https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3424/shard-snb3/igt@gem_tiled_swapping@non-threaded.html https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3425/shard-hsw4/igt@gem_tiled_swapping@non-threaded.html https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3423/shard-glkb2/igt@gem_tiled_swapping@non-threaded.html https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3416/shard-apl6/igt@gem_tiled_swapping@non-threaded.html no indications of softdog/ timeout in run.log. no hints in dmesg.
Also, https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_7387/shards.html
Now also on KBL-shards: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3426/shard-kbl1/igt@gem_tiled_swapping@non-threaded.html
This run: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3434/shard-hsw3/igt@gem_tiled_swapping@non-threaded.html seem to match this pstore: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3445/shard-hsw3/pstore2-1512165009_Panic_1.log
Created attachment 136141 [details] pstore pulled from shard-hsw4 when Ci_DRM_3506 was running Pstore collection appear to be broken see bug 103706 so to get more data on this I managed to pull this pstore log from shards-hsw4 manually when CI_DRM_3506 was running. I believe this pstore show igt@gem_tiled_swapping@non-threaded behaving badly.
Hmm, looking at __lock_page(), I'm reminded of this bug: Commit 50816c48997a ("sched/wait: Standardize internal naming of wait-queue entries") changed the behavior of add_wait_queue() from inserting the wait entry at the head of the wait queue to the tail of the wait queue. This is the relevant hunk: -void add_wait_queue(wait_queue_head_t *q, wait_queue_entry_t *wait) +void add_wait_queue(wait_queue_head_t *q, struct wait_queue_entry *wq_entry) { unsigned long flags; - wait->flags &= ~WQ_FLAG_EXCLUSIVE; + wq_entry->flags &= ~WQ_FLAG_EXCLUSIVE; spin_lock_irqsave(&q->lock, flags); - __add_wait_queue(q, wait); + __add_wait_queue_entry_tail(q, wq_entry); spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL(add_wait_queue); Note the change from __add_wait_queue() to __add_wait_queue_entry_tail(). I'm assuming this was a typo since the commit message doesn't mention any functional changes. This patch restores the old behavior. tip-bot: Commit-ID: c6b9d9a33029014446bd9ed84c1688f6d3d4eab9 Gitweb: https://git.kernel.org/tip/c6b9d9a33029014446bd9ed84c1688f6d3d4eab9 Author: Omar Sandoval <osandov@fb.com> AuthorDate: Tue, 5 Dec 2017 23:15:31 -0800 Committer: Ingo Molnar <mingo@kernel.org> CommitDate: Wed, 6 Dec 2017 19:30:34 +0100 sched/wait: Fix add_wait_queue() behavioral change Drat, which is already in drm-tip.
I will duplicate this on bug 104218, to get better overview of the impact *** This bug has been marked as a duplicate of bug 104218 ***
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.