RE: [PATCH v2 3/4] drm/xe: Wire up dmem cgroup reclaim for VRAM manager

[Upadhyay, Tejas]

> -----Original Message-----
> From: Intel-xe <intel-xe-bounces@xxxxxxxxxxxxxxxxxxxxx> On Behalf Of Thomas
> Hellström
> Sent: 28 April 2026 13:01
> To: intel-xe@xxxxxxxxxxxxxxxxxxxxx
> Cc: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx>; Natalie Vock
> <natalie.vock@xxxxxx>; Johannes Weiner <hannes@xxxxxxxxxxx>; Tejun Heo
> <tj@xxxxxxxxxx>; Michal Koutný <mkoutny@xxxxxxxx>;
> cgroups@xxxxxxxxxxxxxxx; Huang Rui <ray.huang@xxxxxxx>; Brost, Matthew
> <matthew.brost@xxxxxxxxx>; Auld, Matthew <matthew.auld@xxxxxxxxx>;
> Maarten Lankhorst <maarten.lankhorst@xxxxxxxxxxxxxxx>; Maxime Ripard
> <mripard@xxxxxxxxxx>; Thomas Zimmermann <tzimmermann@xxxxxxx>;
> Simona Vetter <simona@xxxxxxxx>; David Airlie <airlied@xxxxxxxxx>; Christian
> König <christian.koenig@xxxxxxx>; Alex Deucher
> <alexander.deucher@xxxxxxx>; Vivi, Rodrigo <rodrigo.vivi@xxxxxxxxx>; dri-
> devel@xxxxxxxxxxxxxxxxxxxxx; amd-gfx@xxxxxxxxxxxxxxxxxxxxx; linux-
> kernel@xxxxxxxxxxxxxxx
> Subject: [PATCH v2 3/4] drm/xe: Wire up dmem cgroup reclaim for VRAM
> manager
> 
> Register the VRAM manager with the dmem cgroup reclaim infrastructure so
> that lowering dmem.max below current VRAM usage triggers TTM eviction
> rather than failing with -EBUSY.
> 
> Assisted-by: GitHub Copilot:claude-sonnet-4.6
> Signed-off-by: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx>
> ---
>  drivers/gpu/drm/xe/xe_ttm_vram_mgr.c | 19 ++++++++++++-------
>  1 file changed, 12 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> index 5fd0d5506a7e..1bdcb3fee901 100644
> --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
> @@ -303,13 +303,6 @@ int __xe_ttm_vram_mgr_init(struct xe_device *xe,
> struct xe_ttm_vram_mgr *mgr,
>  	struct ttm_resource_manager *man = &mgr->manager;
>  	int err;
> 
> -	if (mem_type != XE_PL_STOLEN) {
> -		const char *name = mem_type == XE_PL_VRAM0 ? "vram0" :
> "vram1";
> -		man->cg = drmm_cgroup_register_region(&xe->drm, name,
> size);
> -		if (IS_ERR(man->cg))
> -			return PTR_ERR(man->cg);
> -	}
> -
>  	man->func = &xe_ttm_vram_mgr_func;
>  	mgr->mem_type = mem_type;
>  	mutex_init(&mgr->lock);
> @@ -318,6 +311,18 @@ int __xe_ttm_vram_mgr_init(struct xe_device *xe,
> struct xe_ttm_vram_mgr *mgr,
>  	mgr->visible_avail = io_size;
> 
>  	ttm_resource_manager_init(man, &xe->ttm, size);
> +
> +	if (mem_type != XE_PL_STOLEN) {
> +		const char *name = mem_type == XE_PL_VRAM0 ? "vram0" :
> "vram1";
> +		struct dmem_cgroup_region *cg =
> +			drmm_cgroup_register_region(&xe->drm, name,
> size);
> +
> +		if (IS_ERR(cg))
> +			return PTR_ERR(cg);
> +
> +		ttm_resource_manager_set_dmem_region(man, cg);
> +	}
> +

I think, this reordering is reasonable cleanup, LGTM:
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@xxxxxxxxx>

Tejas
>  	err = gpu_buddy_init(&mgr->mm, man->size, default_page_size);
>  	if (err)
>  		return err;
> --
> 2.53.0