public inbox for drm-ai-reviews@public-inbox.freedesktop.org
 help / color / mirror / Atom feed
From: Natalie Vock <natalie.vock@gmx.de>
To: Maarten Lankhorst <dev@lankhorst.se>,
	Maxime Ripard <mripard@kernel.org>, Tejun Heo <tj@kernel.org>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Michal Koutný <mkoutny@suse.com>,
	Christian Koenig <christian.koenig@amd.com>,
	Huang Rui <ray.huang@amd.com>,
	Matthew Auld <matthew.auld@intel.com>,
	Matthew Brost <matthew.brost@intel.com>,
	Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
	Thomas Zimmermann <tzimmermann@suse.de>,
	David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>,
	Tvrtko Ursulin <tursulin@ursulin.net>
Cc: cgroups@vger.kernel.org, dri-devel@lists.freedesktop.org,
	Natalie Vock <natalie.vock@gmx.de>
Subject: [PATCH v6 5/6] drm/ttm: Be more aggressive when allocating below protection limit
Date: Fri, 13 Mar 2026 12:40:04 +0100	[thread overview]
Message-ID: <20260313-dmemcg-aggressive-protect-v6-5-7c71cc1492db@gmx.de> (raw)
In-Reply-To: <20260313-dmemcg-aggressive-protect-v6-0-7c71cc1492db@gmx.de>

When the cgroup's memory usage is below the low/min limit and allocation
fails, try evicting some unprotected buffers to make space. Otherwise,
application buffers may be forced to go into GTT even though usage is
below the corresponding low/min limit, if other applications filled VRAM
with their allocations first.

Signed-off-by: Natalie Vock <natalie.vock@gmx.de>
---
 drivers/gpu/drm/ttm/ttm_bo.c | 51 +++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 48 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 4adc9b80cba4a..7300b91b77dd3 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -496,6 +496,10 @@ struct ttm_bo_alloc_state {
 	struct dmem_cgroup_pool_state *limit_pool;
 	/** @in_evict: Whether we are currently evicting buffers */
 	bool in_evict;
+	/** @may_try_low: If only unprotected BOs, i.e. BOs whose cgroup
+	 *  is exceeding its dmem low/min protection, should be considered for eviction
+	 */
+	bool may_try_low;
 };
 
 /**
@@ -545,6 +549,42 @@ static int ttm_bo_alloc_at_place(struct ttm_buffer_object *bo,
 		}
 	}
 
+	/*
+	 * cgroup protection plays a special role in eviction.
+	 * Conceptually, protection of memory via the dmem cgroup controller
+	 * entitles the protected cgroup to use a certain amount of memory.
+	 * There are two types of protection - the 'low' limit is a
+	 * "best-effort" protection, whereas the 'min' limit provides a hard
+	 * guarantee that memory within the cgroup's allowance will not be
+	 * evicted under any circumstance.
+	 *
+	 * To faithfully model this concept in TTM, we also need to take cgroup
+	 * protection into account when allocating. When allocation in one
+	 * place fails, TTM will default to trying other places first before
+	 * evicting.
+	 * If the allocation is covered by dmem cgroup protection, however,
+	 * this prevents the allocation from using the memory it is "entitled"
+	 * to. To make sure unprotected allocations cannot push new protected
+	 * allocations out of places they are "entitled" to use, we should
+	 * evict buffers not covered by any cgroup protection, if this
+	 * allocation is covered by cgroup protection.
+	 *
+	 * Buffers covered by 'min' protection are a special case - the 'min'
+	 * limit is a stronger guarantee than 'low', and thus buffers protected
+	 * by 'low' but not 'min' should also be considered for eviction.
+	 * Buffers protected by 'min' will never be considered for eviction
+	 * anyway, so the regular eviction path should be triggered here.
+	 * Buffers protected by 'low' but not 'min' will take a special
+	 * eviction path that only evicts buffers covered by neither 'low' or
+	 * 'min' protections.
+	 */
+	if (!alloc_state->in_evict) {
+		may_evict |= dmem_cgroup_below_min(NULL, alloc_state->charge_pool);
+		alloc_state->may_try_low = may_evict;
+
+		may_evict |= dmem_cgroup_below_low(NULL, alloc_state->charge_pool);
+	}
+
 	ret = ttm_resource_alloc(bo, place, res, alloc_state->charge_pool);
 	if (ret) {
 		if (ret == -ENOSPC && may_evict)
@@ -657,8 +697,12 @@ static int ttm_bo_evict_alloc(struct ttm_device *bdev,
 	evict_walk.walk.arg.trylock_only = true;
 	lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1);
 
