[igt-dev] ✓ Fi.CI.BAT: success for series starting with [v1,1/1] i915/gem_ctx_switch: evenly run 4 child processes

[igt-dev] ✓ Fi.CI.BAT: success for series starting with [v1,1/1] i915/gem_ctx_switch: evenly run 4 child processes

[Patchwork]

== Series Details ==

Series: series starting with [v1,1/1] i915/gem_ctx_switch: evenly run 4 child processes
URL   : https://patchwork.freedesktop.org/series/57120/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_5655 -> IGTPW_2497
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://patchwork.freedesktop.org/api/1.0/series/57120/revisions/1/mbox/

Known issues
------------

  Here are the changes found in IGTPW_2497 that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@gem_exec_suspend@basic-s3:
    - fi-blb-e6850:       PASS -> INCOMPLETE [fdo#107718]

  * igt@kms_busy@basic-flip-c:
    - fi-kbl-7500u:       PASS -> SKIP [fdo#109271] / [fdo#109278] +2

  * igt@kms_chamelium@dp-hpd-fast:
    - fi-kbl-7500u:       PASS -> DMESG-WARN [fdo#102505] / [fdo#103558] / [fdo#105602]

  
#### Possible fixes ####

  * igt@i915_module_load@reload-with-fault-injection:
    - fi-kbl-7567u:       DMESG-WARN [fdo#105602] / [fdo#108529] -> PASS +1

  * igt@i915_pm_rpm@module-reload:
    - fi-kbl-7567u:       DMESG-WARN [fdo#108529] -> PASS
    - {fi-icl-y}:         INCOMPLETE [fdo#108840] -> PASS

  * igt@i915_selftest@live_execlists:
    - fi-apl-guc:         INCOMPLETE [fdo#103927] / [fdo#109720] -> PASS

  * igt@kms_chamelium@common-hpd-after-suspend:
    - fi-kbl-7567u:       DMESG-WARN [fdo#103558] / [fdo#105079] / [fdo#105602] -> PASS

  * igt@kms_chamelium@hdmi-hpd-fast:
    - fi-kbl-7500u:       FAIL [fdo#109485] -> PASS

  * igt@kms_pipe_crc_basic@nonblocking-crc-pipe-c:
    - fi-kbl-7567u:       SKIP [fdo#109271] -> PASS +33

  
#### Warnings ####

  * igt@i915_selftest@live_contexts:
    - fi-icl-u2:          INCOMPLETE [fdo#108569] -> DMESG-FAIL [fdo#108569]

  
  {name}: This element is suppressed. This means it is ignored when computing
          the status of the difference (SUCCESS, WARNING, or FAILURE).

  [fdo#102505]: https://bugs.freedesktop.org/show_bug.cgi?id=102505
  [fdo#103558]: https://bugs.freedesktop.org/show_bug.cgi?id=103558
  [fdo#103927]: https://bugs.freedesktop.org/show_bug.cgi?id=103927
  [fdo#105079]: https://bugs.freedesktop.org/show_bug.cgi?id=105079
  [fdo#105602]: https://bugs.freedesktop.org/show_bug.cgi?id=105602
  [fdo#107718]: https://bugs.freedesktop.org/show_bug.cgi?id=107718
  [fdo#108529]: https://bugs.freedesktop.org/show_bug.cgi?id=108529
  [fdo#108569]: https://bugs.freedesktop.org/show_bug.cgi?id=108569
  [fdo#108840]: https://bugs.freedesktop.org/show_bug.cgi?id=108840
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [fdo#109278]: https://bugs.freedesktop.org/show_bug.cgi?id=109278
  [fdo#109315]: https://bugs.freedesktop.org/show_bug.cgi?id=109315
  [fdo#109485]: https://bugs.freedesktop.org/show_bug.cgi?id=109485
  [fdo#109567]: https://bugs.freedesktop.org/show_bug.cgi?id=109567
  [fdo#109720]: https://bugs.freedesktop.org/show_bug.cgi?id=109720


Participating hosts (45 -> 38)
------------------------------

  Missing    (7): fi-kbl-soraka fi-hsw-4770r fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-skl-6260u fi-bdw-samus 


Build changes
-------------

    * IGT: IGT_4853 -> IGTPW_2497

  CI_DRM_5655: a40729237602fa7454aaf3355ad3058cad5c6ee9 @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_2497: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_2497/
  IGT_4853: 8afdfd8fa9ce17043d9105dedca46ad4555fdcdb @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_2497/
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [igt-dev] [igt PATCH v1 1/1] i915/gem_ctx_switch: evenly run 4 child processes

[Chris Wilson]

Quoting Caz Yokoyama (2019-02-23 06:54:20)
> gem_ctx_switch@basic-all-light test creates 4 child processes for each
> engine and wait. render engine runs slower than others. By this patch,
> 4 child processes run fairly evenly. As a result, parent process runs
> faster. On my NUC, its execution time becomes about half.

A much bigger problem is that there is very little variety in
gem_ctx_switch. There are a few more patterns of context switches that
would be interesting (and need to anticipate different scheduler queue
depths) and a smoketest of those.
-Chris
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [igt-dev] [igt PATCH v1 1/1] i915/gem_ctx_switch: evenly run 4 child processes

[Antonio Argenziano]

On 22/02/19 22:54, Caz Yokoyama wrote:
> gem_ctx_switch@basic-all-light test creates 4 child processes for each
> engine and wait. render engine runs slower than others. By this patch,
> 4 child processes run fairly evenly. As a result, parent process runs
> faster. On my NUC, its execution time becomes about half.
> 
> Signed-off-by: Caz Yokoyama <caz.yokoyama@intel.com>
> ---
>   tests/i915/gem_ctx_switch.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/i915/gem_ctx_switch.c b/tests/i915/gem_ctx_switch.c
> index 1208cb8d..9ed22f35 100644
> --- a/tests/i915/gem_ctx_switch.c
> +++ b/tests/i915/gem_ctx_switch.c
> @@ -247,7 +247,9 @@ static void all(int fd, uint32_t handle, unsigned flags, int timeout)
>   
>   				clock_gettime(CLOCK_MONOTONIC, &start);
>   				do {
> -					for (int loop = 0; loop < qlen; loop++) {
> +					int x = strcmp(name[child], "render") == 0 ? qlen/2 : qlen;
> +
> +					for (int loop = 0; loop < x; loop++) {

Beware that count becomes incorrect. It should be:
count += x;

