From f4359cedfb43b934f38c50d1604db21333abe57b Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Fri, 10 Nov 2017 16:03:01 +0100 Subject: drm/i915: Fix false-positive assert_rpm_wakelock_held in i915_pmic_bus_access_notifier v2 assert_rpm_wakelock_held is triggered from i915_pmic_bus_access_notifier even though it gets unregistered on (runtime) suspend, this is caused by a race happening under the following circumstances: intel_runtime_pm_put does: atomic_dec(&dev_priv->pm.wakeref_count); pm_runtime_mark_last_busy(kdev); pm_runtime_put_autosuspend(kdev); And pm_runtime_put_autosuspend calls intel_runtime_suspend from a workqueue, so there is ample of time between the atomic_dec() and intel_runtime_suspend() unregistering the notifier. If the notifier gets called in this windowd assert_rpm_wakelock_held falsely triggers (at this point we're not runtime-suspended yet). This commit adds disable_rpm_wakeref_asserts and enable_rpm_wakeref_asserts calls around the intel_uncore_forcewake_get(FORCEWAKE_ALL) call in i915_pmic_bus_access_notifier fixing the false-positive WARN_ON. Changes in v2: -Reword comment explaining why disabling the wakeref asserts is ok and necessary Cc: stable@vger.kernel.org Reported-by: FKr Reviewed-by: Imre Deak Signed-off-by: Hans de Goede Link: https://patchwork.freedesktop.org/patch/msgid/20171110150301.9601-2-hdegoede@redhat.com (cherry picked from commit ce30560c80dead91e98a03d90fb8791e57a9b69d) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_uncore.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'drivers') diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 20e3c65c0999..cb48c8b92aa3 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -1240,8 +1240,15 @@ static int i915_pmic_bus_access_notifier(struct notifier_block *nb, * bus, which will be busy after this notification, leading to: * "render: timed out waiting for forcewake ack request." * errors. + * + * The notifier is unregistered during intel_runtime_suspend(), + * so it's ok to access the HW here without holding a RPM + * wake reference -> disable wakeref asserts for the time of + * the access. */ + disable_rpm_wakeref_asserts(dev_priv); intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL); + enable_rpm_wakeref_asserts(dev_priv); break; case MBI_PMIC_BUS_ACCESS_END: intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL); -- cgit v1.2.3 From 294cf1af8cf2eb0d1eced377fdfb9a2d3f0e8b42 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Tue, 14 Nov 2017 14:55:17 +0100 Subject: drm/i915: Re-register PMIC bus access notifier on runtime resume intel_uncore_suspend() unregisters the uncore code's PMIC bus access notifier and gets called on both normal and runtime suspend. intel_uncore_resume_early() re-registers the notifier, but only on normal resume. Add a new intel_uncore_runtime_resume() function which only re-registers the notifier and call that on runtime resume. Cc: stable@vger.kernel.org Reported-by: Imre Deak Reviewed-by: Imre Deak Signed-off-by: Hans de Goede Link: https://patchwork.freedesktop.org/patch/msgid/20171114135518.15981-2-hdegoede@redhat.com (cherry picked from commit bedf4d79c3654921839b62246b0965ddb308b201) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_drv.c | 2 ++ drivers/gpu/drm/i915/intel_uncore.c | 6 ++++++ drivers/gpu/drm/i915/intel_uncore.h | 1 + 3 files changed, 9 insertions(+) (limited to 'drivers') diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 3db5851756f0..34191028bbad 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -2618,6 +2618,8 @@ static int intel_runtime_resume(struct device *kdev) ret = vlv_resume_prepare(dev_priv, true); } + intel_uncore_runtime_resume(dev_priv); + /* * No point of rolling back things in case of an error, as the best * we can do is to hope that things will still work (and disable RPM). diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index cb48c8b92aa3..8c2ce81f01c2 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -434,6 +434,12 @@ void intel_uncore_resume_early(struct drm_i915_private *dev_priv) i915_check_and_clear_faults(dev_priv); } +void intel_uncore_runtime_resume(struct drm_i915_private *dev_priv) +{ + iosf_mbi_register_pmic_bus_access_notifier( + &dev_priv->uncore.pmic_bus_access_nb); +} + void intel_uncore_sanitize(struct drm_i915_private *dev_priv) { i915_modparams.enable_rc6 = diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h index 582771251b57..9ce079b5dd0d 100644 --- a/drivers/gpu/drm/i915/intel_uncore.h +++ b/drivers/gpu/drm/i915/intel_uncore.h @@ -134,6 +134,7 @@ bool intel_uncore_arm_unclaimed_mmio_detection(struct drm_i915_private *dev_priv void intel_uncore_fini(struct drm_i915_private *dev_priv); void intel_uncore_suspend(struct drm_i915_private *dev_priv); void intel_uncore_resume_early(struct drm_i915_private *dev_priv); +void intel_uncore_runtime_resume(struct drm_i915_private *dev_priv); u64 intel_uncore_edram_size(struct drm_i915_private *dev_priv); void assert_forcewakes_inactive(struct drm_i915_private *dev_priv); -- cgit v1.2.3 From dcd1d8302a06b0a0e5d6f9b6851fa4a5579e79b0 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Tue, 19 Sep 2017 16:55:34 +0100 Subject: drm/i915/gvt: ensure -ve return value is handled correctly An earlier fix changed the return type from find_bb_size however the integer return is being assigned to a unsigned int so the -ve error check will never be detected. Make bb_size an int to fix this. Detected by CoverityScan CID#1456886 ("Unsigned compared against 0") Fixes: 1e3197d6ad73 ("drm/i915/gvt: Refine error handling for perform_bb_shadow") Signed-off-by: Colin Ian King Signed-off-by: Zhenyu Wang (cherry picked from commit 24f8a29af4afe7c53e08f4afa0c3fa9eb3791b89) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/gvt/cmd_parser.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index 701a3c6f1669..85d4c57870fb 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -1628,7 +1628,7 @@ static int perform_bb_shadow(struct parser_exec_state *s) struct intel_shadow_bb_entry *entry_obj; struct intel_vgpu *vgpu = s->vgpu; unsigned long gma = 0; - uint32_t bb_size; + int bb_size; void *dst = NULL; int ret = 0; -- cgit v1.2.3 From 6e068270b7883836e4717e262d96d1c43690862a Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 15 Nov 2017 12:14:58 +0000 Subject: drm/i915: Clear breadcrumb node when cancelling signaling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When we call intel_engine_cancel_signaling() to stop reporting when a request is completed via an asynchronous signal, we remove that request from the breadcrumb wait queue. However, we may be concurrently processing that request in the signaler itself, the actual operations on the request's node itself are serialised but we do not