Only call synchronize_rcu_expedited after unlocking struct_mutex to
avoid deadlock because the workqueues depend on struct_mutex.
>From original patch by Andrea:
synchronize_rcu/synchronize_sched/synchronize_rcu_expedited() will
hang until its own workqueues are run. The i915 gem workqueues will
wait on the struct_mutex to be released. So we cannot wait for a
quiescent state using those rcu primitives while holding the
struct_mutex or it creates a circular lock dependency resulting in
kernel hangs (which is reproducible but goes undetected by lockdep).
kswapd0 D 0 700 2 0x00000000
Call Trace:
? __schedule+0x1a5/0x660
? schedule+0x36/0x80
? _synchronize_rcu_expedited.constprop.65+0x2ef/0x300
? wake_up_bit+0x20/0x20
? rcu_stall_kick_kthreads.part.54+0xc0/0xc0
? rcu_exp_wait_wake+0x530/0x530
? i915_gem_shrink+0x34b/0x4b0
? i915_gem_shrinker_scan+0x7c/0x90
? i915_gem_shrinker_scan+0x7c/0x90
? shrink_slab.part.61.constprop.72+0x1c1/0x3a0
? shrink_zone+0x154/0x160
? kswapd+0x40a/0x720
? kthread+0xf4/0x130
? try_to_free_pages+0x450/0x450
? kthread_create_on_node+0x40/0x40
? ret_from_fork+0x23/0x30
plasmashell D 0 4657 4614 0x00000000
Call Trace:
? __schedule+0x1a5/0x660
? schedule+0x36/0x80
? schedule_preempt_disabled+0xe/0x10
? __mutex_lock.isra.4+0x1c9/0x790
? i915_gem_close_object+0x26/0xc0
? i915_gem_close_object+0x26/0xc0
? drm_gem_object_release_handle+0x48/0x90
? drm_gem_handle_delete+0x50/0x80
? drm_ioctl+0x1fa/0x420
? drm_gem_handle_create+0x40/0x40
? pipe_write+0x391/0x410
? __vfs_write+0xc6/0x120
? do_vfs_ioctl+0x8b/0x5d0
? SyS_ioctl+0x3b/0x70
? entry_SYSCALL_64_fastpath+0x13/0x94
kworker/0:0 D 0 29186 2 0x00000000
Workqueue: events __i915_gem_free_work
Call Trace:
? __schedule+0x1a5/0x660
? schedule+0x36/0x80
? schedule_preempt_disabled+0xe/0x10
? __mutex_lock.isra.4+0x1c9/0x790
? del_timer_sync+0x44/0x50
? update_curr+0x57/0x110
? __i915_gem_free_objects+0x31/0x300
? __i915_gem_free_objects+0x31/0x300
? __i915_gem_free_work+0x2d/0x40
? process_one_work+0x13a/0x3b0
? worker_thread+0x4a/0x460
? kthread+0xf4/0x130
? process_one_work+0x3b0/0x3b0
? kthread_create_on_node+0x40/0x40
? ret_from_fork+0x23/0x30
Fixes: 3d3d18f086 ("drm/i915: Avoid rcu_barrier() from reclaim paths (shrinker)")
Reported-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
When we retire the last request on the ring, before we ever access that
ring again we know it will be completely idle and so we can advance the
ring->head fully to the end (i.e. ring->tail) and not just to the start
of the breadcrumb. This allows us to skip re-emitting the breadcrumb
after resetting the GPU if the ring was entirely idle. This prevents us
from overwriting a seqno wraparound by re-executing a stale breadcrumb,
i.e.
submit_request(1)
intel_engine_init_global_seqno(0)
i915_reset()
would then leave 1 in the HWS, but the next request to execute would
also be with seqno 1. The sanity checks upon submission detect this as a
timewarp and explode. By setting the ring as empty, upon reset the HWS
is left as 0, leaving it consistent with the timeline.
v2: Fix check for deleting last element of list. We know that this
request is always the first element of the ring, so only if next
points back to the start will this be the only request in flight.
v3: Remove opencoding of list_is_last()
v4: Move the block to its own function for some clarity.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100144
Testcase: igt/gem_exec_whisper/hang-*
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170406170028.26871-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
As per BSPEC, valid cdclk values for glk are 79.2, 158.4, 316.8 Mhz.
Practically we can achive only 99% of these cdclk values (HW team
checking on this). So cdclk should be calculated for the given pixclk as
per that otherwise it may lead to screen corruption, explained below:
1. For DSI AUO panel(1920x1200 @60) required pixclk is 157100 KHZ
2. glk_calc_cdclk returns 79200 KHZ for this pixclk, For 2PPC it
will be 158400 KHZ
3. Practically 100% of the cdclk can’t be achieved, so 99% of 158400
KHZ = 156816 which is less than the desired pixlclk and causes
panel corruption.
v2: Rebased to new CDLCK code framework
v3: Addressed review comments from Ander/Jani
- Add comment in code about 99% usage of CDCLK
- Calculate max dot clock as well with 99% limit
v4 by Jani:
- drop superfluous whitespace change
- rewrite code comments to clarify
v5: Added details of non-working scenario in commit message
Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Madhav Chauhan <madhav.chauhan@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ander Conselvan de Oliveira <conselvan2@gmail.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1491397463-13637-1-git-send-email-madhav.chauhan@intel.com
There is some conflation related to sink rates, making this change more
complicated than it would otherwise have to be. There are three changes
here that are rather difficult to split up:
1) Use the intel_dp->sink_rates array for all DP, not just eDP 1.4. We
initialize it from DPCD on eDP 1.4 like before, but generate it based
on DP_MAX_LINK_RATE on others. This reduces code complexity when we
need to use the sink rates; they are all always in the sink_rates
array.
2) Update the sink rate array whenever we read DPCD, and use the
information from there. This increases code readability when we need
the sink rates.
3) Disentangle fallback rate limiting from sink rates. In the code, the
max rate is a dynamic property of the *link*, not of the *sink*. Do
the limiting after intersecting the source and sink rates, which are
static properties of the devices.
This paves the way for follow-up refactoring that I've refrained from
doing here to keep this change as simple as it possibly can.
v2: introduce use_rate_select and handle non-confirming eDP (Ville)
v3: don't clobber cached eDP rates on short pulse (Ville)
Cc: Manasi Navare <manasi.d.navare@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Manasi Navare <manasi.d.navare@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/071bad76467f8ab2e73f3f61ad52d5a468004c71.1490712890.git.jani.nikula@intel.com
If the engine is continually completing nops, we can saturate the
signaler and keep it working indefinitely. This angers the NMI watchdog!
A good example is to disable semaphores on snb and run igt/gem_exec_nop -
the parallel, multi-engine workloads are more than sufficient to hog the
CPU, preventing the system from even processing ICMP echo replies.
v2: Tvrtko dug into cond_resched() on x86 and found that it only
depended upon preempt_count and not tif_need_resched() - which means
that we would always call schedule() at that point.
Fixes: c81d46138d ("drm/i915: Convert trace-irq to the breadcrumb waiter")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170404120531.10737-1-chris@chris-wilson.co.uk
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Or rather it is used only by intel_ring_pin() to extract the
drm_i915_private which we can easily pass in. As this is a relatively
rare operation, save the space in the struct, and as such it is even
break even in the extra code for passing around the parameter:
add/remove: 0/0 grow/shrink: 2/3 up/down: 15/-15 (0)
function old new delta
intel_init_ring_buffer 906 918 +12
execlists_context_pin 1308 1311 +3
mock_engine 407 403 -4
intel_engine_create_ring 367 363 -4
intel_ring_pin 326 319 -7
Total: Before=1261794, After=1261794, chg +0.00%
v2: Reorder intel_init_ring_buffer to keep the ring setup together:
add/remove: 0/0 grow/shrink: 2/3 up/down: 9/-15 (-6)
function old new delta
intel_init_ring_buffer 906 912 +6
execlists_context_pin 1308 1311 +3
mock_engine 407 403 -4
intel_engine_create_ring 367 363 -4
intel_ring_pin 326 319 -7
Total: Before=1261794, After=1261788, chg -0.00%
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170403113426.25707-1-chris@chris-wilson.co.uk
Make i915_gem_wait_for_idle() be a little heavier in order to try and
guarantee that the GPU is indeed idle (by checking each engine
individually is idle, i.e. all writes are complete and the rings
stopped) after waiting for in-flight requests to be completed.
v2: And return the final error.
v3: Break the wait_for() out from under the WARN -- the macro expansion
is hideous and unreadable in the warning message
v4: If wait_for_engine() fails the result is catastrophic, mark the
device as wedged and wait for the repair team.
References: https://bugs.freedesktop.org/show_bug.cgi?id=98836
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170330145041.9005-4-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
We pretty print the name of an engine in several places, mostly for
debug, but also in the GPU hang report. Using "ring" in the name is
archaic (we call those engines now to differentiate them from the
multiple rings of commands we execute on each engine), quite verbose and
often tautological. We run out of room in our GPU hang report for
instance if we have more than a couple of engines hung simultaneously.
Bit the bullet and update the strings to reflect the common internal names.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170330134820.12273-1-chris@chris-wilson.co.uk
Michał Winiarski pointed out that the debugging infrastructure (such as
trace_dma_fence_release) likes to pretty print the timeline name, long
after we have freed the timeline. Our timelines currently live as part of
the GTT (due to the strict ordering we currently use through each) which
belong to the context. We aim to free the context and release its
hardware resources as soon as we able to (i.e. when the last
fence/request using it has been signaled and retired). As the
.get_timeline_name is purely a debug feature, rather than extending the
lifetime of the context, or splitting it into many different release
phases just to keep the name around, replace the timeline name with a
constant after the fence has been signaled. This avoids the potential
use-after-free.
Reported-by: Krzysztof Olinski <krzysztof.e.olinski@intel.com>
Fixes: 80b204bce8 ("drm/i915: Enable multiple timelines")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: <stable@vger.kernel.org> # v4.10+
Link: http://patchwork.freedesktop.org/patch/msgid/20170330111614.29757-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
