cc: stable(a)vger.kernel.org Signed-off-by: Lukasz Czapnik <lukasz.czapnik(a)intel.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov(a)intel.com> Signed-off-by: Przemek Kitszel <przemyslaw.kitszel(a)intel.com> Reviewed-by: Simon Horman <horms(a)kernel.org> Tested-by: Kamakshi Nellore <nellorex.kamakshi(a)intel.com> (A Contingent Worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen(a)intel.com> diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 1c4f86221255..b6db4d78c02d 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -2395,7 +2395,7 @@ static int i40e_vc_config_queues_msg(struct i40e_vf *vf, u8 *msg) } if (vf->adq_enabled) { - if (idx >= ARRAY_SIZE(vf->ch)) { + if (idx >= vf->num_tc) { aq_ret = -ENODEV; goto error_param; } @@ -2416,7 +2416,7 @@ static int i40e_vc_config_queues_msg(struct i40e_vf *vf, u8 *msg) * to its appropriate VSIs based on TC mapping */ if (vf->adq_enabled) { - if (idx >= ARRAY_SIZE(vf->ch)) { + if (idx >= vf->num_tc) { aq_ret = -ENODEV; goto error_param; }

2 months, 1 week

2
1
0 0

FAILED: patch "[PATCH] i40e: fix validation of VF state in get resources" 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 877b7e6ffc23766448236e8732254534c518ba42 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025092912-bullfight-urgency-ae9f@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 877b7e6ffc23766448236e8732254534c518ba42 Mon Sep 17 00:00:00 2001 From: Lukasz Czapnik <lukasz.czapnik(a)intel.com> Date: Wed, 13 Aug 2025 12:45:15 +0200 Subject: [PATCH] i40e: fix validation of VF state in get resources VF state I40E_VF_STATE_ACTIVE is not the only state in which VF is actually active so it should not be used to determine if a VF is allowed to obtain resources. Use I40E_VF_STATE_RESOURCES_LOADED that is set only in i40e_vc_get_vf_resources_msg() and cleared during reset. Fixes: 61125b8be85d ("i40e: Fix failed opcode appearing if handling messages from VF") Cc: stable(a)vger.kernel.org Signed-off-by: Lukasz Czapnik <lukasz.czapnik(a)intel.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov(a)intel.com> Signed-off-by: Przemek Kitszel <przemyslaw.kitszel(a)intel.com> Reviewed-by: Simon Horman <horms(a)kernel.org> Tested-by: Rafal Romanowski <rafal.romanowski(a)intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen(a)intel.com> diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index c85715f75435..5ef3dc43a8a0 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -1464,6 +1464,7 @@ static void i40e_trigger_vf_reset(struct i40e_vf *vf, bool flr) * functions that may still be running at this point. */ clear_bit(I40E_VF_STATE_INIT, &vf->vf_states); + clear_bit(I40E_VF_STATE_RESOURCES_LOADED, &vf->vf_states); /* In the case of a VFLR, the HW has already reset the VF and we * just need to clean up, so don't hit the VFRTRIG register. @@ -2130,7 +2131,10 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg) size_t len = 0; int ret; - if (!i40e_sync_vf_state(vf, I40E_VF_STATE_INIT)) { + i40e_sync_vf_state(vf, I40E_VF_STATE_INIT); + + if (!test_bit(I40E_VF_STATE_INIT, &vf->vf_states) || + test_bit(I40E_VF_STATE_RESOURCES_LOADED, &vf->vf_states)) { aq_ret = -EINVAL; goto err; } @@ -2233,6 +2237,7 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg) vf->default_lan_addr.addr); } set_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states); + set_bit(I40E_VF_STATE_RESOURCES_LOADED, &vf->vf_states); err: /* send the response back to the VF */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h index 5cf74f16f433..f558b45725c8 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h @@ -41,7 +41,8 @@ enum i40e_vf_states { I40E_VF_STATE_MC_PROMISC, I40E_VF_STATE_UC_PROMISC, I40E_VF_STATE_PRE_ENABLE, - I40E_VF_STATE_RESETTING + I40E_VF_STATE_RESETTING, + I40E_VF_STATE_RESOURCES_LOADED, }; /* VF capabilities */

2 months, 1 week

2
1
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror September 2025