Linux-stable-mirror

linux-stable-mirror@lists.linaro.org

303 participants
124147 discussions

[PATCH] drm/xe: Tie page count tracepoints to TTM accounting functions

by Matthew Brost

Page accounting can change via the shrinker without calling xe_ttm_tt_unpopulate(), but those paths already update accounting through the xe_ttm_tt_account_*() helpers. Move the page count tracepoints into xe_ttm_tt_account_add() and xe_ttm_tt_account_subtract() so accounting updates are recorded consistently, regardless of whether pages are populated, unpopulated, or reclaimed via the shrinker. This avoids missing page count updates and keeps global accounting balanced across all TT lifecycle paths. Cc: stable(a)vger.kernel.org Fixes: ce3d39fae3d3 ("drm/xe/bo: add GPU memory trace points") Signed-off-by: Matthew Brost <matthew.brost(a)intel.com> --- drivers/gpu/drm/xe/xe_bo.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index 8b6474cd3eaf..33afaee38f48 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -432,6 +432,9 @@ struct sg_table *xe_bo_sg(struct xe_bo *bo) return xe_tt->sg; } +static void update_global_total_pages(struct ttm_device *ttm_dev, + long num_pages); + /* * Account ttm pages against the device shrinker's shrinkable and * purgeable counts. @@ -440,6 +443,7 @@ static void xe_ttm_tt_account_add(struct xe_device *xe, struct ttm_tt *tt) { struct xe_ttm_tt *xe_tt = container_of(tt, struct xe_ttm_tt, ttm); + update_global_total_pages(&xe->ttm, tt->num_pages); if (xe_tt->purgeable) xe_shrinker_mod_pages(xe->mem.shrinker, 0, tt->num_pages); else @@ -450,6 +454,7 @@ static void xe_ttm_tt_account_subtract(struct xe_device *xe, struct ttm_tt *tt) { struct xe_ttm_tt *xe_tt = container_of(tt, struct xe_ttm_tt, ttm); + update_global_total_pages(&xe->ttm, -(long)tt->num_pages); if (xe_tt->purgeable) xe_shrinker_mod_pages(xe->mem.shrinker, 0, -(long)tt->num_pages); else @@ -575,7 +580,6 @@ static int xe_ttm_tt_populate(struct ttm_device *ttm_dev, struct ttm_tt *tt, xe_tt->purgeable = false; xe_ttm_tt_account_add(ttm_to_xe_device(ttm_dev), tt); - update_global_total_pages(ttm_dev, tt->num_pages); return 0; } @@ -592,7 +596,6 @@ static void xe_ttm_tt_unpopulate(struct ttm_device *ttm_dev, struct ttm_tt *tt) ttm_pool_free(&ttm_dev->pool, tt); xe_ttm_tt_account_subtract(xe, tt); - update_global_total_pages(ttm_dev, -(long)tt->num_pages); } static void xe_ttm_tt_destroy(struct ttm_device *ttm_dev, struct ttm_tt *tt) -- 2.34.1

4 days, 13 hours

Bounce probe for stable@vger.kernel.org (no action required)

by stable+owner＠vger.kernel.org

Greetings! This is the mlmmj program managing the <stable(a)vger.kernel.org> mailing list. Your mail host rejected some of the messages we tried to deliver. This usually happens if we have accepted a spam message, but your mail server refused to take it from us (because it's spam). This automated probe is just a test to make sure that there are no other problems with mail delivery to your account. No action is required on your part -- if you are seeing this message, that means everything is normal. (In fact, you can add "Subject: Bounce probe" to your filters to avoid seeing this message in the future.) For internal tracking purposes, here is the list of bounced messages: - 206207

4 days, 14 hours

[PATCH v7 0/9] Error recovery for vfio-pci devices on s390x

by Farhan Ali

Hi, This Linux kernel patch series introduces support for error recovery for passthrough PCI devices on System Z (s390x). Background ---------- For PCI devices on s390x an operating system receives platform specific error events from firmware rather than through AER.Today for passthrough/userspace devices, we don't attempt any error recovery and ignore any error events for the devices. The passthrough/userspace devices are managed by the vfio-pci driver. The driver does register error handling callbacks (error_detected), and on an error trigger an eventfd to userspace. But we need a mechanism to notify userspace (QEMU/guest/userspace drivers) about the error event. Proposal -------- We can expose this error information (currently only the PCI Error Code) via a device feature. Userspace can then obtain the error information via VFIO_DEVICE_FEATURE ioctl and take appropriate actions such as driving a device reset. This is how a typical flow for passthrough devices to a VM would work: For passthrough devices to a VM, the driver bound to the device on the host is vfio-pci. vfio-pci driver does support the error_detected() callback (vfio_pci_core_aer_err_detected()), and on an PCI error s390x recovery code on the host will call the vfio-pci error_detected() callback. The vfio-pci error_detected() callback will notify userspace/QEMU via an eventfd, and return PCI_ERS_RESULT_CAN_RECOVER. At this point the s390x error recovery on the host will skip any further action(see patch 6) and let userspace drive the error recovery. Once userspace/QEMU is notified, it then injects this error into the VM so device drivers in the VM can take recovery actions. For example for a passthrough NVMe device, the VM's OS NVMe driver will access the device. At this point the VM's NVMe driver's error_detected() will drive the recovery by returning PCI_ERS_RESULT_NEED_RESET, and the s390x error recovery in the VM's OS will try to do a reset. Resets are privileged operations and so the VM will need intervention from QEMU to perform the reset. QEMU will invoke the VFIO_DEVICE_RESET ioctl to now notify the host that the VM is requesting a reset of the device. The vfio-pci driver on the host will then perform the reset on the device to recover it. Thanks Farhan ChangeLog --------- v6 series https://lore.kernel.org/all/2c609e61-1861-4bf3-b019-a11c137d26a5@linux.ibm.… v6 -> v7 - Rebase on 6.19-rc4 - Update commit message based on Niklas's suggestion (patch 3). v5 series https://lore.kernel.org/all/20251113183502.2388-1-alifm@linux.ibm.com/ v5 -> v6 - Rebase on 6.18 + Lukas's PCI: Universal error recoverability of devices series (https://lore.kernel.org/all/cover.1763483367.git.lukas@wunner.de/) - Re-work config space accessibility check to pci_dev_save_and_disable() (patch 3). This avoids saving the config space, in the reset path, if the device's config space is corrupted or inaccessible. v4 series https://lore.kernel.org/all/20250924171628.826-1-alifm@linux.ibm.com/ v4 -> v5 - Rebase on 6.18-rc5 - Move bug fixes to the beginning of the series (patch 1 and 2). These patches were posted as a separate fixes series https://lore.kernel.org/all/a14936ac-47d6-461b-816f-0fd66f869b0f@linux.ibm.… - Add matching pci_put_dev() for pci_get_slot() (patch 6). v3 series https://lore.kernel.org/all/20250911183307.1910-1-alifm@linux.ibm.com/ v3 -> v4 - Remove warn messages for each PCI capability not restored (patch 1) - Check PCI_COMMAND and PCI_STATUS register for error value instead of device id (patch 1) - Fix kernel crash in patch 3 - Added reviewed by tags - Address comments from Niklas's (patches 4, 5, 7) - Fix compilation error non s390x system (patch 8) - Explicitly align struct vfio_device_feature_zpci_err (patch 8) v2 series https://lore.kernel.org/all/20250825171226.1602-1-alifm@linux.ibm.com/ v2 -> v3 - Patch 1 avoids saving any config space state if the device is in error (suggested by Alex) - Patch 2 adds additional check only for FLR reset to try other function reset method (suggested by Alex). - Patch 3 fixes a bug in s390 for resetting PCI devices with multiple functions. Creates a new flag pci_slot to allow per function slot. - Patch 4 fixes a bug in s390 for resource to bus address translation. - Rebase on 6.17-rc5 v1 series https://lore.kernel.org/all/20250813170821.1115-1-alifm@linux.ibm.com/ v1 - > v2 - Patches 1 and 2 adds some additional checks for FLR/PM reset to try other function reset method (suggested by Alex). - Patch 3 fixes a bug in s390 for resetting PCI devices with multiple functions. - Patch 7 adds a new device feature for zPCI devices for the VFIO_DEVICE_FEATURE ioctl. The ioctl is used by userspace to retriece any PCI error information for the device (suggested by Alex). - Patch 8 adds a reset_done() callback for the vfio-pci driver, to restore the state of the device after a reset. - Patch 9 removes the pcie check for triggering VFIO_PCI_ERR_IRQ_INDEX. Farhan Ali (9): PCI: Allow per function PCI slots s390/pci: Add architecture specific resource/bus address translation PCI: Avoid saving config space state if inaccessible PCI: Add additional checks for flr reset s390/pci: Update the logic for detecting passthrough device s390/pci: Store PCI error information for passthrough devices vfio-pci/zdev: Add a device feature for error information vfio: Add a reset_done callback for vfio-pci driver vfio: Remove the pcie check for VFIO_PCI_ERR_IRQ_INDEX arch/s390/include/asm/pci.h | 29 ++++++++ arch/s390/pci/pci.c | 75 +++++++++++++++++++++ arch/s390/pci/pci_event.c | 107 +++++++++++++++++------------- drivers/pci/host-bridge.c | 4 +- drivers/pci/pci.c | 19 +++++- drivers/pci/slot.c | 25 ++++++- drivers/vfio/pci/vfio_pci_core.c | 20 ++++-- drivers/vfio/pci/vfio_pci_intrs.c | 3 +- drivers/vfio/pci/vfio_pci_priv.h | 9 +++ drivers/vfio/pci/vfio_pci_zdev.c | 45 ++++++++++++- include/linux/pci.h | 1 + include/uapi/linux/vfio.h | 16 +++++ 12 files changed, 292 insertions(+), 61 deletions(-) -- 2.43.0

4 days, 16 hours

[PATCH v7 1/9] PCI: Allow per function PCI slots

by Farhan Ali

On s390 systems, which use a machine level hypervisor, PCI devices are always accessed through a form of PCI pass-through which fundamentally operates on a per PCI function granularity. This is also reflected in the s390 PCI hotplug driver which creates hotplug slots for individual PCI functions. Its reset_slot() function, which is a wrapper for zpci_hot_reset_device(), thus also resets individual functions. Currently, the kernel's PCI_SLOT() macro assigns the same pci_slot object to multifunction devices. This approach worked fine on s390 systems that only exposed virtual functions as individual PCI domains to the operating system. Since commit 44510d6fa0c0 ("s390/pci: Handling multifunctions") s390 supports exposing the topology of multifunction PCI devices by grouping them in a shared PCI domain. When attempting to reset a function through the hotplug driver, the shared slot assignment causes the wrong function to be reset instead of the intended one. It also leaks memory as we do create a pci_slot object for the function, but don't correctly free it in pci_slot_release(). Add a flag for struct pci_slot to allow per function PCI slots for functions managed through a hypervisor, which exposes individual PCI functions while retaining the topology. Fixes: 44510d6fa0c0 ("s390/pci: Handling multifunctions") Cc: stable(a)vger.kernel.org Suggested-by: Niklas Schnelle <schnelle(a)linux.ibm.com> Reviewed-by: Niklas Schnelle <schnelle(a)linux.ibm.com> Signed-off-by: Farhan Ali <alifm(a)linux.ibm.com> --- drivers/pci/pci.c | 5 +++-- drivers/pci/slot.c | 25 ++++++++++++++++++++++--- include/linux/pci.h | 1 + 3 files changed, 26 insertions(+), 5 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 13dbb405dc31..c105e285cff8 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -4832,8 +4832,9 @@ static int pci_reset_hotplug_slot(struct hotplug_slot *hotplug, bool probe) static int pci_dev_reset_slot_function(struct pci_dev *dev, bool probe) { - if (dev->multifunction || dev->subordinate || !dev->slot || - dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET) + if (dev->subordinate || !dev->slot || + dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET || + (dev->multifunction && !dev->slot->per_func_slot)) return -ENOTTY; return pci_reset_hotplug_slot(dev->slot->hotplug, probe); diff --git a/drivers/pci/slot.c b/drivers/pci/slot.c index 50fb3eb595fe..ed10fa3ae727 100644 --- a/drivers/pci/slot.c +++ b/drivers/pci/slot.c @@ -63,6 +63,22 @@ static ssize_t cur_speed_read_file(struct pci_slot *slot, char *buf) return bus_speed_read(slot->bus->cur_bus_speed, buf); } +static bool pci_dev_matches_slot(struct pci_dev *dev, struct pci_slot *slot) +{ + if (slot->per_func_slot) + return dev->devfn == slot->number; + + return PCI_SLOT(dev->devfn) == slot->number; +} + +static bool pci_slot_enabled_per_func(void) +{ + if (IS_ENABLED(CONFIG_S390)) + return true; + + return false; +} + static void pci_slot_release(struct kobject *kobj) { struct pci_dev *dev; @@ -73,7 +89,7 @@ static void pci_slot_release(struct kobject *kobj) down_read(&pci_bus_sem); list_for_each_entry(dev, &slot->bus->devices, bus_list) - if (PCI_SLOT(dev->devfn) == slot->number) + if (pci_dev_matches_slot(dev, slot)) dev->slot = NULL; up_read(&pci_bus_sem); @@ -166,7 +182,7 @@ void pci_dev_assign_slot(struct pci_dev *dev) mutex_lock(&pci_slot_mutex); list_for_each_entry(slot, &dev->bus->slots, list) - if (PCI_SLOT(dev->devfn) == slot->number) + if (pci_dev_matches_slot(dev, slot)) dev->slot = slot; mutex_unlock(&pci_slot_mutex); } @@ -265,6 +281,9 @@ struct pci_slot *pci_create_slot(struct pci_bus *parent, int slot_nr, slot->bus = pci_bus_get(parent); slot->number = slot_nr; + if (pci_slot_enabled_per_func()) + slot->per_func_slot = 1; + slot->kobj.kset = pci_slots_kset; slot_name = make_slot_name(name); @@ -285,7 +304,7 @@ struct pci_slot *pci_create_slot(struct pci_bus *parent, int slot_nr, down_read(&pci_bus_sem); list_for_each_entry(dev, &parent->devices, bus_list) - if (PCI_SLOT(dev->devfn) == slot_nr) + if (pci_dev_matches_slot(dev, slot)) dev->slot = slot; up_read(&pci_bus_sem); diff --git a/include/linux/pci.h b/include/linux/pci.h index 864775651c6f..08fa57beb7b2 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -78,6 +78,7 @@ struct pci_slot { struct list_head list; /* Node in list of slots */ struct hotplug_slot *hotplug; /* Hotplug info (move here) */ unsigned char number; /* PCI_SLOT(pci_dev->devfn) */ + unsigned int per_func_slot:1; /* Allow per function slot */ struct kobject kobj; }; -- 2.43.0

4 days, 16 hours

[PATCH] drm/vmwgfx: Fix an error return check in vmw_compat_shader_add()

by Haoxiang Li

In vmw_compat_shader_add(), the return value check of vmw_shader_alloc() is not proper. Modify the check for the return pointer 'res'. Found by code review and compiled on ubuntu 20.04. Fixes: 18e4a4669c50 ("drm/vmwgfx: Fix compat shader namespace") Cc: stable(a)vger.kernel.org Signed-off-by: Haoxiang Li <lihaoxiang(a)isrc.iscas.ac.cn> --- drivers/gpu/drm/vmwgfx/vmwgfx_shader.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index 69dfe69ce0f8..7ed938710342 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -923,8 +923,10 @@ int vmw_compat_shader_add(struct vmw_private *dev_priv, ttm_bo_unreserve(&buf->tbo); res = vmw_shader_alloc(dev_priv, buf, size, 0, shader_type); - if (unlikely(ret != 0)) + if (IS_ERR(res)) { + ret = PTR_ERR(res); goto no_reserve; + } ret = vmw_cmdbuf_res_add(man, vmw_cmdbuf_res_shader, vmw_shader_key(user_key, shader_type), -- 2.25.1

4 days, 16 hours

[PATCH] rust: task: restrict Task::group_leader() to current

by Alice Ryhl

The Task::group_leader() method currently allows you to access the group_leader() of any task, for example one you hold a refcount to. But this is not safe in general since the group leader could change when a task exits. See for example commit a15f37a40145c ("kernel/sys.c: fix the racy usage of task_lock(tsk->group_leader) in sys_prlimit64() paths"). All existing users of Task::group_leader() call this method on current, which is guaranteed running, so there's not an actual issue in Rust code today. But to prevent code in the future from making this mistake, restrict Task::group_leader() so that it can only be called on current. There are some other cases where accessing task->group_leader is okay. For example it can be safe if you hold tasklist_lock or rcu_read_lock(). However, only supporting current->group_leader is sufficient for all in-tree Rust users of group_leader right now. Safe Rust functionality for accessing it under rcu or while holding tasklist_lock may be added in the future if required by any future Rust module. Reported-by: Oleg Nesterov <oleg(a)redhat.com> Closes: https://lore.kernel.org/all/aTLnV-5jlgfk1aRK@redhat.com/ Fixes: 313c4281bc9d ("rust: add basic `Task`") Cc: stable(a)vger.kernel.org Signed-off-by: Alice Ryhl <aliceryhl(a)google.com> --- The rust/kernel/task.rs file has had changes land through a few different trees: * Originally task.rs landed through Christian's tree together with file.rs and pid_namespace.rs * The change to add CurrentTask landed through Andrew Morton's tree together with mm.rs * There was a patch to mark some methods #[inline] that landed through tip via Boqun. I don't think there's a clear owner for this file, so to break ambiguity I'm doing to declare that this patch is intended for Andrew Morton's tree. Please let me know if you think a different tree is appropriate. --- rust/kernel/task.rs | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/rust/kernel/task.rs b/rust/kernel/task.rs index 49fad6de06740a9b9ad80b2f4b430cc28cd134fa..9440692a3a6d0d3f908d61d51dcd377a272f6957 100644 --- a/rust/kernel/task.rs +++ b/rust/kernel/task.rs @@ -204,18 +204,6 @@ pub fn as_ptr(&self) -> *mut bindings::task_struct { self.0.get() } - /// Returns the group leader of the given task. - pub fn group_leader(&self) -> &Task { - // SAFETY: The group leader of a task never changes after initialization, so reading this - // field is not a data race. - let ptr = unsafe { *ptr::addr_of!((*self.as_ptr()).group_leader) }; - - // SAFETY: The lifetime of the returned task reference is tied to the lifetime of `self`, - // and given that a task has a reference to its group leader, we know it must be valid for - // the lifetime of the returned task reference. - unsafe { &*ptr.cast() } - } - /// Returns the PID of the given task. pub fn pid(&self) -> Pid { // SAFETY: The pid of a task never changes after initialization, so reading this field is @@ -345,6 +333,18 @@ pub fn active_pid_ns(&self) -> Option<&PidNamespace> { // `release_task()` call. Some(unsafe { PidNamespace::from_ptr(active_ns) }) } + + /// Returns the group leader of the current task. + pub fn group_leader(&self) -> &Task { + // SAFETY: The group leader of a task never changes while the task is running, and `self` + // is the current task, which is guaranteed running. + let ptr = unsafe { (*self.as_ptr()).group_leader }; + + // SAFETY: `current.group_leader` stays valid for at least the duration in which `current` + // is running, and the signature of this function ensures that the returned `&Task` can + // only be used while `current` is still valid, thus still running. + unsafe { &*ptr.cast() } + } } // SAFETY: The type invariants guarantee that `Task` is always refcounted. --- base-commit: 8f0b4cce4481fb22653697cced8d0d04027cb1e8 change-id: 20251218-task-group-leader-a71931ced643 Best regards, -- Alice Ryhl <aliceryhl(a)google.com>

4 days, 17 hours

[PATCH 5.15] net: Remove RTNL dance for SIOCBRADDIF and SIOCBRDELIF.

