All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sergey Senozhatsky <senozhatsky@chromium.org>
To: Sergey Senozhatsky <senozhatsky@chromium.org>,
	Matthew Auld <matthew.william.auld@gmail.com>,
	Jani Nikula <jani.nikula@linux.intel.com>,
	Joonas Lahtinen <joonas.lahtinen@linux.intel.com>,
	David Airlie <airlied@linux.ie>,
	Chris Wilson <chris@chris-wilson.co.uk>,
	intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
	linux-kernel@vger.kernel.org
Subject: Re: drm/i915: __GFP_RETRY_MAYFAIL allocations in stable kernels
Date: Fri, 18 Jun 2021 11:29:30 +0900	[thread overview]
Message-ID: <YMwFCoHzjnuH80p6@google.com> (raw)
In-Reply-To: <YMuGGqs4cDotxuKO@phenom.ffwll.local>

On (21/06/17 19:27), Daniel Vetter wrote:
> > 
> > So can all allocations in gen8_init_scratch() use
> > 	GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOWARN
> 
> Yeah that looks all fairly broken tbh. The only thing I didn't know was
> that GFP_DMA32 wasn't a full gfp mask with reclaim bits set as needed. I
> guess it would be clearer if we use GFP_KERNEL | __GFP_DMA32 for these.

Looks good.

> The commit that introduced a lot of this, including I915_GFP_ALLOW_FAIL
> seems to be
> 
> commit 1abb70f5955d1a9021f96359a2c6502ca569b68d
> Author: Chris Wilson <chris@chris-wilson.co.uk>
> Date:   Tue May 22 09:36:43 2018 +0100
> 
>     drm/i915/gtt: Allow pagedirectory allocations to fail
> 
> which used a selftest as justification, not real world workloads, so looks
> rather dubious.

Exactly, the commit we landed internally partially reverts 1abb70f5955
in 4.19 and 5.4 kernels. I don't mind I915_GFP_ALLOW_FAIL and so on, I
kept those bits, but we need reclaim. I can reproduce cases when order:0
allocation fails with
	__GFP_HIGHMEM|__GFP_RETRY_MAYFAIL
but succeeds with
	GFP_KERNEL|__GFP_HIGHMEM|__GFP_RETRY_MAYFAIL


ON a side note, I'm not very sure if __GFP_RETRY_MAYFAIL is actually
needed. Especially seeing it in syscalls is a bit uncommon:

  drm_ioctl()
   i915_gem_context_create_ioctl()
    i915_gem_create_context()
     i915_ppgtt_create()
      setup_scratch_page()           // __GFP_RETRY_MAYFAIL

But with GFP_KERNEL at least it tries to make some reclaim progress
between retries, so it seems to be good enough.

WARNING: multiple messages have this Message-ID (diff)
From: Sergey Senozhatsky <senozhatsky@chromium.org>
To: Sergey Senozhatsky <senozhatsky@chromium.org>,
	Matthew Auld <matthew.william.auld@gmail.com>,
	Jani Nikula <jani.nikula@linux.intel.com>,
	Joonas Lahtinen <joonas.lahtinen@linux.intel.com>,
	David Airlie <airlied@linux.ie>,
	Chris Wilson <chris@chris-wilson.co.uk>,
	intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
	linux-kernel@vger.kernel.org
Subject: Re: [Intel-gfx] drm/i915: __GFP_RETRY_MAYFAIL allocations in stable kernels
Date: Fri, 18 Jun 2021 11:29:30 +0900	[thread overview]
Message-ID: <YMwFCoHzjnuH80p6@google.com> (raw)
In-Reply-To: <YMuGGqs4cDotxuKO@phenom.ffwll.local>

On (21/06/17 19:27), Daniel Vetter wrote:
> > 
> > So can all allocations in gen8_init_scratch() use
> > 	GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOWARN
> 
> Yeah that looks all fairly broken tbh. The only thing I didn't know was
> that GFP_DMA32 wasn't a full gfp mask with reclaim bits set as needed. I
> guess it would be clearer if we use GFP_KERNEL | __GFP_DMA32 for these.

Looks good.

> The commit that introduced a lot of this, including I915_GFP_ALLOW_FAIL
> seems to be
> 
> commit 1abb70f5955d1a9021f96359a2c6502ca569b68d
> Author: Chris Wilson <chris@chris-wilson.co.uk>
> Date:   Tue May 22 09:36:43 2018 +0100
> 
>     drm/i915/gtt: Allow pagedirectory allocations to fail
> 
> which used a selftest as justification, not real world workloads, so looks
> rather dubious.

Exactly, the commit we landed internally partially reverts 1abb70f5955
in 4.19 and 5.4 kernels. I don't mind I915_GFP_ALLOW_FAIL and so on, I
kept those bits, but we need reclaim. I can reproduce cases when order:0
allocation fails with
	__GFP_HIGHMEM|__GFP_RETRY_MAYFAIL
but succeeds with
	GFP_KERNEL|__GFP_HIGHMEM|__GFP_RETRY_MAYFAIL


ON a side note, I'm not very sure if __GFP_RETRY_MAYFAIL is actually
needed. Especially seeing it in syscalls is a bit uncommon:

  drm_ioctl()
   i915_gem_context_create_ioctl()
    i915_gem_create_context()
     i915_ppgtt_create()
      setup_scratch_page()           // __GFP_RETRY_MAYFAIL

But with GFP_KERNEL at least it tries to make some reclaim progress
between retries, so it seems to be good enough.
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2021-06-18  2:29 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-14 12:45 drm/i915: __GFP_RETRY_MAYFAIL allocations in stable kernels Sergey Senozhatsky
2021-06-14 12:45 ` [Intel-gfx] " Sergey Senozhatsky
2021-06-14 12:45 ` Sergey Senozhatsky
2021-06-14 23:38 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for " Patchwork
2021-06-17 17:27 ` Daniel Vetter
2021-06-17 17:27   ` [Intel-gfx] " Daniel Vetter
2021-06-17 17:27   ` Daniel Vetter
2021-06-18  2:29   ` Sergey Senozhatsky [this message]
2021-06-18  2:29     ` [Intel-gfx] " Sergey Senozhatsky
2021-06-18 15:46   ` Matthew Auld
2021-06-18 15:46     ` [Intel-gfx] " Matthew Auld
2021-06-18 15:46     ` Matthew Auld
2021-06-21 14:10     ` Daniel Vetter
2021-06-21 14:10       ` [Intel-gfx] " Daniel Vetter
2021-06-21 14:10       ` Daniel Vetter

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=YMwFCoHzjnuH80p6@google.com \
    --to=senozhatsky@chromium.org \
    --cc=airlied@linux.ie \
    --cc=chris@chris-wilson.co.uk \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=jani.nikula@linux.intel.com \
    --cc=joonas.lahtinen@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=matthew.william.auld@gmail.com \
    /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 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.