public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
To: John.C.Harrison@Intel.com
Cc: IGT-Dev@Lists.FreeDesktop.Org, Intel-GFX@Lists.FreeDesktop.Org
Subject: Re: [Intel-gfx] [igt-dev] [PATCH i-g-t] tests/sysfs: Update timeslice/preemption for new range limits
Date: Tue, 01 Nov 2022 08:27:11 -0700	[thread overview]
Message-ID: <87tu3id7ao.wl-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <20221031222440.546-1-John.C.Harrison@Intel.com>

On Mon, 31 Oct 2022 15:24:40 -0700, John.C.Harrison@Intel.com wrote:
>
> From: John Harrison <John.C.Harrison@Intel.com>
>
> Guc submission imposes new range limits on certain scheduling
> parameters. The idempotent sections of the timeslice duration and
> pre-emption timeout tests was exceeding those limits and so would fail.
>
> Reduce the excessively large value (654s) to one which does not
> overflow (54s). Also add an assert that the write of the new value
> actually succeeds.
>
> Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
> ---
>  tests/i915/sysfs_preempt_timeout.c    | 4 ++--
>  tests/i915/sysfs_timeslice_duration.c | 4 ++--
>  2 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/tests/i915/sysfs_preempt_timeout.c b/tests/i915/sysfs_preempt_timeout.c
> index 515038281638..5e0a7d96299f 100644
> --- a/tests/i915/sysfs_preempt_timeout.c
> +++ b/tests/i915/sysfs_preempt_timeout.c
> @@ -68,7 +68,7 @@ static void set_preempt_timeout(int engine, unsigned int value)
>  {
>	unsigned int delay;
>
> -	igt_sysfs_printf(engine, ATTR, "%u", value);
> +	igt_assert_lte(0, igt_sysfs_printf(engine, ATTR, "%u", value));
>	igt_sysfs_scanf(engine, ATTR, "%u", &delay);
>	igt_assert_eq(delay, value);
>  }
> @@ -82,7 +82,7 @@ static int wait_for_reset(int fence)
>
>  static void test_idempotent(int i915, int engine)
>  {
> -	unsigned int delays[] = { 0, 1, 1000, 1234, 654321 };
> +	unsigned int delays[] = { 0, 1, 1000, 1234, 54321 };

By way of documenting the difference between GuC and execlists, I think we
should use gem_using_guc_submission and define two different arrays, one
for execlists and the other for GuC?

We could of course go the extra yard and check for errors with excessively
large values but I'm not sure if that's worth it so am ok if we skip that
at this point. Or that's a later patch.

Below too.

Thanks.
--
Ashutosh


>	unsigned int saved;
>
>	/* Quick test that store/show reports the same values */
> diff --git a/tests/i915/sysfs_timeslice_duration.c b/tests/i915/sysfs_timeslice_duration.c
> index 8a2f1c2f2ece..95dc377785a5 100644
> --- a/tests/i915/sysfs_timeslice_duration.c
> +++ b/tests/i915/sysfs_timeslice_duration.c
> @@ -79,7 +79,7 @@ static void set_timeslice(int engine, unsigned int value)
>  {
>	unsigned int delay;
>
> -	igt_sysfs_printf(engine, ATTR, "%u", value);
> +	igt_assert_lte(0, igt_sysfs_printf(engine, ATTR, "%u", value));
>	igt_sysfs_scanf(engine, ATTR, "%u", &delay);
>	igt_assert_eq(delay, value);
>  }
> @@ -93,7 +93,7 @@ static int wait_for_reset(int fence)
>
>  static void test_idempotent(int i915, int engine)
>  {
> -	const unsigned int delays[] = { 0, 1, 1234, 654321 };
> +	const unsigned int delays[] = { 0, 1, 1234, 54321 };
>	unsigned int saved;
>
>	/* Quick test to verify the kernel reports the same values as we write */
> --
> 2.37.3
>

  reply	other threads:[~2022-11-01 15:27 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-31 22:24 [Intel-gfx] [PATCH i-g-t] tests/sysfs: Update timeslice/preemption for new range limits John.C.Harrison
2022-11-01 15:27 ` Dixit, Ashutosh [this message]
2022-11-01 16:22   ` [Intel-gfx] [igt-dev] " John Harrison
2022-11-01 16:25     ` Dixit, Ashutosh

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=87tu3id7ao.wl-ashutosh.dixit@intel.com \
    --to=ashutosh.dixit@intel.com \
    --cc=IGT-Dev@Lists.FreeDesktop.Org \
    --cc=Intel-GFX@Lists.FreeDesktop.Org \
    --cc=John.C.Harrison@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