kernel/linux.git/drivers/pci/hotplug, branch v5.4.50

PCI: pciehp: Prevent deadlock on disconnect

2020-04-29T14:33:04+00:00

[ Upstream commit 87d0f2a5536fdf5053a6d341880f96135549a644 ] This addresses deadlocks in these common cases in hierarchies containing two switches: - All involved ports are runtime suspended and they are unplugged. This can happen easily if the drivers involved automatically enable runtime PM (xHCI for example does that). - System is suspended (e.g., closing the lid on a laptop) with a dock + something else connected, and the dock is unplugged while suspended. These cases lead to the following deadlock: INFO: task irq/126-pciehp:198 blocked for more than 120 seconds. irq/126-pciehp D 0 198 2 0x80000000 Call Trace: schedule+0x2c/0x80 schedule_timeout+0x246/0x350 wait_for_completion+0xb7/0x140 kthread_stop+0x49/0x110 free_irq+0x32/0x70 pcie_shutdown_notification+0x2f/0x50 pciehp_remove+0x27/0x50 pcie_port_remove_service+0x36/0x50 device_release_driver+0x12/0x20 bus_remove_device+0xec/0x160 device_del+0x13b/0x350 device_unregister+0x1a/0x60 remove_iter+0x1e/0x30 device_for_each_child+0x56/0x90 pcie_port_device_remove+0x22/0x40 pcie_portdrv_remove+0x20/0x60 pci_device_remove+0x3e/0xc0 device_release_driver_internal+0x18c/0x250 device_release_driver+0x12/0x20 pci_stop_bus_device+0x6f/0x90 pci_stop_bus_device+0x31/0x90 pci_stop_and_remove_bus_device+0x12/0x20 pciehp_unconfigure_device+0x88/0x140 pciehp_disable_slot+0x6a/0x110 pciehp_handle_presence_or_link_change+0x263/0x400 pciehp_ist+0x1c9/0x1d0 irq_thread_fn+0x24/0x60 irq_thread+0xeb/0x190 kthread+0x120/0x140 INFO: task irq/190-pciehp:2288 blocked for more than 120 seconds. irq/190-pciehp D 0 2288 2 0x80000000 Call Trace: __schedule+0x2a2/0x880 schedule+0x2c/0x80 schedule_preempt_disabled+0xe/0x10 mutex_lock+0x2c/0x30 pci_lock_rescan_remove+0x15/0x20 pciehp_unconfigure_device+0x4d/0x140 pciehp_disable_slot+0x6a/0x110 pciehp_handle_presence_or_link_change+0x263/0x400 pciehp_ist+0x1c9/0x1d0 irq_thread_fn+0x24/0x60 irq_thread+0xeb/0x190 kthread+0x120/0x140 What happens here is that the whole hierarchy is runtime resumed and the parent PCIe downstream port, which got the hot-remove event, starts removing devices below it, taking pci_lock_rescan_remove() lock. When the child PCIe port is runtime resumed it calls pciehp_check_presence() which ends up calling pciehp_card_present() and pciehp_check_link_active(). Both of these use pcie_capability_read_word(), which notices that the underlying device is already gone and returns PCIBIOS_DEVICE_NOT_FOUND with the capability value set to 0. When pciehp gets this value it thinks that its child device is also hot-removed and schedules its IRQ thread to handle the event. The deadlock happens when the child's IRQ thread runs and tries to acquire pci_lock_rescan_remove() which is already taken by the parent and the parent waits for the child's IRQ thread to finish. Prevent this from happening by checking the return value of pcie_capability_read_word() and if it is PCIBIOS_DEVICE_NOT_FOUND stop performing any hot-removal activities. [bhelgaas: add common scenarios to commit log] Link: https://lore.kernel.org/r/20191029170022.57528-2-mika.westerberg@linux.intel.com Tested-by: Kai-Heng Feng Signed-off-by: Mika Westerberg Signed-off-by: Bjorn Helgaas Signed-off-by: Sasha Levin

