[PATCH] drm/xe/guc: Allow MMIO communication even if CTB is enabled

Matthew Brost matthew.brost at intel.com
Fri Jan 12 16:25:48 UTC 2024


On Fri, Jan 12, 2024 at 02:22:48PM +0100, Michal Wajdeczko wrote:
> GuC is still able to communicate over MMIO even after we enable CTB
> communication channel.  While there are only few limited cases when
> we might want to use MMIO, we shouldn't block that with assert.
> 

What is the use case for this? I thought GuC arch team said we should
not use MMIO once CTBs are enabled. Maybe I'm misremebering?

Regardless I think we should probably leave it in until we have use
case.

Matt 

> Signed-off-by: Michal Wajdeczko <michal.wajdeczko at intel.com>
> Cc: Matthew Brost <matthew.brost at intel.com>
> ---
>  drivers/gpu/drm/xe/xe_guc.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
> index 235d27b17ff9..d4db0a34adbf 100644
> --- a/drivers/gpu/drm/xe/xe_guc.c
> +++ b/drivers/gpu/drm/xe/xe_guc.c
> @@ -674,7 +674,6 @@ int xe_guc_mmio_send_recv(struct xe_guc *guc, const u32 *request,
>  
>  	BUILD_BUG_ON(VF_SW_FLAG_COUNT != MED_VF_SW_FLAG_COUNT);
>  
> -	xe_assert(xe, !guc->ct.enabled);
>  	xe_assert(xe, len);
>  	xe_assert(xe, len <= VF_SW_FLAG_COUNT);
>  	xe_assert(xe, len <= MED_VF_SW_FLAG_COUNT);
> 
> base-commit: 7a20187068bed91b523795a18219e570f1e84eb1
> -- 
> 2.25.1
> 


More information about the Intel-xe mailing list