From: "Wang, X" <x.wang@intel.com>
To: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>,
<igt-dev@lists.freedesktop.org>
Subject: Re: [PATCH 4/5] tests/intel/xe_exec_multi_queue: fix expected error for parallel queue creation
Date: Fri, 3 Apr 2026 00:00:09 -0700 [thread overview]
Message-ID: <9a7334ec-27ac-4245-9916-0c993f15506e@intel.com> (raw)
In-Reply-To: <ac8yY4EbPkjA1XnG@nvishwa1-desk>
On 4/2/2026 20:22, Niranjana Vishwanathapura wrote:
> On Thu, Apr 02, 2026 at 06:40:39PM -0700, Xin Wang wrote:
>> Multi-LRC capable engine classes (vcs, vecs) are able to process the
>> multi-queue create request and reject it with -EINVAL when
>> num_placements
>> is greater than 1. Engine classes that do not support multi-LRC cannot
>> form a parallel queue at all and return -EOPNOTSUPP instead.
>>
>> Use xe_engine_class_supports_multi_lrc() to pick the correct expected
>> error code rather than hard-coding -EINVAL for every engine class.
>>
>> Signed-off-by: Xin Wang <x.wang@intel.com>
>> ---
>> tests/intel/xe_exec_multi_queue.c | 3 ++-
>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/tests/intel/xe_exec_multi_queue.c
>> b/tests/intel/xe_exec_multi_queue.c
>> index b5ded0633..dca791fb5 100644
>> --- a/tests/intel/xe_exec_multi_queue.c
>> +++ b/tests/intel/xe_exec_multi_queue.c
>> @@ -113,7 +113,8 @@ __test_sanity(int fd, int gt, int class, bool
>> preempt_mode)
>> /* Multi-Queue can't be a parallel queue */
>> multi_queue.value = DRM_XE_MULTI_GROUP_CREATE;
>> if (n > 1)
>> - igt_assert_eq(__xe_exec_queue_create(fd, vm, 2, 1, eci, ext,
>> &val), -EINVAL);
>> + igt_assert_eq(__xe_exec_queue_create(fd, vm, 2, 1, eci, ext,
>> &val),
>> + xe_engine_class_supports_multi_lrc(fd, class) ?
>> -EINVAL : -EOPNOTSUPP);
>
> We don't want to validate multi-lrc error handling here. We only want
> to validate
> multi-queue error handling. So, I think below would be appropriate.
>
> if (n > 1 && xe_engine_class_supports_multi_lrc(fd, class))
> igt_assert_eq(__xe_exec_queue_create(fd, vm, 2, 1, eci, ext,
> &val), -EINVAL);
>
Good point. I will merge this change to patch 5 in this series.
Thanks,
Xin
> And change commit text saying that we are ensuring validating with
> parallel
> queues only when the engine class supports it.
>
> Thanks,
> Niranjana
>
>>
>> /* Specifying multiple MULTI_GROUP property is invalid */
>> multi_queue.base.next_extension = to_user_pointer(&multi_queue);
>> --
>> 2.43.0
>>
next prev parent reply other threads:[~2026-04-03 7:00 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-03 1:40 [PATCH 0/5] lib/xe: query engine class capabilities from debugfs info Xin Wang
2026-04-03 1:40 ` [PATCH 1/5] lib/xe: cache engine class masks " Xin Wang
2026-04-03 1:40 ` [PATCH 2/5] lib/xe: add xe_engine_class_supports_multi_lrc() Xin Wang
2026-04-03 1:40 ` [PATCH 3/5] lib/xe: use debugfs info to implement xe_engine_class_supports_multi_queue() Xin Wang
2026-04-03 1:40 ` [PATCH 4/5] tests/intel/xe_exec_multi_queue: fix expected error for parallel queue creation Xin Wang
2026-04-03 3:22 ` Niranjana Vishwanathapura
2026-04-03 7:00 ` Wang, X [this message]
2026-04-03 1:40 ` [PATCH 5/5] tests/intel: skip multi-LRC tests for engine classes that do not support it Xin Wang
2026-04-03 2:42 ` ✓ Xe.CI.BAT: success for lib/xe: query engine class capabilities from debugfs info Patchwork
2026-04-03 3:03 ` ✓ i915.CI.BAT: " Patchwork
2026-04-03 12:57 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-04-03 22:34 ` ✗ i915.CI.Full: " 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=9a7334ec-27ac-4245-9916-0c993f15506e@intel.com \
--to=x.wang@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=niranjana.vishwanathapura@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