* [Intel-gfx] [PATCH] drm/i915/gt: Autotune idle timeouts
@ 2020-02-25 16:23 Chris Wilson
2020-02-26 17:39 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for " Patchwork
2020-02-26 18:34 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
0 siblings, 2 replies; 3+ messages in thread
From: Chris Wilson @ 2020-02-25 16:23 UTC (permalink / raw)
To: intel-gfx
As we measure how long it takes for each heartbeat when idling the
system, we have a reasonable expectation for the baseline latency when
idling. We can use this baseline to estimate how long we expect it
should take to idle, and so provide a more precise upper bound for
declaring a problem.
References: b81e4d9b5941 ("drm/i915/gt: Track engine round-trip times")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Stuart Summers <stuart.summers@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
---
drivers/gpu/drm/i915/gt/intel_gt.c | 4 +--
drivers/gpu/drm/i915/gt/intel_gt_pm.c | 2 +-
drivers/gpu/drm/i915/gt/intel_gt_requests.c | 27 +++++++++++++++++++
drivers/gpu/drm/i915/gt/selftest_rc6.c | 2 +-
drivers/gpu/drm/i915/gt/selftest_timeline.c | 2 +-
drivers/gpu/drm/i915/i915_gem.h | 2 --
.../gpu/drm/i915/selftests/igt_flush_test.c | 2 +-
7 files changed, 33 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c
index 3dea8881e915..4057ac4d350a 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt.c
@@ -440,7 +440,7 @@ static int __engines_record_defaults(struct intel_gt *gt)
}
/* Flush the default context image to memory, and enable powersaving. */
- if (intel_gt_wait_for_idle(gt, I915_GEM_IDLE_TIMEOUT) == -ETIME) {
+ if (intel_gt_wait_for_idle(gt, 10) == -ETIME) {
err = -EIO;
goto out;
}
@@ -543,7 +543,7 @@ static int __engines_verify_workarounds(struct intel_gt *gt)
}
/* Flush and restore the kernel context for safety */
- if (intel_gt_wait_for_idle(gt, I915_GEM_IDLE_TIMEOUT) == -ETIME)
+ if (intel_gt_wait_for_idle(gt, 10) == -ETIME)
err = -EIO;
return err;
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_pm.c
index 8b653c0f5e5f..202550c06139 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.c
@@ -248,7 +248,7 @@ static void wait_for_suspend(struct intel_gt *gt)
if (!intel_gt_pm_is_awake(gt))
return;
- if (intel_gt_wait_for_idle(gt, I915_GEM_IDLE_TIMEOUT) == -ETIME) {
+ if (intel_gt_wait_for_idle(gt, 10) == -ETIME) {
/*
* Forcibly cancel outstanding work and leave
* the gpu quiet.
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_requests.c b/drivers/gpu/drm/i915/gt/intel_gt_requests.c
index 8a5054f21bf8..5c9797cb3d0b 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_requests.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_requests.c
@@ -180,12 +180,39 @@ long intel_gt_retire_requests_timeout(struct intel_gt *gt, long timeout)
return active_count ? timeout : 0;
}
+static long
+intel_gt_timeout(struct intel_gt *gt, intel_engine_mask_t mask, int factor)
+{
+ struct intel_engine_cs *engine;
+ intel_engine_mask_t tmp;
+ unsigned long max;
+
+ max = 0;
+ for_each_engine_masked(engine, gt, mask, tmp) {
+ unsigned long latency;
+
+ latency = ewma__engine_latency_read(&engine->latency);
+ if (latency > max)
+ max = latency;
+
+ factor++; /* allow each engine to flush pm sequentially */
+ }
+ if (max == 0) /* no latency measured yet */
+ return MAX_SCHEDULE_TIMEOUT;
+
+ return usecs_to_jiffies(max * factor) + 1;
+}
+
int intel_gt_wait_for_idle(struct intel_gt *gt, long timeout)
{
/* If the device is asleep, we have no requests outstanding */
if (!intel_gt_pm_is_awake(gt))
return 0;
+ /* Adjust our expected jiffie timeout based on historical latency */
+ if (timeout < MAX_SCHEDULE_TIMEOUT)
+ timeout = intel_gt_timeout(gt, ALL_ENGINES, timeout);
+
while ((timeout = intel_gt_retire_requests_timeout(gt, timeout)) > 0) {
cond_resched();
if (signal_pending(current))
