Re: [PATCH v3] cgroup/dmem: implement dmem.high soft limit via prioritized eviction

[Maarten Lankhorst <maarten.lankhorst@linux.intel.com>]

Hello,

Den 2026-05-28 kl. 14:03, skrev Qiliang Yuan:
> The dmem cgroup v2 controller currently only provides a hard "max"
> limit, which causes immediate allocation failures when a cgroup's
> device memory usage reaches its quota.  GPU-bound AI workloads need
> smoother over-subscription support: a soft limit that temporarily
> allows excess usage while applying backpressure through reclaim
> rather than outright failure.
> 
> Add dmem.high, a soft limit that penalizes over-limit cgroups by
> evicting their buffer objects first when eviction is triggered (e.g.
> due to a "max" limit hit).  Unlike the rejected v1 approach which
> used sleep-on-allocation throttling, this version provides a
> meaningful recovery action through prioritized reclaim.
> 
> Expose "high" as a new cgroupfs control file per region via
> set_resource_high() and get_resource_high(), and initialize it to
> PAGE_COUNTER_MAX in reset_all_resource_limits().  Like get_resource_max(),
> get_resource_high() returns PAGE_COUNTER_MAX when the pool is NULL.
> 
> Extend dmem_cgroup_state_evict_valuable() with a "try_high"
> parameter.  When set, the function walks the page_counter parent
> chain to check whether any ancestor exceeds its high limit, then
> verifies that the pool is above its effective minimum to respect
> dmem.min protection.  Only pools meeting both criteria are evicted.
> 
> Refactor ttm_bo_evict_alloc() into a 3-pass eviction strategy.
> Pass 1 uses trylock and targets only BOs whose cgroup exceeds
> dmem.high.  Pass 2 falls back to the standard above-elow eviction.
> Pass 3 begins with a properly-locked high-priority pass in case
> Pass 1 failed due to trylock contention, then proceeds with the
> standard repeat-while-making-progress loop with low-watermark
> fallback.
> 
> Signed-off-by: Qiliang Yuan <realwujing@gmail.com>
> ---
> Introduce a "high" soft limit for the dmem cgroup v2 controller.
> When a "max" limit is hit and eviction is triggered, buffer objects
> belonging to cgroups that exceed their dmem.high limit are targeted
> first, providing a meaningful recovery action through reclaim.
> 
> The dmem cgroup currently only supports hard "max" limits, which
> cause immediate allocation failures for GPU-bound workloads. A soft
> limit enables smoother over-subscription by penalizing over-limit
> cgroups via prioritized eviction rather than outright rejection.
> 
> The implementation adds a "high" cgroupfs control file per region,
> a try_high parameter to dmem_cgroup_state_evict_valuable() for
> tier-1 eviction, and a 3-pass strategy in ttm_bo_evict_alloc().
> ---
> V2 -> V3:
> - Walk the page_counter parent chain in the try_high pass to prevent
>   child cgroups from evading the penalty when a parent cgroup exceeds
>   its dmem.high limit.
> - Check dmem.min protection in the try_high pass to avoid evicting
>   BOs below the effective minimum.
> - Add a properly-locked high-priority retry at the beginning of Pass 3
>   so that actively-used over-limit BOs (which failed trylock in Pass 1)
>   are not skipped while innocent cgroups are evicted.
> - Fix get_resource_high(NULL) returning 0 instead of PAGE_COUNTER_MAX
>   to match the behavior of get_resource_max().
> 
> V1 -> V2:
> - Replace sleep-on-allocation throttling with prioritized eviction.
>   When a "max" limit is hit, BOs from cgroups exceeding dmem.high are
>   evicted first in a dedicated pass. No throttling or sleeping is
>   performed in the charge path.
> - Remove task throttling (schedule_timeout_killable, TIF_NOTIFY_RESUME,
>   resume_user_mode_work() integration) entirely.
> - Add dmem.high cgroupfs control file per region.
> - Extend dmem_cgroup_state_evict_valuable() with try_high parameter
>   to target over-limit cgroups as tier-1 eviction.
> - Refactor ttm_bo_evict_alloc() into a 3-pass eviction strategy:
>   (1) trylock: evict only BOs exceeding dmem.high
>   (2) trylock: above-elow
>   (3) proper-lock: repeat with low fallback.
> - Initialize high to PAGE_COUNTER_MAX in reset_all_resource_limits().
> 
> v1: https://lore.kernel.org/all/20260520-feature-dmem-high-v1-1-97ca0cb7f95a@gmail.com
> v2: https://lore.kernel.org/all/20260522-feature-dmem-high-v2-1-d805deddecbb@gmail.com
> ---
>  drivers/gpu/drm/ttm/ttm_bo.c | 35 ++++++++++++++++++++----
>  include/linux/cgroup_dmem.h  |  4 +--
>  kernel/cgroup/dmem.c         | 65 ++++++++++++++++++++++++++++++++++++++++++--
>  3 files changed, 94 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
> index bcd76f6bb7f02..2f2b428f1d30a 100644
> --- a/drivers/gpu/drm/ttm/ttm_bo.c
> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
> @@ -505,6 +505,8 @@ struct ttm_bo_evict_walk {
>  
>  	/** @limit_pool: Which pool limit we should test against */
>  	struct dmem_cgroup_pool_state *limit_pool;
> +	/** @try_high: Whether to only evict BO's above the high watermark (first pass) */
> +	bool try_high;
>  	/** @try_low: Whether we should attempt to evict BO's with low watermark threshold */
>  	bool try_low;
>  	/** @hit_low: If we cannot evict a bo when @try_low is false (first pass) */
> @@ -518,7 +520,8 @@ static s64 ttm_bo_evict_cb(struct ttm_lru_walk *walk, struct ttm_buffer_object *
>  	s64 lret;
>  
>  	if (!dmem_cgroup_state_evict_valuable(evict_walk->limit_pool, bo->resource->css,
> -					      evict_walk->try_low, &evict_walk->hit_low))
> +					      evict_walk->try_high, evict_walk->try_low,
> +					      &evict_walk->hit_low))
>  		return 0;
>  
>  	if (bo->pin_count || !bo->bdev->funcs->eviction_valuable(bo, evict_walk->place))
> @@ -577,31 +580,51 @@ static int ttm_bo_evict_alloc(struct ttm_device *bdev,
>  	};
>  	s64 lret;
>  
> +	/*
> +	 * Pass 1 (trylock): Only evict BOs whose cgroup is above its
> +	 * dmem.high soft limit. This penalizes over-limit cgroups first.
> +	 */
>  	evict_walk.walk.arg.trylock_only = true;
> +	evict_walk.try_high = true;
>  	lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1);
> +	evict_walk.try_high = false;
> +	if (lret)
> +		goto out;

I believe the first pass for 'high' should not be trylock only. High needs to be
preferentially evicted, even if the objects are locked elsewhere.


> -	/* One more attempt if we hit low limit? */
> +	/*
> +	 * Pass 2 (trylock): Evict BOs above the effective low watermark.
> +	 * Falls back to low-priority eviction if needed.
> +	 */
> +	lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1);
>  	if (!lret && evict_walk.hit_low) {
>  		evict_walk.try_low = true;
>  		lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1);
>  	}
> +
>  	if (lret || !ticket)
>  		goto out;
>  
> -	/* Reset low limit */
> +	/*
> +	 * Pass 3+ (properly locked): Evict while making progress.
> +	 * First retry the high-priority pass with proper locking in case
> +	 * Pass 1 failed due to trylock contention on over-limit BOs.
> +	 * If that still fails, fall back to the standard low-priority eviction.
> +	 */
>  	evict_walk.try_low = evict_walk.hit_low = false;
> -	/* If ticket-locking, repeat while making progress. */
>  	evict_walk.walk.arg.trylock_only = false;
> +	evict_walk.try_high = true;
> +	lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1);
> +	evict_walk.try_high = false;
> +	if (lret)
> +		goto out;
>  
>  retry:
>  	do {
> -		/* The walk may clear the evict_walk.walk.ticket field */
>  		evict_walk.walk.arg.ticket = ticket;
>  		evict_walk.evicted = 0;
>  		lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1);
>  	} while (!lret && evict_walk.evicted);
>  
> -	/* We hit the low limit? Try once more */
>  	if (!lret && evict_walk.hit_low && !evict_walk.try_low) {
>  		evict_walk.try_low = true;
>  		goto retry;
> diff --git a/include/linux/cgroup_dmem.h b/include/linux/cgroup_dmem.h
> index dd4869f1d736e..06115d35509b1 100644
> --- a/include/linux/cgroup_dmem.h
> +++ b/include/linux/cgroup_dmem.h
> @@ -23,7 +23,7 @@ int dmem_cgroup_try_charge(struct dmem_cgroup_region *region, u64 size,
>  void dmem_cgroup_uncharge(struct dmem_cgroup_pool_state *pool, u64 size);
>  bool dmem_cgroup_state_evict_valuable(struct dmem_cgroup_pool_state *limit_pool,
>  				      struct dmem_cgroup_pool_state *test_pool,
> -				      bool ignore_low, bool *ret_hit_low);
> +				      bool try_high, bool ignore_low, bool *ret_hit_low);
>  
>  void dmem_cgroup_pool_state_put(struct dmem_cgroup_pool_state *pool);
>  #else
> @@ -54,7 +54,7 @@ static inline void dmem_cgroup_uncharge(struct dmem_cgroup_pool_state *pool, u64
>  static inline
>  bool dmem_cgroup_state_evict_valuable(struct dmem_cgroup_pool_state *limit_pool,
>  				      struct dmem_cgroup_pool_state *test_pool,
> -				      bool ignore_low, bool *ret_hit_low)
> +				      bool try_high, bool ignore_low, bool *ret_hit_low)
>  {
>  	return true;
>  }
> diff --git a/kernel/cgroup/dmem.c b/kernel/cgroup/dmem.c
> index 4753a67d0f0f2..c80444c0da177 100644
> --- a/kernel/cgroup/dmem.c
> +++ b/kernel/cgroup/dmem.c
> @@ -156,6 +156,12 @@ set_resource_low(struct dmem_cgroup_pool_state *pool, u64 val)
>  	page_counter_set_low(&pool->cnt, val);
>  }
>  
> +static void
> +set_resource_high(struct dmem_cgroup_pool_state *pool, u64 val)
> +{
> +	page_counter_set_high(&pool->cnt, val);
> +}
> +
>  static void
>  set_resource_max(struct dmem_cgroup_pool_state *pool, u64 val)
>  {
> @@ -167,6 +173,11 @@ static u64 get_resource_low(struct dmem_cgroup_pool_state *pool)
>  	return pool ? READ_ONCE(pool->cnt.low) : 0;
>  }
>  
> +static u64 get_resource_high(struct dmem_cgroup_pool_state *pool)
> +{
> +	return pool ? READ_ONCE(pool->cnt.high) : PAGE_COUNTER_MAX;
> +}
> +
>  static u64 get_resource_min(struct dmem_cgroup_pool_state *pool)
>  {
>  	return pool ? READ_ONCE(pool->cnt.min) : 0;
> @@ -186,6 +197,7 @@ static void reset_all_resource_limits(struct dmem_cgroup_pool_state *rpool)
>  {
>  	set_resource_min(rpool, 0);
>  	set_resource_low(rpool, 0);
> +	set_resource_high(rpool, PAGE_COUNTER_MAX);
>  	set_resource_max(rpool, PAGE_COUNTER_MAX);
>  }
>  
> @@ -289,10 +301,13 @@ dmem_cgroup_calculate_protection(struct dmem_cgroup_pool_state *limit_pool,
>   * dmem_cgroup_state_evict_valuable() - Check if we should evict from test_pool
>   * @limit_pool: The pool for which we hit limits
>   * @test_pool: The pool for which to test
> + * @try_high: Only evict BOs whose usage exceeds the high limit (first pass)
>   * @ignore_low: Whether we have to respect low watermarks.
>   * @ret_hit_low: Pointer to whether it makes sense to consider low watermark.
>   *
>   * This function returns true if we can evict from @test_pool, false if not.
> + * When @try_high is set, only pools with usage above their high limit are
> + * evictable, enabling prioritized eviction of over-limit cgroups.
>   * When returning false and @ignore_low is false, @ret_hit_low may
>   * be set to true to indicate this function can be retried with @ignore_low
>   * set to true.
> @@ -301,7 +316,7 @@ dmem_cgroup_calculate_protection(struct dmem_cgroup_pool_state *limit_pool,
>   */
>  bool dmem_cgroup_state_evict_valuable(struct dmem_cgroup_pool_state *limit_pool,
>  				      struct dmem_cgroup_pool_state *test_pool,
> -				      bool ignore_low, bool *ret_hit_low)
> +				      bool try_high, bool ignore_low, bool *ret_hit_low)
>  {
>  	struct dmem_cgroup_pool_state *pool = test_pool;
>  	struct page_counter *ctest;
> @@ -331,9 +346,38 @@ bool dmem_cgroup_state_evict_valuable(struct dmem_cgroup_pool_state *limit_pool,
>  
>  	ctest = &test_pool->cnt;
>  
> +	used = page_counter_read(ctest);
> +
> +	if (try_high) {
> +		struct page_counter *c;
> +
> +		/*
> +		 * Walk the page_counter parent chain to check whether any
> +		 * ancestor cgroup exceeds its dmem.high limit.  This prevents
> +		 * child cgroups from evading the penalty when a parent cgroup
> +		 * is over its high limit.
> +		 */
> +		if (used <= READ_ONCE(ctest->high)) {
> +			for (c = ctest->parent; c; c = c->parent) {
> +				if (page_counter_read(c) > READ_ONCE(c->high))
> +					break;
> +			}
> +			if (!c)
> +				return false;
> +		}
> +
> +		/*
> +		 * Respect dmem.min protection: do not evict BOs below the
> +		 * effective minimum even during the high-priority pass.
> +		 */
> +		dmem_cgroup_calculate_protection(limit_pool, test_pool);
> +		min = READ_ONCE(ctest->emin);
> +
> +		return used > min;
> +	}
> +
>  	dmem_cgroup_calculate_protection(limit_pool, test_pool);
>  
> -	used = page_counter_read(ctest);
>  	min = READ_ONCE(ctest->emin);
>  
>  	if (used <= min)
> @@ -835,6 +879,17 @@ static ssize_t dmem_cgroup_region_low_write(struct kernfs_open_file *of,
>  	return dmemcg_limit_write(of, buf, nbytes, off, set_resource_low);
>  }
>  
> +static int dmem_cgroup_region_high_show(struct seq_file *sf, void *v)
> +{
> +	return dmemcg_limit_show(sf, v, get_resource_high);
> +}
> +
> +static ssize_t dmem_cgroup_region_high_write(struct kernfs_open_file *of,
> +					  char *buf, size_t nbytes, loff_t off)
> +{
> +	return dmemcg_limit_write(of, buf, nbytes, off, set_resource_high);
> +}
> +
>  static int dmem_cgroup_region_max_show(struct seq_file *sf, void *v)
>  {
>  	return dmemcg_limit_show(sf, v, get_resource_max);
> @@ -868,6 +923,12 @@ static struct cftype files[] = {
>  		.seq_show = dmem_cgroup_region_low_show,
>  		.flags = CFTYPE_NOT_ON_ROOT,
>  	},
> +	{
> +		.name = "high",
> +		.write = dmem_cgroup_region_high_write,
> +		.seq_show = dmem_cgroup_region_high_show,
> +		.flags = CFTYPE_NOT_ON_ROOT,
> +	},
>  	{
>  		.name = "max",
>  		.write = dmem_cgroup_region_max_write,

The rest of the patch looks good.