[Intel-gfx] [PATCH 1/2] drm/i915: Move priority bumping for flips earlier

Tue Nov 29 08:50:11 UTC 2016

On 28/11/2016 14:36, Chris Wilson wrote:
> David found another issue with priority bumping from mmioflips, where we
> are accessing the requests concurrently to them being retired and freed.
> Whilst we are skipping the dependency if has been submitted, that is not
> sufficient to stop the dependency from disappearing if another thread
> retires that request. To prevent we can either employ the struct_mutex (or a
> request mutex in the future) to serialise retiring before it is freed.
> Alternatively, we need to keep the dependencies alive using RCU whilst
> they are being accessed via the DFS.
>
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.698111] general protection fault: 0000 [#1] PREEMPT SMP
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.698305] Modules linked in: snd_hda_intel snd_hda_codec snd_hwdep x86_pkg_temp_thermal snd_hda_core coretemp crct10dif_pclmul crc32_pclmul snd_pcm ghash_clmulni_intel mei_me mei i915 e1000e ptp pps_core i2c_hid
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.698750] CPU: 1 PID: 6716 Comm: kworker/u8:2 Not tainted 4.9.0-rc6-CI-Nightly_816+ #1
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.698871] Hardware name: GIGABYTE GB-BKi7A-7500/MFLP7AP-00, BIOS F1 07/27/2016
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.699125] Workqueue: events_unbound intel_mmio_flip_work_func [i915]
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.699266] task: ffff880260a5e800 task.stack: ffffc90000f6c000
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.699361] RIP: 0010:[<ffffffffa006595d>]  [<ffffffffa006595d>] execlists_schedule+0x8d/0x300 [i915]
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.699632] RSP: 0018:ffffc90000f6fcd8  EFLAGS: 00010206
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.699724] RAX: dead0000000000f8 RBX: ffff8801f64b2bf0 RCX: ffff8801f64b2c10
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.699842] RDX: dead000000000100 RSI: 0000000000000000 RDI: ffff8801f64b0458
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.699972] RBP: ffffc90000f6fd68 R08: ffff88026488dc00 R09: 0000000000000002
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700090] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000400
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700195] R13: ffffc90000f6fcf0 R14: ffff88020955aa40 R15: ffff88020955aa68
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700307] FS:  0000000000000000(0000) GS:ffff88026dc80000(0000) knlGS:0000000000000000
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700435] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700532] CR2: 0000000002a69e90 CR3: 0000000002c07000 CR4: 00000000003406e0
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700635] Stack:
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700682]  ffff880260a5e880 ffffc90000f6fd50 ffffffff810af69a ffffc90000f6fd28
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700827]  ffff88020955a628 ffff8801e1eaebf0 0000000000000020 0000000000000000
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.700947]  00000196af1edc96 ffff88025dfa4000 ffff8801f0b030a8 ffffc90000f6fcf0
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701071] Call Trace:
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701117]  [<ffffffff810af69a>] ? dequeue_entity+0x25a/0xb50
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701260]  [<ffffffffa00516be>] fence_set_priority+0x7e/0x80 [i915]
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701406]  [<ffffffffa0051a15>] i915_gem_object_wait_priority+0x85/0x160 [i915]
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701599]  [<ffffffffa008ccd7>] intel_mmio_flip_work_func+0x47/0x2b0 [i915]
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701717]  [<ffffffff81094c4d>] process_one_work+0x14d/0x470
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701809]  [<ffffffff81094fb3>] worker_thread+0x43/0x4e0
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701888]  [<ffffffff81094f70>] ? process_one_work+0x470/0x470
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.701969]  [<ffffffff81094f70>] ? process_one_work+0x470/0x470
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.702072]  [<ffffffff8109a4d5>] kthread+0xc5/0xe0
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.702152]  [<ffffffff81771c59>] ? _raw_spin_unlock_irq+0x9/0x10
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.702234]  [<ffffffff8109a410>] ? kthread_park+0x60/0x60
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.702318]  [<ffffffff81772272>] ret_from_fork+0x22/0x30
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.702387] Code: 89 42 08 48 8b 45 88 48 89 55 c0 4c 89 6d c8 4c 8d 70 d8 4d 8d 7e 28 4d 39 ef 74 72 49 8b 1e 48 8b 13 48 39 d3 48 8d 42 f8 74 3e <48> 8b 10 8b 52 38 41 39 d4 7e 26 48 8b 50 30 48 8b 78 28 48 8d
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.702921] RIP  [<ffffffffa006595d>] execlists_schedule+0x8d/0x300 [i915]
> Nov 25 21:42:54 kbl-gbbki7 kernel: [ 1746.703027]  RSP <ffffc90000f6fcd8>
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711015] ---[ end trace 4ecf3ae63087e670 ]---
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711023] BUG: unable to handle kernel NULL pointer dereference at 000000000000000b
