kernel/git/stable/linux.git/drivers/pci, branch linux-4.19.y

PCI: Fix use-after-free of slot->bus on hot remove

2024-12-05T09:59:39+00:00

commit c7acef99642b763ba585f4a43af999fcdbcc3dc4 upstream. Dennis reports a boot crash on recent Lenovo laptops with a USB4 dock. Since commit 0fc70886569c ("thunderbolt: Reset USB4 v2 host router") and commit 59a54c5f3dbd ("thunderbolt: Reset topology created by the boot firmware"), USB4 v2 and v1 Host Routers are reset on probe of the thunderbolt driver. The reset clears the Presence Detect State and Data Link Layer Link Active bits at the USB4 Host Router's Root Port and thus causes hot removal of the dock. The crash occurs when pciehp is unbound from one of the dock's Downstream Ports: pciehp creates a pci_slot on bind and destroys it on unbind. The pci_slot contains a pointer to the pci_bus below the Downstream Port, but a reference on that pci_bus is never acquired. The pci_bus is destroyed before the pci_slot, so a use-after-free ensues when pci_slot_release() accesses slot->bus. In principle this should not happen because pci_stop_bus_device() unbinds pciehp (and therefore destroys the pci_slot) before the pci_bus is destroyed by pci_remove_bus_device(). However the stacktrace provided by Dennis shows that pciehp is unbound from pci_remove_bus_device() instead of pci_stop_bus_device(). To understand the significance of this, one needs to know that the PCI core uses a two step process to remove a portion of the hierarchy: It first unbinds all drivers in the sub-hierarchy in pci_stop_bus_device() and then actually removes the devices in pci_remove_bus_device(). There is no precaution to prevent driver binding in-between pci_stop_bus_device() and pci_remove_bus_device(). In Dennis' case, it seems removal of the hierarchy by pciehp races with driver binding by pci_bus_add_devices(). pciehp is bound to the Downstream Port after pci_stop_bus_device() has run, so it is unbound by pci_remove_bus_device() instead of pci_stop_bus_device(). Because the pci_bus has already been destroyed at that point, accesses to it result in a use-after-free. One might conclude that driver binding needs to be prevented after pci_stop_bus_device() has run. However it seems risky that pci_slot points to pci_bus without holding a reference. Solely relying on correct ordering of driver unbind versus pci_bus destruction is certainly not defensive programming. If pci_slot has a need to access data in pci_bus, it ought to acquire a reference. Amend pci_create_slot() accordingly. Dennis reports that the crash is not reproducible with this change. Abridged stacktrace: pcieport 0000:00:07.0: PME: Signaling with IRQ 156 pcieport 0000:00:07.0: pciehp: Slot #12 AttnBtn- PwrCtrl- MRL- AttnInd- PwrInd- HotPlug+ Surprise+ Interlock- NoCompl+ IbPresDis- LLActRep+ pci_bus 0000:20: dev 00, created physical slot 12 pcieport 0000:00:07.0: pciehp: Slot(12): Card not present ... pcieport 0000:21:02.0: pciehp: pcie_disable_notification: SLOTCTRL d8 write cmd 0 Oops: general protection fault, probably for non-canonical address 0x6b6b6b6b6b6b6b6b: 0000 [#1] PREEMPT SMP NOPTI CPU: 13 UID: 0 PID: 134 Comm: irq/156-pciehp Not tainted 6.11.0-devel+ #1 RIP: 0010:dev_driver_string+0x12/0x40 pci_destroy_slot pciehp_remove pcie_port_remove_service device_release_driver_internal bus_remove_device device_del device_unregister remove_iter device_for_each_child pcie_portdrv_remove pci_device_remove device_release_driver_internal bus_remove_device device_del pci_remove_bus_device (recursive invocation) pci_remove_bus_device pciehp_unconfigure_device pciehp_disable_slot pciehp_handle_presence_or_link_change pciehp_ist Link: https://lore.kernel.org/r/4bfd4c0e976c1776cd08e76603903b338cf25729.1728579288.git.lukas@wunner.de Reported-by: Dennis Wassenberg Closes: https://lore.kernel.org/r/6de4b45ff2b32dd91a805ec02ec8ec73ef411bf6.camel@secunet.com/ Tested-by: Dennis Wassenberg Signed-off-by: Lukas Wunner Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

PCI: cpqphp: Fix PCIBIOS_* return value confusion

2024-12-05T09:59:36+00:00

[ Upstream commit e2226dbc4a4919d9c8bd9293299b532090bdf020 ] Code in and related to PCI_RefinedAccessConfig() has three types of return type confusion: - PCI_RefinedAccessConfig() tests pci_bus_read_config_dword() return value against -1. - PCI_RefinedAccessConfig() returns both -1 and PCIBIOS_* return codes. - Callers of PCI_RefinedAccessConfig() only test for -1. Make PCI_RefinedAccessConfig() return PCIBIOS_* codes consistently and adapt callers accordingly. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Link: https://lore.kernel.org/r/20241022091140.3504-2-ilpo.jarvinen@linux.intel.com Signed-off-by: Ilpo Järvinen Signed-off-by: Bjorn Helgaas Signed-off-by: Sasha Levin

PCI: cpqphp: Use PCI_POSSIBLE_ERROR() to check config reads

2024-12-05T09:59:35+00:00

[ Upstream commit a18a025c2fb5fbf2d1d0606ea0d7441ac90e9c39 ] When config pci_ops.read() can detect failed PCI transactions, the data returned to the CPU is PCI_ERROR_RESPONSE (~0 or 0xffffffff). Obviously a successful PCI config read may *also* return that data if a config register happens to contain ~0, so it doesn't definitively indicate an error unless we know the register cannot contain ~0. Use PCI_POSSIBLE_ERROR() to check the response we get when we read data from hardware. This unifies PCI error response checking and makes error checks consistent and easier to find. Link: https://lore.kernel.org/r/b12005c0d57bb9d4c8b486724d078b7bd92f8321.1637243717.git.naveennaidu479@gmail.com Signed-off-by: Naveen Naidu Signed-off-by: Bjorn Helgaas Stable-dep-of: e2226dbc4a49 ("PCI: cpqphp: Fix PCIBIOS_* return value confusion") Signed-off-by: Sasha Levin

PCI: Add function 0 DMA alias quirk for Glenfly Arise chip

2024-11-08T15:19:19+00:00

commit 9246b487ab3c3b5993aae7552b7a4c541cc14a49 upstream. Add DMA support for audio function of Glenfly Arise chip, which uses Requester ID of function 0. Link: https://lore.kernel.org/r/CA2BBD087345B6D1+20240823095708.3237375-1-wangyuli@uniontech.com Signed-off-by: SiyuLi Signed-off-by: WangYuli [bhelgaas: lower-case hex to match local code, drop unused Device IDs] Signed-off-by: Bjorn Helgaas Reviewed-by: Takashi Iwai Signed-off-by: Greg Kroah-Hartman

PCI: Mark Creative Labs EMU20k2 INTx masking as broken

2024-11-08T15:19:17+00:00

[ Upstream commit 2910306655a7072640021563ec9501bfa67f0cb1 ] Per user reports, the Creative Labs EMU20k2 (Sound Blaster X-Fi Titanium Series) generates spurious interrupts when used with vfio-pci unless DisINTx masking support is disabled. Thus, quirk the device to mark INTx masking as broken. Closes: https://lore.kernel.org/all/VI1PR10MB8207C507DB5420AB4C7281E0DB9A2@VI1PR10MB8207.EURPRD10.PROD.OUTLOOK.COM Link: https://lore.kernel.org/linux-pci/20240912215331.839220-1-alex.williamson@redhat.com Reported-by: zdravko delineshev Signed-off-by: Alex Williamson [kwilczynski: commit log] Signed-off-by: Krzysztof Wilczyński Signed-off-by: Sasha Levin

PCI: xilinx-nwl: Fix off-by-one in INTx IRQ handler

2024-11-08T15:19:09+00:00

[ Upstream commit 0199d2f2bd8cd97b310f7ed82a067247d7456029 ] MSGF_LEG_MASK is laid out with INTA in bit 0, INTB in bit 1, INTC in bit 2, and INTD in bit 3. Hardware IRQ numbers start at 0, and we register PCI_NUM_INTX IRQs. So to enable INTA (aka hwirq 0) we should set bit 0. Remove the subtraction of one. This bug would cause INTx interrupts not to be delivered, as enabling INTB would actually enable INTA, and enabling INTA wouldn't enable anything at all. It is likely that this got overlooked for so long since most PCIe hardware uses MSIs. This fixes the following UBSAN error: UBSAN: shift-out-of-bounds in ../drivers/pci/controller/pcie-xilinx-nwl.c:389:11 shift exponent 18446744073709551615 is too large for 32-bit type 'int' CPU: 1 PID: 61 Comm: kworker/u10:1 Not tainted 6.6.20+ #268 Hardware name: xlnx,zynqmp (DT) Workqueue: events_unbound deferred_probe_work_func Call trace: dump_backtrace (arch/arm64/kernel/stacktrace.c:235) show_stack (arch/arm64/kernel/stacktrace.c:242) dump_stack_lvl (lib/dump_stack.c:107) dump_stack (lib/dump_stack.c:114) __ubsan_handle_shift_out_of_bounds (lib/ubsan.c:218 lib/ubsan.c:387) nwl_unmask_leg_irq (drivers/pci/controller/pcie-xilinx-nwl.c:389 (discriminator 1)) irq_enable (kernel/irq/internals.h:234 kernel/irq/chip.c:170 kernel/irq/chip.c:439 kernel/irq/chip.c:432 kernel/irq/chip.c:345) __irq_startup (kernel/irq/internals.h:239 kernel/irq/chip.c:180 kernel/irq/chip.c:250) irq_startup (kernel/irq/chip.c:270) __setup_irq (kernel/irq/manage.c:1800) request_threaded_irq (kernel/irq/manage.c:2206) pcie_pme_probe (include/linux/interrupt.h:168 drivers/pci/pcie/pme.c:348) Fixes: 9a181e1093af ("PCI: xilinx-nwl: Modify IRQ chip for legacy interrupts") Link: https://lore.kernel.org/r/20240531161337.864994-3-sean.anderson@linux.dev Signed-off-by: Sean Anderson Signed-off-by: Bjorn Helgaas Cc: stable@vger.kernel.org Signed-off-by: Sasha Levin

PCI: xilinx-nwl: Use irq_data_get_irq_chip_data()

2024-11-08T15:19:09+00:00

[ Upstream commit e56427068a8d796bb7b8e297f2b6e947380e383f ] Going through a full irq descriptor lookup instead of just using the proper helper function which provides direct access is suboptimal. In fact it _is_ wrong because the chip callback needs to get the chip data which is relevant for the chip while using the irq descriptor variant returns the irq chip data of the top level chip of a hierarchy. It does not matter in this case because the chip is the top level chip, but that doesn't make it more correct. Signed-off-by: Thomas Gleixner Reviewed-by: Rob Herring Cc: Bjorn Helgaas Link: https://lore.kernel.org/r/20201210194044.364211860@linutronix.de Stable-dep-of: 0199d2f2bd8c ("PCI: xilinx-nwl: Fix off-by-one in INTx IRQ handler") Signed-off-by: Sasha Levin

PCI: xilinx-nwl: Fix register misspelling

2024-11-08T15:19:07+00:00

[ Upstream commit a437027ae1730b8dc379c75fa0dd7d3036917400 ] MSIC -> MISC Fixes: c2a7ff18edcd ("PCI: xilinx-nwl: Expand error logging") Link: https://lore.kernel.org/r/20240531161337.864994-4-sean.anderson@linux.dev Signed-off-by: Sean Anderson Signed-off-by: Bjorn Helgaas Signed-off-by: Sasha Levin

PCI: Add missing bridge lock to pci_bus_lock()

2024-09-12T09:02:53+00:00

[ Upstream commit a4e772898f8bf2e7e1cf661a12c60a5612c4afab ] One of the true positives that the cfg_access_lock lockdep effort identified is this sequence: WARNING: CPU: 14 PID: 1 at drivers/pci/pci.c:4886 pci_bridge_secondary_bus_reset+0x5d/0x70 RIP: 0010:pci_bridge_secondary_bus_reset+0x5d/0x70 Call Trace: ? __warn+0x8c/0x190 ? pci_bridge_secondary_bus_reset+0x5d/0x70 ? report_bug+0x1f8/0x200 ? handle_bug+0x3c/0x70 ? exc_invalid_op+0x18/0x70 ? asm_exc_invalid_op+0x1a/0x20 ? pci_bridge_secondary_bus_reset+0x5d/0x70 pci_reset_bus+0x1d8/0x270 vmd_probe+0x778/0xa10 pci_device_probe+0x95/0x120 Where pci_reset_bus() users are triggering unlocked secondary bus resets. Ironically pci_bus_reset(), several calls down from pci_reset_bus(), uses pci_bus_lock() before issuing the reset which locks everything *but* the bridge itself. For the same motivation as adding: bridge = pci_upstream_bridge(dev); if (bridge) pci_dev_lock(bridge); to pci_reset_function() for the "bus" and "cxl_bus" reset cases, add pci_dev_lock() for @bus->self to pci_bus_lock(). Link: https://lore.kernel.org/r/171711747501.1628941.15217746952476635316.stgit@dwillia2-xfh.jf.intel.com Reported-by: Imre Deak Closes: http://lore.kernel.org/r/6657833b3b5ae_14984b29437@dwillia2-xfh.jf.intel.com.notmuch Signed-off-by: Dan Williams Signed-off-by: Keith Busch [bhelgaas: squash in recursive locking deadlock fix from Keith Busch: https://lore.kernel.org/r/20240711193650.701834-1-kbusch@meta.com] Signed-off-by: Bjorn Helgaas Tested-by: Hans de Goede Tested-by: Kalle Valo Reviewed-by: Dave Jiang Signed-off-by: Sasha Levin

pci/hotplug/pnv_php: Fix hotplug driver crash on Powernv

2024-09-12T09:02:53+00:00

[ Upstream commit 335e35b748527f0c06ded9eebb65387f60647fda ] The hotplug driver for powerpc (pci/hotplug/pnv_php.c) causes a kernel crash when we try to hot-unplug/disable the PCIe switch/bridge from the PHB. The crash occurs because although the MSI data structure has been released during disable/hot-unplug path and it has been assigned with NULL, still during unregistration the code was again trying to explicitly disable the MSI which causes the NULL pointer dereference and kernel crash. The patch fixes the check during unregistration path to prevent invoking pci_disable_msi/msix() since its data structure is already freed. Reported-by: Timothy Pearson Closes: https://lore.kernel.org/all/1981605666.2142272.1703742465927.JavaMail.zimbra@raptorengineeringinc.com/ Acked-by: Bjorn Helgaas Tested-by: Shawn Anastasio Signed-off-by: Krishna Kumar Signed-off-by: Michael Ellerman Link: https://msgid.link/20240701074513.94873-2-krishnak@linux.ibm.com Signed-off-by: Sasha Levin