PCI: pciehp: Fix indefinite wait on sysfs requests

2020-04-17T08:50:10+00:00

commit 3e487d2e4aa466decd226353755c9d423e8fbacc upstream. David Hoyer reports that powering pciehp slots up or down via sysfs may hang: The call to wait_event() in pciehp_sysfs_enable_slot() and _disable_slot() does not return because ctrl->ist_running remains true. This flag, which was introduced by commit 157c1062fcd8 ("PCI: pciehp: Avoid returning prematurely from sysfs requests"), signifies that the IRQ thread pciehp_ist() is running. It is set to true at the top of pciehp_ist() and reset to false at the end. However there are two additional return statements in pciehp_ist() before which the commit neglected to reset the flag to false and wake up waiters for the flag. That omission opens up the following race when powering up the slot: * pciehp_ist() runs because a PCI_EXP_SLTSTA_PDC event was requested by pciehp_sysfs_enable_slot() * pciehp_ist() turns on slot power via the following call stack: pciehp_handle_presence_or_link_change() -> pciehp_enable_slot() -> __pciehp_enable_slot() -> board_added() -> pciehp_power_on_slot() * after slot power is turned on, the link comes up, resulting in a PCI_EXP_SLTSTA_DLLSC event * the IRQ handler pciehp_isr() stores the event in ctrl->pending_events and returns IRQ_WAKE_THREAD * the IRQ thread is already woken (it's bringing up the slot), but the genirq code remembers to re-run the IRQ thread after it has finished (such that it can deal with the new event) by setting IRQTF_RUNTHREAD via __handle_irq_event_percpu() -> __irq_wake_thread() * the IRQ thread removes PCI_EXP_SLTSTA_DLLSC from ctrl->pending_events via board_added() -> pciehp_check_link_status() in order to deal with presence and link flaps per commit 6c35a1ac3da6 ("PCI: pciehp: Tolerate initially unstable link") * after pciehp_ist() has successfully brought up the slot, it resets ctrl->ist_running to false and wakes up the sysfs requester * the genirq code re-runs pciehp_ist(), which sets ctrl->ist_running to true but then returns with IRQ_NONE because ctrl->pending_events is empty * pciehp_sysfs_enable_slot() is finally woken but notices that ctrl->ist_running is true, hence continues waiting The only way to get the hung task going again is to trigger a hotplug event which brings down the slot, e.g. by yanking out the card. The same race exists when powering down the slot because remove_board() likewise clears link or presence changes in ctrl->pending_events per commit 3943af9d01e9 ("PCI: pciehp: Ignore Link State Changes after powering off a slot") and thereby may cause a re-run of pciehp_ist() which returns with IRQ_NONE without resetting ctrl->ist_running to false. Fix by adding a goto label before the teardown steps at the end of pciehp_ist() and jumping to that label from the two return statements which currently neglect to reset the ctrl->ist_running flag. Fixes: 157c1062fcd8 ("PCI: pciehp: Avoid returning prematurely from sysfs requests") Link: https://lore.kernel.org/r/cca1effa488065cb055120aa01b65719094bdcb5.1584530321.git.lukas@wunner.de Reported-by: David Hoyer Signed-off-by: Lukas Wunner Signed-off-by: Bjorn Helgaas Reviewed-by: Keith Busch Cc: stable@vger.kernel.org # v4.19+ Signed-off-by: Greg Kroah-Hartman

PCI: pciehp: Do not disable interrupt twice on suspend

2020-01-17T18:48:51+00:00

commit 75fcc0ce72e5cea2e357cdde858216c5bad40442 upstream. We try to keep PCIe hotplug ports runtime suspended when entering system suspend. Because the PCIe portdrv sets the DPM_FLAG_NEVER_SKIP flag, the PM core always calls system suspend/resume hooks even if the device is left runtime suspended. Since PCIe hotplug driver re-used the same function for both runtime suspend and system suspend, it ended up disabling hotplug interrupt twice and the second time following was printed: pciehp 0000:03:01.0:pcie204: pcie_do_write_cmd: no response from device Prevent this from happening by checking whether the device is already runtime suspended when the system suspend hook is called. Fixes: 9c62f0bfb832 ("PCI: pciehp: Implement runtime PM callbacks") Link: https://lore.kernel.org/r/20191029170022.57528-1-mika.westerberg@linux.intel.com Reported-by: Kai-Heng Feng Tested-by: Kai-Heng Feng Signed-off-by: Mika Westerberg Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki Signed-off-by: Greg Kroah-Hartman

PCI: rpaphp: Correctly match ibm, my-drc-index to drc-name when using drc-info

2020-01-04T18:18:00+00:00

[ Upstream commit 4f9f2d3d7a434b7f882b72550194c9278f4a3925 ] The newer ibm,drc-info property is a condensed description of the old ibm,drc-* properties (ie. names, types, indexes, and power-domains). When matching a drc-index to a drc-name we need to verify that the index is within the start and last drc-index range and map it to a drc-name using the drc-name-prefix and logical index. Fix the mapping by checking that the index is within the range of the current drc-info entry, and build the name from the drc-name-prefix concatenated with the starting drc-name-suffix value and the sequential index obtained by subtracting ibm,my-drc-index from this entries drc-start-index. Signed-off-by: Tyrel Datwyler Acked-by: Bjorn Helgaas Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/1573449697-5448-10-git-send-email-tyreld@linux.ibm.com Signed-off-by: Sasha Levin

PCI: rpaphp: Annotate and correctly byte swap DRC properties

2020-01-04T18:17:57+00:00

[ Upstream commit 0737686778c6dbe0908d684dd5b9c05b127526ba ] The device tree is in big endian format and any properties directly retrieved using OF helpers that don't explicitly byte swap should be annotated. In particular there are several places where we grab the opaque property value for the old ibm,drc-* properties and the ibm,my-drc-index property. Fix this for better static checking by annotating values we know to explicitly big endian, and byte swap where appropriate. Signed-off-by: Tyrel Datwyler Acked-by: Bjorn Helgaas Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/1573449697-5448-9-git-send-email-tyreld@linux.ibm.com Signed-off-by: Sasha Levin

PCI: rpaphp: Don't rely on firmware feature to imply drc-info support

2020-01-04T18:17:55+00:00

[ Upstream commit 52e2b0f16574afd082cff0f0e8567b2d9f68c033 ] In the event that the partition is migrated to a platform with older firmware that doesn't support the ibm,drc-info property the device tree is modified to remove the ibm,drc-info property and replace it with the older style ibm,drc-* properties for types, names, indexes, and power-domains. One of the requirements of the drc-info firmware feature is that the client is able to handle both the new property, and old style properties at runtime. Therefore we can't rely on the firmware feature alone to dictate which property is currently present in the device tree. Fix this short coming by checking explicitly for the ibm,drc-info property, and falling back to the older ibm,drc-* properties if it doesn't exist. Signed-off-by: Tyrel Datwyler Acked-by: Bjorn Helgaas Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/1573449697-5448-6-git-send-email-tyreld@linux.ibm.com Signed-off-by: Sasha Levin

PCI: rpaphp: Fix up pointer to first drc-info entry

2020-01-04T18:17:44+00:00

[ Upstream commit 9723c25f99aff0451cfe6392e1b9fdd99d0bf9f0 ] The first entry of the ibm,drc-info property is an int encoded count of the number of drc-info entries that follow. The "value" pointer returned by of_prop_next_u32() is still pointing at the this value when we call of_read_drc_info_cell(), but the helper function expects that value to be pointing at the first element of an entry. Fix up by incrementing the "value" pointer to point at the first element of the first drc-info entry prior. Signed-off-by: Tyrel Datwyler Acked-by: Bjorn Helgaas Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/1573449697-5448-5-git-send-email-tyreld@linux.ibm.com Signed-off-by: Sasha Levin

PCI: pciehp: Avoid returning prematurely from sysfs requests

2019-12-21T10:04:27+00:00

commit 157c1062fcd86ade3c674503705033051fd3d401 upstream. A sysfs request to enable or disable a PCIe hotplug slot should not return before it has been carried out. That is sought to be achieved by waiting until the controller's "pending_events" have been cleared. However the IRQ thread pciehp_ist() clears the "pending_events" before it acts on them. If pciehp_sysfs_enable_slot() / _disable_slot() happen to check the "pending_events" after they have been cleared but while pciehp_ist() is still running, the functions may return prematurely with an incorrect return value. Fix by introducing an "ist_running" flag which must be false before a sysfs request is allowed to return. Fixes: 32a8cef274fe ("PCI: pciehp: Enable/disable exclusively from IRQ thread") Link: https://lore.kernel.org/linux-pci/1562226638-54134-1-git-send-email-wangxiongfeng2@huawei.com Link: https://lore.kernel.org/r/4174210466e27eb7e2243dd1d801d5f75baaffd8.1565345211.git.lukas@wunner.de Reported-and-tested-by: Xiongfeng Wang Signed-off-by: Lukas Wunner Signed-off-by: Bjorn Helgaas Cc: stable@vger.kernel.org # v4.19+ Signed-off-by: Greg Kroah-Hartman

ACPI / hotplug / PCI: Allocate resources directly under the non-hotplug bridge

2019-12-17T18:56:32+00:00

commit 77adf9355304f8dcf09054280af5e23fc451ab3d upstream. Valerio and others reported that commit 84c8b58ed3ad ("ACPI / hotplug / PCI: Don't scan bridges managed by native hotplug") prevents some recent LG and HP laptops from booting with endless loop of: ACPI Error: No handler or method for GPE 08, disabling event (20190215/evgpe-835) ACPI Error: No handler or method for GPE 09, disabling event (20190215/evgpe-835) ACPI Error: No handler or method for GPE 0A, disabling event (20190215/evgpe-835) ... What seems to happen is that during boot, after the initial PCI enumeration when EC is enabled the platform triggers ACPI Notify() to one of the root ports. The root port itself looks like this: pci 0000:00:1b.0: PCI bridge to [bus 02-3a] pci 0000:00:1b.0: bridge window [mem 0xc4000000-0xda0fffff] pci 0000:00:1b.0: bridge window [mem 0x80000000-0xa1ffffff 64bit pref] The BIOS has configured the root port so that it does not have I/O bridge window. Now when the ACPI Notify() is triggered ACPI hotplug handler calls acpiphp_native_scan_bridge() for each non-hotplug bridge (as this system is using native PCIe hotplug) and pci_assign_unassigned_bridge_resources() to allocate resources. The device connected to the root port is a PCIe switch (Thunderbolt controller) with two hotplug downstream ports. Because of the hotplug ports __pci_bus_size_bridges() tries to add "additional I/O" of 256 bytes to each (DEFAULT_HOTPLUG_IO_SIZE). This gets further aligned to 4k as that's the minimum I/O window size so each hotplug port gets 4k I/O window and the same happens for the root port (which is also hotplug port). This means 3 * 4k = 12k I/O window. Because of this pci_assign_unassigned_bridge_resources() ends up opening a I/O bridge window for the root port at first available I/O address which seems to be in range 0x1000 - 0x3fff. Normally this range is used for ACPI stuff such as GPE bits (below is part of /proc/ioports): 1800-1803 : ACPI PM1a_EVT_BLK 1804-1805 : ACPI PM1a_CNT_BLK 1808-180b : ACPI PM_TMR 1810-1815 : ACPI CPU throttle 1850-1850 : ACPI PM2_CNT_BLK 1854-1857 : pnp 00:05 1860-187f : ACPI GPE0_BLK However, when the ACPI Notify() happened this range was not yet reserved for ACPI/PNP (that happens later) so PCI gets it. It then starts writing to this range and accidentally stomps over GPE bits among other things causing the endless stream of messages about missing GPE handler. This problem does not happen if "pci=hpiosize=0" is passed in the kernel command line. The reason is that then the kernel does not try to allocate the additional 256 bytes for each hotplug port. Fix this by allocating resources directly below the non-hotplug bridges where a new device may appear as a result of ACPI Notify(). This avoids the hotplug bridges and prevents opening the additional I/O window. Fixes: 84c8b58ed3ad ("ACPI / hotplug / PCI: Don't scan bridges managed by native hotplug") Link: https://bugzilla.kernel.org/show_bug.cgi?id=203617 Link: https://lore.kernel.org/r/20191030150545.19885-1-mika.westerberg@linux.intel.com Reported-by: Valerio Passini Signed-off-by: Mika Westerberg Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

Merge tag 'pci-v5.4-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci

2019-09-24T02:16:01+00:00

Pull PCI updates from Bjorn Helgaas: "Enumeration: - Consolidate _HPP/_HPX stuff in pci-acpi.c and simplify it (Krzysztof Wilczynski) - Fix incorrect PCIe device types and remove dev->has_secondary_link to simplify code that deals with upstream/downstream ports (Mika Westerberg) - After suspend, restore Resizable BAR size bits correctly for 1MB BARs (Sumit Saxena) - Enable PCI_MSI_IRQ_DOMAIN support for RISC-V (Wesley Terpstra) Virtualization: - Add ACS quirks for iProc PAXB (Abhinav Ratna), Amazon Annapurna Labs (Ali Saidi) - Move sysfs SR-IOV functions to iov.c (Kelsey Skunberg) - Remove group write permissions from sysfs sriov_numvfs, sriov_drivers_autoprobe (Kelsey Skunberg) Hotplug: - Simplify pciehp indicator control (Denis Efremov) Peer-to-peer DMA: - Allow P2P DMA between root ports for whitelisted bridges (Logan Gunthorpe) - Whitelist some Intel host bridges for P2P DMA (Logan Gunthorpe) - DMA map P2P DMA requests that traverse host bridge (Logan Gunthorpe) Amazon Annapurna Labs host bridge driver: - Add DT binding and controller driver (Jonathan Chocron) Hyper-V host bridge driver: - Fix hv_pci_dev->pci_slot use-after-free (Dexuan Cui) - Fix PCI domain number collisions (Haiyang Zhang) - Use instance ID bytes 4 & 5 as PCI domain numbers (Haiyang Zhang) - Fix build errors on non-SYSFS config (Randy Dunlap) i.MX6 host bridge driver: - Limit DBI register length (Stefan Agner) Intel VMD host bridge driver: - Fix config addressing issues (Jon Derrick) Layerscape host bridge driver: - Add bar_fixed_64bit property to endpoint driver (Xiaowei Bao) - Add CONFIG_PCI_LAYERSCAPE_EP to build EP/RC drivers separately (Xiaowei Bao) Mediatek host bridge driver: - Add MT7629 controller support (Jianjun Wang) Mobiveil host bridge driver: - Fix CPU base address setup (Hou Zhiqiang) - Make "num-lanes" property optional (Hou Zhiqiang) Tegra host bridge driver: - Fix OF node reference leak (Nishka Dasgupta) - Disable MSI for root ports to work around design problem (Vidya Sagar) - Add Tegra194 DT binding and controller support (Vidya Sagar) - Add support for sideband pins and slot regulators (Vidya Sagar) - Add PIPE2UPHY support (Vidya Sagar) Misc: - Remove unused pci_block_cfg_access() et al (Kelsey Skunberg) - Unexport pci_bus_get(), etc (Kelsey Skunberg) - Hide PM, VC, link speed, ATS, ECRC, PTM constants and interfaces in the PCI core (Kelsey Skunberg) - Clean up sysfs DEVICE_ATTR() usage (Kelsey Skunberg) - Mark expected switch fall-through (Gustavo A. R. Silva) - Propagate errors for optional regulators and PHYs (Thierry Reding) - Fix kernel command line resource_alignment parameter issues (Logan Gunthorpe)" * tag 'pci-v5.4-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci: (112 commits) PCI: Add pci_irq_vector() and other stubs when !CONFIG_PCI arm64: tegra: Add PCIe slot supply information in p2972-0000 platform arm64: tegra: Add configuration for PCIe C5 sideband signals PCI: tegra: Add support to enable slot regulators PCI: tegra: Add support to configure sideband pins PCI: vmd: Fix shadow offsets to reflect spec changes PCI: vmd: Fix config addressing when using bus offsets PCI: dwc: Add validation that PCIe core is set to correct mode PCI: dwc: al: Add Amazon Annapurna Labs PCIe controller driver dt-bindings: PCI: Add Amazon's Annapurna Labs PCIe host bridge binding PCI: Add quirk to disable MSI-X support for Amazon's Annapurna Labs Root Port PCI/VPD: Prevent VPD access for Amazon's Annapurna Labs Root Port PCI: Add ACS quirk for Amazon Annapurna Labs root ports PCI: Add Amazon's Annapurna Labs vendor ID MAINTAINERS: Add PCI native host/endpoint controllers designated reviewer PCI: hv: Use bytes 4 and 5 from instance ID as the PCI domain numbers dt-bindings: PCI: tegra: Add PCIe slot supplies regulator entries dt-bindings: PCI: tegra: Add sideband pins configuration entries PCI: tegra: Add Tegra194 PCIe support PCI: Get rid of dev->has_secondary_link flag ...