Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Gustavo Sousa <gustavo.sousa@intel.com>
To: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>,
	<igt-dev@lists.freedesktop.org>
Cc: Kamil Konieczny <kamil.konieczny@linux.intel.com>,
	Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
Subject: Re: [PATCH v2 i-g-t 1/1] tests/intel/xe_pmu: Provision VFs inside dynamic subtests
Date: Tue, 13 Oct 2026 15:30:47 -0300	[thread overview]
Message-ID: <87jynla1wo.fsf@intel.com> (raw)
In-Reply-To: <5553bf027e3bdb9253a18fce5477a42ef10f0b25.1791900914.git.marcin.bernatowicz@linux.intel.com>

Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com> writes:

> VF setup helpers can call igt_assert(), which is not allowed in the outer
> igt_subtest_with_dynamic() scope. Set up VFs in the first dynamic case,
> reuse them across engine cases, and clean up after the group. This keeps
> VF setup out of unrelated subtests.
>
> Assisted-by: Copilot:Claude-Opus-5
> Signed-off-by: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
> Cc: Gustavo Sousa <gustavo.sousa@intel.com>
> Cc: Kamil Konieczny <kamil.konieczny@linux.intel.com>
> Reviewed-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>
> ---
>  tests/intel/xe_pmu.c | 48 +++++++++++++++++++++++++++++++++-----------
>  1 file changed, 36 insertions(+), 12 deletions(-)
>
> diff --git a/tests/intel/xe_pmu.c b/tests/intel/xe_pmu.c
> index d41ebba1f..159a7f8a2 100644
> --- a/tests/intel/xe_pmu.c
> +++ b/tests/intel/xe_pmu.c
> @@ -117,6 +117,8 @@ static char xe_device[NAME_MAX];
>  static bool autoprobe;
>  static int total_exec_quantum;
>  static bool has_engine_active_ticks;
> +static unsigned int num_functions;

I feel like num_functions as a name for a global variable is a bit too
generic for what its purpose is.

Maybe sriov_num_fns (or the longer sriov_num_functions).

And then, for consistency, sriov_vfs_setup_started?

Anyway, the patch looks good to me. Thanks for this!

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>

> +static bool vfs_setup_started;
>  
>  #define test_each_engine(test, fd, hwe) \
>  	igt_subtest_with_dynamic(test) \
> @@ -999,6 +1001,7 @@ static unsigned int enable_and_provision_vfs(int fd)
>  	num_vfs = total_vfs > 2 ? 2 : total_vfs;
>  
>  	/* Enable VF's */
> +	vfs_setup_started = true;
>  	igt_sriov_disable_driver_autoprobe(fd);
>  	igt_sriov_enable_vfs(fd, num_vfs);
>  	igt_require(igt_sriov_get_enabled_vfs(fd) == num_vfs);
> @@ -1021,6 +1024,16 @@ static unsigned int enable_and_provision_vfs(int fd)
>  	return num_vfs;
>  }
>  
> +static void ensure_vfs_enabled(int fd)
> +{
> +	if (num_functions)
> +		return;
> +
> +	igt_require_f(!vfs_setup_started,
> +		      "VF setup failed in an earlier dynamic subtest; see its log\n");
> +	num_functions = enable_and_provision_vfs(fd) + 1;
> +}
> +
>  static void unprovision_and_disable_vfs(int fd)
>  {
>  	int ret;
> @@ -1039,6 +1052,16 @@ static void unprovision_and_disable_vfs(int fd)
>  	igt_sriov_clear_exit_handler();
>  }
>  
> +static void cleanup_vfs_setup(int fd)
> +{
> +	if (!vfs_setup_started)
> +		return;
> +
> +	unprovision_and_disable_vfs(fd);
> +	vfs_setup_started = false;
> +	num_functions = 0;
> +}
> +
>  static void stash_gt_freq(int fd, uint32_t **stash_min, uint32_t **stash_max)
>  {
>  	int max_gt, gt;
> @@ -1190,28 +1213,29 @@ 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)
> -			engine_activity_all_fn(fd, eci, num_fns);
> +		test_each_engine("all-fn-engine-activity-load", fd, eci) {
> +			ensure_vfs_enabled(fd);
> +			engine_activity_all_fn(fd, eci, num_functions);
> +		}
>  
>  		igt_describe("Validate per-function engine activity");
> -		test_each_engine("fn-engine-activity-load", fd, eci)
> -			for (int fn = 0; fn < num_fns; fn++)
> -				engine_activity_fn(fd, eci, fn, XE_SRIOV_SCHED_PRIORITY_LOW);
> +		test_each_engine("fn-engine-activity-load", fd, eci) {
> +			ensure_vfs_enabled(fd);
> +			for (int fn = 0; fn < num_functions; 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) {
> -			for (int fn = 0; fn < num_fns; fn++)
> +			ensure_vfs_enabled(fd);
> +			for (int fn = 0; fn < num_functions; fn++)
>  				engine_activity_fn(fd, eci, fn, XE_SRIOV_SCHED_PRIORITY_NORMAL);
>  		}
>  
>  		igt_fixture()
> -			unprovision_and_disable_vfs(fd);
> +			cleanup_vfs_setup(fd);
>  	}
>  
>  	igt_subtest_group() {
> -- 
> 2.43.0

  reply	other threads:[~2026-10-13 18:32 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-13 14:36 [PATCH v2 i-g-t 0/1] xe_pmu: Do not enable VFs in unrelated subtests Marcin Bernatowicz
2026-10-13 14:36 ` [PATCH v2 i-g-t 1/1] tests/intel/xe_pmu: Provision VFs inside dynamic subtests Marcin Bernatowicz
2026-10-13 18:30   ` Gustavo Sousa [this message]
2026-10-13 17:46 ` ✓ i915.CI.BAT: success for xe_pmu: Do not enable VFs in unrelated subtests Patchwork
2026-10-13 22:04 ` ✓ Xe.CI.BAT: " Patchwork
2026-10-13 23:17 ` ✗ i915.CI.Full: failure " 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=87jynla1wo.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