Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Matthew Brost <matthew.brost@intel.com>
To: Sanjay Yadav <sanjay.kumar.yadav@intel.com>
Cc: <intel-xe@lists.freedesktop.org>, <matthew.auld@intel.com>
Subject: Re: [PATCH] drm/xe: Use DRM_BUDDY_CONTIGUOUS_ALLOCATION for contiguous allocations
Date: Fri, 5 Dec 2025 08:19:29 -0800	[thread overview]
Message-ID: <aTMGEVO1mRJXSXrB@lstrano-desk.jf.intel.com> (raw)
In-Reply-To: <20251205031112.3590252-2-sanjay.kumar.yadav@intel.com>

On Fri, Dec 05, 2025 at 08:41:13AM +0530, Sanjay Yadav wrote:
> The VRAM/stolen memory managers do not currently set
> DRM_BUDDY_CONTIGUOUS_ALLOCATION for contiguous allocations. Enabling
> this flag activates the buddy allocator's try_harder path, which helps
> handle fragmented memory scenarios.
> 
> This enables the __alloc_contig_try_harder fallback in the buddy
> allocator, allowing contiguous allocation requests to succeed even when
> memory is fragmented by combining allocations from both(RHS and LHS)
> sides of a large free block.
> 
> Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6713
> Suggested-by: Matthew Auld <matthew.auld@intel.com>
> Signed-off-by: Sanjay Yadav <sanjay.kumar.yadav@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_ttm_vram_mgr.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> index 9f70802fce92..ef6cfb8af0ac 100644
> --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> @@ -82,6 +82,9 @@ static int xe_ttm_vram_mgr_new(struct ttm_resource_manager *man,
>  	if (place->flags & TTM_PL_FLAG_TOPDOWN)
>  		vres->flags |= DRM_BUDDY_TOPDOWN_ALLOCATION;
>  
> +	if (place->flags & TTM_PL_FLAG_CONTIGUOUS)
> +		vres->flags |= DRM_BUDDY_CONTIGUOUS_ALLOCATION;

If we set this flag can we remove the logic which rounds the allocation
size to a pow2 and trims if TTM_PL_FLAG_CONTIGUOUS is set? From a quick
look it looks we can as drm_buddy has roundup / trim logic if
DRM_BUDDY_CONTIGUOUS_ALLOCATION is set.

Matt

> +
>  	if (place->fpfn || lpfn != man->size >> PAGE_SHIFT)
>  		vres->flags |= DRM_BUDDY_RANGE_ALLOCATION;
>  
> -- 
> 2.52.0
> 

  parent reply	other threads:[~2025-12-05 16:19 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-05  3:11 [PATCH] drm/xe: Use DRM_BUDDY_CONTIGUOUS_ALLOCATION for contiguous allocations Sanjay Yadav
2025-12-05  3:20 ` ✓ CI.KUnit: success for " Patchwork
2025-12-05  4:21 ` ✓ Xe.CI.BAT: " Patchwork
2025-12-05  5:08 ` ✓ Xe.CI.Full: " Patchwork
2025-12-05 16:19 ` Matthew Brost [this message]
2025-12-09  4:37   ` [PATCH] " Yadav, Sanjay Kumar

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=aTMGEVO1mRJXSXrB@lstrano-desk.jf.intel.com \
    --to=matthew.brost@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.auld@intel.com \
    --cc=sanjay.kumar.yadav@intel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox