linux_dsm_epyc7002/drivers/gpu/drm/i915
Chris Wilson 3e817471a3 drm/i915/gem: Take runtime-pm wakeref prior to unbinding
Some machines require ACPI for runtime resume, and ACPI is quite kmalloc
happy. We cannot handle kmalloc from inside the vm->mutex, as they are
used by the shrinker, and so we must ensure the global runtime-pm is
awake prior to unbinding to avoid the potential inversion.

<4> [57.121748] ======================================================
<4> [57.121750] WARNING: possible circular locking dependency detected
<4> [57.121753] 5.4.0-rc8-CI-CI_DRM_7466+ #1 Tainted: G     U
<4> [57.121754] ------------------------------------------------------
<4> [57.121756] i915_pm_rpm/1105 is trying to acquire lock:
<4> [57.121758] ffffffff82263a40 (fs_reclaim){+.+.}, at: fs_reclaim_acquire.part.117+0x0/0x30
<4> [57.121766]
but task is already holding lock:
<4> [57.121768] ffff888475a593c0 (&vm->mutex){+.+.}, at: i915_vma_unbind+0x21/0x50 [i915]
<4> [57.121868]
which lock already depends on the new lock.

<4> [57.121869]
the existing dependency chain (in reverse order) is:
<4> [57.121871]
-> #1 (&vm->mutex){+.+.}:
<4> [57.121951]        i915_gem_shrinker_taints_mutex+0xa2/0xd0 [i915]
<4> [57.122028]        i915_address_space_init+0xa9/0x170 [i915]
<4> [57.122104]        i915_ggtt_init_hw+0x47/0x130 [i915]
<4> [57.122150]        i915_driver_probe+0xbb4/0x15f0 [i915]
<4> [57.122197]        i915_pci_probe+0x43/0x1c0 [i915]
<4> [57.122202]        pci_device_probe+0x9e/0x120
<4> [57.122206]        really_probe+0xea/0x420
<4> [57.122209]        driver_probe_device+0x10b/0x120
<4> [57.122212]        device_driver_attach+0x4a/0x50
<4> [57.122214]        __driver_attach+0x97/0x130
<4> [57.122217]        bus_for_each_dev+0x74/0xc0
<4> [57.122220]        bus_add_driver+0x142/0x220
<4> [57.122222]        driver_register+0x56/0xf0
<4> [57.122226]        do_one_initcall+0x58/0x2ff
<4> [57.122230]        do_init_module+0x56/0x1f8
<4> [57.122233]        load_module+0x243e/0x29f0
<4> [57.122236]        __do_sys_finit_module+0xe9/0x110
<4> [57.122239]        do_syscall_64+0x4f/0x210
<4> [57.122242]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
<4> [57.122244]
-> #0 (fs_reclaim){+.+.}:
<4> [57.122249]        __lock_acquire+0x1328/0x15d0
<4> [57.122251]        lock_acquire+0xa7/0x1c0
<4> [57.122254]        fs_reclaim_acquire.part.117+0x24/0x30
<4> [57.122257]        __kmalloc+0x48/0x320
<4> [57.122261]        acpi_ns_internalize_name+0x44/0x9b
<4> [57.122264]        acpi_ns_get_node_unlocked+0x6b/0xd3
<4> [57.122267]        acpi_ns_get_node+0x3b/0x50
<4> [57.122271]        acpi_get_handle+0x8a/0xb4
<4> [57.122274]        acpi_has_method+0x1c/0x40
<4> [57.122278]        acpi_pci_set_power_state+0x40/0xe0
<4> [57.122281]        pci_platform_power_transition+0x3e/0x90
<4> [57.122284]        pci_set_power_state+0x83/0xf0
<4> [57.122287]        pci_restore_standard_config+0x22/0x40
<4> [57.122289]        pci_pm_runtime_resume+0x23/0xc0
<4> [57.122293]        __rpm_callback+0xb1/0x110
<4> [57.122296]        rpm_callback+0x1a/0x70
<4> [57.122299]        rpm_resume+0x50e/0x790
<4> [57.122302]        __pm_runtime_resume+0x42/0x80
<4> [57.122357]        __intel_runtime_pm_get+0x15/0x60 [i915]
<4> [57.122435]        ggtt_unbind_vma+0x24/0x60 [i915]
<4> [57.122514]        __i915_vma_unbind.part.39+0xb5/0x500 [i915]
<4> [57.122593]        i915_vma_unbind+0x2d/0x50 [i915]
<4> [57.122668]        i915_gem_object_unbind+0x11c/0x260 [i915]
<4> [57.122740]        i915_gem_object_set_cache_level+0x32/0x90 [i915]
<4> [57.122810]        i915_gem_set_caching_ioctl+0x1f7/0x2f0 [i915]
<4> [57.122815]        drm_ioctl_kernel+0xa7/0xf0
<4> [57.122818]        drm_ioctl+0x2e1/0x390
<4> [57.122822]        do_vfs_ioctl+0xa0/0x6f0
<4> [57.122825]        ksys_ioctl+0x35/0x60
<4> [57.122828]        __x64_sys_ioctl+0x11/0x20
<4> [57.122830]        do_syscall_64+0x4f/0x210
<4> [57.122833]        entry_SYSCALL_64_after_hwframe+0x49/0xbe

Closes: https://gitlab.freedesktop.org/drm/intel/issues/711
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191203101347.2836057-1-chris@chris-wilson.co.uk
2019-12-03 13:09:53 +00:00
..
