mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>
To: christian.koenig@amd.com, dakr@kernel.org, ecourtney@nvidia.com,
	 simona@ffwll.ch, matthew.brost@intel.com, nat@pixelcluster.dev,
	airlied@gmail.com, 	dri-devel@lists.freedesktop.org,
	linux-kernel@vger.kernel.org, 	intel-gfx@lists.freedesktop.org,
	intel-xe@lists.freedesktop.org, 	amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH 11/12] drm/ttm: support using drm_exec during eviction v4
Date: Mon, 13 Jul 2026 13:59:17 +0200	[thread overview]
Message-ID: <d09a8c6871494c34af9cdf8f99d3739e78427223.camel@linux.intel.com> (raw)
In-Reply-To: <20260710190752.2355-12-christian.koenig@amd.com>

On Fri, 2026-07-10 at 20:52 +0200, Christian König wrote:
> Allow specifying a drm_exec object in TTMs operation context which is
> used to lock objects during eviction.
> 
> This allows to handle deadlocks much more gracefully and with that
> avoid returning -ENOMEM on heavily contended domains.
> 
> v2: rebased on top of Thomas work
> v3: rebased again
> v4: adjust to dma_resv changes
> 
> Signed-off-by: Christian König <christian.koenig@amd.com>

Same thing here. Use a ww context at the dma-buf level.

Thanks,
Thomas


> ---
>  drivers/gpu/drm/ttm/ttm_bo_util.c | 12 ++++++++----
>  include/drm/ttm/ttm_bo.h          |  5 +++++
>  2 files changed, 13 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c
> b/drivers/gpu/drm/ttm/ttm_bo_util.c
> index 1bdd69643c313..570640ae79d84 100644
> --- a/drivers/gpu/drm/ttm/ttm_bo_util.c
> +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
> @@ -38,6 +38,7 @@
>  #include <drm/ttm/ttm_tt.h>
>  
>  #include <drm/drm_cache.h>
> +#include <drm/drm_exec.h>
>  
>  struct ttm_transfer_obj {
>  	struct ttm_buffer_object base;
> @@ -839,7 +840,9 @@ static int ttm_lru_walk_ticketlock(struct
> ttm_bo_lru_cursor *curs,
>  	struct ttm_lru_walk_arg *arg = curs->arg;
>  	int ret;
>  
> -	if (arg->ctx->interruptible)
> +	if (arg->ctx->exec)
> +		ret = drm_exec_lock_resv(arg->ctx->exec, resv);
> +	else if (arg->ctx->interruptible)
>  		ret = dma_resv_lock_interruptible(resv, arg-
> >ticket);
>  	else
>  		ret = dma_resv_lock(resv, arg->ticket);
> @@ -853,7 +856,8 @@ static int ttm_lru_walk_ticketlock(struct
> ttm_bo_lru_cursor *curs,
>  		 * trylocking for this walk.
>  		 */
>  		arg->ticket = NULL;
> -	} else if (ret == -EDEADLK) {
> +
> +	} else if (!arg->ctx->exec && ret == -EDEADLK) {
>  		/* Caller needs to exit the ww transaction. */
>  		ret = -ENOSPC;
>  	}
> @@ -993,8 +997,8 @@ __ttm_bo_lru_cursor_iter(struct ttm_bo_lru_cursor
> *curs, bool first)
>  		if (!ttm_lru_walk_trylock(curs, res->bo->base.resv))
> {
>  			struct dma_resv *resv;
>  
> -			if (!arg->ticket || arg->ctx->no_wait_gpu ||
> -			    arg->trylock_only) {
> +			if ((!arg->ticket || arg->ctx->no_wait_gpu
> ||
> +			     arg->trylock_only) && !arg->ctx->exec)
> {
>  				spin_unlock(lru_lock);
>  				continue;
>  			}
> diff --git a/include/drm/ttm/ttm_bo.h b/include/drm/ttm/ttm_bo.h
> index e1221e3be7bda..0ffa84a5caa65 100644
> --- a/include/drm/ttm/ttm_bo.h
> +++ b/include/drm/ttm/ttm_bo.h
> @@ -186,6 +186,11 @@ struct ttm_operation_ctx {
>  	 * @bytes_moved: Statistics on how many bytes have been
> moved.
>  	 */
>  	uint64_t bytes_moved;
> +	/*
> +	 * @exec: optional drm_exec object to use for locking and
> contention
> +	 * handling
> +	 */
> +	struct drm_exec *exec;
>  };
>  
>  struct ttm_lru_walk;

  reply	other threads:[~2026-07-13 11:59 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-10 18:52 Refcounting dma_resv and using that for drm_exec support in TTM Christian König
2026-07-10 18:52 ` [PATCH 01/12] dma-buf: Add reference counting to dma_resv Christian König
2026-07-11  1:27   ` Matthew Brost
2026-07-11 13:10     ` Danilo Krummrich
2026-07-12 19:51       ` Matthew Brost
2026-07-13  9:25         ` Christian König
2026-07-13 14:57           ` Matthew Brost
2026-07-10 18:52 ` [PATCH 02/12] dma-buf/tests: Convert st-dma-resv tests to use dma_resv_alloc Christian König
2026-07-10 18:52 ` [PATCH 03/12] drm/gem: Add helper for drm_gem_object resv assignment Christian König
2026-07-10 18:52 ` [PATCH 04/12] drm/ttm: Switch LRU cursor to track dma_resv instead of buffer objects Christian König
2026-07-13 13:40   ` Thomas Hellström
2026-07-13 15:15     ` Matthew Brost
2026-07-10 18:52 ` [PATCH 05/12] drm/ttm: switch to ttm_bo_lru_for_each_reserved_guarded for swapout Christian König
2026-07-10 18:52 ` [PATCH 06/12] drm/ttm: move delete handling into ttm_bo_evict Christian König
2026-07-10 18:52 ` [PATCH 07/12] drm/ttm: use ttm_bo_lru_for_each_reserved_guarded in evict_all Christian König
2026-07-10 18:52 ` [PATCH 08/12] drm/ttm: use dma_resv reference in ttm_device_clear_lru_dma_mappings Christian König
2026-07-10 18:52 ` [PATCH 09/12] drm/ttm: nuke buffer refcounting Christian König
2026-07-11 13:26   ` Danilo Krummrich
2026-07-13 15:06     ` Matthew Brost
2026-07-10 18:52 ` [PATCH 10/12] drm/exec: add drm_exec_lock_resv function Christian König
2026-07-13 11:57   ` Thomas Hellström
2026-07-13 12:07     ` Christian König
2026-07-13 12:19       ` Thomas Hellström
2026-07-13 13:14         ` Christian König
2026-07-13 13:37           ` Thomas Hellström
2026-07-10 18:52 ` [PATCH 11/12] drm/ttm: support using drm_exec during eviction v4 Christian König
2026-07-13 11:59   ` Thomas Hellström [this message]
2026-07-10 18:52 ` [PATCH 12/12] drm/amdgpu: use drm_exec during BO validation Christian König
2026-07-13 11:32 ` Refcounting dma_resv and using that for drm_exec support in TTM Thomas Hellström
2026-07-13 15:03   ` Matthew Brost

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=d09a8c6871494c34af9cdf8f99d3739e78427223.camel@linux.intel.com \
    --to=thomas.hellstrom@linux.intel.com \
    --cc=airlied@gmail.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=christian.koenig@amd.com \
    --cc=dakr@kernel.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=ecourtney@nvidia.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=matthew.brost@intel.com \
    --cc=nat@pixelcluster.dev \
    --cc=simona@ffwll.ch \
    /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