Antonio

>   						execbuf.rsvd1 = contexts[loop % nctx];
>   						gem_execbuf(fd, &execbuf);
>   					}
> 
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

[PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Chris Wilson]

Not all engines are created equal, and our weighting ends up favouring
the many faster xCS rings at the expense of RCS. Our qlen estimation
also failed to factor in the context switch overhead, which is a
significant factor for nop batches. So we oversubscribe the number of
batches submitted to RCS and end up waiting for those to complete at the
end of our subtest timeslice.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Caz Yokoyama <caz.yokoyama@intel.com>
---
 tests/i915/gem_ctx_switch.c | 39 +++++++++++++++++++++++++++++--------
 1 file changed, 31 insertions(+), 8 deletions(-)

diff --git a/tests/i915/gem_ctx_switch.c b/tests/i915/gem_ctx_switch.c
index 1208cb8d7..87e13b915 100644
--- a/tests/i915/gem_ctx_switch.c
+++ b/tests/i915/gem_ctx_switch.c
@@ -26,6 +26,7 @@
  */
 
 #include "igt.h"
+#include <limits.h>
 #include <unistd.h>
 #include <stdlib.h>
 #include <stdint.h>
@@ -58,29 +59,50 @@ static int measure_qlen(int fd,
 {
 	const struct drm_i915_gem_exec_object2 * const obj =
 		(struct drm_i915_gem_exec_object2 *)(uintptr_t)execbuf->buffers_ptr;
-	int qlen = 64;
+	uint32_t ctx[64];
+	int min = INT_MAX, max = 0;
+
+	for (int i = 0; i < ARRAY_SIZE(ctx); i++)
+		ctx[i] = gem_context_create(fd);
 
 	for (unsigned int n = 0; n < nengine; n++) {
 		uint64_t saved = execbuf->flags;
 		struct timespec tv = {};
+		int q;
 
 		execbuf->flags |= engine[n];
 
-		igt_nsec_elapsed(&tv);
-		for (int loop = 0; loop < qlen; loop++)
+		for (int i = 0; i < ARRAY_SIZE(ctx); i++) {
+			execbuf->rsvd1 = ctx[i];
 			gem_execbuf(fd, execbuf);
+		}
 		gem_sync(fd, obj->handle);
 
-		execbuf->flags = saved;
+		igt_nsec_elapsed(&tv);
+		for (int i = 0; i < ARRAY_SIZE(ctx); i++) {
+			execbuf->rsvd1 = ctx[i];
+			gem_execbuf(fd, execbuf);
+		}
+		gem_sync(fd, obj->handle);
 
 		/*
 		 * Be conservative and aim not to overshoot timeout, so scale
 		 * down by 8 for hopefully a max of 12.5% error.
 		 */
-		qlen = qlen * timeout * 1e9 / igt_nsec_elapsed(&tv) / 8 + 1;
+		q = ARRAY_SIZE(ctx) * timeout * 1e9 / igt_nsec_elapsed(&tv) / 8 + 1;
+		if (q < min)
+			min = q;
+		if (q > max)
+			max = q;
+
+		execbuf->flags = saved;
 	}
 
-	return qlen;
+	for (int i = 0; i < ARRAY_SIZE(ctx); i++)
+		gem_context_destroy(fd, ctx[i]);
+
+	igt_debug("Estimated qlen: {min:%d, max:%d}\n", min, max);
+	return min;
 }
 
 static void single(int fd, uint32_t handle,
@@ -259,9 +281,10 @@ static void all(int fd, uint32_t handle, unsigned flags, int timeout)
 				clock_gettime(CLOCK_MONOTONIC, &now);
 				gem_close(fd, obj[0].handle);
 
-				igt_info("[%d:%d] %s: %'u cycles: %.3fus%s\n",
+				igt_info("[%d:%d] %s: %'u cycles: %.3fus%s (elapsed: %.3fs)\n",
 					 nctx, child, name[child], count, elapsed(&start, &now)*1e6 / count,
-					 flags & INTERRUPTIBLE ? " (interruptible)" : "");
+					 flags & INTERRUPTIBLE ? " (interruptible)" : "",
+					 elapsed(&start, &now));
 			}
 			igt_waitchildren();
 		}
-- 
2.20.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[igt-dev] [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Chris Wilson]

Not all engines are created equal, and our weighting ends up favouring
the many faster xCS rings at the expense of RCS. Our qlen estimation
also failed to factor in the context switch overhead, which is a
significant factor for nop batches. So we oversubscribe the number of
batches submitted to RCS and end up waiting for those to complete at the
end of our subtest timeslice.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Caz Yokoyama <caz.yokoyama@intel.com>
---
 tests/i915/gem_ctx_switch.c | 39 +++++++++++++++++++++++++++++--------
 1 file changed, 31 insertions(+), 8 deletions(-)

diff --git a/tests/i915/gem_ctx_switch.c b/tests/i915/gem_ctx_switch.c
index 1208cb8d7..87e13b915 100644
--- a/tests/i915/gem_ctx_switch.c
+++ b/tests/i915/gem_ctx_switch.c
@@ -26,6 +26,7 @@
  */
 
 #include "igt.h"
+#include <limits.h>
 #include <unistd.h>
 #include <stdlib.h>
 #include <stdint.h>
@@ -58,29 +59,50 @@ static int measure_qlen(int fd,
 {
 	const struct drm_i915_gem_exec_object2 * const obj =
 		(struct drm_i915_gem_exec_object2 *)(uintptr_t)execbuf->buffers_ptr;
-	int qlen = 64;
+	uint32_t ctx[64];
+	int min = INT_MAX, max = 0;
+
+	for (int i = 0; i < ARRAY_SIZE(ctx); i++)
+		ctx[i] = gem_context_create(fd);
 
 	for (unsigned int n = 0; n < nengine; n++) {
 		uint64_t saved = execbuf->flags;
 		struct timespec tv = {};
+		int q;
 
 		execbuf->flags |= engine[n];
 
-		igt_nsec_elapsed(&tv);
-		for (int loop = 0; loop < qlen; loop++)
+		for (int i = 0; i < ARRAY_SIZE(ctx); i++) {
+			execbuf->rsvd1 = ctx[i];
 			gem_execbuf(fd, execbuf);
+		}
 		gem_sync(fd, obj->handle);
 
-		execbuf->flags = saved;
+		igt_nsec_elapsed(&tv);
+		for (int i = 0; i < ARRAY_SIZE(ctx); i++) {
+			execbuf->rsvd1 = ctx[i];
+			gem_execbuf(fd, execbuf);
+		}
+		gem_sync(fd, obj->handle);
 
 		/*
 		 * Be conservative and aim not to overshoot timeout, so scale
 		 * down by 8 for hopefully a max of 12.5% error.
 		 */
-		qlen = qlen * timeout * 1e9 / igt_nsec_elapsed(&tv) / 8 + 1;
+		q = ARRAY_SIZE(ctx) * timeout * 1e9 / igt_nsec_elapsed(&tv) / 8 + 1;
+		if (q < min)
+			min = q;
+		if (q > max)
+			max = q;
+
+		execbuf->flags = saved;
 	}
 
-	return qlen;
+	for (int i = 0; i < ARRAY_SIZE(ctx); i++)
+		gem_context_destroy(fd, ctx[i]);
+
+	igt_debug("Estimated qlen: {min:%d, max:%d}\n", min, max);
+	return min;
 }
 
 static void single(int fd, uint32_t handle,
@@ -259,9 +281,10 @@ static void all(int fd, uint32_t handle, unsigned flags, int timeout)
 				clock_gettime(CLOCK_MONOTONIC, &now);
 				gem_close(fd, obj[0].handle);
 
-				igt_info("[%d:%d] %s: %'u cycles: %.3fus%s\n",
+				igt_info("[%d:%d] %s: %'u cycles: %.3fus%s (elapsed: %.3fs)\n",
 					 nctx, child, name[child], count, elapsed(&start, &now)*1e6 / count,
-					 flags & INTERRUPTIBLE ? " (interruptible)" : "");
+					 flags & INTERRUPTIBLE ? " (interruptible)" : "",
+					 elapsed(&start, &now));
 			}
 			igt_waitchildren();
 		}
-- 
2.20.1

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

[igt-dev] ✓ Fi.CI.BAT: success for series starting with [i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches (rev2)

[Patchwork]

== Series Details ==

Series: series starting with [i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches (rev2)
URL   : https://patchwork.freedesktop.org/series/57120/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_5655 -> IGTPW_2499
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://patchwork.freedesktop.org/api/1.0/series/57120/revisions/2/mbox/

Known issues
------------

  Here are the changes found in IGTPW_2499 that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@amdgpu/amd_basic@userptr:
    - fi-kbl-8809g:       PASS -> DMESG-WARN [fdo#108965]

  * igt@gem_exec_suspend@basic-s3:
    - fi-blb-e6850:       PASS -> INCOMPLETE [fdo#107718]

  * igt@i915_pm_rpm@module-reload:
    - fi-skl-6770hq:      PASS -> FAIL [fdo#108511]

  
#### Possible fixes ####

  * igt@i915_module_load@reload-with-fault-injection:
    - fi-kbl-7567u:       DMESG-WARN [fdo#105602] / [fdo#108529] -> PASS +1

  * igt@i915_pm_rpm@basic-pci-d3-state:
    - fi-bsw-kefka:       SKIP [fdo#109271] -> PASS

  * igt@i915_pm_rpm@basic-rte:
    - fi-bsw-kefka:       FAIL [fdo#108800] -> PASS

  * igt@i915_pm_rpm@module-reload:
    - fi-kbl-7567u:       DMESG-WARN [fdo#108529] -> PASS
    - {fi-icl-y}:         INCOMPLETE [fdo#108840] -> PASS

  * igt@i915_selftest@live_execlists:
    - fi-apl-guc:         INCOMPLETE [fdo#103927] / [fdo#109720] -> PASS

  * igt@kms_chamelium@common-hpd-after-suspend:
    - fi-kbl-7567u:       DMESG-WARN [fdo#103558] / [fdo#105079] / [fdo#105602] -> PASS

  * igt@kms_pipe_crc_basic@nonblocking-crc-pipe-c:
    - fi-kbl-7567u:       SKIP [fdo#109271] -> PASS +33

  
#### Warnings ####

  * igt@i915_selftest@live_contexts:
    - fi-icl-u2:          INCOMPLETE [fdo#108569] -> DMESG-FAIL [fdo#108569]

  
  {name}: This element is suppressed. This means it is ignored when computing
          the status of the difference (SUCCESS, WARNING, or FAILURE).

  [fdo#103558]: https://bugs.freedesktop.org/show_bug.cgi?id=103558
  [fdo#103927]: https://bugs.freedesktop.org/show_bug.cgi?id=103927
  [fdo#105079]: https://bugs.freedesktop.org/show_bug.cgi?id=105079
  [fdo#105602]: https://bugs.freedesktop.org/show_bug.cgi?id=105602
  [fdo#107718]: https://bugs.freedesktop.org/show_bug.cgi?id=107718
  [fdo#108511]: https://bugs.freedesktop.org/show_bug.cgi?id=108511
  [fdo#108529]: https://bugs.freedesktop.org/show_bug.cgi?id=108529
  [fdo#108569]: https://bugs.freedesktop.org/show_bug.cgi?id=108569
  [fdo#108800]: https://bugs.freedesktop.org/show_bug.cgi?id=108800
  [fdo#108840]: https://bugs.freedesktop.org/show_bug.cgi?id=108840
  [fdo#108965]: https://bugs.freedesktop.org/show_bug.cgi?id=108965
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [fdo#109315]: https://bugs.freedesktop.org/show_bug.cgi?id=109315
  [fdo#109720]: https://bugs.freedesktop.org/show_bug.cgi?id=109720


Participating hosts (45 -> 40)
------------------------------

  Missing    (5): fi-kbl-soraka fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-bdw-samus 


Build changes
-------------

    * IGT: IGT_4853 -> IGTPW_2499

  CI_DRM_5655: a40729237602fa7454aaf3355ad3058cad5c6ee9 @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_2499: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_2499/
  IGT_4853: 8afdfd8fa9ce17043d9105dedca46ad4555fdcdb @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_2499/
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

[igt-dev] ✓ Fi.CI.IGT: success for series starting with [v1,1/1] i915/gem_ctx_switch: evenly run 4 child processes

[Patchwork]

== Series Details ==

Series: series starting with [v1,1/1] i915/gem_ctx_switch: evenly run 4 child processes
URL   : https://patchwork.freedesktop.org/series/57120/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_5655_full -> IGTPW_2497_full
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://patchwork.freedesktop.org/api/1.0/series/57120/revisions/1/mbox/

Known issues
------------

  Here are the changes found in IGTPW_2497_full that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@gem_busy@busy-vebox:
    - shard-snb:          NOTRUN -> SKIP [fdo#109271] +31

  * igt@gem_ctx_param@invalid-param-set:
    - shard-kbl:          NOTRUN -> FAIL [fdo#109674]

  * igt@i915_query@query-topology-unsupported:
    - shard-apl:          NOTRUN -> SKIP [fdo#109271] +28

  * igt@kms_atomic_transition@1x-modeset-transitions-nonblocking-fencing:
    - shard-apl:          PASS -> FAIL [fdo#109660]

  * igt@kms_busy@basic-flip-f:
    - shard-apl:          NOTRUN -> SKIP [fdo#109271] / [fdo#109278]

  * igt@kms_busy@basic-modeset-d:
    - shard-glk:          NOTRUN -> SKIP [fdo#109271] / [fdo#109278]

  * igt@kms_busy@basic-modeset-e:
    - shard-kbl:          NOTRUN -> SKIP [fdo#109271] / [fdo#109278] +1

  * igt@kms_busy@extended-modeset-hang-newfb-with-reset-render-a:
    - shard-kbl:          NOTRUN -> DMESG-WARN [fdo#107956]

  * igt@kms_ccs@pipe-a-crc-sprite-planes-basic:
    - shard-apl:          PASS -> FAIL [fdo#106510] / [fdo#108145]

  * igt@kms_chv_cursor_fail@pipe-b-256x256-left-edge:
    - shard-snb:          PASS -> INCOMPLETE [fdo#105411]

  * igt@kms_content_protection@legacy:
    - shard-glk:          NOTRUN -> SKIP [fdo#109271] +11

  * igt@kms_cursor_crc@cursor-128x42-sliding:
    - shard-kbl:          PASS -> FAIL [fdo#103232] +2

  * igt@kms_cursor_crc@cursor-256x256-random:
    - shard-apl:          PASS -> FAIL [fdo#103232] +4

  * igt@kms_cursor_crc@cursor-64x64-suspend:
    - shard-apl:          PASS -> FAIL [fdo#103191] / [fdo#103232]

  * igt@kms_cursor_crc@cursor-size-change:
    - shard-glk:          PASS -> FAIL [fdo#103232]

  * igt@kms_cursor_legacy@2x-long-nonblocking-modeset-vs-cursor-atomic:
    - shard-glk:          NOTRUN -> FAIL [fdo#105454] / [fdo#106509]

  * igt@kms_flip@flip-vs-blocking-wf-vblank:
    - shard-hsw:          PASS -> INCOMPLETE [fdo#103540]

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-blt:
    - shard-apl:          NOTRUN -> FAIL [fdo#103167]

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-cpu:
    - shard-apl:          PASS -> FAIL [fdo#103167] +3

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-move:
    - shard-kbl:          PASS -> FAIL [fdo#103167]

  * igt@kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-onoff:
    - shard-glk:          PASS -> FAIL [fdo#103167] +5

  * igt@kms_frontbuffer_tracking@psr-1p-primscrn-indfb-plflip-blt:
    - shard-kbl:          NOTRUN -> SKIP [fdo#109271] +47

  * igt@kms_plane@plane-position-covered-pipe-a-planes:
    - shard-glk:          PASS -> FAIL [fdo#103166] +2
    - shard-apl:          PASS -> FAIL [fdo#103166] +2

  * igt@kms_plane@plane-position-covered-pipe-b-planes:
    - shard-kbl:          PASS -> FAIL [fdo#103166]

  * igt@kms_plane_alpha_blend@pipe-a-alpha-transparant-fb:
    - shard-kbl:          NOTRUN -> FAIL [fdo#108145]

  * igt@kms_plane_alpha_blend@pipe-b-alpha-transparant-fb:
    - shard-apl:          NOTRUN -> FAIL [fdo#108145]
    - shard-snb:          NOTRUN -> SKIP [fdo#109271] / [fdo#109278] +2

  * igt@kms_plane_multiple@atomic-pipe-a-tiling-none:
    - shard-glk:          NOTRUN -> FAIL [fdo#103166]

  * igt@kms_plane_multiple@atomic-pipe-a-tiling-y:
    - shard-apl:          NOTRUN -> FAIL [fdo#103166]

  * igt@kms_rotation_crc@multiplane-rotation-cropping-top:
    - shard-kbl:          PASS -> FAIL [fdo#109016] +1

  * igt@kms_vblank@pipe-b-ts-continuation-suspend:
    - shard-apl:          PASS -> FAIL [fdo#104894]

  
#### Possible fixes ####

  * igt@i915_pm_rc6_residency@rc6-accuracy:
    - shard-snb:          SKIP [fdo#109271] -> PASS

  * igt@i915_suspend@fence-restore-tiled2untiled:
    - shard-kbl:          INCOMPLETE [fdo#103665] -> PASS

  * igt@kms_atomic_transition@1x-modeset-transitions-nonblocking:
    - shard-apl:          FAIL [fdo#109660] -> PASS

  * igt@kms_ccs@pipe-b-crc-sprite-planes-basic:
    - shard-kbl:          FAIL [fdo#107725] / [fdo#108145] -> PASS

  * igt@kms_color@pipe-c-legacy-gamma:
    - shard-glk:          FAIL [fdo#104782] -> PASS

  * igt@kms_cursor_crc@cursor-128x128-random:
    - shard-apl:          FAIL [fdo#103232] -> PASS +2

  * igt@kms_cursor_crc@cursor-64x64-random:
    - shard-kbl:          FAIL [fdo#103232] -> PASS

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-pwrite:
    - shard-apl:          FAIL [fdo#103167] -> PASS +2

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen:
    - shard-glk:          FAIL [fdo#103167] -> PASS +1
    - shard-kbl:          FAIL [fdo#103167] -> PASS +1

  * igt@kms_plane@pixel-format-pipe-c-planes:
    - shard-glk:          FAIL [fdo#103166] -> PASS +1

  * igt@kms_plane_alpha_blend@pipe-a-alpha-opaque-fb:
    - shard-apl:          FAIL [fdo#108145] -> PASS
    - shard-glk:          FAIL [fdo#108145] -> PASS +1
    - shard-kbl:          FAIL [fdo#108145] -> PASS

  * igt@kms_plane_multiple@atomic-pipe-c-tiling-none:
    - shard-apl:          FAIL [fdo#103166] -> PASS

  * igt@kms_setmode@basic:
    - shard-kbl:          FAIL [fdo#99912] -> PASS

  
  [fdo#103166]: https://bugs.freedesktop.org/show_bug.cgi?id=103166
  [fdo#103167]: https://bugs.freedesktop.org/show_bug.cgi?id=103167
  [fdo#103191]: https://bugs.freedesktop.org/show_bug.cgi?id=103191
  [fdo#103232]: https://bugs.freedesktop.org/show_bug.cgi?id=103232
  [fdo#103540]: https://bugs.freedesktop.org/show_bug.cgi?id=103540
  [fdo#103665]: https://bugs.freedesktop.org/show_bug.cgi?id=103665
  [fdo#104782]: https://bugs.freedesktop.org/show_bug.cgi?id=104782
  [fdo#104894]: https://bugs.freedesktop.org/show_bug.cgi?id=104894
  [fdo#105411]: https://bugs.freedesktop.org/show_bug.cgi?id=105411
  [fdo#105454]: https://bugs.freedesktop.org/show_bug.cgi?id=105454
  [fdo#106509]: https://bugs.freedesktop.org/show_bug.cgi?id=106509
  [fdo#106510]: https://bugs.freedesktop.org/show_bug.cgi?id=106510
  [fdo#107725]: https://bugs.freedesktop.org/show_bug.cgi?id=107725
  [fdo#107956]: https://bugs.freedesktop.org/show_bug.cgi?id=107956
  [fdo#108145]: https://bugs.freedesktop.org/show_bug.cgi?id=108145
  [fdo#109016]: https://bugs.freedesktop.org/show_bug.cgi?id=109016
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [fdo#109278]: https://bugs.freedesktop.org/show_bug.cgi?id=109278
  [fdo#109660]: https://bugs.freedesktop.org/show_bug.cgi?id=109660
  [fdo#109674]: https://bugs.freedesktop.org/show_bug.cgi?id=109674
  [fdo#99912]: https://bugs.freedesktop.org/show_bug.cgi?id=99912


Participating hosts (6 -> 5)
------------------------------

  Missing    (1): shard-skl 


Build changes
-------------

    * IGT: IGT_4853 -> IGTPW_2497
    * Piglit: piglit_4509 -> None

  CI_DRM_5655: a40729237602fa7454aaf3355ad3058cad5c6ee9 @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_2497: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_2497/
  IGT_4853: 8afdfd8fa9ce17043d9105dedca46ad4555fdcdb @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_2497/
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

[igt-dev] [igt PATCH v1 1/1] i915/gem_ctx_switch: evenly run 4 child processes

[Caz Yokoyama]

gem_ctx_switch@basic-all-light test creates 4 child processes for each
engine and wait. render engine runs slower than others. By this patch,
4 child processes run fairly evenly. As a result, parent process runs
faster. On my NUC, its execution time becomes about half.

Signed-off-by: Caz Yokoyama <caz.yokoyama@intel.com>
---
 tests/i915/gem_ctx_switch.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tests/i915/gem_ctx_switch.c b/tests/i915/gem_ctx_switch.c
index 1208cb8d..9ed22f35 100644
--- a/tests/i915/gem_ctx_switch.c
+++ b/tests/i915/gem_ctx_switch.c
@@ -247,7 +247,9 @@ static void all(int fd, uint32_t handle, unsigned flags, int timeout)
 
 				clock_gettime(CLOCK_MONOTONIC, &start);
 				do {
-					for (int loop = 0; loop < qlen; loop++) {
+					int x = strcmp(name[child], "render") == 0 ? qlen/2 : qlen;
+
+					for (int loop = 0; loop < x; loop++) {
 						execbuf.rsvd1 = contexts[loop % nctx];
 						gem_execbuf(fd, &execbuf);
 					}
-- 
2.17.1

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

[igt-dev] ✓ Fi.CI.IGT: success for series starting with [i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches (rev2)

[Patchwork]

== Series Details ==

Series: series starting with [i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches (rev2)
URL   : https://patchwork.freedesktop.org/series/57120/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_5655_full -> IGTPW_2499_full
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://patchwork.freedesktop.org/api/1.0/series/57120/revisions/2/mbox/

Known issues
------------

  Here are the changes found in IGTPW_2499_full that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@gem_busy@busy-vebox:
    - shard-snb:          NOTRUN -> SKIP [fdo#109271] +34

  * igt@gem_ctx_param@invalid-param-set:
    - shard-kbl:          NOTRUN -> FAIL [fdo#109674]

  * igt@i915_query@query-topology-unsupported:
    - shard-apl:          NOTRUN -> SKIP [fdo#109271] +29

  * igt@kms_busy@basic-flip-f:
    - shard-apl:          NOTRUN -> SKIP [fdo#109271] / [fdo#109278] +1

  * igt@kms_busy@basic-modeset-d:
    - shard-glk:          NOTRUN -> SKIP [fdo#109271] / [fdo#109278]

  * igt@kms_busy@basic-modeset-e:
    - shard-kbl:          NOTRUN -> SKIP [fdo#109271] / [fdo#109278] +1

  * igt@kms_busy@extended-modeset-hang-newfb-with-reset-render-a:
    - shard-kbl:          NOTRUN -> DMESG-WARN [fdo#107956]

  * igt@kms_color@pipe-b-degamma:
    - shard-apl:          PASS -> FAIL [fdo#104782] +1

  * igt@kms_content_protection@legacy:
    - shard-glk:          NOTRUN -> SKIP [fdo#109271] +11

  * igt@kms_cursor_crc@cursor-128x42-onscreen:
    - shard-apl:          NOTRUN -> FAIL [fdo#103232]

  * igt@kms_cursor_crc@cursor-256x256-random:
    - shard-kbl:          PASS -> FAIL [fdo#103232] +1

  * igt@kms_cursor_crc@cursor-256x256-suspend:
    - shard-kbl:          PASS -> FAIL [fdo#103191] / [fdo#103232]

  * igt@kms_cursor_crc@cursor-256x85-random:
    - shard-apl:          PASS -> FAIL [fdo#103232] +5

  * igt@kms_cursor_crc@cursor-64x64-suspend:
    - shard-apl:          PASS -> FAIL [fdo#103191] / [fdo#103232] +1

  * igt@kms_cursor_crc@cursor-alpha-opaque:
    - shard-kbl:          NOTRUN -> FAIL [fdo#109350]
    - shard-apl:          PASS -> FAIL [fdo#109350]

  * igt@kms_cursor_crc@cursor-size-change:
    - shard-glk:          PASS -> FAIL [fdo#103232]

  * igt@kms_cursor_legacy@2x-long-nonblocking-modeset-vs-cursor-atomic:
    - shard-glk:          NOTRUN -> FAIL [fdo#105454] / [fdo#106509]

  * igt@kms_flip@flip-vs-expired-vblank-interruptible:
    - shard-glk:          PASS -> FAIL [fdo#105363]

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-blt:
    - shard-apl:          NOTRUN -> FAIL [fdo#103167]

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-gtt:
    - shard-kbl:          PASS -> FAIL [fdo#103167]

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-render:
    - shard-apl:          PASS -> FAIL [fdo#103167]

  * igt@kms_frontbuffer_tracking@fbc-1p-rte:
    - shard-glk:          PASS -> FAIL [fdo#103167] / [fdo#105682] +1

  * igt@kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-onoff:
    - shard-glk:          PASS -> FAIL [fdo#103167] +9

  * igt@kms_frontbuffer_tracking@fbc-suspend:
    - shard-snb:          PASS -> DMESG-WARN [fdo#102365]

  * igt@kms_frontbuffer_tracking@psr-1p-primscrn-indfb-plflip-blt:
    - shard-kbl:          NOTRUN -> SKIP [fdo#109271] +47

  * igt@kms_plane@pixel-format-pipe-a-planes-source-clamping:
    - shard-glk:          PASS -> FAIL [fdo#108948]
    - shard-apl:          PASS -> FAIL [fdo#108948]
    - shard-kbl:          NOTRUN -> FAIL [fdo#108948]

  * igt@kms_plane_alpha_blend@pipe-a-alpha-transparant-fb:
    - shard-kbl:          NOTRUN -> FAIL [fdo#108145]

  * igt@kms_plane_alpha_blend@pipe-b-alpha-transparant-fb:
    - shard-apl:          NOTRUN -> FAIL [fdo#108145]
    - shard-snb:          NOTRUN -> SKIP [fdo#109271] / [fdo#109278] +3

  * igt@kms_plane_multiple@atomic-pipe-a-tiling-none:
    - shard-glk:          NOTRUN -> FAIL [fdo#103166]

  * igt@kms_plane_multiple@atomic-pipe-b-tiling-none:
    - shard-glk:          PASS -> FAIL [fdo#103166] +7

  * igt@kms_plane_multiple@atomic-pipe-b-tiling-x:
    - shard-apl:          PASS -> FAIL [fdo#103166] +1

  * igt@kms_setmode@basic:
    - shard-apl:          PASS -> FAIL [fdo#99912]

  * igt@kms_vblank@pipe-c-ts-continuation-dpms-suspend:
    - shard-apl:          PASS -> FAIL [fdo#104894]

  
#### Possible fixes ####

  * igt@i915_suspend@fence-restore-tiled2untiled:
    - shard-kbl:          INCOMPLETE [fdo#103665] -> PASS

  * igt@kms_atomic_transition@1x-modeset-transitions-nonblocking:
    - shard-apl:          FAIL [fdo#109660] -> PASS

  * igt@kms_ccs@pipe-b-crc-sprite-planes-basic:
    - shard-kbl:          FAIL [fdo#107725] / [fdo#108145] -> PASS

  * igt@kms_cursor_crc@cursor-128x128-random:
    - shard-apl:          FAIL [fdo#103232] -> PASS +2

  * igt@kms_cursor_crc@cursor-64x64-random:
    - shard-kbl:          FAIL [fdo#103232] -> PASS

  * igt@kms_cursor_crc@cursor-alpha-opaque:
    - shard-glk:          FAIL [fdo#109350] -> PASS

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-onoff:
    - shard-apl:          FAIL [fdo#103167] -> PASS +1

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen:
    - shard-glk:          FAIL [fdo#103167] -> PASS +1
    - shard-kbl:          FAIL [fdo#103167] -> PASS +1

  * igt@kms_plane@pixel-format-pipe-c-planes:
    - shard-glk:          FAIL [fdo#103166] -> PASS

  * igt@kms_plane_alpha_blend@pipe-a-alpha-opaque-fb:
    - shard-apl:          FAIL [fdo#108145] -> PASS
    - shard-glk:          FAIL [fdo#108145] -> PASS +1
    - shard-kbl:          FAIL [fdo#108145] -> PASS

  * igt@kms_plane_multiple@atomic-pipe-a-tiling-none:
    - shard-kbl:          FAIL [fdo#103166] -> PASS
    - shard-apl:          FAIL [fdo#103166] -> PASS +1

  * igt@kms_setmode@basic:
    - shard-kbl:          FAIL [fdo#99912] -> PASS

  
  [fdo#102365]: https://bugs.freedesktop.org/show_bug.cgi?id=102365
  [fdo#103166]: https://bugs.freedesktop.org/show_bug.cgi?id=103166
  [fdo#103167]: https://bugs.freedesktop.org/show_bug.cgi?id=103167
  [fdo#103191]: https://bugs.freedesktop.org/show_bug.cgi?id=103191
  [fdo#103232]: https://bugs.freedesktop.org/show_bug.cgi?id=103232
  [fdo#103665]: https://bugs.freedesktop.org/show_bug.cgi?id=103665
  [fdo#104782]: https://bugs.freedesktop.org/show_bug.cgi?id=104782
  [fdo#104894]: https://bugs.freedesktop.org/show_bug.cgi?id=104894
  [fdo#105363]: https://bugs.freedesktop.org/show_bug.cgi?id=105363
  [fdo#105454]: https://bugs.freedesktop.org/show_bug.cgi?id=105454
  [fdo#105682]: https://bugs.freedesktop.org/show_bug.cgi?id=105682
  [fdo#106509]: https://bugs.freedesktop.org/show_bug.cgi?id=106509
  [fdo#107725]: https://bugs.freedesktop.org/show_bug.cgi?id=107725
  [fdo#107956]: https://bugs.freedesktop.org/show_bug.cgi?id=107956
  [fdo#108145]: https://bugs.freedesktop.org/show_bug.cgi?id=108145
  [fdo#108948]: https://bugs.freedesktop.org/show_bug.cgi?id=108948
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [fdo#109278]: https://bugs.freedesktop.org/show_bug.cgi?id=109278
  [fdo#109350]: https://bugs.freedesktop.org/show_bug.cgi?id=109350
  [fdo#109660]: https://bugs.freedesktop.org/show_bug.cgi?id=109660
  [fdo#109674]: https://bugs.freedesktop.org/show_bug.cgi?id=109674
  [fdo#99912]: https://bugs.freedesktop.org/show_bug.cgi?id=99912


Participating hosts (6 -> 5)
------------------------------

  Missing    (1): shard-skl 


Build changes
-------------

    * IGT: IGT_4853 -> IGTPW_2499
    * Piglit: piglit_4509 -> None

  CI_DRM_5655: a40729237602fa7454aaf3355ad3058cad5c6ee9 @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_2499: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_2499/
  IGT_4853: 8afdfd8fa9ce17043d9105dedca46ad4555fdcdb @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_2499/
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Caz Yokoyama]

Chris,
By your patch, measure_qlen() reports how many gem_execbuf() can be
executed(queue length) within timeout of the slowest engine, correct?