actually clear the waiter after removing it from the tree allowing both parties to attempt to do so and corrupting the rbtree. (Previously removing from the breadcrumb wait queue could only be done on behalf of i915_wait_request, so this race could not happen). Reported-by: "He, Bo" Fixes: 9eb143bbec7d ("drm/i915: Allow a request to be cancelled") Signed-off-by: Chris Wilson Cc: "He, Bo" Cc: Tvrtko Ursulin Cc: Michał Winiarski Cc: Joonas Lahtinen Link: https://patchwork.freedesktop.org/patch/msgid/20171115121458.24655-1-chris@chris-wilson.co.uk Reviewed-by: Joonas Lahtinen (cherry picked from commit c534612e780c4a2c8ef5bfc11583c7d58436baca) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/intel_breadcrumbs.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers') diff --git a/drivers/gpu/drm/i915/intel_breadcrumbs.c b/drivers/gpu/drm/i915/intel_breadcrumbs.c index 48e1ba01ccf8..5f8b9f1f40f1 100644 --- a/drivers/gpu/drm/i915/intel_breadcrumbs.c +++ b/drivers/gpu/drm/i915/intel_breadcrumbs.c @@ -517,6 +517,7 @@ static void __intel_engine_remove_wait(struct intel_engine_cs *engine, GEM_BUG_ON(RB_EMPTY_NODE(&wait->node)); rb_erase(&wait->node, &b->waiters); + RB_CLEAR_NODE(&wait->node); out: GEM_BUG_ON(b->irq_wait == wait); -- cgit v1.2.3 From 457db89b538ea4eb2a188c75f8f3a83469395ee0 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Tue, 14 Nov 2017 17:35:20 +0000 Subject: drm/i915: Mark the userptr invalidate workqueue as WQ_MEM_RECLAIM MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 21cc6431e0c2 ("drm/i915: Mark the userptr invalidate workqueue as WQ_MEM_RECLAIM") tried to fixup the check_flush_dependency warning for hitting i915_gem_userptr_mn_invalidate_range_start from within the shrinker, but I failed to notice userptr has 2 similarly named workqueues. I marked up i915-userptr-acquire as WQ_MEM_RECLAIM whereas we only wait upon i915-userptr-release from inside the reclaim paths. [62530.869510] workqueue: PF_MEMALLOC task 7983(gem_shrink) is flushing !WQ_MEM_RECLAIM i915-userptr-release: (null) [62530.869515] ------------[ cut here ]------------ [62530.869519] WARNING: CPU: 1 PID: 7983 at kernel/workqueue.c:2434 check_flush_dependency+0x7f/0x110 [62530.869519] Modules linked in: pegasus mii ip6table_filter ip6_tables bnep iptable_filter snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic binfmt_misc nls_iso8859_1 intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp snd_hda_intel snd_hda_codec kvm_intel snd_hda_core snd_hwdep kvm snd_pcm irqbypass snd_seq_midi snd_seq_midi_event snd_rawmidi crct10dif_pclmul crc32_pclmul 8250_dw ghash_clmulni_intel snd_seq pcbc snd_seq_device snd_timer btusb aesni_intel btrtl btbcm aes_x86_64 iwlwifi btintel crypto_simd glue_helper cryptd bluetooth snd intel_cstate input_leds idma64 intel_rapl_perf ecdh_generic serio_raw soundcore cfg80211 wmi_bmof virt_dma intel_lpss_pci intel_lpss acpi_als kfifo_buf industrialio winbond_cir soc_button_array rc_core spidev tpm_crb intel_hid acpi_pad mac_hid sparse_keymap [62530.869546] parport_pc ppdev lp parport ip_tables x_tables autofs4 hid_generic usbhid i915 i2c_algo_bit prime_numbers drm_kms_helper syscopyarea e1000e sysfillrect sysimgblt fb_sys_fops ahci ptp pps_core libahci drm wmi video i2c_hid hid [62530.869557] CPU: 1 PID: 7983 Comm: gem_shrink Tainted: G U W L 4.14.0-rc8-drm-tip-ww45-commit-1342299+ #1 [62530.869558] Hardware name: Intel Corporation CoffeeLake Client Platform/CoffeeLake H DDR4 RVP, BIOS CNLSFWR1.R00.X098.A00.1707301945 07/30/2017 [62530.869559] task: ffffa1049dbeec80 task.stack: ffffae7d05c44000 [62530.869560] RIP: 0010:check_flush_dependency+0x7f/0x110 [62530.869561] RSP: 0018:ffffae7d05c473a0 EFLAGS: 00010286 [62530.869562] RAX: 000000000000006e RBX: ffffa1049540f400 RCX: ffffffffa3e55788 [62530.869562] RDX: 0000000000000000 RSI: 0000000000000092 RDI: 0000000000000202 [62530.869563] RBP: ffffae7d05c473c0 R08: 000000000000006e R09: 000000000038bb0e [62530.869563] R10: 0000000000000000 R11: 000000000000006e R12: ffffa1049dbeec80 [62530.869564] R13: 0000000000000000 R14: 0000000000000000 R15: ffffae7d05c473e0 [62530.869565] FS: 00007f621b129880(0000) GS:ffffa1050b240000(0000) knlGS:0000000000000000 [62530.869566] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [62530.869566] CR2: 00007f6214400000 CR3: 0000000353a17003 CR4: 00000000003606e0 [62530.869567] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [62530.869567] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [62530.869568] Call Trace: [62530.869570] flush_workqueue+0x115/0x3d0 [62530.869573] ? wake_up_process+0x15/0x20 [62530.869596] i915_gem_userptr_mn_invalidate_range_start+0x12f/0x160 [i915] [62530.869614] ? i915_gem_userptr_mn_invalidate_range_start+0x12f/0x160 [i915] [62530.869616] __mmu_notifier_invalidate_range_start+0x55/0x80 [62530.869618] try_to_unmap_one+0x791/0x8b0 [62530.869620] ? call_rwsem_down_read_failed+0x18/0x30 [62530.869622] rmap_walk_anon+0x10b/0x260 [62530.869624] rmap_walk+0x48/0x60 [62530.869625] try_to_unmap+0x93/0xf0 [62530.869626] ? page_remove_rmap+0x2a0/0x2a0 [62530.869627] ? page_not_mapped+0x20/0x20 [62530.869629] ? page_get_anon_vma+0x90/0x90 [62530.869630] ? invalid_mkclean_vma+0x20/0x20 [62530.869631] migrate_pages+0x946/0xaa0 [62530.869633] ? __ClearPageMovable+0x10/0x10 [62530.869635] ? isolate_freepages_block+0x3c0/0x3c0 [62530.869636] compact_zone+0x22f/0x970 [62530.869638] compact_zone_order+0xa3/0xd0 [62530.869640] try_to_compact_pages+0x1a5/0x2a0 [62530.869641] ? try_to_compact_pages+0x1a5/0x2a0 [62530.869643] __alloc_pages_direct_compact+0x50/0x110 [62530.869644] __alloc_pages_slowpath+0x4da/0xf30 [62530.869646] __alloc_pages_nodemask+0x262/0x280 [62530.869648] alloc_pages_vma+0x165/0x1e0 [62530.869649] shmem_alloc_hugepage+0xd0/0x130 [62530.869651] ? __radix_tree_insert+0x45/0x230 [62530.869652] ? __vm_enough_memory+0x29/0x130 [62530.869654] shmem_alloc_and_acct_page+0x10d/0x1e0 [62530.869655] shmem_getpage_gfp+0x426/0xc00 [62530.869657] shmem_fault+0xa0/0x1e0 [62530.869659] ? file_update_time+0x60/0x110 [62530.869660] __do_fault+0x1e/0xc0 [62530.869661] __handle_mm_fault+0xa35/0x1170 [62530.869662] handle_mm_fault+0xcc/0x1c0 [62530.869664] __do_page_fault+0x262/0x4f0 [62530.869666] do_page_fault+0x2e/0xe0 [62530.869667] page_fault+0x22/0x30 [62530.869668] RIP: 0033:0x404335 [62530.869669] RSP: 002b:00007fff7829e420 EFLAGS: 00010216 [62530.869670] RAX: 00007f6210400000 RBX: 0000000000000004 RCX: 0000000000b80000 [62530.869670] RDX: 0000000000002e01 RSI: 0000000000008000 RDI: 0000000000000004 [62530.869671] RBP: 0000000000000019 R08: 0000000000000002 R09: 0000000000000000 [62530.869671] R10: 0000000000000559 R11: 0000000000000246 R12: 0000000008000000 [62530.869672] R13: 00000000004042f0 R14: 0000000000000004 R15: 000000000000007e [62530.869673] Code: 00 8b b0 18 05 00 00 48 8d 8b b0 00 00 00 48 8d 90 c0 06 00 00 4d 89 f0 48 c7 c7 40 c0 c8 a3 c6 05 68 c5 e8 00 01 e8 c2 68 04 00 <0f> ff 4d 85 ed 74 18 49 8b 45 20 48 8b 70 08 8b 86 00 01 00 00 [62530.869691] ---[ end trace 01e01ad0ff5781f8 ]--- Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103739 Fixes: 21cc6431e0c2 ("drm/i915: Mark the userptr invalidate workqueue as WQ_MEM_RECLAIM") Signed-off-by: Chris Wilson Cc: Michał Winiarski Link: https://patchwork.freedesktop.org/patch/msgid/20171114173520.8829-1-chris@chris-wilson.co.uk Reviewed-by: Matthew Auld (cherry picked from commit 41729bf2248bc8593e5103d43974079cc269524c) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_gem_userptr.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c index e26b23171b56..ccde12d9e5f5 100644 --- a/drivers/gpu/drm/i915/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c @@ -172,7 +172,9 @@ i915_mmu_notifier_create(struct mm_struct *mm) spin_lock_init(&mn->lock); mn->mn.ops = &i915_gem_userptr_notifier; mn->objects = RB_ROOT_CACHED; - mn->wq = alloc_workqueue("i915-userptr-release", WQ_UNBOUND, 0); + mn->wq = alloc_workqueue("i915-userptr-release", + WQ_UNBOUND | WQ_MEM_RECLAIM, + 0); if (mn->wq == NULL) { kfree(mn); return ERR_PTR(-ENOMEM); @@ -827,7 +829,7 @@ int i915_gem_init_userptr(struct drm_i915_private *dev_priv) dev_priv->mm.userptr_wq = alloc_workqueue("i915-userptr-acquire", - WQ_HIGHPRI | WQ_MEM_RECLAIM, + WQ_HIGHPRI | WQ_UNBOUND, 0); if (!dev_priv->mm.userptr_wq) return -ENOMEM; -- cgit v1.2.3 From 3572f04c69ed4369da5d3c65d84fb18774aa60b6 Mon Sep 17 00:00:00 2001 From: Ville Syrjälä Date: Thu, 16 Nov 2017 18:02:15 +0200 Subject: drm/i915: Fix init_clock_gating for resume MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Moving the init_clock_gating() call from intel_modeset_init_hw() to intel_modeset_gem_init() had an unintended effect of not applying some workarounds on resume. This, for example, cause some kind of corruption to appear at the top of my IVB Thinkpad X1 Carbon LVDS screen after hibernation. Fix the problem by explicitly calling init_clock_gating() from the resume path. I really hope this doesn't break something else again. At least the problems reported at https://bugs.freedesktop.org/show_bug.cgi?id=103549 didn't make a comeback, even after a hibernate cycle. v2: Reorder the init_clock_gating vs. modeset_init_hw to match the display reset path (Rodrigo) Cc: stable@vger.kernel.org Cc: Chris Wilson Cc: Rodrigo Vivi Fixes: 6ac43272768c ("drm/i915: Move init_clock_gating() back to where it was") Reviewed-by: Rodrigo Vivi Reviewed-by: Chris Wilson Link: https://patchwork.freedesktop.org/patch/msgid/20171116160215.25715-1-ville.syrjala@linux.intel.com Signed-off-by: Ville Syrjälä (cherry picked from commit 675f7ff35bd256e65d3d0f52718d8babf5d1002a) Signed-off-by: Jani Nikula --- drivers/gpu/drm/i915/i915_drv.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers') diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 34191028bbad..7d9b07df32fa 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -1714,6 +1714,7 @@ static int i915_drm_resume(struct drm_device *dev) intel_guc_resume(dev_priv); intel_modeset_init_hw(dev); + intel_init_clock_gating(dev_priv); spin_lock_irq(&dev_priv->irq_lock); if (dev_priv->display.hpd_irq_setup) -- cgit v1.2.3