From: Gustavo Sousa <gustavo.sousa@intel.com>
To: Kamil Konieczny <kamil.konieczny@linux.intel.com>
Cc: <igt-dev@lists.freedesktop.org>,
Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
Subject: Re: [PATCH i-g-t v2] tests/intel/xe_pmu: Do not enable VFs in unrelated subtests
Date: Tue, 6 Oct 2026 13:05:48 -0300 [thread overview]
Message-ID: <87qzi2ddb7.fsf@intel.com> (raw)
In-Reply-To: <20261006144440.alu37vjplb5igvxx@kamilkon-DESK.igk.intel.com>
Kamil Konieczny <kamil.konieczny@linux.intel.com> writes:
> Hi Gustavo,
> On 2026-10-06 at 11:29:12 -0300, Gustavo Sousa wrote:
>> Kamil Konieczny <kamil.konieczny@linux.intel.com> writes:
>>
>> > Hi Gustavo,
>> > On 2026-10-06 at 11:11:19 -0300, Gustavo Sousa wrote:
>> >> The test xe_pmu has a subtest group that enables SRIOV VFs as part of
>> >> its fixture. Even though the fixture is part of a subtest group, the
>> >> enabling/disabling will happen regardless of whether those subests are
>> >> selected or not (via --run-subtest).
>> >>
>> >> As most subtests in xe_pmu don't need that, avoid doing that as a
>> >> fixture and rather incorporate the enabling/disabling as part of a new
>> >> macro called test_each_engine_with_vfs().
>> >>
>> >> v2:
>> >> - Add "intel" to commit subject prefix. (Kamil)
>> >> - Fix argument name of test_each_engine_with_vfs():
>> >> s/num_vfs/num_fns/. (Kamil)
>> >> - Enable/disable VFs only once per macro call by moving the call to
>> >> xe_for_each_engine() inside the for-loop. (Marcin)
>> >>
>> >> Cc: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
>> >> Cc: Kamil Konieczny <kamil.konieczny@linux.intel.com>
>> >> Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
>> >> ---
>> >> tests/intel/xe_pmu.c | 21 ++++++++++++---------
>> >> 1 file changed, 12 insertions(+), 9 deletions(-)
>> >>
>> >> diff --git a/tests/intel/xe_pmu.c b/tests/intel/xe_pmu.c
>> >> index d41ebba1faef..eda56c078653 100644
>> >> --- a/tests/intel/xe_pmu.c
>> >> +++ b/tests/intel/xe_pmu.c
>> >> @@ -124,6 +124,15 @@ static bool has_engine_active_ticks;
>> >> igt_dynamic_f("engine-%s%d", xe_engine_class_string(hwe->engine_class), \
>> >> hwe->engine_instance)
>> >>
>> >> +#define test_each_engine_with_vfs(test, fd, hwe, num_fns) \
>> >> + igt_subtest_with_dynamic(test) \
>> >> + for (num_fns = enable_and_provision_vfs(fd) + 1; \
>> >> + num_fns; \
>> >> + unprovision_and_disable_vfs(fd), num_fns = 0) \
>> >> + xe_for_each_engine(fd, hwe) \
>> >> + igt_dynamic_f("engine-%s%d", xe_engine_class_string(hwe->engine_class), \
>> >> + hwe->engine_instance)
>> >> +
>> >> static bool has_event(const char *device, const char *event)
>> >> {
>> >> char buf[512];
>> >> @@ -1192,26 +1201,20 @@ int igt_main()
>> >> igt_subtest_group() {
>> >> unsigned int num_fns;
>> >>
>> >> - igt_fixture()
>> >> - num_fns = enable_and_provision_vfs(fd) + 1;
>> >> -
>> >> igt_describe("Validate engine activity on all functions");
>> >> - test_each_engine("all-fn-engine-activity-load", fd, eci)
>> >> + test_each_engine_with_vfs("all-fn-engine-activity-load", fd, eci, num_fns)
>> >> engine_activity_all_fn(fd, eci, num_fns);
>> >>
>> >> igt_describe("Validate per-function engine activity");
>> >> - test_each_engine("fn-engine-activity-load", fd, eci)
>> >> + test_each_engine_with_vfs("fn-engine-activity-load", fd, eci, num_fns)
>> >> for (int fn = 0; fn < num_fns; fn++)
>> >> engine_activity_fn(fd, eci, fn, XE_SRIOV_SCHED_PRIORITY_LOW);
>> >>
>> >> igt_describe("Validate per-function engine activity when sched-if-idle is set");
>> >> - test_each_engine("fn-engine-activity-sched-if-idle", fd, eci) {
>> >> + test_each_engine_with_vfs("fn-engine-activity-sched-if-idle", fd, eci, num_fns) {
>> >> for (int fn = 0; fn < num_fns; fn++)
>> >> engine_activity_fn(fd, eci, fn, XE_SRIOV_SCHED_PRIORITY_NORMAL);
>> >> }
>> >> -
>> >> - igt_fixture()
>> >> - unprovision_and_disable_vfs(fd);
>> >
>> > Hmm, maybe that last one should stay, test should do a cleanup,
>> > after it ends with igt_assert() or skip, for example:
>> >
>> > igt_fixture() {
>> > if (num_nfs)
>> > unprovision_and_disable_vfs(fd);
>> > }
>>
>> The cleanup is already done in the for-loop generated by the macro. Do
>> you see a case where we really need this extra cleanup call?
>
> When test fails due to igt_assert(), code execution jumps to
> next igt_fixture, so also no cleanup will happen in for-loop.
I think skip and assertion failures actually make the program jump
back to where the igt_dynamic_f() was called, no?
Here is an example code (saved as tests/foo.c):
#include "igt_core.h"
static void run_dynamic_test_logic(int index)
{
igt_info("Running dynamic work for iteration %d...\n", index + 1);
igt_assert_f(index != 1, "Intentional IGT assertion failure on the 2nd dynamic subtest!\n");
igt_skip_on_f(index == 2, "Intentional IGT skip on the 3rd dynamic subtest!\n");
igt_info("Dynamic work for iteration %d completed successfully.\n", index + 1);
}
int igt_main()
{
igt_subtest_with_dynamic("container-subtest") {
for (int i = 0; i < 3; i++) {
igt_info("\nBefore calling igt_dynamic_f().\n");
igt_dynamic_f("dynamic-subtest-%d", i) {
run_dynamic_test_logic(i);
}
igt_info("After calling igt_dynamic_f().\n\n");
}
}
}
, and here is the output of compiling and running it:
$ ninja -C build tests/foo && ./build/tests/foo
ninja: Entering directory `build'
[1/9] Compiling C object tests/foo.p/foo.c.o
[2/9] Generating lib/version.h with a custom command
[3/3] Linking target tests/foo
IGT-Version: 2.6-g095481e279a9 (x86_64) (Linux: 7.2.7-200.fc44.x86_64 x86_64)
Using IGT_SRANDOM=1791302421 for randomisation
Starting subtest: container-subtest
Before calling igt_dynamic_f().
Starting dynamic subtest: dynamic-subtest-0
Running dynamic work for iteration 1...
Dynamic work for iteration 1 completed successfully.
Dynamic subtest dynamic-subtest-0: SUCCESS (0.000s)
After calling igt_dynamic_f().
Before calling igt_dynamic_f().
Starting dynamic subtest: dynamic-subtest-1
Running dynamic work for iteration 2...
[61375.969517] (foo:1196168) CRITICAL: Test assertion failure function run_dynamic_test_logic, file ../tests/foo.c:6:
[61375.969520] (foo:1196168) CRITICAL: Failed assertion: index != 1
[61375.969527] (foo:1196168) CRITICAL: Last errno: 13, Permission denied
[61375.969528] (foo:1196168) CRITICAL: Intentional IGT assertion failure on the 2nd dynamic subtest!
Stack trace: not implemented
Dynamic subtest dynamic-subtest-1 failed.
**** DEBUG ****
[61375.969512] (foo:1196168) INFO: Running dynamic work for iteration 2...
[61375.969517] (foo:1196168) CRITICAL: Test assertion failure function run_dynamic_test_logic, file ../tests/foo.c:6:
[61375.969520] (foo:1196168) CRITICAL: Failed assertion: index != 1
[61375.969527] (foo:1196168) CRITICAL: Last errno: 13, Permission denied
[61375.969528] (foo:1196168) CRITICAL: Intentional IGT assertion failure on the 2nd dynamic subtest!
[61375.969530] (foo:1196168) igt_core-INFO: Stack trace: not implemented
**** END ****
Dynamic subtest dynamic-subtest-1: FAIL (0.000s)
After calling igt_dynamic_f().
Before calling igt_dynamic_f().
Starting dynamic subtest: dynamic-subtest-2
Running dynamic work for iteration 3...
Test requirement not met in function run_dynamic_test_logic, file ../tests/foo.c:7:
Test requirement: !(index == 2)
Intentional IGT skip on the 3rd dynamic subtest!
Last errno: 13, Permission denied
Dynamic subtest dynamic-subtest-2: SKIP (0.000s)
After calling igt_dynamic_f().
Subtest container-subtest: FAIL (0.000s)
--
Gustavo Sousa
>
> Regards,
> Kamil
>
>>
>> --
>> Gustavo Sousa
>>
>> >
>> >
>> > Regards,
>> > Kamil
>> >
>> >> }
>> >>
>> >> igt_subtest_group() {
>> >>
>> >> ---
>> >> base-commit: faad2cd0e3e8eb2e18111c2358823a7bdfb56d94
>> >> change-id: 20261005-xe_pmu-do-not-enable-vfs-in-unrelated-subtests-e24c1cd95d71
>> >>
>> >> Best regards,
>> >> --
>> >> Gustavo Sousa <gustavo.sousa@intel.com>
>> >>
next prev parent reply other threads:[~2026-10-06 16:06 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-06 14:11 [PATCH i-g-t v2] tests/intel/xe_pmu: Do not enable VFs in unrelated subtests Gustavo Sousa
2026-10-06 14:23 ` Kamil Konieczny
2026-10-06 14:29 ` Gustavo Sousa
2026-10-06 14:44 ` Kamil Konieczny
2026-10-06 16:05 ` Gustavo Sousa [this message]
2026-10-06 15:39 ` ✓ i915.CI.BAT: success for " Patchwork
2026-10-06 15:48 ` ✓ Xe.CI.BAT: " Patchwork
2026-10-06 20:57 ` ✗ i915.CI.Full: failure " Patchwork
2026-10-06 21:44 ` ✗ Xe.CI.FULL: " Patchwork
2026-10-07 9:39 ` [PATCH i-g-t v2] " Bernatowicz, Marcin
2026-10-07 12:46 ` Gustavo Sousa
2026-10-07 14:50 ` [PATCH i-g-t] tests/intel/xe_pmu: Provision VFs inside dynamic subtests Marcin Bernatowicz
2026-10-08 10:03 ` Kamil Konieczny
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=87qzi2ddb7.fsf@intel.com \
--to=gustavo.sousa@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=kamil.konieczny@linux.intel.com \
--cc=marcin.bernatowicz@linux.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