Run time becomes 95 sec which is less than half.
-caz

On Sat, 2019-02-23 at 01:34 +0000, Chris Wilson wrote:
> Not all engines are created equal, and our weighting ends up
> favouring
> the many faster xCS rings at the expense of RCS. Our qlen estimation
> also failed to factor in the context switch overhead, which is a
> significant factor for nop batches. So we oversubscribe the number of
> batches submitted to RCS and end up waiting for those to complete at
> the
> end of our subtest timeslice.
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Caz Yokoyama <caz.yokoyama@intel.com>
> ---
>  tests/i915/gem_ctx_switch.c | 39 +++++++++++++++++++++++++++++----
> ----
>  1 file changed, 31 insertions(+), 8 deletions(-)
> 
> diff --git a/tests/i915/gem_ctx_switch.c
> b/tests/i915/gem_ctx_switch.c
> index 1208cb8d7..87e13b915 100644
> --- a/tests/i915/gem_ctx_switch.c
> +++ b/tests/i915/gem_ctx_switch.c
> @@ -26,6 +26,7 @@
>   */
>  
>  #include "igt.h"
> +#include <limits.h>
>  #include <unistd.h>
>  #include <stdlib.h>
>  #include <stdint.h>
> @@ -58,29 +59,50 @@ static int measure_qlen(int fd,
>  {
>  	const struct drm_i915_gem_exec_object2 * const obj =
>  		(struct drm_i915_gem_exec_object2 *)(uintptr_t)execbuf-
> >buffers_ptr;
> -	int qlen = 64;
> +	uint32_t ctx[64];
> +	int min = INT_MAX, max = 0;
> +
> +	for (int i = 0; i < ARRAY_SIZE(ctx); i++)
> +		ctx[i] = gem_context_create(fd);
>  
>  	for (unsigned int n = 0; n < nengine; n++) {
>  		uint64_t saved = execbuf->flags;
>  		struct timespec tv = {};
> +		int q;
>  
>  		execbuf->flags |= engine[n];
>  
> -		igt_nsec_elapsed(&tv);
> -		for (int loop = 0; loop < qlen; loop++)
> +		for (int i = 0; i < ARRAY_SIZE(ctx); i++) {
> +			execbuf->rsvd1 = ctx[i];
>  			gem_execbuf(fd, execbuf);
> +		}
>  		gem_sync(fd, obj->handle);
>  
> -		execbuf->flags = saved;
> +		igt_nsec_elapsed(&tv);
> +		for (int i = 0; i < ARRAY_SIZE(ctx); i++) {
> +			execbuf->rsvd1 = ctx[i];
> +			gem_execbuf(fd, execbuf);
> +		}
> +		gem_sync(fd, obj->handle);
>  
>  		/*
>  		 * Be conservative and aim not to overshoot timeout, so
> scale
>  		 * down by 8 for hopefully a max of 12.5% error.
>  		 */
> -		qlen = qlen * timeout * 1e9 / igt_nsec_elapsed(&tv) / 8
> + 1;
> +		q = ARRAY_SIZE(ctx) * timeout * 1e9 /
> igt_nsec_elapsed(&tv) / 8 + 1;
> +		if (q < min)
> +			min = q;
> +		if (q > max)
> +			max = q;
> +
> +		execbuf->flags = saved;
>  	}
>  
> -	return qlen;
> +	for (int i = 0; i < ARRAY_SIZE(ctx); i++)
> +		gem_context_destroy(fd, ctx[i]);
> +
> +	igt_debug("Estimated qlen: {min:%d, max:%d}\n", min, max);
> +	return min;
>  }
>  
>  static void single(int fd, uint32_t handle,
> @@ -259,9 +281,10 @@ static void all(int fd, uint32_t handle,
> unsigned flags, int timeout)
>  				clock_gettime(CLOCK_MONOTONIC, &now);
>  				gem_close(fd, obj[0].handle);
>  
> -				igt_info("[%d:%d] %s: %'u cycles:
> %.3fus%s\n",
> +				igt_info("[%d:%d] %s: %'u cycles:
> %.3fus%s (elapsed: %.3fs)\n",
>  					 nctx, child, name[child],
> count, elapsed(&start, &now)*1e6 / count,
> -					 flags & INTERRUPTIBLE ? "
> (interruptible)" : "");
> +					 flags & INTERRUPTIBLE ? "
> (interruptible)" : "",
> +					 elapsed(&start, &now));
>  			}
>  			igt_waitchildren();
>  		}

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [igt-dev] [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Caz Yokoyama]

Chris,
By your patch, measure_qlen() reports how many gem_execbuf() can be
executed(queue length) within timeout of the slowest engine, correct?