diff --git a/drivers/gpu/drm/i915/gt/selftest_rc6.c b/drivers/gpu/drm/i915/gt/selftest_rc6.c
index 5f7e2dcf5686..70d040b39685 100644
--- a/drivers/gpu/drm/i915/gt/selftest_rc6.c
+++ b/drivers/gpu/drm/i915/gt/selftest_rc6.c
@@ -176,7 +176,7 @@ int live_rc6_ctx_wa(void *arg)
goto out;
}
- if (intel_gt_wait_for_idle(gt, HZ / 5) == -ETIME) {
+ if (intel_gt_wait_for_idle(gt, 2) == -ETIME) {
intel_gt_set_wedged(gt);
err = -ETIME;
goto out;
diff --git a/drivers/gpu/drm/i915/gt/selftest_timeline.c b/drivers/gpu/drm/i915/gt/selftest_timeline.c
index c2578a0f2f14..31f6ca3e6d76 100644
--- a/drivers/gpu/drm/i915/gt/selftest_timeline.c
+++ b/drivers/gpu/drm/i915/gt/selftest_timeline.c
@@ -789,7 +789,7 @@ static int live_hwsp_rollover_kernel(void *arg)
int i;
engine_heartbeat_disable(engine, &heartbeat);
- if (intel_gt_wait_for_idle(gt, HZ / 2)) {
+ if (intel_gt_wait_for_idle(gt, 5)) {
err = -EIO;
goto out;
}
diff --git a/drivers/gpu/drm/i915/i915_gem.h b/drivers/gpu/drm/i915/i915_gem.h
index 1753c84d6c0d..51c64ae7833f 100644
--- a/drivers/gpu/drm/i915/i915_gem.h
+++ b/drivers/gpu/drm/i915/i915_gem.h
@@ -82,8 +82,6 @@ struct drm_i915_private;
#define GEM_TRACE_DUMP_ON(expr) BUILD_BUG_ON_INVALID(expr)
#endif
-#define I915_GEM_IDLE_TIMEOUT (HZ / 5)
-
static inline void tasklet_lock(struct tasklet_struct *t)
{
while (!tasklet_trylock(t))
diff --git a/drivers/gpu/drm/i915/selftests/igt_flush_test.c b/drivers/gpu/drm/i915/selftests/igt_flush_test.c
index 7b0939e3f007..44700f29a8e7 100644
--- a/drivers/gpu/drm/i915/selftests/igt_flush_test.c
+++ b/drivers/gpu/drm/i915/selftests/igt_flush_test.c
@@ -19,7 +19,7 @@ int igt_flush_test(struct drm_i915_private *i915)
cond_resched();
- if (intel_gt_wait_for_idle(gt, HZ / 5) == -ETIME) {
+ if (intel_gt_wait_for_idle(gt, 5) == -ETIME) {
pr_err("%pS timed out, cancelling all further testing.\n",
__builtin_return_address(0));
--
2.25.1
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply related [flat|nested] 3+ messages in thread
* [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/gt: Autotune idle timeouts
2020-02-25 16:23 [Intel-gfx] [PATCH] drm/i915/gt: Autotune idle timeouts Chris Wilson
@ 2020-02-26 17:39 ` Patchwork
2020-02-26 18:34 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
1 sibling, 0 replies; 3+ messages in thread
From: Patchwork @ 2020-02-26 17:39 UTC (permalink / raw)
To: Chris Wilson; +Cc: intel-gfx
== Series Details ==
Series: drm/i915/gt: Autotune idle timeouts
URL : https://patchwork.freedesktop.org/series/73919/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
251374a10d4d drm/i915/gt: Autotune idle timeouts
-:12: ERROR:GIT_COMMIT_ID: Please use git commit description style 'commit <12+ chars of sha1> ("<title line>")' - ie: 'commit b81e4d9b5941 ("drm/i915/gt: Track engine round-trip times")'
#12:
References: b81e4d9b5941 ("drm/i915/gt: Track engine round-trip times")
total: 1 errors, 0 warnings, 0 checks, 95 lines checked
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/gt: Autotune idle timeouts
2020-02-25 16:23 [Intel-gfx] [PATCH] drm/i915/gt: Autotune idle timeouts Chris Wilson
2020-02-26 17:39 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for " Patchwork
@ 2020-02-26 18:34 ` Patchwork
1 sibling, 0 replies; 3+ messages in thread
From: Patchwork @ 2020-02-26 18:34 UTC (permalink / raw)
To: Chris Wilson; +Cc: intel-gfx
== Series Details ==
Series: drm/i915/gt: Autotune idle timeouts
URL : https://patchwork.freedesktop.org/series/73919/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8008 -> Patchwork_16709
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with Patchwork_16709 absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in Patchwork_16709, please notify your bug team to allow them
to document this new failure mode, which will reduce false positives in CI.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16709/index.html
