From: Nirmoy Das <nirmoy.das@intel.com>
To: <janga.rahul.kumar@intel.com>, <igt-dev@lists.freedesktop.org>,
<ramadevi.gandi@intel.com>
Subject: Re: [PATCH i-g-t] lib/intel_compute: Add documentation of global workgroup size
Date: Wed, 20 Mar 2024 12:56:15 +0100 [thread overview]
Message-ID: <2f4bb0d7-2fe9-4373-9995-9308dc5088b3@intel.com> (raw)
In-Reply-To: <20240318085227.2624098-1-janga.rahul.kumar@intel.com>
[-- Attachment #1: Type: text/plain, Size: 1800 bytes --]
On 3/18/2024 9:52 AM, janga.rahul.kumar@intel.com wrote:
> From: Janga Rahul Kumar<janga.rahul.kumar@intel.com>
>
> Global workgroup size is calculated based on local X/Y/Z maximum
> and thread group id X dimension.
>
> Cc: Nirmoy Das<nirmoy.das@intel.com>
> Signed-off-by: Janga Rahul Kumar<janga.rahul.kumar@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
> ---
> lib/intel_compute.c | 12 ++++++++----
> 1 file changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/lib/intel_compute.c b/lib/intel_compute.c
> index 9d3b97efe..72025a481 100644
> --- a/lib/intel_compute.c
> +++ b/lib/intel_compute.c
> @@ -1243,15 +1243,19 @@ static void xe2lpg_compute_exec_compute(uint32_t *addr_bo_buffer_batch,
> addr_bo_buffer_batch[b++] = offset_indirect_data_start;
> addr_bo_buffer_batch[b++] = 0xbe040000;
> addr_bo_buffer_batch[b++] = 0xffffffff;
> - addr_bo_buffer_batch[b++] = 0x000003ff;
> + addr_bo_buffer_batch[b++] = 0x000003ff; // Local X/Y/Z Dimension
>
> if (threadgroup_preemption)
> - addr_bo_buffer_batch[b++] = 0x00200000; // Global workgroup size
> + /*
> + * Create multiple threadgroups using higher gloabl workgroup size
> + * Global Workgroup size = Local X * Thread Group X + Local Y * Thread Group Y + Local Z * Thread Group Z
> + */
> + addr_bo_buffer_batch[b++] = 0x00200000; // Thread Group ID X Dimension
> else
> addr_bo_buffer_batch[b++] = 0x00000002;
>
> - addr_bo_buffer_batch[b++] = 0x00000001;
> - addr_bo_buffer_batch[b++] = 0x00000001;
> + addr_bo_buffer_batch[b++] = 0x00000001; // Thread Group ID Y Dimension
> + addr_bo_buffer_batch[b++] = 0x00000001; // Thread Group ID Z Dimension
> addr_bo_buffer_batch[b++] = 0x00000000;
> addr_bo_buffer_batch[b++] = 0x00000000;
> addr_bo_buffer_batch[b++] = 0x00000000;
[-- Attachment #2: Type: text/html, Size: 2697 bytes --]
next prev parent reply other threads:[~2024-03-20 11:56 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-18 8:52 [PATCH i-g-t] lib/intel_compute: Add documentation of global workgroup size janga.rahul.kumar
2024-03-18 17:32 ` ✓ Fi.CI.BAT: success for " Patchwork
2024-03-18 17:35 ` ✓ CI.xeBAT: " Patchwork
2024-03-18 22:02 ` ✗ Fi.CI.IGT: failure " Patchwork
2024-03-21 18:26 ` Kamil Konieczny
2024-03-20 11:56 ` Nirmoy Das [this message]
2024-03-25 11:33 ` ✓ Fi.CI.IGT: success " 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=2f4bb0d7-2fe9-4373-9995-9308dc5088b3@intel.com \
--to=nirmoy.das@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=janga.rahul.kumar@intel.com \
--cc=ramadevi.gandi@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