Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Francois Dugast <francois.dugast@intel.com>
To: "Zbigniew Kempczyński" <zbigniew.kempczynski@intel.com>
Cc: <igt-dev@lists.freedesktop.org>
Subject: Re: [PATCH i-g-t] lib/intel_compute: Fix localgroup size passed in indirect data for Xe3p
Date: Mon, 4 May 2026 15:24:44 +0200	[thread overview]
Message-ID: <afieEvWrJLfRZsrg@fdugast-desk> (raw)
In-Reply-To: <20260504113814.3720157-2-zbigniew.kempczynski@intel.com>

On Mon, May 04, 2026 at 01:38:15PM +0200, Zbigniew Kempczyński wrote:
> Previously hardcoded constants are not correct for handling larger buffers
> in compute square tests in Xe3p. Replace these values to commonly used
> localgroup size definitions.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> Cc: Francois Dugast <francois.dugast@intel.com>

Reviewed-by: Francois Dugast <francois.dugast@intel.com>

> ---
>  lib/intel_compute.c | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/lib/intel_compute.c b/lib/intel_compute.c
> index 7dda423c80..680675982a 100644
> --- a/lib/intel_compute.c
> +++ b/lib/intel_compute.c
> @@ -2054,7 +2054,7 @@ static void xe3p_compute_exec_compute(int fd,
>  	addr_bo_buffer_batch[b++] = 0x008004A0; /* dw3 */
>  	addr_bo_buffer_batch[b++] = 0xBE040000; /* dw4 */
>  	addr_bo_buffer_batch[b++] = 0xFFFFFFFF; /* dw5 */
> -	addr_bo_buffer_batch[b++] = 0x000001FF; /* dw6 */
> +	addr_bo_buffer_batch[b++] = 0x000003FF; /* dw6 */ // Local X/Y/Z Dimension
>  
>  	if (threadgroup_preemption)
>  		addr_bo_buffer_batch[b++] = XE2_THREADGROUP_PREEMPT_XDIM; // Thread Group ID X Dimension
> @@ -2081,7 +2081,7 @@ static void xe3p_compute_exec_compute(int fd,
>  
>  	addr_bo_buffer_batch[b++] = 0x00000000; /* dw22 */
>  	addr_bo_buffer_batch[b++] = 0x00000000; /* dw23 */
> -	addr_bo_buffer_batch[b++] = 0x0C000010; /* dw24 */
> +	addr_bo_buffer_batch[b++] = 0x0C000000 | THREADS_PER_GROUP; /* dw24 */
>  	addr_bo_buffer_batch[b++] = 0x00000000; /* dw25 */
>  	addr_bo_buffer_batch[b++] = 0x00000400; /* dw26 */
>  
> @@ -2135,9 +2135,9 @@ static void xe3p_create_indirect_data(uint32_t *addr_bo_buffer_batch,
>  	addr_bo_buffer_batch[b++] = addr_output & 0xffffffff;
>  	addr_bo_buffer_batch[b++] = addr_output >> 32;
>  	addr_bo_buffer_batch[b++] = loop_count;
> -	addr_bo_buffer_batch[b++] = 0x00000200;
> -	addr_bo_buffer_batch[b++] = 0x00000001;
> -	addr_bo_buffer_batch[b++] = 0x00000001;
> +	addr_bo_buffer_batch[b++] = ENQUEUED_LOCAL_SIZE_X;
> +	addr_bo_buffer_batch[b++] = ENQUEUED_LOCAL_SIZE_Y;
> +	addr_bo_buffer_batch[b++] = ENQUEUED_LOCAL_SIZE_Z;
>  	addr_bo_buffer_batch[b++] = 0x00000000;
>  	addr_bo_buffer_batch[b++] = 0x00000000;
>  	addr_bo_buffer_batch[b++] = 0x00000000;
> -- 
> 2.43.0
> 

  reply	other threads:[~2026-05-04 13:25 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-04 11:38 [PATCH i-g-t] lib/intel_compute: Fix localgroup size passed in indirect data for Xe3p Zbigniew Kempczyński
2026-05-04 13:24 ` Francois Dugast [this message]
2026-05-04 17:36 ` ✗ Fi.CI.BAT: failure for " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=afieEvWrJLfRZsrg@fdugast-desk \
    --to=francois.dugast@intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=zbigniew.kempczynski@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox