[PATCH] drm/i915: Skip idling an idle engine

[PATCH] drm/i915: Skip idling an idle engine

[Chris Wilson]

During suspend (or module unload), if we have never accessed the engine
(i.e. userspace never submitted a batch to it), the engine is idle. Then
we attempt to idle the engine by forcing it to the default context,
which actually means we submit a render batch to setup the golden
context state and then wait for it to complete. We can skip this
entirely as we know the engine is idle.

Testcase: igt/drm_module_reload_basic #byt
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95634
---
 drivers/gpu/drm/i915/i915_gem.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index b3c15698d611..8fb4fe6ae98c 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3428,6 +3428,9 @@ int i915_gpu_idle(struct drm_device *dev)
 
 	/* Flush everything onto the inactive list. */
 	for_each_engine(engine, dev_priv) {
+		if (engine->last_context == NULL)
+			continue;
+
 		if (!i915.enable_execlists) {
 			struct drm_i915_gem_request *req;
 
-- 
2.8.1

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

✗ Ro.CI.BAT: failure for drm/i915: Skip idling an idle engine

[Patchwork]

== Series Details ==

Series: drm/i915: Skip idling an idle engine
URL   : https://patchwork.freedesktop.org/series/7628/
State : failure

== Summary ==

Series 7628v1 drm/i915: Skip idling an idle engine
http://patchwork.freedesktop.org/api/1.0/series/7628/revisions/1/mbox

Test gem_busy:
        Subgroup basic-parallel-vebox:
                dmesg-warn -> PASS       (ro-skl-i7-6700hq)
        Subgroup basic-render:
                pass       -> DMESG-WARN (ro-skl-i7-6700hq)
Test gem_exec_basic:
        Subgroup readonly-blt:
                pass       -> DMESG-WARN (ro-skl-i7-6700hq)
Test gem_exec_flush:
        Subgroup basic-batch-kernel-default-cmd:
                pass       -> FAIL       (ro-byt-n2820)
        Subgroup basic-wb-prw-default:
                pass       -> DMESG-WARN (ro-skl-i7-6700hq)
Test gem_linear_blits:
        Subgroup basic:
                pass       -> DMESG-WARN (ro-skl-i7-6700hq)
Test gem_ringfill:
        Subgroup basic-default-interruptible:
                dmesg-warn -> PASS       (ro-skl-i7-6700hq)
Test gem_storedw_loop:
        Subgroup basic-default:
                dmesg-warn -> PASS       (ro-skl-i7-6700hq)
Test kms_pipe_crc_basic:
        Subgroup bad-pipe:
                dmesg-warn -> PASS       (ro-skl-i7-6700hq)
Test kms_setmode:
        Subgroup basic-clone-single-crtc:
                pass       -> DMESG-WARN (ro-skl-i7-6700hq)
Test kms_sink_crc_basic:
                pass       -> SKIP       (ro-skl-i7-6700hq)
Test pm_rpm:
        Subgroup basic-pci-d3-state:
                fail       -> DMESG-WARN (ro-skl-i7-6700hq)
Test pm_rps:
        Subgroup basic-api:
                dmesg-warn -> PASS       (ro-skl-i7-6700hq)

ro-bdw-i5-5250u  total:209  pass:172  dwarn:0   dfail:0   fail:0   skip:37 
ro-bdw-i7-5557U  total:209  pass:197  dwarn:0   dfail:0   fail:0   skip:12 
ro-bdw-i7-5600u  total:209  pass:180  dwarn:0   dfail:0   fail:1   skip:28 
ro-bsw-n3050     total:209  pass:168  dwarn:0   dfail:0   fail:2   skip:39 
ro-byt-n2820     total:209  pass:169  dwarn:0   dfail:0   fail:3   skip:37 
ro-hsw-i3-4010u  total:209  pass:186  dwarn:0   dfail:0   fail:0   skip:23 
ro-hsw-i7-4770r  total:209  pass:186  dwarn:0   dfail:0   fail:0   skip:23 
ro-ilk-i7-620lm  total:209  pass:146  dwarn:0   dfail:0   fail:1   skip:62 
ro-ilk1-i5-650   total:204  pass:146  dwarn:0   dfail:0   fail:1   skip:57 
ro-ivb-i7-3770   total:209  pass:177  dwarn:0   dfail:0   fail:0   skip:32 
ro-ivb2-i7-3770  total:209  pass:181  dwarn:0   dfail:0   fail:0   skip:28 
ro-skl-i7-6700hq total:204  pass:174  dwarn:8   dfail:0   fail:0   skip:22 
ro-snb-i7-2620M  total:209  pass:170  dwarn:0   dfail:0   fail:1   skip:38 

Results at /archive/results/CI_IGT_test/RO_Patchwork_990/

8621fb5 drm-intel-nightly: 2016y-05m-23d-18h-18m-33s UTC integration manifest
af35cac drm/i915: Skip idling an idle engine

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

Re: [PATCH] drm/i915: Skip idling an idle engine

[Joonas Lahtinen]

On ti, 2016-05-24 at 11:56 +0100, Chris Wilson wrote:
> During suspend (or module unload), if we have never accessed the engine
> (i.e. userspace never submitted a batch to it), the engine is idle. Then
> we attempt to idle the engine by forcing it to the default context,
> which actually means we submit a render batch to setup the golden
> context state and then wait for it to complete. We can skip this
> entirely as we know the engine is idle.
> 

Seems reasonable.

By quick look we do not have an inactive engine list, though? So the
comment above could be nuked.

Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

> Testcase: igt/drm_module_reload_basic #byt
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95634
> ---
>  drivers/gpu/drm/i915/i915_gem.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> index b3c15698d611..8fb4fe6ae98c 100644
> --- a/drivers/gpu/drm/i915/i915_gem.c
> +++ b/drivers/gpu/drm/i915/i915_gem.c
> @@ -3428,6 +3428,9 @@ int i915_gpu_idle(struct drm_device *dev)
>  
>  	/* Flush everything onto the inactive list. */
>  	for_each_engine(engine, dev_priv) {
> +		if (engine->last_context == NULL)
> +			continue;
> +
>  		if (!i915.enable_execlists) {
>  			struct drm_i915_gem_request *req;
>  
-- 
Joonas Lahtinen
Open Source Technology Center
Intel Corporation
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915: Skip idling an idle engine

[Chris Wilson]

On Wed, May 25, 2016 at 11:50:22AM +0300, Joonas Lahtinen wrote:
> On ti, 2016-05-24 at 11:56 +0100, Chris Wilson wrote:
> > During suspend (or module unload), if we have never accessed the engine
> > (i.e. userspace never submitted a batch to it), the engine is idle. Then
> > we attempt to idle the engine by forcing it to the default context,
> > which actually means we submit a render batch to setup the golden
> > context state and then wait for it to complete. We can skip this
> > entirely as we know the engine is idle.
> > 
> 
> Seems reasonable.
> 
> By quick look we do not have an inactive engine list, though? So the
> comment above could be nuked.

It dates from when i915_gpu_idle() also did retire-requests as well, and
the inactive list on the engine rather than the vm.

I've a few more in this re bug 95634, just need to finish rebasing.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx