From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 05/12] drm/i915: Trim NEWCLIENT boosting
Date: Mon, 4 Feb 2019 12:42:06 +0000 [thread overview]
Message-ID: <4bb3a10e-046a-7e23-5170-47002bd1111b@linux.intel.com> (raw)
In-Reply-To: <154928319347.14784.4960498142443840794@skylake-alporthouse-com>
On 04/02/2019 12:26, Chris Wilson wrote:
> Quoting Tvrtko Ursulin (2019-02-04 12:11:41)
>>
>> On 04/02/2019 08:41, Chris Wilson wrote:
>>> Limit the NEWCLIENT boost to only give its small priority boost to fresh
>>> clients only that have no dependencies.
>>
>> Needs some actual explaining/documenting on why we would want this.
>>
>>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>>> ---
>>> drivers/gpu/drm/i915/i915_request.c | 2 +-
>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
>>> index 7968875d0bed..69bc549e95d8 100644
>>> --- a/drivers/gpu/drm/i915/i915_request.c
>>> +++ b/drivers/gpu/drm/i915/i915_request.c
>>> @@ -979,7 +979,7 @@ void i915_request_add(struct i915_request *request)
>>> * Allow interactive/synchronous clients to jump ahead of
>>> * the bulk clients. (FQ_CODEL)
>>> */
>>> - if (!prev || i915_request_completed(prev))
>>> + if (list_empty(&request->sched.signalers_list))
>>
>> Why are the !prev and completed checks gone? Seems in disagreement with
>> the commit message text - they wouldn't have to be new clients any more,
>> just have no dependencies.
>
> What? signalers_list is a superset of !prev || i915_request_completed.
Ok my bad - I forgot about adding the previous unfinished request as a
dependency.
Regards,
Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2019-02-04 12:42 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-04 8:41 [PATCH 01/12] drm/i915: Allow normal clients to always preempt idle priority clients Chris Wilson
2019-02-04 8:41 ` [PATCH 02/12] drm/i915/execlists: Suppress mere WAIT preemption Chris Wilson
2019-02-04 10:06 ` Tvrtko Ursulin
2019-02-04 10:18 ` Chris Wilson
2019-02-04 12:08 ` Tvrtko Ursulin
2019-02-04 12:19 ` Chris Wilson
2019-02-04 12:29 ` Tvrtko Ursulin
2019-02-04 10:49 ` [PATCH] " Chris Wilson
2019-02-04 8:41 ` [PATCH 03/12] drm/i915/execlists: Suppress redundant preemption Chris Wilson
2019-02-04 12:05 ` Tvrtko Ursulin
2019-02-04 12:25 ` Chris Wilson
2019-02-04 8:41 ` [PATCH 04/12] drm/i915/selftests: Exercise some AB...BA preemption chains Chris Wilson
2019-02-04 8:41 ` [PATCH 05/12] drm/i915: Trim NEWCLIENT boosting Chris Wilson
2019-02-04 12:11 ` Tvrtko Ursulin
2019-02-04 12:26 ` Chris Wilson
2019-02-04 12:42 ` Tvrtko Ursulin [this message]
2019-02-04 12:27 ` Chris Wilson
2019-02-04 8:41 ` [PATCH 06/12] drm/i915: Show support for accurate sw PMU busyness tracking Chris Wilson
2019-02-04 12:14 ` Tvrtko Ursulin
2019-02-04 12:28 ` Chris Wilson
2019-02-04 12:29 ` Chris Wilson
2019-02-04 12:37 ` Tvrtko Ursulin
2019-02-04 12:43 ` Chris Wilson
2019-02-04 8:41 ` [PATCH 07/12] drm/i915: Revoke mmaps and prevent access to fence registers across reset Chris Wilson
2019-02-04 13:33 ` Mika Kuoppala
2019-02-04 13:47 ` Chris Wilson
2019-02-04 8:41 ` [PATCH 08/12] drm/i915: Force the GPU reset upon wedging Chris Wilson
2019-02-04 8:41 ` [PATCH 09/12] drm/i915: Uninterruptibly drain the timelines on unwedging Chris Wilson
2019-02-04 8:41 ` [PATCH 10/12] drm/i915: Wait for old resets before applying debugfs/i915_wedged Chris Wilson
2019-02-04 8:41 ` [PATCH 11/12] drm/i915: Serialise resets with wedging Chris Wilson
2019-02-04 8:41 ` [PATCH 12/12] drm/i915: Don't claim an unstarted request was guilty Chris Wilson
2019-02-04 9:20 ` [PATCH 01/12] drm/i915: Allow normal clients to always preempt idle priority clients Tvrtko Ursulin
2019-02-04 10:19 ` ✗ Fi.CI.CHECKPATCH: warning for series starting with [01/12] " Patchwork
2019-02-04 10:23 ` ✗ Fi.CI.SPARSE: " Patchwork
2019-02-04 10:48 ` ✓ Fi.CI.BAT: success " Patchwork
2019-02-04 11:27 ` ✗ Fi.CI.BAT: failure for series starting with [01/12] drm/i915: Allow normal clients to always preempt idle priority clients (rev2) Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4bb3a10e-046a-7e23-5170-47002bd1111b@linux.intel.com \
--to=tvrtko.ursulin@linux.intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=intel-gfx@lists.freedesktop.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox