Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/i915/gem: Reduce flush_ggtt() from a wait-for-idle to a mere barrier flush
Date: Wed, 20 Nov 2019 16:14:40 +0000	[thread overview]
Message-ID: <d132f808-7a5a-016a-9111-59da21d984b3@linux.intel.com> (raw)
In-Reply-To: <157426575648.13839.14923384484472849836@skylake-alporthouse-com>


On 20/11/2019 16:02, Chris Wilson wrote:
> Quoting Tvrtko Ursulin (2019-11-20 15:58:49)
>>
>> On 20/11/2019 13:41, Chris Wilson wrote:
>>> Since we use barriers, we need only explicitly flush those barriers to
>>> ensure tha we can reclaim the available ggtt for ourselves. The barrier
>>> flush was implicit inside the intel_gt_wait_for_idle() -- except because
>>> we use i915_gem_evict from inside an active timeline during execbuf, we
>>> could easily end up waiting upon ourselves.
>>>
>>> Fixes: 7936a22dd466 ("drm/i915/gt: Wait for new requests in intel_gt_retire_requests()")
>>> Fixes: a46bfdc83fee ("drm/i915/gt: Wait for new requests in intel_gt_retire_requests()")
>>> Testcase: igt/gem_exec_reloc/basic-range
>>
>> Bugzilla: ?
> 
> It's been in CI since before the w/e (the test itself is much, much
> older), I guess it hasn't been vetted yet as no bug has been filed.
>   
>> This test gets permanently stuck on some platforms?
> 
> All !full-ppgtt platforms.

How it will cope with actual ggtt pressure? Wait for presumably there 
for a reason and now it will only retire what's already done and send an 
idle pulse down the engines.

Regards,

Tvrtko


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

WARNING: multiple messages have this Message-ID (diff)
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [Intel-gfx] [PATCH] drm/i915/gem: Reduce flush_ggtt() from a wait-for-idle to a mere barrier flush
Date: Wed, 20 Nov 2019 16:14:40 +0000	[thread overview]
Message-ID: <d132f808-7a5a-016a-9111-59da21d984b3@linux.intel.com> (raw)
Message-ID: <20191120161440.oLj6ebUTFswWEqsSR_N3H5v_Yl5YaBWwhYTwdx-kY28@z> (raw)
In-Reply-To: <157426575648.13839.14923384484472849836@skylake-alporthouse-com>


On 20/11/2019 16:02, Chris Wilson wrote:
> Quoting Tvrtko Ursulin (2019-11-20 15:58:49)
>>
>> On 20/11/2019 13:41, Chris Wilson wrote:
>>> Since we use barriers, we need only explicitly flush those barriers to
>>> ensure tha we can reclaim the available ggtt for ourselves. The barrier
>>> flush was implicit inside the intel_gt_wait_for_idle() -- except because
>>> we use i915_gem_evict from inside an active timeline during execbuf, we
>>> could easily end up waiting upon ourselves.
>>>
>>> Fixes: 7936a22dd466 ("drm/i915/gt: Wait for new requests in intel_gt_retire_requests()")
>>> Fixes: a46bfdc83fee ("drm/i915/gt: Wait for new requests in intel_gt_retire_requests()")
>>> Testcase: igt/gem_exec_reloc/basic-range
>>
>> Bugzilla: ?
> 
> It's been in CI since before the w/e (the test itself is much, much
> older), I guess it hasn't been vetted yet as no bug has been filed.
>   
>> This test gets permanently stuck on some platforms?
> 
> All !full-ppgtt platforms.

How it will cope with actual ggtt pressure? Wait for presumably there 
for a reason and now it will only retire what's already done and send an 
idle pulse down the engines.

Regards,

Tvrtko


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

  parent reply	other threads:[~2019-11-20 16:14 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-20 13:41 [PATCH] drm/i915/gem: Reduce flush_ggtt() from a wait-for-idle to a mere barrier flush Chris Wilson
2019-11-20 13:41 ` [Intel-gfx] " Chris Wilson
2019-11-20 15:58 ` Tvrtko Ursulin
2019-11-20 15:58   ` [Intel-gfx] " Tvrtko Ursulin
2019-11-20 16:02   ` Chris Wilson
2019-11-20 16:02     ` [Intel-gfx] " Chris Wilson
2019-11-20 16:14     ` Tvrtko Ursulin [this message]
2019-11-20 16:14       ` Tvrtko Ursulin
2019-11-20 16:28       ` Chris Wilson
2019-11-20 16:28         ` [Intel-gfx] " Chris Wilson
2019-11-20 16:42 ` [PATCH] drm/i915/gem: Reduce ggtt_flush() " Chris Wilson
2019-11-20 16:42   ` [Intel-gfx] " Chris Wilson
2019-11-20 16:49   ` Chris Wilson
2019-11-20 16:49     ` [Intel-gfx] " Chris Wilson
2019-11-20 19:05 ` [PATCH] drm/i915/gem: Reduce flush_ggtt() " Chris Wilson
2019-11-20 19:05   ` [Intel-gfx] " Chris Wilson
2019-11-20 19:20 ` [PATCH] Revert "drm/i915/gt: Wait for new requests in intel_gt_retire_requests()" Chris Wilson
2019-11-20 19:20   ` [Intel-gfx] " Chris Wilson
2019-11-21  0:37 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915/gem: Reduce flush_ggtt() from a wait-for-idle to a mere barrier flush (rev3) Patchwork
2019-11-21  0:37   ` [Intel-gfx] " Patchwork
2019-11-21  0:59 ` ✓ Fi.CI.BAT: success " Patchwork
2019-11-21  0:59   ` [Intel-gfx] " Patchwork
2019-11-22  0:16 ` ✗ Fi.CI.IGT: failure " Patchwork
2019-11-22  0:16   ` [Intel-gfx] " 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=d132f808-7a5a-016a-9111-59da21d984b3@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