Re: [PATCH v7 2/3] drm/xe/guc_ads: allocate UM queues in VRAM on dGFX

Matthew Auld <[email protected]> Tue, 4 Aug 2026 08:45:00 +0100
Newsgroups org.freedesktop.lists.intel-xe,org.kernel.vger.stable
Message-ID <[email protected]>
On 03/08/2026 21:08, Jia Yao wrote:
> On iGPU, the UM queue BO is allocated in system memory. On dGFX, the BO
> was previously created in system memory and later reallocated in
> xe_guc_realloc_post_hwconfig().  Allocate the UM queue BO directly in
> VRAM on dGFX, where it is ultimately required.
> 
> Fixes: 9c57bc08652a ("drm/xe/lnl: Drop force_probe requirement")
> Cc: Gwan-gyeong Mun <[email protected]>
> Cc: Matthew Auld <[email protected]>
> Cc: <[email protected]> # v6.12+
> Signed-off-by: Jia Yao <[email protected]>

Reviewed-by: Matthew Auld <[email protected]>


> ---
>   drivers/gpu/drm/xe/xe_guc.c     |  6 ------
>   drivers/gpu/drm/xe/xe_guc_ads.c | 17 ++++++++++-------
>   2 files changed, 10 insertions(+), 13 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
> index 4f132bf8539d..4286bd05c686 100644
> --- a/drivers/gpu/drm/xe/xe_guc.c
> +++ b/drivers/gpu/drm/xe/xe_guc.c
> @@ -732,12 +732,6 @@ static int xe_guc_realloc_post_hwconfig(struct xe_guc *guc)
>   	if (ret)
>   		return ret;
>   
> -	if (guc->ads.um_queue_bo) {
> -		ret = xe_managed_bo_reinit_in_vram(xe, tile, &guc->ads.um_queue_bo);
> -		if (ret)
> -			return ret;
> -	}
> -
>   	return 0;
>   }
>   
> diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c
> index ef8e9cc35c8c..a69ec584cc0f 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ads.c
> +++ b/drivers/gpu/drm/xe/xe_guc_ads.c
> @@ -446,19 +446,22 @@ int xe_guc_ads_init(struct xe_guc_ads *ads)
>   		/*
>   		 * Allocate a separate BO for the HW fault ring (UM queues).
>   		 *
> -		 * Round the size up to the next power of two so that
> +		 * Round the size up to the next power of two so that on iGPU
>   		 * (system memory, no IOMMU) the TTM pool issues a single
>   		 * alloc_pages(order=N) call, maximising the chance of getting
>   		 * a physically contiguous block.  GuC requires contiguous DPA.
>   		 */
> -		size_t um_size = roundup_pow_of_two(GUC_UM_QUEUE_SIZE *
> +		size_t um_size = IS_DGFX(xe) ?
> +				 GUC_UM_QUEUE_SIZE * GUC_UM_HW_QUEUE_MAX :
> +				 roundup_pow_of_two(GUC_UM_QUEUE_SIZE *
>   						    GUC_UM_HW_QUEUE_MAX);
>   
> -		bo = xe_managed_bo_create_pin_map(xe, tile, um_size,
> -						  XE_BO_FLAG_SYSTEM |
> -						  XE_BO_FLAG_GGTT |
> -						  XE_BO_FLAG_GGTT_INVALIDATE |
> -						  XE_BO_FLAG_PINNED_NORESTORE);
> +		u32 um_flags = XE_BO_FLAG_VRAM_IF_DGFX(tile) |
> +			       XE_BO_FLAG_GGTT |
> +			       XE_BO_FLAG_GGTT_INVALIDATE |
> +			       XE_BO_FLAG_PINNED_NORESTORE;
> +
> +		bo = xe_managed_bo_create_pin_map(xe, tile, um_size, um_flags);
>   		if (IS_ERR(bo))
>   			return PTR_ERR(bo);
>