display drm/i915/bios: assume vbt is 4-byte aligned into oprom 2019-12-02 13:54:23 -08:00
gem drm/i915/gem: Unbind all current vma on changing cache-level 2019-12-02 21:57:22 +00:00
gt drm/i915/gt: Simplify rc6 w/a application 2019-12-02 21:57:22 +00:00
gvt drm/i915: make more headers self-contained 2019-11-08 10:16:13 +00:00
oa drm/i915/tgl: Add perf support on TGL 2019-10-29 12:53:54 +02:00
selftests drm/i915/selftests: Always lock the drm_mm around insert/remove 2019-11-29 14:23:53 +00:00
i915_active_types.h drm/i915: Serialise i915_active_fence_set() with itself 2019-11-27 17:02:14 +00:00
i915_active.c drm/i915: Serialise i915_active_wait() with its retirement 2019-12-02 21:57:23 +00:00
i915_active.h drm/i915: Specialise i915_active.work lock classes 2019-12-02 21:57:23 +00:00
i915_buddy.c
i915_buddy.h
i915_cmd_parser.c drm/i915/gem: Excise the per-batch whitelist from the context 2019-11-28 11:39:50 +00:00
i915_debugfs.c drm/i915: Stop using connector->encoder and encoder->crtc links in i915_display_info 2019-12-02 18:38:40 +02:00
i915_debugfs.h
i915_drv.c drm/i915/gt: Simplify rc6 w/a application 2019-12-02 21:57:22 +00:00
i915_drv.h drm/i915: Support more QGV points 2019-11-26 18:27:37 +02:00
i915_fixed.h
i915_gem_evict.c drm/i915: Move request runtime management onto gt 2019-10-04 15:39:26 +01:00
i915_gem_fence_reg.c drm/i915: set num_fence_regs to 0 if there is no aperture 2019-10-29 10:35:47 +00:00
i915_gem_fence_reg.h drm/i915: Store i915_ggtt as the backpointer on fence registers 2019-10-16 19:41:36 +01:00
i915_gem_gtt.c drm/i915: Refactor gen6_flush_pd() 2019-12-01 15:18:19 +00:00
i915_gem_gtt.h drm/i915/gen7: Re-enable full-ppgtt for ivb & hsw 2019-11-30 09:21:12 +00:00
i915_gem.c drm/i915/gem: Take runtime-pm wakeref prior to unbinding 2019-12-03 13:09:53 +00:00
i915_gem.h drm/i915/gem: Manually dump the debug trace on GEM_BUG_ON 2019-11-19 17:44:39 +00:00
i915_getparam.c Backmerge i915 security patches from commit 'ea0b163b13ff' into drm-next 2019-11-14 11:09:06 +10:00
i915_globals.c
i915_globals.h
i915_gpu_error.c drm/i915: Switch kunmap() to take the page not vaddr 2019-11-25 09:24:31 +00:00
i915_gpu_error.h drm/i915/tgl: Add gam instdone 2019-10-30 09:15:27 +00:00
i915_ioc32.c
i915_irq.c drm/i915: Program SHPD_FILTER_CNT on CNP+ 2019-12-02 08:18:54 -08:00
i915_irq.h drm/i915: Extract the GuC interrupt handlers 2019-10-26 19:28:59 +01:00
i915_memcpy.c
i915_memcpy.h
i915_mm.c
i915_params.c drm/i915/lmem: add the fake lmem region 2019-10-31 20:41:47 +00:00
i915_params.h drm/i915/lmem: add the fake lmem region 2019-10-31 20:41:47 +00:00
i915_pci.c drm/i915/gen7: Re-enable full-ppgtt for ivb & hsw 2019-11-30 09:21:12 +00:00
i915_perf_types.h drm/i915/perf: Describe structure members in documentation 2019-10-24 15:01:35 +01:00
i915_perf.c drm/i915: Serialise with engine-pm around requests on the kernel_context 2019-11-25 13:17:18 +00:00
i915_perf.h drm/i915/perf: introduce a versioning of the i915-perf uapi 2019-10-14 21:30:25 +01:00
i915_pmu.c drm/i915: Mark up the calling context for intel_wakeref_put() 2019-11-20 15:59:23 +00:00
i915_pmu.h drm/i915/pmu: Support multiple GPUs 2019-10-17 10:50:47 +01:00
i915_priolist_types.h drm/i915/gt: Replace hangcheck by heartbeats 2019-10-23 23:52:10 +01:00
i915_pvinfo.h
i915_query.c drm/i915/query: Align flavour of engine data lookup 2019-11-23 19:33:12 +00:00
i915_query.h
i915_reg.h drm/i915: Program SHPD_FILTER_CNT on CNP+ 2019-12-02 08:18:54 -08:00
i915_request.c drm/i915: Use a ctor for TYPESAFE_BY_RCU i915_request 2019-11-22 10:47:38 +00:00
i915_request.h drm/i915: Mark up "sentinel" requests 2019-10-12 08:51:17 +01:00
i915_scatterlist.c
i915_scatterlist.h
i915_scheduler_types.h drm/i915: Note the addition of timeslicing to the pretend scheduler 2019-10-11 09:36:16 +01:00
i915_scheduler.c drm/i915: Use a ctor for TYPESAFE_BY_RCU i915_request 2019-11-22 10:47:38 +00:00
i915_scheduler.h drm/i915: Use a ctor for TYPESAFE_BY_RCU i915_request 2019-11-22 10:47:38 +00:00
i915_selftest.h drm/i915/selftests: Perform some basic cycle counting of MI ops 2019-11-11 18:30:13 +00:00
i915_suspend.c drm/i915: Drop struct_mutex from suspend state save/restore 2019-10-04 15:39:41 +01:00
i915_suspend.h
i915_sw_fence_work.c
i915_sw_fence_work.h
i915_sw_fence.c drm/i915: Use a ctor for TYPESAFE_BY_RCU i915_request 2019-11-22 10:47:38 +00:00
i915_sw_fence.h drm/i915: Use a ctor for TYPESAFE_BY_RCU i915_request 2019-11-22 10:47:38 +00:00
i915_switcheroo.c drm/i915: split out i915_switcheroo.[ch] from i915_drv.c 2019-10-06 11:25:00 +03:00
i915_switcheroo.h drm/i915: split out i915_switcheroo.[ch] from i915_drv.c 2019-10-06 11:25:00 +03:00
i915_syncmap.c
i915_syncmap.h
i915_sysfs.c drm/i915: Extract GT render power state management 2019-10-26 19:28:59 +01:00
i915_sysfs.h
i915_trace_points.c
i915_trace.h drm/i915: Move context management under GEM 2019-10-04 15:39:34 +01:00
i915_user_extensions.c
i915_user_extensions.h
i915_utils.c drm/i915: Rename "inject_load_failure" module parameter 2019-10-29 15:37:57 +00:00
i915_utils.h drm/i915: Fix i915_inject_load_error() name to read *_probe_* 2019-10-29 15:37:57 +00:00
i915_vgpu.c
i915_vgpu.h
i915_vma.c drm/i915: Split detaching and removing the vma 2019-10-31 14:52:19 +00:00
i915_vma.h drm/i915: Lift i915_vma_parked() onto the gt 2019-10-21 21:07:56 +01:00
intel_csr.c drm/i915/dmc: Update ICL DMC version to v1.09 2019-09-27 10:20:39 -07:00
intel_csr.h
intel_device_info.c drm/i915/display/cnl+: Handle fused off DSC 2019-10-29 12:12:49 -07:00
intel_device_info.h drm/i915/display/cnl+: Handle fused off DSC 2019-10-29 12:12:49 -07:00
intel_gvt.c
intel_gvt.h
intel_memory_region.c drm/i915/lmem: add the fake lmem region 2019-10-31 20:41:47 +00:00
intel_memory_region.h drm/i915/lmem: add the fake lmem region 2019-10-31 20:41:47 +00:00
intel_pch.c drm/i915: Fix detection for a CMP-V PCH 2019-11-13 14:03:15 +02:00
intel_pch.h drm/i915: Fix detection for a CMP-V PCH 2019-11-13 14:03:15 +02:00
intel_pm.c drm/i915: Don't set undefined bits in dirty_pipes 2019-11-29 21:49:16 +02:00
intel_pm.h drm/i915/gen8+: Add RC6 CTX corruption WA 2019-11-05 11:43:07 -08:00
intel_region_lmem.c drm/i915/lmem: fixup fake lmem teardown 2019-11-06 14:34:10 +00:00
intel_region_lmem.h drm/i915/lmem: add the fake lmem region 2019-10-31 20:41:47 +00:00
intel_runtime_pm.c drm/i915/display: abstract all vgaarb access to intel_vga.[ch] 2019-10-02 13:31:54 +03:00
intel_runtime_pm.h
intel_sideband.c
intel_sideband.h
intel_uncore.c
intel_uncore.h
intel_wakeref.c drm/i915: Mark up the calling context for intel_wakeref_put() 2019-11-20 15:59:23 +00:00
intel_wakeref.h drm/i915: Mark intel_wakeref_get() as a sleeper 2019-11-21 13:22:04 +00:00
intel_wopcm.c
intel_wopcm.h
Kconfig drm/i915: Put future HW and their uAPIs under STAGING & BROKEN 2019-10-27 15:47:10 +00:00
Kconfig.debug Merge drm/drm-next into drm-intel-next-queued 2019-11-15 13:17:39 +02:00
Kconfig.profile drm/i915: Default to a more lenient forced preemption timeout 2019-11-26 09:38:44 +00:00
Kconfig.unstable drm/i915/lmem: add the fake lmem region 2019-10-31 20:41:47 +00:00
Makefile drm/i915/selftests: Wrap vm_mmap() around GEM objects 2019-11-07 21:22:58 +00:00