-	/* One more attempt if we hit low limit? */
-	if (!lret && evict_walk.hit_low) {
+	/* If we failed to find enough BOs to evict, but we skipped over
+	 * some BOs because they were covered by dmem low protection, retry
+	 * evicting these protected BOs too, except if we're told not to
+	 * consider protected BOs at all.
+	 */
+	if (!lret && evict_walk.hit_low && state->may_try_low) {
 		evict_walk.try_low = true;
 		lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1);
 	}
@@ -679,7 +723,8 @@ static int ttm_bo_evict_alloc(struct ttm_device *bdev,
 	} while (!lret && evict_walk.evicted);
 
 	/* We hit the low limit? Try once more */
-	if (!lret && evict_walk.hit_low && !evict_walk.try_low) {
+	if (!lret && evict_walk.hit_low && !evict_walk.try_low &&
+			state->may_try_low) {
 		evict_walk.try_low = true;
 		goto retry;
 	}

-- 
2.53.0


  parent reply	other threads:[~2026-03-13 11:41 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-13 11:39 [PATCH v6 0/6] cgroup/dmem,drm/ttm: Improve protection in contended cases Natalie Vock
2026-03-13 11:40 ` [PATCH v6 1/6] cgroup/dmem: Add queries for protection values Natalie Vock
2026-03-13 21:10   ` Claude review: " Claude Code Review Bot
2026-03-13 11:40 ` [PATCH v6 2/6] cgroup,cgroup/dmem: Add (dmem_)cgroup_common_ancestor helper Natalie Vock
2026-03-13 14:16   ` Michal Koutný
2026-03-13 21:10   ` Claude review: " Claude Code Review Bot
2026-03-13 11:40 ` [PATCH v6 3/6] drm/ttm: Extract code for attempting allocation in a place Natalie Vock
2026-03-13 21:10   ` Claude review: " Claude Code Review Bot
2026-03-13 11:40 ` [PATCH v6 4/6] drm/ttm: Split cgroup charge and resource allocation Natalie Vock
2026-03-13 12:53   ` Tvrtko Ursulin
2026-03-13 21:10   ` Claude review: " Claude Code Review Bot
2026-03-13 11:40 ` Natalie Vock [this message]
2026-03-13 13:29   ` [PATCH v6 5/6] drm/ttm: Be more aggressive when allocating below protection limit Tvrtko Ursulin
2026-03-13 21:10   ` Claude review: " Claude Code Review Bot
2026-03-13 11:40 ` [PATCH v6 6/6] drm/ttm: Use common ancestor of evictor and evictee as limit pool Natalie Vock
2026-03-13 14:11   ` Tvrtko Ursulin
2026-03-13 14:16   ` Michal Koutný
2026-03-13 21:10   ` Claude review: " Claude Code Review Bot
2026-03-13 21:10 ` Claude review: cgroup/dmem,drm/ttm: Improve protection in contended cases Claude Code Review Bot

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=20260313-dmemcg-aggressive-protect-v6-5-7c71cc1492db@gmx.de \
    --to=natalie.vock@gmx.de \
    --cc=airlied@gmail.com \
    --cc=cgroups@vger.kernel.org \
    --cc=christian.koenig@amd.com \
    --cc=dev@lankhorst.se \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=hannes@cmpxchg.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=matthew.auld@intel.com \
    --cc=matthew.brost@intel.com \
    --cc=mkoutny@suse.com \
    --cc=mripard@kernel.org \
    --cc=ray.huang@amd.com \
    --cc=simona@ffwll.ch \
    --cc=tj@kernel.org \
    --cc=tursulin@ursulin.net \
    --cc=tzimmermann@suse.de \
    /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