PCI: Fix use-after-free in pci_bus_release_domain_nr()
PATCH 1/1 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:
[ 19.285870] ================================================================== [ 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 [ 19.309764] really_probe+0xb8/0x298 [ 19.309777] __driver_probe_device+0x78/0xd8 [ 19.309788] driver_probe_device+0x38/0x120 [ 19.309799] __device_attach_driver+0x94/0xe0 [ 19.309812] bus_for_each_drv+0x70/0xc8 [ 19.309822] __device_attach+0xfc/0x188 [ 19.309833] device_initial_probe+0x10/0x18 [ 19.309844] bus_probe_device+0x94/0xa0 [ 19.309854] deferred_probe_work_func+0x80/0xb8 [ 19.309864] process_one_work+0x1e0/0x348 [ 19.309882] worker_thread+0x48/0x410 [ 19.309891] kthread+0xf4/0x110 [ 19.309904] ret_from_fork+0x10/0x20
[ 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 [ 19.311653] really_probe+0xb8/0x298 [ 19.311663] __driver_probe_device+0x78/0xd8 [ 19.311672] driver_probe_device+0x38/0x120 [ 19.311682] __device_attach_driver+0x94/0xe0 [ 19.311694] bus_for_each_drv+0x70/0xc8 [ 19.311702] __device_attach+0xfc/0x188 [ 19.311713] device_initial_probe+0x10/0x18 [ 19.311724] bus_probe_device+0x94/0xa0 [ 19.311733] deferred_probe_work_func+0x80/0xb8 [ 19.311743] process_one_work+0x1e0/0x348 [ 19.311753] worker_thread+0x48/0x410 [ 19.311763] kthread+0xf4/0x110 [ 19.311771] ret_from_fork+0x10/0x20
[ 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 [ 19.311906] really_probe+0xb8/0x298 [ 19.311916] __driver_probe_device+0x78/0xd8 [ 19.311926] driver_probe_device+0x38/0x120 [ 19.311936] __device_attach_driver+0x94/0xe0 [ 19.311947] bus_for_each_drv+0x70/0xc8 [ 19.311956] __device_attach+0xfc/0x188 [ 19.311966] device_initial_probe+0x10/0x18 [ 19.311976] bus_probe_device+0x94/0xa0 [ 19.311985] deferred_probe_work_func+0x80/0xb8 [ 19.311995] process_one_work+0x1e0/0x348 [ 19.312005] worker_thread+0x48/0x410 [ 19.312014] kthread+0xf4/0x110 [ 19.312022] ret_from_fork+0x10/0x20
[ 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 [ 19.331919] ==================================================================
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.
--- drivers/pci/remove.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/pci/remove.c +3 -2
@@ -157,8 +157,6 @@ void pci_remove_root_bus(struct pci_bus *bus)list_for_each_entry_safe (child, tmp, &bus->devices, bus_list)pci_remove_bus_device (child); -pci_remove_bus (bus); - host_bridge->bus =NULL ;#ifdef CONFIG_PCI_DOMAINS_GENERIC /* Release domain_nr if it was dynamically allocated */
@@ -166,6 +164,9 @@ void pci_remove_root_bus(struct pci_bus *bus)pci_bus_release_domain_nr (bus, host_bridge->dev.parent);#endif +pci_remove_bus (bus); + host_bridge->bus =NULL ; +/* remove the host bridge */ device_del (&host_bridge->dev); }
Signature · 1 line
2.39.2
Quoted 83 lines · Rob Herring
Commit
c14f7ccc9f5d("PCI: Assign PCI domain IDs byida_alloc()") introduced a use-after-free bug in the bus removal cleanup. The issue was found with kfence:[ 19.285870] ================================================================== [ 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 [ 19.309764] really_probe+0xb8/0x298 [ 19.309777] __driver_probe_device+0x78/0xd8 [ 19.309788] driver_probe_device+0x38/0x120 [ 19.309799] __device_attach_driver+0x94/0xe0 [ 19.309812] bus_for_each_drv+0x70/0xc8 [ 19.309822] __device_attach+0xfc/0x188 [ 19.309833] device_initial_probe+0x10/0x18 [ 19.309844] bus_probe_device+0x94/0xa0 [ 19.309854] deferred_probe_work_func+0x80/0xb8 [ 19.309864] process_one_work+0x1e0/0x348 [ 19.309882] worker_thread+0x48/0x410 [ 19.309891] kthread+0xf4/0x110 [ 19.309904] ret_from_fork+0x10/0x20[ 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 [ 19.311653] really_probe+0xb8/0x298 [ 19.311663] __driver_probe_device+0x78/0xd8 [ 19.311672] driver_probe_device+0x38/0x120 [ 19.311682] __device_attach_driver+0x94/0xe0 [ 19.311694] bus_for_each_drv+0x70/0xc8 [ 19.311702] __device_attach+0xfc/0x188 [ 19.311713] device_initial_probe+0x10/0x18 [ 19.311724] bus_probe_device+0x94/0xa0 [ 19.311733] deferred_probe_work_func+0x80/0xb8 [ 19.311743] process_one_work+0x1e0/0x348 [ 19.311753] worker_thread+0x48/0x410 [ 19.311763] kthread+0xf4/0x110 [ 19.311771] ret_from_fork+0x10/0x20[ 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 [ 19.311906] really_probe+0xb8/0x298 [ 19.311916] __driver_probe_device+0x78/0xd8 [ 19.311926] driver_probe_device+0x38/0x120 [ 19.311936] __device_attach_driver+0x94/0xe0 [ 19.311947] bus_for_each_drv+0x70/0xc8 [ 19.311956] __device_attach+0xfc/0x188 [ 19.311966] device_initial_probe+0x10/0x18 [ 19.311976] bus_probe_device+0x94/0xa0 [ 19.311985] deferred_probe_work_func+0x80/0xb8 [ 19.311995] process_one_work+0x1e0/0x348 [ 19.312005] worker_thread+0x48/0x410 [ 19.312014] kthread+0xf4/0x110 [ 19.312022] ret_from_fork+0x10/0x20[ 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 [ 19.331919] ==================================================================
Quoted 6 lines
The stack trace is a bit misleading as
dw_pcie_host_deinit()doesn't directly callpci_bus_release_domain_nr(). The issue turns out to be inpci_remove_root_bus()which first callspci_remove_bus()which frees the structpci_buswhen its struct device is released. Thenpci_bus_release_domain_nr()is called and accesses the freed structpci_bus. Reordering these fixes the issue.
Since time log is not useful, remove it and just add stack trace.
Change looks good to me. But I am not sure whether pci_remove_bus() directly frees struct pci_bus or just removes the bus resources?
Quoted 33 lines
Fixes c14f7ccc9f5d ("PCI: Assign PCI domain IDs by ida_alloc()")Reported-by Jon Hunter <[email protected]>Tested-by Jon Hunter <[email protected]>Cc Pali Rohár <[email protected]>Signed-off-by Rob Herring <[email protected]>--- drivers/pci/remove.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)drivers/pci/remove.c +3
-2@@ -157,8 +157,6 @@ void pci_remove_root_bus(struct pci_bus *bus)list_for_each_entry_safe (child, tmp, &bus->devices, bus_list)pci_remove_bus_device (child); -pci_remove_bus (bus); - host_bridge->bus =NULL ;#ifdef CONFIG_PCI_DOMAINS_GENERIC /* Release domain_nr if it was dynamically allocated */ @@ -166,6 +164,9 @@ void pci_remove_root_bus(struct pci_bus *bus)pci_bus_release_domain_nr (bus, host_bridge->dev.parent);#endif +pci_remove_bus (bus); + host_bridge->bus =NULL ; +/* remove the host bridge */ device_del (&host_bridge->dev); }
Signature · 2 lines
Sathyanarayanan Kuppuswamy Linux Kernel Developer
Quoted 87 lines · Sathyanarayanan Kuppuswamy
Quoted 83 lines · Rob Herring
Commit
c14f7ccc9f5d("PCI: Assign PCI domain IDs byida_alloc()") introduced a use-after-free bug in the bus removal cleanup. The issue was found with kfence:[ 19.285870] ================================================================== [ 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 [ 19.309764] really_probe+0xb8/0x298 [ 19.309777] __driver_probe_device+0x78/0xd8 [ 19.309788] driver_probe_device+0x38/0x120 [ 19.309799] __device_attach_driver+0x94/0xe0 [ 19.309812] bus_for_each_drv+0x70/0xc8 [ 19.309822] __device_attach+0xfc/0x188 [ 19.309833] device_initial_probe+0x10/0x18 [ 19.309844] bus_probe_device+0x94/0xa0 [ 19.309854] deferred_probe_work_func+0x80/0xb8 [ 19.309864] process_one_work+0x1e0/0x348 [ 19.309882] worker_thread+0x48/0x410 [ 19.309891] kthread+0xf4/0x110 [ 19.309904] ret_from_fork+0x10/0x20[ 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 [ 19.311653] really_probe+0xb8/0x298 [ 19.311663] __driver_probe_device+0x78/0xd8 [ 19.311672] driver_probe_device+0x38/0x120 [ 19.311682] __device_attach_driver+0x94/0xe0 [ 19.311694] bus_for_each_drv+0x70/0xc8 [ 19.311702] __device_attach+0xfc/0x188 [ 19.311713] device_initial_probe+0x10/0x18 [ 19.311724] bus_probe_device+0x94/0xa0 [ 19.311733] deferred_probe_work_func+0x80/0xb8 [ 19.311743] process_one_work+0x1e0/0x348 [ 19.311753] worker_thread+0x48/0x410 [ 19.311763] kthread+0xf4/0x110 [ 19.311771] ret_from_fork+0x10/0x20[ 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 [ 19.311906] really_probe+0xb8/0x298 [ 19.311916] __driver_probe_device+0x78/0xd8 [ 19.311926] driver_probe_device+0x38/0x120 [ 19.311936] __device_attach_driver+0x94/0xe0 [ 19.311947] bus_for_each_drv+0x70/0xc8 [ 19.311956] __device_attach+0xfc/0x188 [ 19.311966] device_initial_probe+0x10/0x18 [ 19.311976] bus_probe_device+0x94/0xa0 [ 19.311985] deferred_probe_work_func+0x80/0xb8 [ 19.311995] process_one_work+0x1e0/0x348 [ 19.312005] worker_thread+0x48/0x410 [ 19.312014] kthread+0xf4/0x110 [ 19.312022] ret_from_fork+0x10/0x20[ 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 [ 19.331919] ==================================================================
Quoted 8 lines
Quoted 6 lines
The stack trace is a bit misleading as
dw_pcie_host_deinit()doesn't directly callpci_bus_release_domain_nr(). The issue turns out to be inpci_remove_root_bus()which first callspci_remove_bus()which frees the structpci_buswhen its struct device is released. Thenpci_bus_release_domain_nr()is called and accesses the freed structpci_bus. Reordering these fixes the issue.Since time log is not useful, remove it and just add stack trace.
Actually, it was useful. It told me that the use-after-free happened pretty much right after the free.
Bjorn likes to edit commit messages so I'll leave it up to him to leave or change.
Quoted 2 lines
Change looks good to me. But I am not sure whether
pci_remove_bus()directly frees structpci_busor just removes the bus resources?
struct pci_bus embeds a struct device. The struct device is put which causes the free of the struct pci_bus. So indirect in that the refcounting triggers the free, but direct in that pci_remove_bus() causes it.
Rob
Sathyanarayanan Kuppuswamy
Reviewed-byQuoted 104 lines · Rob Herring
Quoted 87 lines · Sathyanarayanan Kuppuswamy
Quoted 83 lines · Rob Herring
Commit
c14f7ccc9f5d("PCI: Assign PCI domain IDs byida_alloc()") introduced a use-after-free bug in the bus removal cleanup. The issue was found with kfence:[ 19.285870] ================================================================== [ 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 [ 19.309764] really_probe+0xb8/0x298 [ 19.309777] __driver_probe_device+0x78/0xd8 [ 19.309788] driver_probe_device+0x38/0x120 [ 19.309799] __device_attach_driver+0x94/0xe0 [ 19.309812] bus_for_each_drv+0x70/0xc8 [ 19.309822] __device_attach+0xfc/0x188 [ 19.309833] device_initial_probe+0x10/0x18 [ 19.309844] bus_probe_device+0x94/0xa0 [ 19.309854] deferred_probe_work_func+0x80/0xb8 [ 19.309864] process_one_work+0x1e0/0x348 [ 19.309882] worker_thread+0x48/0x410 [ 19.309891] kthread+0xf4/0x110 [ 19.309904] ret_from_fork+0x10/0x20[ 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 [ 19.311653] really_probe+0xb8/0x298 [ 19.311663] __driver_probe_device+0x78/0xd8 [ 19.311672] driver_probe_device+0x38/0x120 [ 19.311682] __device_attach_driver+0x94/0xe0 [ 19.311694] bus_for_each_drv+0x70/0xc8 [ 19.311702] __device_attach+0xfc/0x188 [ 19.311713] device_initial_probe+0x10/0x18 [ 19.311724] bus_probe_device+0x94/0xa0 [ 19.311733] deferred_probe_work_func+0x80/0xb8 [ 19.311743] process_one_work+0x1e0/0x348 [ 19.311753] worker_thread+0x48/0x410 [ 19.311763] kthread+0xf4/0x110 [ 19.311771] ret_from_fork+0x10/0x20[ 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 [ 19.311906] really_probe+0xb8/0x298 [ 19.311916] __driver_probe_device+0x78/0xd8 [ 19.311926] driver_probe_device+0x38/0x120 [ 19.311936] __device_attach_driver+0x94/0xe0 [ 19.311947] bus_for_each_drv+0x70/0xc8 [ 19.311956] __device_attach+0xfc/0x188 [ 19.311966] device_initial_probe+0x10/0x18 [ 19.311976] bus_probe_device+0x94/0xa0 [ 19.311985] deferred_probe_work_func+0x80/0xb8 [ 19.311995] process_one_work+0x1e0/0x348 [ 19.312005] worker_thread+0x48/0x410 [ 19.312014] kthread+0xf4/0x110 [ 19.312022] ret_from_fork+0x10/0x20[ 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 [ 19.331919] ==================================================================Quoted 8 lines
Quoted 6 lines
The stack trace is a bit misleading as
dw_pcie_host_deinit()doesn't directly callpci_bus_release_domain_nr(). The issue turns out to be inpci_remove_root_bus()which first callspci_remove_bus()which frees the structpci_buswhen its struct device is released. Thenpci_bus_release_domain_nr()is called and accesses the freed structpci_bus. Reordering these fixes the issue.Since time log is not useful, remove it and just add stack trace.
Actually, it was useful. It told me that the use-after-free happened pretty much right after the free.
Bjorn likes to edit commit messages so I'll leave it up to him to leave or change.
Quoted 7 lines
Quoted 2 lines
Change looks good to me. But I am not sure whether
pci_remove_bus()directly frees structpci_busor just removes the bus resources?struct
pci_busembeds a struct device. The struct device is put which causes the free of the structpci_bus. So indirect in that the refcounting triggers the free, but direct in thatpci_remove_bus()causes it.
Got it.
Quoted 2 lines
Rob
Signature · 2 lines
Sathyanarayanan Kuppuswamy Linux Kernel Developer
Bjorn Helgaas
AppliedQuoted 90 lines · Rob Herring
Commit
c14f7ccc9f5d("PCI: Assign PCI domain IDs byida_alloc()") introduced a use-after-free bug in the bus removal cleanup. The issue was found with kfence:[ 19.285870] ================================================================== [ 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 [ 19.309764] really_probe+0xb8/0x298 [ 19.309777] __driver_probe_device+0x78/0xd8 [ 19.309788] driver_probe_device+0x38/0x120 [ 19.309799] __device_attach_driver+0x94/0xe0 [ 19.309812] bus_for_each_drv+0x70/0xc8 [ 19.309822] __device_attach+0xfc/0x188 [ 19.309833] device_initial_probe+0x10/0x18 [ 19.309844] bus_probe_device+0x94/0xa0 [ 19.309854] deferred_probe_work_func+0x80/0xb8 [ 19.309864] process_one_work+0x1e0/0x348 [ 19.309882] worker_thread+0x48/0x410 [ 19.309891] kthread+0xf4/0x110 [ 19.309904] ret_from_fork+0x10/0x20[ 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 [ 19.311653] really_probe+0xb8/0x298 [ 19.311663] __driver_probe_device+0x78/0xd8 [ 19.311672] driver_probe_device+0x38/0x120 [ 19.311682] __device_attach_driver+0x94/0xe0 [ 19.311694] bus_for_each_drv+0x70/0xc8 [ 19.311702] __device_attach+0xfc/0x188 [ 19.311713] device_initial_probe+0x10/0x18 [ 19.311724] bus_probe_device+0x94/0xa0 [ 19.311733] deferred_probe_work_func+0x80/0xb8 [ 19.311743] process_one_work+0x1e0/0x348 [ 19.311753] worker_thread+0x48/0x410 [ 19.311763] kthread+0xf4/0x110 [ 19.311771] ret_from_fork+0x10/0x20[ 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 [ 19.311906] really_probe+0xb8/0x298 [ 19.311916] __driver_probe_device+0x78/0xd8 [ 19.311926] driver_probe_device+0x38/0x120 [ 19.311936] __device_attach_driver+0x94/0xe0 [ 19.311947] bus_for_each_drv+0x70/0xc8 [ 19.311956] __device_attach+0xfc/0x188 [ 19.311966] device_initial_probe+0x10/0x18 [ 19.311976] bus_probe_device+0x94/0xa0 [ 19.311985] deferred_probe_work_func+0x80/0xb8 [ 19.311995] process_one_work+0x1e0/0x348 [ 19.312005] worker_thread+0x48/0x410 [ 19.312014] kthread+0xf4/0x110 [ 19.312022] ret_from_fork+0x10/0x20[ 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 [ 19.331919] ==================================================================The stack trace is a bit misleading as
dw_pcie_host_deinit()doesn't directly callpci_bus_release_domain_nr(). The issue turns out to be inpci_remove_root_bus()which first callspci_remove_bus()which frees the structpci_buswhen its struct device is released. Thenpci_bus_release_domain_nr()is called and accesses the freed structpci_bus. Reordering these fixes the issue.
Quoted 5 lines
Fixes c14f7ccc9f5d ("PCI: Assign PCI domain IDs by ida_alloc()")Reported-by Jon Hunter <[email protected]>Tested-by Jon Hunter <[email protected]>Cc Pali Rohár <[email protected]>Signed-off-by Rob Herring <[email protected]>
Applied to for-linus for v6.3, thanks! I also added a stable tag since c14f7ccc9f5d appeared in v6.2.
Quoted 30 lines
--- drivers/pci/remove.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)drivers/pci/remove.c +3
-2@@ -157,8 +157,6 @@ void pci_remove_root_bus(struct pci_bus *bus)list_for_each_entry_safe (child, tmp, &bus->devices, bus_list)pci_remove_bus_device (child); -pci_remove_bus (bus); - host_bridge->bus =NULL ;#ifdef CONFIG_PCI_DOMAINS_GENERIC /* Release domain_nr if it was dynamically allocated */ @@ -166,6 +164,9 @@ void pci_remove_root_bus(struct pci_bus *bus)pci_bus_release_domain_nr (bus, host_bridge->dev.parent);#endif +pci_remove_bus (bus); + host_bridge->bus =NULL ; +/* remove the host bridge */ device_del (&host_bridge->dev); }Signature · 1 line
2.39.2