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711070] IP: [<ffffffff810baf76>] __wake_up_common+0x26/0x80
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711104] PGD 25df92067
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711119] PUD 25b1f0067
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711134] PMD 0
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711140]
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711151] Oops: 0000 [#2] PREEMPT SMP
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711172] Modules linked in: snd_hda_intel snd_hda_codec snd_hwdep x86_pkg_temp_thermal snd_hda_core coretemp crct10dif_pclmul crc32_pclmul snd_pcm ghash_clmulni_intel mei_me mei i915 e1000e ptp pps_core i2c_hid
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711298] CPU: 1 PID: 6716 Comm: kworker/u8:2 Tainted: G      D         4.9.0-rc6-CI-Nightly_816+ #1
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711342] Hardware name: GIGABYTE GB-BKi7A-7500/MFLP7AP-00, BIOS F1 07/27/2016
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711386] task: ffff880260a5e800 task.stack: ffffc90000f6c000
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711415] RIP: 0010:[<ffffffff810baf76>]  [<ffffffff810baf76>] __wake_up_common+0x26/0x80
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711449] RSP: 0018:ffffc90000f6fe50  EFLAGS: 00010092
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711464] RAX: 0000000000000282 RBX: ffffc90000f6ff18 RCX: 0000000000000000
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711484] RDX: 000000000000000b RSI: 0000000000000003 RDI: ffffc90000f6ff18
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711503] RBP: ffffc90000f6fe88 R08: 0000000000000000 R09: 0000000000000005
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711522] R10: 00000000ffffffff R11: 00000000000002cb R12: ffffc90000f6ff20
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711541] R13: 0000000000000282 R14: 0000000000000000 R15: 0000000000000003
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711561] FS:  0000000000000000(0000) GS:ffff88026dc80000(0000) knlGS:0000000000000000
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711583] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711599] CR2: 0000000000000028 CR3: 0000000263b53000 CR4: 00000000003406e0
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711617] Stack:
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711624]  0000000100f6fe60 0000000000000000 ffffc90000f6ff18 ffffc90000f6ff10
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711650]  0000000000000282 0000000000000000 ffff88020955aa68 ffffc90000f6fe98
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711674]  ffffffff810bb02e ffffc90000f6fec0 ffffffff810bb8d2 ffff880260a5ee40
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711701] Call Trace:
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711710]  [<ffffffff810bb02e>] __wake_up_locked+0xe/0x10
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711726]  [<ffffffff810bb8d2>] complete+0x32/0x50
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711757]  [<ffffffff8107a1a8>] mm_release+0xb8/0x140
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711773]  [<ffffffff81080e6b>] do_exit+0x28b/0xae0
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711789]  [<ffffffff81773f17>] rewind_stack_do_exit+0x17/0x20
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711806] Code: 5d c3 0f 1f 00 55 48 89 e5 41 57 41 56 41 55 41 54 4c 8d 67 08 53 41 89 f7 48 83 ec 10 89 55 cc 48 8b 57 08 4c 89 45 d0 49 39 d4 <48> 8b 32 74 45 41 89 ce 48 8d 42 e8 4c 8d 6e e8 eb 03 49 89 d5
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.711937] RIP  [<ffffffff810baf76>] __wake_up_common+0x26/0x80
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.713142]  RSP <ffffc90000f6fe50>
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.714340] CR2: 000000000000000b
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.715439] ---[ end trace 4ecf3ae63087e671 ]---
> Nov 25 21:44:11 kbl-gbbki7 kernel: [ 1746.715441] Fixing recursive fault but reboot is needed!
>
> Fixes: 27745e829a5c ("drm/i915/execlists: Use a local lock for dfs_link access")
> Fixes: 9a151987d709 ("drm/i915: Add execution priority boosting for mmioflips")
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: David Weinehall <david.weinehall at linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> ---
>  drivers/gpu/drm/i915/intel_display.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index 77c4ff9efbe3..7a4f7730877a 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -12022,7 +12022,6 @@ static void intel_mmio_flip_work_func(struct work_struct *w)
>  		to_intel_framebuffer(crtc->base.primary->fb);
>  	struct drm_i915_gem_object *obj = intel_fb->obj;
>
> -	i915_gem_object_wait_priority(obj, 0, I915_PRIORITY_DISPLAY);
>  	WARN_ON(i915_gem_object_wait(obj, 0, MAX_SCHEDULE_TIMEOUT, NULL) < 0);
>
>  	intel_pipe_update_start(crtc);
> @@ -12278,6 +12277,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
>  		i915_add_request_no_flush(request);
>  	}
>
> +	i915_gem_object_wait_priority(obj, 0, I915_PRIORITY_DISPLAY);
>  	i915_gem_track_fb(intel_fb_obj(old_fb), obj,
>  			  to_intel_plane(primary)->frontbuffer_bit);
>  	mutex_unlock(&dev->struct_mutex);
>

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>

Regards,

Tvrtko