[PATCH 2/4] drm/xe: Do not grab forcewakes when issuing GGTT TLB invalidation via GuC

Matthew Brost matthew.brost at intel.com
Tue Mar 5 18:57:35 UTC 2024


On Tue, Mar 05, 2024 at 02:12:48PM +0100, Maarten Lankhorst wrote:
> Forcewakes are not required for communication with the GuC via CTB
> as it is a memory based interfaced. Acquring forcewakes takes
> considerable time. With that, do not grab a forcewake when issuing a
> GGTT TLB invalidation via the GuC.
> 
> Cc: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
> Cc: Lucas De Marchi <lucas.demarchi at intel.com>
> Cc: Matt Roper <matthew.d.roper at intel.com>
> Signed-off-by: Matthew Brost <matthew.brost at intel.com>

I am the author too.

Matt

> Signed-off-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
> ---
>  drivers/gpu/drm/xe/xe_ggtt.c                | 7 -------
>  drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 2 ++
>  2 files changed, 2 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
> index 0435e1202c46..325337c38961 100644
> --- a/drivers/gpu/drm/xe/xe_ggtt.c
> +++ b/drivers/gpu/drm/xe/xe_ggtt.c
> @@ -256,16 +256,9 @@ static void ggtt_invalidate_gt_tlb(struct xe_gt *gt)
>  	if (!gt)
>  		return;
>  
> -	/*
> -	 * Invalidation can happen when there's no in-flight work keeping the
> -	 * GT awake.  We need to explicitly grab forcewake to ensure the GT
> -	 * and GuC are accessible.
> -	 */
> -	xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
>  	err = xe_gt_tlb_invalidation_ggtt(gt);
>  	if (err)
>  		drm_warn(&gt_to_xe(gt)->drm, "xe_gt_tlb_invalidation_ggtt error=%d", err);
> -	xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
>  }
>  
>  void xe_ggtt_invalidate(struct xe_ggtt *ggtt)
> diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
> index f29ee1ccfa71..a3c4ffba679d 100644
> --- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
> +++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
> @@ -247,6 +247,7 @@ int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt)
>  
>  		xe_gt_tlb_invalidation_wait(gt, seqno);
>  	} else if (xe_device_uc_enabled(xe)) {
> +		xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
>  		if (xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20) {
>  			xe_mmio_write32(gt, PVC_GUC_TLB_INV_DESC1,
>  					PVC_GUC_TLB_INV_DESC1_INVALIDATE);
> @@ -256,6 +257,7 @@ int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt)
>  			xe_mmio_write32(gt, GUC_TLB_INV_CR,
>  					GUC_TLB_INV_CR_INVALIDATE);
>  		}
> +		xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
>  	}
>  
>  	return 0;
> -- 
> 2.43.0
> 


More information about the Intel-xe mailing list