Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Summers, Stuart" <stuart.summers@intel.com>
To: "igt-dev@lists.freedesktop.org" <igt-dev@lists.freedesktop.org>,
	"Wang, X" <x.wang@intel.com>
Subject: Re: [PATCH] tests/xe_exec_threads: relax data check in multi-queue hang mode
Date: Thu, 7 May 2026 19:05:56 +0000	[thread overview]
Message-ID: <704dd473ea9651a2911b59e3abd12e5051e673a9.camel@intel.com> (raw)
In-Reply-To: <20260506205650.20503-1-x.wang@intel.com>

On Wed, 2026-05-06 at 13:56 -0700, Xin Wang wrote:
> In multi-queue hang mode, a reset of any queue in a queue group will
> cause the entire group to reset. And since GuC only submits on the
> primary queue, without explicit barriers or priority differences
> between queues, we can't guarantee which queue hardware will pick up
> on submission. As such, we don't know which queues will complete
> prior to the group reset and need to check for all possible values.
> 
> Suggested-by: Stuart Summers <stuart.summers@intel.com>
> Signed-off-by: Xin Wang <x.wang@intel.com>

Reviewed-by: Stuart Summers <stuart.summers@intel.com>

> ---
>  tests/intel/xe_exec_threads.c | 11 ++++++++---
>  1 file changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/tests/intel/xe_exec_threads.c
> b/tests/intel/xe_exec_threads.c
> index 40a6c0c6d..9331a3b88 100644
> --- a/tests/intel/xe_exec_threads.c
> +++ b/tests/intel/xe_exec_threads.c
> @@ -707,10 +707,15 @@ test_legacy_mode(int fd, uint32_t vm, uint64_t
> addr, uint64_t userptr,
>              i < n_execs; i++) {
>                 int e = i % n_exec_queues;
>  
> -               if (flags & HANG && e == hang_exec_queue)
> +               if (flags & HANG && e == hang_exec_queue) {
>                         igt_assert_eq(data[i].data, 0x0);
> -               else
> -                       igt_assert_eq(data[i].data, 0xc0ffee);
> +               } else {
> +                       /* Multi-queue reset affects all queues; data
> outcome is non-deterministic */
> +                       if (flags & HANG && flags & MULTI_QUEUE)
> +                               igt_assert(data[i].data == 0x0 ||
> data[i].data == 0xc0ffee);
> +                       else
> +                               igt_assert_eq(data[i].data,
> 0xc0ffee);
> +               }
>         }
>  
>         syncobj_destroy(fd, sync[0].handle);


      parent reply	other threads:[~2026-05-07 19:06 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-06 20:56 [PATCH] tests/xe_exec_threads: relax data check in multi-queue hang mode Xin Wang
2026-05-06 22:08 ` ✓ i915.CI.BAT: success for " Patchwork
2026-05-06 22:17 ` ✓ Xe.CI.BAT: " Patchwork
2026-05-06 23:38 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-05-07  4:25 ` ✗ i915.CI.Full: " Patchwork
2026-05-07 19:05 ` Summers, Stuart [this message]

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=704dd473ea9651a2911b59e3abd12e5051e673a9.camel@intel.com \
    --to=stuart.summers@intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=x.wang@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