Run time becomes 95 sec which is less than half.
-caz

On Sat, 2019-02-23 at 01:34 +0000, Chris Wilson wrote:
> Not all engines are created equal, and our weighting ends up
> favouring
> the many faster xCS rings at the expense of RCS. Our qlen estimation
> also failed to factor in the context switch overhead, which is a
> significant factor for nop batches. So we oversubscribe the number of
> batches submitted to RCS and end up waiting for those to complete at
> the
> end of our subtest timeslice.
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Caz Yokoyama <caz.yokoyama@intel.com>
> ---
>  tests/i915/gem_ctx_switch.c | 39 +++++++++++++++++++++++++++++----
> ----
>  1 file changed, 31 insertions(+), 8 deletions(-)
> 
> diff --git a/tests/i915/gem_ctx_switch.c
> b/tests/i915/gem_ctx_switch.c
> index 1208cb8d7..87e13b915 100644
> --- a/tests/i915/gem_ctx_switch.c
> +++ b/tests/i915/gem_ctx_switch.c
> @@ -26,6 +26,7 @@
>   */
>  
>  #include "igt.h"
> +#include <limits.h>
>  #include <unistd.h>
>  #include <stdlib.h>
>  #include <stdint.h>
> @@ -58,29 +59,50 @@ static int measure_qlen(int fd,
>  {
>  	const struct drm_i915_gem_exec_object2 * const obj =
>  		(struct drm_i915_gem_exec_object2 *)(uintptr_t)execbuf-
> >buffers_ptr;
> -	int qlen = 64;
> +	uint32_t ctx[64];
> +	int min = INT_MAX, max = 0;
> +
> +	for (int i = 0; i < ARRAY_SIZE(ctx); i++)
> +		ctx[i] = gem_context_create(fd);
>  
>  	for (unsigned int n = 0; n < nengine; n++) {
>  		uint64_t saved = execbuf->flags;
>  		struct timespec tv = {};
> +		int q;
>  
>  		execbuf->flags |= engine[n];
>  
> -		igt_nsec_elapsed(&tv);
> -		for (int loop = 0; loop < qlen; loop++)
> +		for (int i = 0; i < ARRAY_SIZE(ctx); i++) {
> +			execbuf->rsvd1 = ctx[i];
>  			gem_execbuf(fd, execbuf);
> +		}
>  		gem_sync(fd, obj->handle);
>  
> -		execbuf->flags = saved;
> +		igt_nsec_elapsed(&tv);
> +		for (int i = 0; i < ARRAY_SIZE(ctx); i++) {
> +			execbuf->rsvd1 = ctx[i];
> +			gem_execbuf(fd, execbuf);
> +		}
> +		gem_sync(fd, obj->handle);
>  
>  		/*
>  		 * Be conservative and aim not to overshoot timeout, so
> scale
>  		 * down by 8 for hopefully a max of 12.5% error.
>  		 */
> -		qlen = qlen * timeout * 1e9 / igt_nsec_elapsed(&tv) / 8
> + 1;
> +		q = ARRAY_SIZE(ctx) * timeout * 1e9 /
> igt_nsec_elapsed(&tv) / 8 + 1;
> +		if (q < min)
> +			min = q;
> +		if (q > max)
> +			max = q;
> +
> +		execbuf->flags = saved;
>  	}
>  
> -	return qlen;
> +	for (int i = 0; i < ARRAY_SIZE(ctx); i++)
> +		gem_context_destroy(fd, ctx[i]);
> +
> +	igt_debug("Estimated qlen: {min:%d, max:%d}\n", min, max);
> +	return min;
>  }
>  
>  static void single(int fd, uint32_t handle,
> @@ -259,9 +281,10 @@ static void all(int fd, uint32_t handle,
> unsigned flags, int timeout)
>  				clock_gettime(CLOCK_MONOTONIC, &now);
>  				gem_close(fd, obj[0].handle);
>  
> -				igt_info("[%d:%d] %s: %'u cycles:
> %.3fus%s\n",
> +				igt_info("[%d:%d] %s: %'u cycles:
> %.3fus%s (elapsed: %.3fs)\n",
>  					 nctx, child, name[child],
> count, elapsed(&start, &now)*1e6 / count,
> -					 flags & INTERRUPTIBLE ? "
> (interruptible)" : "");
> +					 flags & INTERRUPTIBLE ? "
> (interruptible)" : "",
> +					 elapsed(&start, &now));
>  			}
>  			igt_waitchildren();
>  		}

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Chris Wilson]

Quoting Caz Yokoyama (2019-02-25 18:28:34)
> Chris,
> By your patch, measure_qlen() reports how many gem_execbuf() can be
> executed(queue length) within timeout of the slowest engine, correct?

More or less, yes.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Chris Wilson]

Quoting Caz Yokoyama (2019-02-25 18:28:34)
> Chris,
> By your patch, measure_qlen() reports how many gem_execbuf() can be
> executed(queue length) within timeout of the slowest engine, correct?

More or less, yes.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [igt-dev] [igt PATCH v1 1/1] i915/gem_ctx_switch: evenly run 4 child processes

[Caz Yokoyama]

Drop this patch. Chris submitted the better patch.
-caz

On Sat, 2019-02-23 at 00:41 +0000, Chris Wilson wrote:
> Quoting Caz Yokoyama (2019-02-23 06:54:20)
> > gem_ctx_switch@basic-all-light test creates 4 child processes for
> > each
> > engine and wait. render engine runs slower than others. By this
> > patch,
> > 4 child processes run fairly evenly. As a result, parent process
> > runs
> > faster. On my NUC, its execution time becomes about half.
> 
> A much bigger problem is that there is very little variety in
> gem_ctx_switch. There are a few more patterns of context switches
> that
> would be interesting (and need to anticipate different scheduler
> queue
> depths) and a smoketest of those.
> -Chris

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Caz Yokoyama]

Reviewed-by:  Caz Yokoyama <caz.yokoyama@intel.com>
-caz

On Mon, 2019-02-25 at 18:29 +0000, Chris Wilson wrote:
> Quoting Caz Yokoyama (2019-02-25 18:28:34)
> > Chris,
> > By your patch, measure_qlen() reports how many gem_execbuf() can be
> > executed(queue length) within timeout of the slowest engine,
> > correct?
> 
> More or less, yes.
> -Chris

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [igt-dev] [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Caz Yokoyama]

Reviewed-by:  Caz Yokoyama <caz.yokoyama@intel.com>
-caz

On Mon, 2019-02-25 at 18:29 +0000, Chris Wilson wrote:
> Quoting Caz Yokoyama (2019-02-25 18:28:34)
> > Chris,
> > By your patch, measure_qlen() reports how many gem_execbuf() can be
> > executed(queue length) within timeout of the slowest engine,
> > correct?
> 
> More or less, yes.
> -Chris

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Chris Wilson]

Quoting Caz Yokoyama (2019-02-26 16:14:18)
> Reviewed-by:  Caz Yokoyama <caz.yokoyama@intel.com>

And pushed, thanks for raising the issue and review.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [igt-dev] [PATCH i-g-t] i915/gem_ctx_switch: Use minimum qlen over all engines and measure switches

[Chris Wilson]

Quoting Caz Yokoyama (2019-02-26 16:14:18)
> Reviewed-by:  Caz Yokoyama <caz.yokoyama@intel.com>

And pushed, thanks for raising the issue and review.
-Chris
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev