linux/drivers/pci/remove.c

196 lines
4.3 KiB

PCI/pwrctrl: Unregister platform device only if one actually exists If a PCI device has an associated device_node with power supplies, pci_bus_add_device() creates platform devices for use by pwrctrl. When the PCI device is removed, pci_stop_dev() uses of_find_device_by_node() to locate the related platform device, then unregisters it. But when we remove a PCI device with no associated device node, dev_of_node(dev) is NULL, and of_find_device_by_node(NULL) returns the first device with "dev->of_node == NULL". The result is that we (a) mistakenly unregister a completely unrelated platform device, leading to issues like the first trace below, and (b) dereference the NULL pointer from dev_of_node() when clearing OF_POPULATED, as in the second trace. Unregister a platform device only if there is one associated with this PCI device. This resolves issues seen when doing: # echo 1 > /sys/bus/pci/devices/.../remove Sample issue from unregistering the wrong platform device: WARNING: CPU: 0 PID: 5095 at drivers/regulator/core.c:5885 regulator_unregister+0x140/0x160 Call trace: regulator_unregister+0x140/0x160 devm_rdev_release+0x1c/0x30 release_nodes+0x68/0x100 devres_release_all+0x98/0xf8 device_unbind_cleanup+0x20/0x70 device_release_driver_internal+0x1f4/0x240 device_release_driver+0x20/0x40 bus_remove_device+0xd8/0x170 device_del+0x154/0x380 device_unregister+0x28/0x88 of_device_unregister+0x1c/0x30 pci_stop_bus_device+0x154/0x1b0 pci_stop_and_remove_bus_device_locked+0x28/0x48 remove_store+0xa0/0xb8 dev_attr_store+0x20/0x40 sysfs_kf_write+0x4c/0x68 Later NULL pointer dereference for of_node_clear_flag(NULL, OF_POPULATED): Unable to handle kernel NULL pointer dereference at virtual address 00000000000000c0 Call trace: pci_stop_bus_device+0x190/0x1b0 pci_stop_and_remove_bus_device_locked+0x28/0x48 remove_store+0xa0/0xb8 dev_attr_store+0x20/0x40 sysfs_kf_write+0x4c/0x68 Link: https://lore.kernel.org/r/20241126210443.4052876-1-briannorris@chromium.org Fixes: 681725afb6b9 ("PCI/pwrctl: Remove pwrctl device without iterating over all children of pwrctl parent") Reported-by: Saurabh Sengar <ssengar@linux.microsoft.com> Closes: https://lore.kernel.org/r/1732890621-19656-1-git-send-email-ssengar@linux.microsoft.com Signed-off-by: Brian Norris <briannorris@chromium.org> [bhelgaas: commit log] Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2024-11-26 13:04:34 -08:00
PCI/pwrctrl: Unregister platform device only if one actually exists If a PCI device has an associated device_node with power supplies, pci_bus_add_device() creates platform devices for use by pwrctrl. When the PCI device is removed, pci_stop_dev() uses of_find_device_by_node() to locate the related platform device, then unregisters it. But when we remove a PCI device with no associated device node, dev_of_node(dev) is NULL, and of_find_device_by_node(NULL) returns the first device with "dev->of_node == NULL". The result is that we (a) mistakenly unregister a completely unrelated platform device, leading to issues like the first trace below, and (b) dereference the NULL pointer from dev_of_node() when clearing OF_POPULATED, as in the second trace. Unregister a platform device only if there is one associated with this PCI device. This resolves issues seen when doing: # echo 1 > /sys/bus/pci/devices/.../remove Sample issue from unregistering the wrong platform device: WARNING: CPU: 0 PID: 5095 at drivers/regulator/core.c:5885 regulator_unregister+0x140/0x160 Call trace: regulator_unregister+0x140/0x160 devm_rdev_release+0x1c/0x30 release_nodes+0x68/0x100 devres_release_all+0x98/0xf8 device_unbind_cleanup+0x20/0x70 device_release_driver_internal+0x1f4/0x240 device_release_driver+0x20/0x40 bus_remove_device+0xd8/0x170 device_del+0x154/0x380 device_unregister+0x28/0x88 of_device_unregister+0x1c/0x30 pci_stop_bus_device+0x154/0x1b0 pci_stop_and_remove_bus_device_locked+0x28/0x48 remove_store+0xa0/0xb8 dev_attr_store+0x20/0x40 sysfs_kf_write+0x4c/0x68 Later NULL pointer dereference for of_node_clear_flag(NULL, OF_POPULATED): Unable to handle kernel NULL pointer dereference at virtual address 00000000000000c0 Call trace: pci_stop_bus_device+0x190/0x1b0 pci_stop_and_remove_bus_device_locked+0x28/0x48 remove_store+0xa0/0xb8 dev_attr_store+0x20/0x40 sysfs_kf_write+0x4c/0x68 Link: https://lore.kernel.org/r/20241126210443.4052876-1-briannorris@chromium.org Fixes: 681725afb6b9 ("PCI/pwrctl: Remove pwrctl device without iterating over all children of pwrctl parent") Reported-by: Saurabh Sengar <ssengar@linux.microsoft.com> Closes: https://lore.kernel.org/r/1732890621-19656-1-git-send-email-ssengar@linux.microsoft.com Signed-off-by: Brian Norris <briannorris@chromium.org> [bhelgaas: commit log] Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2024-11-26 13:04:34 -08:00
PCI/pwrctrl: Unregister platform device only if one actually exists If a PCI device has an associated device_node with power supplies, pci_bus_add_device() creates platform devices for use by pwrctrl. When the PCI device is removed, pci_stop_dev() uses of_find_device_by_node() to locate the related platform device, then unregisters it. But when we remove a PCI device with no associated device node, dev_of_node(dev) is NULL, and of_find_device_by_node(NULL) returns the first device with "dev->of_node == NULL". The result is that we (a) mistakenly unregister a completely unrelated platform device, leading to issues like the first trace below, and (b) dereference the NULL pointer from dev_of_node() when clearing OF_POPULATED, as in the second trace. Unregister a platform device only if there is one associated with this PCI device. This resolves issues seen when doing: # echo 1 > /sys/bus/pci/devices/.../remove Sample issue from unregistering the wrong platform device: WARNING: CPU: 0 PID: 5095 at drivers/regulator/core.c:5885 regulator_unregister+0x140/0x160 Call trace: regulator_unregister+0x140/0x160 devm_rdev_release+0x1c/0x30 release_nodes+0x68/0x100 devres_release_all+0x98/0xf8 device_unbind_cleanup+0x20/0x70 device_release_driver_internal+0x1f4/0x240 device_release_driver+0x20/0x40 bus_remove_device+0xd8/0x170 device_del+0x154/0x380 device_unregister+0x28/0x88 of_device_unregister+0x1c/0x30 pci_stop_bus_device+0x154/0x1b0 pci_stop_and_remove_bus_device_locked+0x28/0x48 remove_store+0xa0/0xb8 dev_attr_store+0x20/0x40 sysfs_kf_write+0x4c/0x68 Later NULL pointer dereference for of_node_clear_flag(NULL, OF_POPULATED): Unable to handle kernel NULL pointer dereference at virtual address 00000000000000c0 Call trace: pci_stop_bus_device+0x190/0x1b0 pci_stop_and_remove_bus_device_locked+0x28/0x48 remove_store+0xa0/0xb8 dev_attr_store+0x20/0x40 sysfs_kf_write+0x4c/0x68 Link: https://lore.kernel.org/r/20241126210443.4052876-1-briannorris@chromium.org Fixes: 681725afb6b9 ("PCI/pwrctl: Remove pwrctl device without iterating over all children of pwrctl parent") Reported-by: Saurabh Sengar <ssengar@linux.microsoft.com> Closes: https://lore.kernel.org/r/1732890621-19656-1-git-send-email-ssengar@linux.microsoft.com Signed-off-by: Brian Norris <briannorris@chromium.org> [bhelgaas: commit log] Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2024-11-26 13:04:34 -08:00
PCI/NPEM: Add Native PCIe Enclosure Management support Native PCIe Enclosure Management (NPEM, PCIe r6.1 sec 6.28) allows managing LEDs in storage enclosures. NPEM is indication oriented and it does not give direct access to LEDs. Although each indication *could* represent an individual LED, multiple indications could also be represented as a single, multi-color LED or a single LED blinking in a specific interval. The specification leaves that open. Each enabled indication (capability register bit on) is represented as a ledclass_dev which can be controlled through sysfs. For every ledclass device only 2 brightness states are allowed: LED_ON (1) or LED_OFF (0). This corresponds to the NPEM control register (Indication bit on/off). Ledclass devices appear in sysfs as child devices (subdirectory) of PCI device which has an NPEM Extended Capability and indication is enabled in NPEM capability register. For example, these are LEDs created for pcieport "10000:02:05.0" on my setup: leds/ ├── 10000:02:05.0:enclosure:fail ├── 10000:02:05.0:enclosure:locate ├── 10000:02:05.0:enclosure:ok └── 10000:02:05.0:enclosure:rebuild They can be also found in "/sys/class/leds" directory. The parent PCIe device domain/bus/device/function address is used to guarantee uniqueness across leds subsystem. To enable/disable a "fail" indication, the "brightness" file can be edited: echo 1 > ./leds/10000:02:05.0:enclosure:fail/brightness echo 0 > ./leds/10000:02:05.0:enclosure:fail/brightness PCIe r6.1, sec 7.9.19.2 defines the possible indications. Multiple indications for same parent PCIe device can conflict and hardware may update them when processing new request. To avoid issues, driver refresh all indications by reading back control register. This driver expects to be the exclusive NPEM extended capability manager. It waits up to 1 second after imposing new request, it doesn't verify if controller is busy before write, and it assumes the mutex lock gives protection from concurrent updates. If _DSM LED management is available, we assume the platform may be using NPEM for its own purposes (see PCI Firmware Spec r3.3 sec 4.7), so the driver does not use NPEM. A future patch will add _DSM support; an info message notes whether NPEM or _DSM is being used. NPEM is a PCIe extended capability so it should be registered in pcie_init_capabilities() but it is not possible due to LED dependency. The parent pci_device must be added earlier for led_classdev_register() to be successful. NPEM does not require configuration on kernel side, so it is safe to register LED devices later. Link: https://lore.kernel.org/r/20240904104848.23480-3-mariusz.tkaczyk@linux.intel.com Suggested-by: Lukas Wunner <lukas@wunner.de> Signed-off-by: Mariusz Tkaczyk <mariusz.tkaczyk@linux.intel.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Tested-by: Stuart Hayes <stuart.w.hayes@gmail.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
2024-09-04 12:48:47 +02:00
PCI: Move device_del() from pci_stop_dev() to pci_destroy_dev() After commit bcdde7e221a8 (sysfs: make __sysfs_remove_dir() recursive) I'm seeing traces analogous to the one below in Thunderbolt testing: WARNING: CPU: 3 PID: 76 at /scratch/rafael/work/linux-pm/fs/sysfs/group.c:214 sysfs_remove_group+0x59/0xe0() sysfs group ffffffff81c6c500 not found for kobject '0000:08' Modules linked in: ... CPU: 3 PID: 76 Comm: kworker/u16:7 Not tainted 3.13.0-rc1+ #76 Hardware name: Acer Aspire S5-391/Venus , BIOS V1.02 05/29/2012 Workqueue: kacpi_hotplug acpi_hotplug_work_fn 0000000000000009 ffff8801644b9ac8 ffffffff816b23bf 0000000000000007 ffff8801644b9b18 ffff8801644b9b08 ffffffff81046607 ffff88016925b800 0000000000000000 ffffffff81c6c500 ffff88016924f928 ffff88016924f800 Call Trace: [<ffffffff816b23bf>] dump_stack+0x4e/0x71 [<ffffffff81046607>] warn_slowpath_common+0x87/0xb0 [<ffffffff810466d1>] warn_slowpath_fmt+0x41/0x50 [<ffffffff811e42ef>] ? sysfs_get_dirent_ns+0x6f/0x80 [<ffffffff811e5389>] sysfs_remove_group+0x59/0xe0 [<ffffffff8149f00b>] dpm_sysfs_remove+0x3b/0x50 [<ffffffff81495818>] device_del+0x58/0x1c0 [<ffffffff814959c8>] device_unregister+0x48/0x60 [<ffffffff813254fe>] pci_remove_bus+0x6e/0x80 [<ffffffff81325548>] pci_remove_bus_device+0x38/0x110 [<ffffffff8132555d>] pci_remove_bus_device+0x4d/0x110 [<ffffffff81325639>] pci_stop_and_remove_bus_device+0x19/0x20 [<ffffffff813418d0>] disable_slot+0x20/0xe0 [<ffffffff81341a38>] acpiphp_check_bridge+0xa8/0xd0 [<ffffffff813427ad>] hotplug_event+0x17d/0x220 [<ffffffff81342880>] hotplug_event_work+0x30/0x70 [<ffffffff8136d665>] acpi_hotplug_work_fn+0x18/0x24 [<ffffffff81061331>] process_one_work+0x261/0x450 [<ffffffff81061a7e>] worker_thread+0x21e/0x370 [<ffffffff81061860>] ? rescuer_thread+0x300/0x300 [<ffffffff81068342>] kthread+0xd2/0xe0 [<ffffffff81068270>] ? flush_kthread_worker+0x70/0x70 [<ffffffff816c19bc>] ret_from_fork+0x7c/0xb0 [<ffffffff81068270>] ? flush_kthread_worker+0x70/0x70 (Mika Westerberg sees them too in his tests). Some investigation documented in kernel bug #65281 led me to the conclusion that the source of the problem is the device_del() in pci_stop_dev() as it now causes the sysfs directory of the device to be removed recursively along with all of its subdirectories. That includes the sysfs directory of the device's subordinate bus (dev->subordinate) and its "power" group. Consequently, when pci_remove_bus() is called for dev->subordinate in pci_remove_bus_device(), it calls device_unregister(&bus->dev), but at this point the sysfs directory of bus->dev doesn't exist any more and its "power" group doesn't exist either. Thus, when dpm_sysfs_remove() called from device_del() tries to remove that group, it triggers the above warning. That indicates a logical mistake in the design of pci_stop_and_remove_bus_device(), which causes bus device objects to be left behind their parents (bridge device objects) and can be fixed by moving the device_del() from pci_stop_dev() into pci_destroy_dev(), so pci_remove_bus() can be called for the device's subordinate bus before the device itself is unregistered from the hierarchy. Still, the driver, if any, should be detached from the device in pci_stop_dev(), so use device_release_driver() directly from there. References: https://bugzilla.kernel.org/show_bug.cgi?id=65281#c6 Reported-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2013-12-18 13:53:32 -07:00
PCI/DOE: Create mailboxes on device enumeration Currently a DOE instance cannot be shared by multiple drivers because each driver creates its own pci_doe_mb struct for a given DOE instance. For the same reason a DOE instance cannot be shared between the PCI core and a driver. Moreover, finding out which protocols a DOE instance supports requires creating a pci_doe_mb for it. If a device has multiple DOE instances, a driver looking for a specific protocol may need to create a pci_doe_mb for each of the device's DOE instances and then destroy those which do not support the desired protocol. That's obviously an inefficient way to do things. Overcome these issues by creating mailboxes in the PCI core on device enumeration. Provide a pci_find_doe_mailbox() API call to allow drivers to get a pci_doe_mb for a given (pci_dev, vendor, protocol) triple. This API is modeled after pci_find_capability() and can later be amended with a pci_find_next_doe_mailbox() call to iterate over all mailboxes of a given pci_dev which support a specific protocol. On removal, destroy the mailboxes in pci_destroy_dev(), after the driver is unbound. This allows drivers to use DOE in their ->remove() hook. On surprise removal, cancel ongoing DOE exchanges and prevent new ones from being scheduled. Thereby ensure that a hot-removed device doesn't needlessly wait for a running exchange to time out. Tested-by: Ira Weiny <ira.weiny@intel.com> Signed-off-by: Lukas Wunner <lukas@wunner.de> Reviewed-by: Ming Li <ming4.li@intel.com> Reviewed-by: Ira Weiny <ira.weiny@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Acked-by: Bjorn Helgaas <bhelgaas@google.com> Link: https://lore.kernel.org/r/40a6f973f72ef283d79dd55e7e6fddc7481199af.1678543498.git.lukas@wunner.de Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2023-03-11 15:40:12 +01:00
2018-09-04 12:34:18 -05:00
PCI: Fix use-after-free in pci_bus_release_domain_nr() Commit c14f7ccc9f5d ("PCI: Assign PCI domain IDs by ida_alloc()") introduced a use-after-free bug in the bus removal cleanup. The issue was found with kfence: [ 19.293351] BUG: KFENCE: use-after-free read in pci_bus_release_domain_nr+0x10/0x70 [ 19.302817] Use-after-free read at 0x000000007f3b80eb (in kfence-#115): [ 19.309677] pci_bus_release_domain_nr+0x10/0x70 [ 19.309691] dw_pcie_host_deinit+0x28/0x78 [ 19.309702] tegra_pcie_deinit_controller+0x1c/0x38 [pcie_tegra194] [ 19.309734] tegra_pcie_dw_probe+0x648/0xb28 [pcie_tegra194] [ 19.309752] platform_probe+0x90/0xd8 ... [ 19.311457] kfence-#115: 0x00000000063a155a-0x00000000ba698da8, size=1072, cache=kmalloc-2k [ 19.311469] allocated by task 96 on cpu 10 at 19.279323s: [ 19.311562] __kmem_cache_alloc_node+0x260/0x278 [ 19.311571] kmalloc_trace+0x24/0x30 [ 19.311580] pci_alloc_bus+0x24/0xa0 [ 19.311590] pci_register_host_bridge+0x48/0x4b8 [ 19.311601] pci_scan_root_bus_bridge+0xc0/0xe8 [ 19.311613] pci_host_probe+0x18/0xc0 [ 19.311623] dw_pcie_host_init+0x2c0/0x568 [ 19.311630] tegra_pcie_dw_probe+0x610/0xb28 [pcie_tegra194] [ 19.311647] platform_probe+0x90/0xd8 ... [ 19.311782] freed by task 96 on cpu 10 at 19.285833s: [ 19.311799] release_pcibus_dev+0x30/0x40 [ 19.311808] device_release+0x30/0x90 [ 19.311814] kobject_put+0xa8/0x120 [ 19.311832] device_unregister+0x20/0x30 [ 19.311839] pci_remove_bus+0x78/0x88 [ 19.311850] pci_remove_root_bus+0x5c/0x98 [ 19.311860] dw_pcie_host_deinit+0x28/0x78 [ 19.311866] tegra_pcie_deinit_controller+0x1c/0x38 [pcie_tegra194] [ 19.311883] tegra_pcie_dw_probe+0x648/0xb28 [pcie_tegra194] [ 19.311900] platform_probe+0x90/0xd8 ... [ 19.313579] CPU: 10 PID: 96 Comm: kworker/u24:2 Not tainted 6.2.0 #4 [ 19.320171] Hardware name: /, BIOS 1.0-d7fb19b 08/10/2022 [ 19.325852] Workqueue: events_unbound deferred_probe_work_func The stack trace is a bit misleading as dw_pcie_host_deinit() doesn't directly call pci_bus_release_domain_nr(). The issue turns out to be in pci_remove_root_bus() which first calls pci_remove_bus() which frees the struct pci_bus when its struct device is released. Then pci_bus_release_domain_nr() is called and accesses the freed struct pci_bus. Reordering these fixes the issue. Fixes: c14f7ccc9f5d ("PCI: Assign PCI domain IDs by ida_alloc()") Link: https://lore.kernel.org/r/20230329123835.2724518-1-robh@kernel.org Link: https://lore.kernel.org/r/b529cb69-0602-9eed-fc02-2f068707a006@nvidia.com Reported-by: Jon Hunter <jonathanh@nvidia.com> Tested-by: Jon Hunter <jonathanh@nvidia.com> Signed-off-by: Rob Herring <robh@kernel.org> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com> Cc: stable@vger.kernel.org # v6.2+ Cc: Pali Rohár <pali@kernel.org>
2023-03-29 07:38:35 -05:00
PCI: Fix pci_host_bridge struct device release/free handling The PCI code has several paths where the struct pci_host_bridge is freed directly. This is wrong because it contains a struct device which is refcounted and should be freed using put_device(). This can result in use-after-free errors. I think this problem has existed since 2012 with commit 7b5436635800 ("PCI: add generic device into pci_host_bridge struct"). It generally hasn't mattered as most host bridge drivers are still built-in and can't unbind. The problem is a struct device should never be freed directly once device_initialize() is called and a ref is held, but that doesn't happen until pci_register_host_bridge(). There's then a window between allocating the host bridge and pci_register_host_bridge() where kfree should be used. This is fragile and requires callers to do the right thing. To fix this, we need to split device_register() into device_initialize() and device_add() calls, so that the host bridge struct is always freed by using a put_device(). devm_pci_alloc_host_bridge() is using devm_kzalloc() to allocate struct pci_host_bridge which will be freed directly. Instead, we can use a custom devres action to call put_device(). Link: https://lore.kernel.org/r/20200513223859.11295-2-robh@kernel.org Reported-by: Anders Roxell <anders.roxell@linaro.org> Tested-by: Anders Roxell <anders.roxell@linaro.org> Signed-off-by: Rob Herring <robh@kernel.org> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Reviewed-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com> Acked-by: Arnd Bergmann <arnd@arndb.de>
2020-05-13 17:38:59 -05:00