by Thadeu Lima de Souza Cascardo

commit ed3ba9b6e280e14cc3148c1b226ba453f02fa76c upstream. SIOCBRDELIF is passed to dev_ioctl() first and later forwarded to br_ioctl_call(), which causes unnecessary RTNL dance and the splat below [0] under RTNL pressure. Let's say Thread A is trying to detach a device from a bridge and Thread B is trying to remove the bridge. In dev_ioctl(), Thread A bumps the bridge device's refcnt by netdev_hold() and releases RTNL because the following br_ioctl_call() also re-acquires RTNL. In the race window, Thread B could acquire RTNL and try to remove the bridge device. Then, rtnl_unlock() by Thread B will release RTNL and wait for netdev_put() by Thread A. Thread A, however, must hold RTNL after the unlock in dev_ifsioc(), which may take long under RTNL pressure, resulting in the splat by Thread B. Thread A (SIOCBRDELIF) Thread B (SIOCBRDELBR) ---------------------- ---------------------- sock_ioctl sock_ioctl `- sock_do_ioctl `- br_ioctl_call `- dev_ioctl `- br_ioctl_stub |- rtnl_lock | |- dev_ifsioc ' ' |- dev = __dev_get_by_name(...) |- netdev_hold(dev, ...) . / |- rtnl_unlock ------. | | |- br_ioctl_call `---> |- rtnl_lock Race | | `- br_ioctl_stub |- br_del_bridge Window | | | |- dev = __dev_get_by_name(...) | | | May take long | `- br_dev_delete(dev, ...) | | | under RTNL pressure | `- unregister_netdevice_queue(dev, ...) | | | | `- rtnl_unlock \ | |- rtnl_lock <-' `- netdev_run_todo | |- ... `- netdev_run_todo | `- rtnl_unlock |- __rtnl_unlock | |- netdev_wait_allrefs_any |- netdev_put(dev, ...) <----------------' Wait refcnt decrement and log splat below To avoid blocking SIOCBRDELBR unnecessarily, let's not call dev_ioctl() for SIOCBRADDIF and SIOCBRDELIF. In the dev_ioctl() path, we do the following: 1. Copy struct ifreq by get_user_ifreq in sock_do_ioctl() 2. Check CAP_NET_ADMIN in dev_ioctl() 3. Call dev_load() in dev_ioctl() 4. Fetch the master dev from ifr.ifr_name in dev_ifsioc() 3. can be done by request_module() in br_ioctl_call(), so we move 1., 2., and 4. to br_ioctl_stub(). Note that 2. is also checked later in add_del_if(), but it's better performed before RTNL. SIOCBRADDIF and SIOCBRDELIF have been processed in dev_ioctl() since the pre-git era, and there seems to be no specific reason to process them there. [0]: unregister_netdevice: waiting for wpan3 to become free. Usage count = 2 ref_tracker: wpan3@ffff8880662d8608 has 1/1 users at __netdev_tracker_alloc include/linux/netdevice.h:4282 [inline] netdev_hold include/linux/netdevice.h:4311 [inline] dev_ifsioc+0xc6a/0x1160 net/core/dev_ioctl.c:624 dev_ioctl+0x255/0x10c0 net/core/dev_ioctl.c:826 sock_do_ioctl+0x1ca/0x260 net/socket.c:1213 sock_ioctl+0x23a/0x6c0 net/socket.c:1318 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:906 [inline] __se_sys_ioctl fs/ioctl.c:892 [inline] __x64_sys_ioctl+0x1a4/0x210 fs/ioctl.c:892 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcb/0x250 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 893b19587534 ("net: bridge: fix ioctl locking") Reported-by: syzkaller <syzkaller(a)googlegroups.com> Reported-by: yan kang <kangyan91(a)outlook.com> Reported-by: yue sun <samsun1006219(a)gmail.com> Closes: https://lore.kernel.org/netdev/SY8P300MB0421225D54EB92762AE8F0F2A1D32@SY8P3… Signed-off-by: Kuniyuki Iwashima <kuniyu(a)amazon.com> Acked-by: Stanislav Fomichev <sdf(a)fomichev.me> Reviewed-by: Ido Schimmel <idosch(a)nvidia.com> Acked-by: Nikolay Aleksandrov <razor(a)blackwall.org> Link: https://patch.msgid.link/20250316192851.19781-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [cascardo: fixed conflict at dev_ifsioc] Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo(a)igalia.com> --- include/linux/if_bridge.h | 6 ++---- net/bridge/br_ioctl.c | 36 +++++++++++++++++++++++++++++++++--- net/bridge/br_private.h | 3 +-- net/core/dev_ioctl.c | 15 --------------- net/socket.c | 19 +++++++++---------- 5 files changed, 45 insertions(+), 34 deletions(-) diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h index 509e18c7e740..37082feae336 100644 --- a/include/linux/if_bridge.h +++ b/include/linux/if_bridge.h @@ -62,11 +62,9 @@ struct br_ip_list { #define BR_DEFAULT_AGEING_TIME (300 * HZ) struct net_bridge; -void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +void brioctl_set(int (*hook)(struct net *net, unsigned int cmd, void __user *uarg)); -int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg); +int br_ioctl_call(struct net *net, unsigned int cmd, void __user *uarg); #if IS_ENABLED(CONFIG_BRIDGE) && IS_ENABLED(CONFIG_BRIDGE_IGMP_SNOOPING) int br_multicast_list_adjacent(struct net_device *dev, diff --git a/net/bridge/br_ioctl.c b/net/bridge/br_ioctl.c index 9922497e59f8..85c23a38bdb2 100644 --- a/net/bridge/br_ioctl.c +++ b/net/bridge/br_ioctl.c @@ -368,10 +368,26 @@ static int old_deviceless(struct net *net, void __user *uarg) return -EOPNOTSUPP; } -int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg) +int br_ioctl_stub(struct net *net, unsigned int cmd, void __user *uarg) { int ret = -EOPNOTSUPP; + struct ifreq ifr; + + if (cmd == SIOCBRADDIF || cmd == SIOCBRDELIF) { + void __user *data; + char *colon; + + if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + + if (get_user_ifreq(&ifr, &data, uarg)) + return -EFAULT; + + ifr.ifr_name[IFNAMSIZ - 1] = 0; + colon = strchr(ifr.ifr_name, ':'); + if (colon) + *colon = 0; + } rtnl_lock(); @@ -404,7 +420,21 @@ int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, break; case SIOCBRADDIF: case SIOCBRDELIF: - ret = add_del_if(br, ifr->ifr_ifindex, cmd == SIOCBRADDIF); + { + struct net_device *dev; + + dev = __dev_get_by_name(net, ifr.ifr_name); + if (!dev || !netif_device_present(dev)) { + ret = -ENODEV; + break; + } + if (!netif_is_bridge_master(dev)) { + ret = -EOPNOTSUPP; + break; + } + + ret = add_del_if(netdev_priv(dev), ifr.ifr_ifindex, cmd == SIOCBRADDIF); + } break; } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 8acb427ae6de..0cf756677953 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -874,8 +874,7 @@ br_port_get_check_rtnl(const struct net_device *dev) /* br_ioctl.c */ int br_dev_siocdevprivate(struct net_device *dev, struct ifreq *rq, void __user *data, int cmd); -int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg); +int br_ioctl_stub(struct net *net, unsigned int cmd, void __user *uarg); /* br_multicast.c */ #ifdef CONFIG_BRIDGE_IGMP_SNOOPING diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 6ddfd7bfc512..30fca446f58a 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -375,19 +375,6 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, case SIOCWANDEV: return dev_siocwandev(dev, &ifr->ifr_settings); - case SIOCBRADDIF: - case SIOCBRDELIF: - if (!netif_device_present(dev)) - return -ENODEV; - if (!netif_is_bridge_master(dev)) - return -EOPNOTSUPP; - dev_hold(dev); - rtnl_unlock(); - err = br_ioctl_call(net, netdev_priv(dev), cmd, ifr, NULL); - dev_put(dev); - rtnl_lock(); - return err; - case SIOCSHWTSTAMP: err = net_hwtstamp_validate(ifr); if (err) @@ -574,8 +561,6 @@ int dev_ioctl(struct net *net, unsigned int cmd, struct ifreq *ifr, case SIOCBONDRELEASE: case SIOCBONDSETHWADDR: case SIOCBONDCHANGEACTIVE: - case SIOCBRADDIF: - case SIOCBRDELIF: case SIOCSHWTSTAMP: if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) return -EPERM; diff --git a/net/socket.c b/net/socket.c index 1d71fa44ace4..f3d0a8d66cce 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1097,12 +1097,10 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from) */ static DEFINE_MUTEX(br_ioctl_mutex); -static int (*br_ioctl_hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +static int (*br_ioctl_hook)(struct net *net, unsigned int cmd, void __user *uarg); -void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +void brioctl_set(int (*hook)(struct net *net, unsigned int cmd, void __user *uarg)) { mutex_lock(&br_ioctl_mutex); @@ -1111,8 +1109,7 @@ void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, } EXPORT_SYMBOL(brioctl_set); -int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg) +int br_ioctl_call(struct net *net, unsigned int cmd, void __user *uarg) { int err = -ENOPKG; @@ -1121,7 +1118,7 @@ int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, mutex_lock(&br_ioctl_mutex); if (br_ioctl_hook) - err = br_ioctl_hook(net, br, cmd, ifr, uarg); + err = br_ioctl_hook(net, cmd, uarg); mutex_unlock(&br_ioctl_mutex); return err; @@ -1218,7 +1215,9 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg) case SIOCSIFBR: case SIOCBRADDBR: case SIOCBRDELBR: - err = br_ioctl_call(net, NULL, cmd, NULL, argp); + case SIOCBRADDIF: + case SIOCBRDELIF: + err = br_ioctl_call(net, cmd, argp); break; case SIOCGIFVLAN: case SIOCSIFVLAN: @@ -3321,6 +3320,8 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGPGRP: case SIOCBRADDBR: case SIOCBRDELBR: + case SIOCBRADDIF: + case SIOCBRDELIF: case SIOCGIFVLAN: case SIOCSIFVLAN: case SIOCGSKNS: @@ -3358,8 +3359,6 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGIFPFLAGS: case SIOCGIFTXQLEN: case SIOCSIFTXQLEN: - case SIOCBRADDIF: - case SIOCBRDELIF: case SIOCGIFNAME: case SIOCSIFNAME: case SIOCGMIIPHY: -- 2.47.3

4 days, 17 hours

[PATCH 6.1] net: Remove RTNL dance for SIOCBRADDIF and SIOCBRDELIF.

by Thadeu Lima de Souza Cascardo

commit ed3ba9b6e280e14cc3148c1b226ba453f02fa76c upstream. SIOCBRDELIF is passed to dev_ioctl() first and later forwarded to br_ioctl_call(), which causes unnecessary RTNL dance and the splat below [0] under RTNL pressure. Let's say Thread A is trying to detach a device from a bridge and Thread B is trying to remove the bridge. In dev_ioctl(), Thread A bumps the bridge device's refcnt by netdev_hold() and releases RTNL because the following br_ioctl_call() also re-acquires RTNL. In the race window, Thread B could acquire RTNL and try to remove the bridge device. Then, rtnl_unlock() by Thread B will release RTNL and wait for netdev_put() by Thread A. Thread A, however, must hold RTNL after the unlock in dev_ifsioc(), which may take long under RTNL pressure, resulting in the splat by Thread B. Thread A (SIOCBRDELIF) Thread B (SIOCBRDELBR) ---------------------- ---------------------- sock_ioctl sock_ioctl `- sock_do_ioctl `- br_ioctl_call `- dev_ioctl `- br_ioctl_stub |- rtnl_lock | |- dev_ifsioc ' ' |- dev = __dev_get_by_name(...) |- netdev_hold(dev, ...) . / |- rtnl_unlock ------. | | |- br_ioctl_call `---> |- rtnl_lock Race | | `- br_ioctl_stub |- br_del_bridge Window | | | |- dev = __dev_get_by_name(...) | | | May take long | `- br_dev_delete(dev, ...) | | | under RTNL pressure | `- unregister_netdevice_queue(dev, ...) | | | | `- rtnl_unlock \ | |- rtnl_lock <-' `- netdev_run_todo | |- ... `- netdev_run_todo | `- rtnl_unlock |- __rtnl_unlock | |- netdev_wait_allrefs_any |- netdev_put(dev, ...) <----------------' Wait refcnt decrement and log splat below To avoid blocking SIOCBRDELBR unnecessarily, let's not call dev_ioctl() for SIOCBRADDIF and SIOCBRDELIF. In the dev_ioctl() path, we do the following: 1. Copy struct ifreq by get_user_ifreq in sock_do_ioctl() 2. Check CAP_NET_ADMIN in dev_ioctl() 3. Call dev_load() in dev_ioctl() 4. Fetch the master dev from ifr.ifr_name in dev_ifsioc() 3. can be done by request_module() in br_ioctl_call(), so we move 1., 2., and 4. to br_ioctl_stub(). Note that 2. is also checked later in add_del_if(), but it's better performed before RTNL. SIOCBRADDIF and SIOCBRDELIF have been processed in dev_ioctl() since the pre-git era, and there seems to be no specific reason to process them there. [0]: unregister_netdevice: waiting for wpan3 to become free. Usage count = 2 ref_tracker: wpan3@ffff8880662d8608 has 1/1 users at __netdev_tracker_alloc include/linux/netdevice.h:4282 [inline] netdev_hold include/linux/netdevice.h:4311 [inline] dev_ifsioc+0xc6a/0x1160 net/core/dev_ioctl.c:624 dev_ioctl+0x255/0x10c0 net/core/dev_ioctl.c:826 sock_do_ioctl+0x1ca/0x260 net/socket.c:1213 sock_ioctl+0x23a/0x6c0 net/socket.c:1318 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:906 [inline] __se_sys_ioctl fs/ioctl.c:892 [inline] __x64_sys_ioctl+0x1a4/0x210 fs/ioctl.c:892 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcb/0x250 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 893b19587534 ("net: bridge: fix ioctl locking") Reported-by: syzkaller <syzkaller(a)googlegroups.com> Reported-by: yan kang <kangyan91(a)outlook.com> Reported-by: yue sun <samsun1006219(a)gmail.com> Closes: https://lore.kernel.org/netdev/SY8P300MB0421225D54EB92762AE8F0F2A1D32@SY8P3… Signed-off-by: Kuniyuki Iwashima <kuniyu(a)amazon.com> Acked-by: Stanislav Fomichev <sdf(a)fomichev.me> Reviewed-by: Ido Schimmel <idosch(a)nvidia.com> Acked-by: Nikolay Aleksandrov <razor(a)blackwall.org> Link: https://patch.msgid.link/20250316192851.19781-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [cascardo: fixed conflict at dev_ifsioc] Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo(a)igalia.com> --- include/linux/if_bridge.h | 6 ++---- net/bridge/br_ioctl.c | 36 +++++++++++++++++++++++++++++++++--- net/bridge/br_private.h | 3 +-- net/core/dev_ioctl.c | 16 ---------------- net/socket.c | 19 +++++++++---------- 5 files changed, 45 insertions(+), 35 deletions(-) diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h index d62ef428e3aa..3963f0af4dad 100644 --- a/include/linux/if_bridge.h +++ b/include/linux/if_bridge.h @@ -63,11 +63,9 @@ struct br_ip_list { #define BR_DEFAULT_AGEING_TIME (300 * HZ) struct net_bridge; -void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +void brioctl_set(int (*hook)(struct net *net, unsigned int cmd, void __user *uarg)); -int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg); +int br_ioctl_call(struct net *net, unsigned int cmd, void __user *uarg); #if IS_ENABLED(CONFIG_BRIDGE) && IS_ENABLED(CONFIG_BRIDGE_IGMP_SNOOPING) int br_multicast_list_adjacent(struct net_device *dev, diff --git a/net/bridge/br_ioctl.c b/net/bridge/br_ioctl.c index f213ed108361..6bc0a11f2ed3 100644 --- a/net/bridge/br_ioctl.c +++ b/net/bridge/br_ioctl.c @@ -394,10 +394,26 @@ static int old_deviceless(struct net *net, void __user *data) return -EOPNOTSUPP; } -int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg) +int br_ioctl_stub(struct net *net, unsigned int cmd, void __user *uarg) { int ret = -EOPNOTSUPP; + struct ifreq ifr; + + if (cmd == SIOCBRADDIF || cmd == SIOCBRDELIF) { + void __user *data; + char *colon; + + if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + + if (get_user_ifreq(&ifr, &data, uarg)) + return -EFAULT; + + ifr.ifr_name[IFNAMSIZ - 1] = 0; + colon = strchr(ifr.ifr_name, ':'); + if (colon) + *colon = 0; + } rtnl_lock(); @@ -430,7 +446,21 @@ int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, break; case SIOCBRADDIF: case SIOCBRDELIF: - ret = add_del_if(br, ifr->ifr_ifindex, cmd == SIOCBRADDIF); + { + struct net_device *dev; + + dev = __dev_get_by_name(net, ifr.ifr_name); + if (!dev || !netif_device_present(dev)) { + ret = -ENODEV; + break; + } + if (!netif_is_bridge_master(dev)) { + ret = -EOPNOTSUPP; + break; + } + + ret = add_del_if(netdev_priv(dev), ifr.ifr_ifindex, cmd == SIOCBRADDIF); + } break; } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 901b9f609b0c..2d3859c2979a 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -894,8 +894,7 @@ br_port_get_check_rtnl(const struct net_device *dev) /* br_ioctl.c */ int br_dev_siocdevprivate(struct net_device *dev, struct ifreq *rq, void __user *data, int cmd); -int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg); +int br_ioctl_stub(struct net *net, unsigned int cmd, void __user *uarg); /* br_multicast.c */ #ifdef CONFIG_BRIDGE_IGMP_SNOOPING diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 5cdbfbf9a7dc..def973eb496a 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -315,7 +315,6 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, int err; struct net_device *dev = __dev_get_by_name(net, ifr->ifr_name); const struct net_device_ops *ops; - netdevice_tracker dev_tracker; if (!dev) return -ENODEV; @@ -378,19 +377,6 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, case SIOCWANDEV: return dev_siocwandev(dev, &ifr->ifr_settings); - case SIOCBRADDIF: - case SIOCBRDELIF: - if (!netif_device_present(dev)) - return -ENODEV; - if (!netif_is_bridge_master(dev)) - return -EOPNOTSUPP; - netdev_hold(dev, &dev_tracker, GFP_KERNEL); - rtnl_unlock(); - err = br_ioctl_call(net, netdev_priv(dev), cmd, ifr, NULL); - netdev_put(dev, &dev_tracker); - rtnl_lock(); - return err; - case SIOCSHWTSTAMP: err = net_hwtstamp_validate(ifr); if (err) @@ -575,8 +561,6 @@ int dev_ioctl(struct net *net, unsigned int cmd, struct ifreq *ifr, case SIOCBONDRELEASE: case SIOCBONDSETHWADDR: case SIOCBONDCHANGEACTIVE: - case SIOCBRADDIF: - case SIOCBRDELIF: case SIOCSHWTSTAMP: if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) return -EPERM; diff --git a/net/socket.c b/net/socket.c index bd438b89e698..701389e2f22b 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1151,12 +1151,10 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from) */ static DEFINE_MUTEX(br_ioctl_mutex); -static int (*br_ioctl_hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +static int (*br_ioctl_hook)(struct net *net, unsigned int cmd, void __user *uarg); -void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +void brioctl_set(int (*hook)(struct net *net, unsigned int cmd, void __user *uarg)) { mutex_lock(&br_ioctl_mutex); @@ -1165,8 +1163,7 @@ void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, } EXPORT_SYMBOL(brioctl_set); -int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg) +int br_ioctl_call(struct net *net, unsigned int cmd, void __user *uarg) { int err = -ENOPKG; @@ -1175,7 +1172,7 @@ int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, mutex_lock(&br_ioctl_mutex); if (br_ioctl_hook) - err = br_ioctl_hook(net, br, cmd, ifr, uarg); + err = br_ioctl_hook(net, cmd, uarg); mutex_unlock(&br_ioctl_mutex); return err; @@ -1272,7 +1269,9 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg) case SIOCSIFBR: case SIOCBRADDBR: case SIOCBRDELBR: - err = br_ioctl_call(net, NULL, cmd, NULL, argp); + case SIOCBRADDIF: + case SIOCBRDELIF: + err = br_ioctl_call(net, cmd, argp); break; case SIOCGIFVLAN: case SIOCSIFVLAN: @@ -3376,6 +3375,8 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGPGRP: case SIOCBRADDBR: case SIOCBRDELBR: + case SIOCBRADDIF: + case SIOCBRDELIF: case SIOCGIFVLAN: case SIOCSIFVLAN: case SIOCGSKNS: @@ -3415,8 +3416,6 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGIFPFLAGS: case SIOCGIFTXQLEN: case SIOCSIFTXQLEN: - case SIOCBRADDIF: - case SIOCBRDELIF: case SIOCGIFNAME: case SIOCSIFNAME: case SIOCGMIIPHY: -- 2.47.3