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in Patchwork_16709:
### IGT changes ###
#### Possible regressions ####
* igt@i915_selftest@live_gt_lrc:
- fi-bwr-2160: [PASS][1] -> [FAIL][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8008/fi-bwr-2160/igt@i915_selftest@live_gt_lrc.html
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16709/fi-bwr-2160/igt@i915_selftest@live_gt_lrc.html
Known issues
------------
Here are the changes found in Patchwork_16709 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@gem_mmap@basic:
- fi-tgl-y: [PASS][3] -> [DMESG-WARN][4] ([CI#94] / [i915#402])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8008/fi-tgl-y/igt@gem_mmap@basic.html
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16709/fi-tgl-y/igt@gem_mmap@basic.html
* igt@i915_selftest@live_gem_contexts:
- fi-tgl-y: [PASS][5] -> [INCOMPLETE][6] ([CI#94] / [i915#455])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8008/fi-tgl-y/igt@i915_selftest@live_gem_contexts.html
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16709/fi-tgl-y/igt@i915_selftest@live_gem_contexts.html
#### Possible fixes ####
* igt@i915_selftest@live_execlists:
- fi-icl-y: [DMESG-FAIL][7] ([fdo#108569]) -> [PASS][8]
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8008/fi-icl-y/igt@i915_selftest@live_execlists.html
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16709/fi-icl-y/igt@i915_selftest@live_execlists.html
* igt@kms_addfb_basic@bad-pitch-0:
- fi-tgl-y: [DMESG-WARN][9] ([CI#94] / [i915#402]) -> [PASS][10]
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8008/fi-tgl-y/igt@kms_addfb_basic@bad-pitch-0.html
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16709/fi-tgl-y/igt@kms_addfb_basic@bad-pitch-0.html
#### Warnings ####
* igt@kms_chamelium@hdmi-hpd-fast:
- fi-kbl-7500u: [FAIL][11] ([fdo#111407]) -> [FAIL][12] ([fdo#111096] / [i915#323])
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8008/fi-kbl-7500u/igt@kms_chamelium@hdmi-hpd-fast.html
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16709/fi-kbl-7500u/igt@kms_chamelium@hdmi-hpd-fast.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[CI#94]: https://gitlab.freedesktop.org/gfx-ci/i915-infra/issues/94
[fdo#108569]: https://bugs.freedesktop.org/show_bug.cgi?id=108569
[fdo#111096]: https://bugs.freedesktop.org/show_bug.cgi?id=111096
[fdo#111407]: https://bugs.freedesktop.org/show_bug.cgi?id=111407
[i915#1233]: https://gitlab.freedesktop.org/drm/intel/issues/1233
[i915#323]: https://gitlab.freedesktop.org/drm/intel/issues/323
[i915#402]: https://gitlab.freedesktop.org/drm/intel/issues/402
[i915#455]: https://gitlab.freedesktop.org/drm/intel/issues/455
Participating hosts (52 -> 41)
------------------------------
Additional (1): fi-byt-n2820
Missing (12): fi-ilk-m540 fi-hsw-4200u fi-skl-6770hq fi-byt-squawks fi-bsw-cyan fi-snb-2520m fi-ctg-p8600 fi-cfl-8109u fi-skl-lmem fi-byt-clapper fi-bdw-samus fi-snb-2600
Build changes
-------------
* CI: CI-20190529 -> None
* Linux: CI_DRM_8008 -> Patchwork_16709
CI-20190529: 20190529
CI_DRM_8008: 13b6e2575f2c05722679bc1c9d0b97c13bde49a1 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_5469: 4f875016eb1ebc211b8aadb280ae16c7e6cdc8ba @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_16709: 251374a10d4da8da6dc216851ce57cf13c0fe6c0 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
251374a10d4d drm/i915/gt: Autotune idle timeouts
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16709/index.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-02-26 18:34 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-02-25 16:23 [Intel-gfx] [PATCH] drm/i915/gt: Autotune idle timeouts Chris Wilson
2020-02-26 17:39 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for " Patchwork
2020-02-26 18:34 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.