4 days, 17 hours

[PATCH 6.6] net: Remove RTNL dance for SIOCBRADDIF and SIOCBRDELIF.

by Thadeu Lima de Souza Cascardo

commit ed3ba9b6e280e14cc3148c1b226ba453f02fa76c upstream. SIOCBRDELIF is passed to dev_ioctl() first and later forwarded to br_ioctl_call(), which causes unnecessary RTNL dance and the splat below [0] under RTNL pressure. Let's say Thread A is trying to detach a device from a bridge and Thread B is trying to remove the bridge. In dev_ioctl(), Thread A bumps the bridge device's refcnt by netdev_hold() and releases RTNL because the following br_ioctl_call() also re-acquires RTNL. In the race window, Thread B could acquire RTNL and try to remove the bridge device. Then, rtnl_unlock() by Thread B will release RTNL and wait for netdev_put() by Thread A. Thread A, however, must hold RTNL after the unlock in dev_ifsioc(), which may take long under RTNL pressure, resulting in the splat by Thread B. Thread A (SIOCBRDELIF) Thread B (SIOCBRDELBR) ---------------------- ---------------------- sock_ioctl sock_ioctl `- sock_do_ioctl `- br_ioctl_call `- dev_ioctl `- br_ioctl_stub |- rtnl_lock | |- dev_ifsioc ' ' |- dev = __dev_get_by_name(...) |- netdev_hold(dev, ...) . / |- rtnl_unlock ------. | | |- br_ioctl_call `---> |- rtnl_lock Race | | `- br_ioctl_stub |- br_del_bridge Window | | | |- dev = __dev_get_by_name(...) | | | May take long | `- br_dev_delete(dev, ...) | | | under RTNL pressure | `- unregister_netdevice_queue(dev, ...) | | | | `- rtnl_unlock \ | |- rtnl_lock <-' `- netdev_run_todo | |- ... `- netdev_run_todo | `- rtnl_unlock |- __rtnl_unlock | |- netdev_wait_allrefs_any |- netdev_put(dev, ...) <----------------' Wait refcnt decrement and log splat below To avoid blocking SIOCBRDELBR unnecessarily, let's not call dev_ioctl() for SIOCBRADDIF and SIOCBRDELIF. In the dev_ioctl() path, we do the following: 1. Copy struct ifreq by get_user_ifreq in sock_do_ioctl() 2. Check CAP_NET_ADMIN in dev_ioctl() 3. Call dev_load() in dev_ioctl() 4. Fetch the master dev from ifr.ifr_name in dev_ifsioc() 3. can be done by request_module() in br_ioctl_call(), so we move 1., 2., and 4. to br_ioctl_stub(). Note that 2. is also checked later in add_del_if(), but it's better performed before RTNL. SIOCBRADDIF and SIOCBRDELIF have been processed in dev_ioctl() since the pre-git era, and there seems to be no specific reason to process them there. [0]: unregister_netdevice: waiting for wpan3 to become free. Usage count = 2 ref_tracker: wpan3@ffff8880662d8608 has 1/1 users at __netdev_tracker_alloc include/linux/netdevice.h:4282 [inline] netdev_hold include/linux/netdevice.h:4311 [inline] dev_ifsioc+0xc6a/0x1160 net/core/dev_ioctl.c:624 dev_ioctl+0x255/0x10c0 net/core/dev_ioctl.c:826 sock_do_ioctl+0x1ca/0x260 net/socket.c:1213 sock_ioctl+0x23a/0x6c0 net/socket.c:1318 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:906 [inline] __se_sys_ioctl fs/ioctl.c:892 [inline] __x64_sys_ioctl+0x1a4/0x210 fs/ioctl.c:892 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcb/0x250 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 893b19587534 ("net: bridge: fix ioctl locking") Reported-by: syzkaller <syzkaller(a)googlegroups.com> Reported-by: yan kang <kangyan91(a)outlook.com> Reported-by: yue sun <samsun1006219(a)gmail.com> Closes: https://lore.kernel.org/netdev/SY8P300MB0421225D54EB92762AE8F0F2A1D32@SY8P3… Signed-off-by: Kuniyuki Iwashima <kuniyu(a)amazon.com> Acked-by: Stanislav Fomichev <sdf(a)fomichev.me> Reviewed-by: Ido Schimmel <idosch(a)nvidia.com> Acked-by: Nikolay Aleksandrov <razor(a)blackwall.org> Link: https://patch.msgid.link/20250316192851.19781-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [cascardo: fixed conflict at dev_ifsioc] Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo(a)igalia.com> --- include/linux/if_bridge.h | 6 ++---- net/bridge/br_ioctl.c | 36 +++++++++++++++++++++++++++++++++--- net/bridge/br_private.h | 3 +-- net/core/dev_ioctl.c | 16 ---------------- net/socket.c | 19 +++++++++---------- 5 files changed, 45 insertions(+), 35 deletions(-) diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h index 3ff96ae31bf6..c5fe3b2a53e8 100644 --- a/include/linux/if_bridge.h +++ b/include/linux/if_bridge.h @@ -65,11 +65,9 @@ struct br_ip_list { #define BR_DEFAULT_AGEING_TIME (300 * HZ) struct net_bridge; -void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +void brioctl_set(int (*hook)(struct net *net, unsigned int cmd, void __user *uarg)); -int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg); +int br_ioctl_call(struct net *net, unsigned int cmd, void __user *uarg); #if IS_ENABLED(CONFIG_BRIDGE) && IS_ENABLED(CONFIG_BRIDGE_IGMP_SNOOPING) int br_multicast_list_adjacent(struct net_device *dev, diff --git a/net/bridge/br_ioctl.c b/net/bridge/br_ioctl.c index f213ed108361..6bc0a11f2ed3 100644 --- a/net/bridge/br_ioctl.c +++ b/net/bridge/br_ioctl.c @@ -394,10 +394,26 @@ static int old_deviceless(struct net *net, void __user *data) return -EOPNOTSUPP; } -int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg) +int br_ioctl_stub(struct net *net, unsigned int cmd, void __user *uarg) { int ret = -EOPNOTSUPP; + struct ifreq ifr; + + if (cmd == SIOCBRADDIF || cmd == SIOCBRDELIF) { + void __user *data; + char *colon; + + if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + + if (get_user_ifreq(&ifr, &data, uarg)) + return -EFAULT; + + ifr.ifr_name[IFNAMSIZ - 1] = 0; + colon = strchr(ifr.ifr_name, ':'); + if (colon) + *colon = 0; + } rtnl_lock(); @@ -430,7 +446,21 @@ int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, break; case SIOCBRADDIF: case SIOCBRDELIF: - ret = add_del_if(br, ifr->ifr_ifindex, cmd == SIOCBRADDIF); + { + struct net_device *dev; + + dev = __dev_get_by_name(net, ifr.ifr_name); + if (!dev || !netif_device_present(dev)) { + ret = -ENODEV; + break; + } + if (!netif_is_bridge_master(dev)) { + ret = -EOPNOTSUPP; + break; + } + + ret = add_del_if(netdev_priv(dev), ifr.ifr_ifindex, cmd == SIOCBRADDIF); + } break; } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index c8a4e3b39b0e..7427ee6264b4 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -947,8 +947,7 @@ br_port_get_check_rtnl(const struct net_device *dev) /* br_ioctl.c */ int br_dev_siocdevprivate(struct net_device *dev, struct ifreq *rq, void __user *data, int cmd); -int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg); +int br_ioctl_stub(struct net *net, unsigned int cmd, void __user *uarg); /* br_multicast.c */ #ifdef CONFIG_BRIDGE_IGMP_SNOOPING diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index b46aedc36939..541d42bf1e40 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -517,7 +517,6 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, int err; struct net_device *dev = __dev_get_by_name(net, ifr->ifr_name); const struct net_device_ops *ops; - netdevice_tracker dev_tracker; if (!dev) return -ENODEV; @@ -580,19 +579,6 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, case SIOCWANDEV: return dev_siocwandev(dev, &ifr->ifr_settings); - case SIOCBRADDIF: - case SIOCBRDELIF: - if (!netif_device_present(dev)) - return -ENODEV; - if (!netif_is_bridge_master(dev)) - return -EOPNOTSUPP; - netdev_hold(dev, &dev_tracker, GFP_KERNEL); - rtnl_unlock(); - err = br_ioctl_call(net, netdev_priv(dev), cmd, ifr, NULL); - netdev_put(dev, &dev_tracker); - rtnl_lock(); - return err; - case SIOCDEVPRIVATE ... SIOCDEVPRIVATE + 15: return dev_siocdevprivate(dev, ifr, data, cmd); @@ -773,8 +759,6 @@ int dev_ioctl(struct net *net, unsigned int cmd, struct ifreq *ifr, case SIOCBONDRELEASE: case SIOCBONDSETHWADDR: case SIOCBONDCHANGEACTIVE: - case SIOCBRADDIF: - case SIOCBRDELIF: case SIOCSHWTSTAMP: if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) return -EPERM; diff --git a/net/socket.c b/net/socket.c index bad58f23f307..94c38c73fc69 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1168,12 +1168,10 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from) */ static DEFINE_MUTEX(br_ioctl_mutex); -static int (*br_ioctl_hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +static int (*br_ioctl_hook)(struct net *net, unsigned int cmd, void __user *uarg); -void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +void brioctl_set(int (*hook)(struct net *net, unsigned int cmd, void __user *uarg)) { mutex_lock(&br_ioctl_mutex); @@ -1182,8 +1180,7 @@ void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, } EXPORT_SYMBOL(brioctl_set); -int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg) +int br_ioctl_call(struct net *net, unsigned int cmd, void __user *uarg) { int err = -ENOPKG; @@ -1192,7 +1189,7 @@ int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, mutex_lock(&br_ioctl_mutex); if (br_ioctl_hook) - err = br_ioctl_hook(net, br, cmd, ifr, uarg); + err = br_ioctl_hook(net, cmd, uarg); mutex_unlock(&br_ioctl_mutex); return err; @@ -1292,7 +1289,9 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg) case SIOCSIFBR: case SIOCBRADDBR: case SIOCBRDELBR: - err = br_ioctl_call(net, NULL, cmd, NULL, argp); + case SIOCBRADDIF: + case SIOCBRDELIF: + err = br_ioctl_call(net, cmd, argp); break; case SIOCGIFVLAN: case SIOCSIFVLAN: @@ -3454,6 +3453,8 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGPGRP: case SIOCBRADDBR: case SIOCBRDELBR: + case SIOCBRADDIF: + case SIOCBRDELIF: case SIOCGIFVLAN: case SIOCSIFVLAN: case SIOCGSKNS: @@ -3493,8 +3494,6 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGIFPFLAGS: case SIOCGIFTXQLEN: case SIOCSIFTXQLEN: - case SIOCBRADDIF: - case SIOCBRDELIF: case SIOCGIFNAME: case SIOCSIFNAME: case SIOCGMIIPHY: -- 2.47.3

4 days, 17 hours

[PATCH 6.12] net: Remove RTNL dance for SIOCBRADDIF and SIOCBRDELIF.

by Thadeu Lima de Souza Cascardo

commit ed3ba9b6e280e14cc3148c1b226ba453f02fa76c upstream. SIOCBRDELIF is passed to dev_ioctl() first and later forwarded to br_ioctl_call(), which causes unnecessary RTNL dance and the splat below [0] under RTNL pressure. Let's say Thread A is trying to detach a device from a bridge and Thread B is trying to remove the bridge. In dev_ioctl(), Thread A bumps the bridge device's refcnt by netdev_hold() and releases RTNL because the following br_ioctl_call() also re-acquires RTNL. In the race window, Thread B could acquire RTNL and try to remove the bridge device. Then, rtnl_unlock() by Thread B will release RTNL and wait for netdev_put() by Thread A. Thread A, however, must hold RTNL after the unlock in dev_ifsioc(), which may take long under RTNL pressure, resulting in the splat by Thread B. Thread A (SIOCBRDELIF) Thread B (SIOCBRDELBR) ---------------------- ---------------------- sock_ioctl sock_ioctl `- sock_do_ioctl `- br_ioctl_call `- dev_ioctl `- br_ioctl_stub |- rtnl_lock | |- dev_ifsioc ' ' |- dev = __dev_get_by_name(...) |- netdev_hold(dev, ...) . / |- rtnl_unlock ------. | | |- br_ioctl_call `---> |- rtnl_lock Race | | `- br_ioctl_stub |- br_del_bridge Window | | | |- dev = __dev_get_by_name(...) | | | May take long | `- br_dev_delete(dev, ...) | | | under RTNL pressure | `- unregister_netdevice_queue(dev, ...) | | | | `- rtnl_unlock \ | |- rtnl_lock <-' `- netdev_run_todo | |- ... `- netdev_run_todo | `- rtnl_unlock |- __rtnl_unlock | |- netdev_wait_allrefs_any |- netdev_put(dev, ...) <----------------' Wait refcnt decrement and log splat below To avoid blocking SIOCBRDELBR unnecessarily, let's not call dev_ioctl() for SIOCBRADDIF and SIOCBRDELIF. In the dev_ioctl() path, we do the following: 1. Copy struct ifreq by get_user_ifreq in sock_do_ioctl() 2. Check CAP_NET_ADMIN in dev_ioctl() 3. Call dev_load() in dev_ioctl() 4. Fetch the master dev from ifr.ifr_name in dev_ifsioc() 3. can be done by request_module() in br_ioctl_call(), so we move 1., 2., and 4. to br_ioctl_stub(). Note that 2. is also checked later in add_del_if(), but it's better performed before RTNL. SIOCBRADDIF and SIOCBRDELIF have been processed in dev_ioctl() since the pre-git era, and there seems to be no specific reason to process them there. [0]: unregister_netdevice: waiting for wpan3 to become free. Usage count = 2 ref_tracker: wpan3@ffff8880662d8608 has 1/1 users at __netdev_tracker_alloc include/linux/netdevice.h:4282 [inline] netdev_hold include/linux/netdevice.h:4311 [inline] dev_ifsioc+0xc6a/0x1160 net/core/dev_ioctl.c:624 dev_ioctl+0x255/0x10c0 net/core/dev_ioctl.c:826 sock_do_ioctl+0x1ca/0x260 net/socket.c:1213 sock_ioctl+0x23a/0x6c0 net/socket.c:1318 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:906 [inline] __se_sys_ioctl fs/ioctl.c:892 [inline] __x64_sys_ioctl+0x1a4/0x210 fs/ioctl.c:892 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcb/0x250 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 893b19587534 ("net: bridge: fix ioctl locking") Reported-by: syzkaller <syzkaller(a)googlegroups.com> Reported-by: yan kang <kangyan91(a)outlook.com> Reported-by: yue sun <samsun1006219(a)gmail.com> Closes: https://lore.kernel.org/netdev/SY8P300MB0421225D54EB92762AE8F0F2A1D32@SY8P3… Signed-off-by: Kuniyuki Iwashima <kuniyu(a)amazon.com> Acked-by: Stanislav Fomichev <sdf(a)fomichev.me> Reviewed-by: Ido Schimmel <idosch(a)nvidia.com> Acked-by: Nikolay Aleksandrov <razor(a)blackwall.org> Link: https://patch.msgid.link/20250316192851.19781-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [cascardo: fixed conflict at dev_ifsioc] Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo(a)igalia.com> --- include/linux/if_bridge.h | 6 ++---- net/bridge/br_ioctl.c | 36 +++++++++++++++++++++++++++++++++--- net/bridge/br_private.h | 3 +-- net/core/dev_ioctl.c | 16 ---------------- net/socket.c | 19 +++++++++---------- 5 files changed, 45 insertions(+), 35 deletions(-) diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h index 3ff96ae31bf6..c5fe3b2a53e8 100644 --- a/include/linux/if_bridge.h +++ b/include/linux/if_bridge.h @@ -65,11 +65,9 @@ struct br_ip_list { #define BR_DEFAULT_AGEING_TIME (300 * HZ) struct net_bridge; -void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +void brioctl_set(int (*hook)(struct net *net, unsigned int cmd, void __user *uarg)); -int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg); +int br_ioctl_call(struct net *net, unsigned int cmd, void __user *uarg); #if IS_ENABLED(CONFIG_BRIDGE) && IS_ENABLED(CONFIG_BRIDGE_IGMP_SNOOPING) int br_multicast_list_adjacent(struct net_device *dev, diff --git a/net/bridge/br_ioctl.c b/net/bridge/br_ioctl.c index f213ed108361..6bc0a11f2ed3 100644 --- a/net/bridge/br_ioctl.c +++ b/net/bridge/br_ioctl.c @@ -394,10 +394,26 @@ static int old_deviceless(struct net *net, void __user *data) return -EOPNOTSUPP; } -int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg) +int br_ioctl_stub(struct net *net, unsigned int cmd, void __user *uarg) { int ret = -EOPNOTSUPP; + struct ifreq ifr; + + if (cmd == SIOCBRADDIF || cmd == SIOCBRDELIF) { + void __user *data; + char *colon; + + if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + + if (get_user_ifreq(&ifr, &data, uarg)) + return -EFAULT; + + ifr.ifr_name[IFNAMSIZ - 1] = 0; + colon = strchr(ifr.ifr_name, ':'); + if (colon) + *colon = 0; + } rtnl_lock(); @@ -430,7 +446,21 @@ int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, break; case SIOCBRADDIF: case SIOCBRDELIF: - ret = add_del_if(br, ifr->ifr_ifindex, cmd == SIOCBRADDIF); + { + struct net_device *dev; + + dev = __dev_get_by_name(net, ifr.ifr_name); + if (!dev || !netif_device_present(dev)) { + ret = -ENODEV; + break; + } + if (!netif_is_bridge_master(dev)) { + ret = -EOPNOTSUPP; + break; + } + + ret = add_del_if(netdev_priv(dev), ifr.ifr_ifindex, cmd == SIOCBRADDIF); + } break; } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 741b0b8c4bab..8c47af5c1dab 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -952,8 +952,7 @@ br_port_get_check_rtnl(const struct net_device *dev) /* br_ioctl.c */ int br_dev_siocdevprivate(struct net_device *dev, struct ifreq *rq, void __user *data, int cmd); -int br_ioctl_stub(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg); +int br_ioctl_stub(struct net *net, unsigned int cmd, void __user *uarg); /* br_multicast.c */ #ifdef CONFIG_BRIDGE_IGMP_SNOOPING diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 473c437b6b53..81cd8df798c0 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -514,7 +514,6 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, int err; struct net_device *dev = __dev_get_by_name(net, ifr->ifr_name); const struct net_device_ops *ops; - netdevice_tracker dev_tracker; if (!dev) return -ENODEV; @@ -577,19 +576,6 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, case SIOCWANDEV: return dev_siocwandev(dev, &ifr->ifr_settings); - case SIOCBRADDIF: - case SIOCBRDELIF: - if (!netif_device_present(dev)) - return -ENODEV; - if (!netif_is_bridge_master(dev)) - return -EOPNOTSUPP; - netdev_hold(dev, &dev_tracker, GFP_KERNEL); - rtnl_unlock(); - err = br_ioctl_call(net, netdev_priv(dev), cmd, ifr, NULL); - netdev_put(dev, &dev_tracker); - rtnl_lock(); - return err; - case SIOCDEVPRIVATE ... SIOCDEVPRIVATE + 15: return dev_siocdevprivate(dev, ifr, data, cmd); @@ -770,8 +756,6 @@ int dev_ioctl(struct net *net, unsigned int cmd, struct ifreq *ifr, case SIOCBONDRELEASE: case SIOCBONDSETHWADDR: case SIOCBONDCHANGEACTIVE: - case SIOCBRADDIF: - case SIOCBRDELIF: case SIOCSHWTSTAMP: if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) return -EPERM; diff --git a/net/socket.c b/net/socket.c index 042451f01c65..a0f6f8b3376d 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1173,12 +1173,10 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from) */ static DEFINE_MUTEX(br_ioctl_mutex); -static int (*br_ioctl_hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +static int (*br_ioctl_hook)(struct net *net, unsigned int cmd, void __user *uarg); -void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, - unsigned int cmd, struct ifreq *ifr, +void brioctl_set(int (*hook)(struct net *net, unsigned int cmd, void __user *uarg)) { mutex_lock(&br_ioctl_mutex); @@ -1187,8 +1185,7 @@ void brioctl_set(int (*hook)(struct net *net, struct net_bridge *br, } EXPORT_SYMBOL(brioctl_set); -int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, - struct ifreq *ifr, void __user *uarg) +int br_ioctl_call(struct net *net, unsigned int cmd, void __user *uarg) { int err = -ENOPKG; @@ -1197,7 +1194,7 @@ int br_ioctl_call(struct net *net, struct net_bridge *br, unsigned int cmd, mutex_lock(&br_ioctl_mutex); if (br_ioctl_hook) - err = br_ioctl_hook(net, br, cmd, ifr, uarg); + err = br_ioctl_hook(net, cmd, uarg); mutex_unlock(&br_ioctl_mutex); return err; @@ -1297,7 +1294,9 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg) case SIOCSIFBR: case SIOCBRADDBR: case SIOCBRDELBR: - err = br_ioctl_call(net, NULL, cmd, NULL, argp); + case SIOCBRADDIF: + case SIOCBRDELIF: + err = br_ioctl_call(net, cmd, argp); break; case SIOCGIFVLAN: case SIOCSIFVLAN: @@ -3466,6 +3465,8 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGPGRP: case SIOCBRADDBR: case SIOCBRDELBR: + case SIOCBRADDIF: + case SIOCBRDELIF: case SIOCGIFVLAN: case SIOCSIFVLAN: case SIOCGSKNS: @@ -3505,8 +3506,6 @@ static int compat_sock_ioctl_trans(struct file *file, struct socket *sock, case SIOCGIFPFLAGS: case SIOCGIFTXQLEN: case SIOCSIFTXQLEN: - case SIOCBRADDIF: - case SIOCBRDELIF: case SIOCGIFNAME: case SIOCSIFNAME: case SIOCGMIIPHY: -- 2.47.3

4 days, 17 hours

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror