From: "Belgaumkar, Vinay" <vinay.belgaumkar@intel.com>
To: Ashutosh Dixit <ashutosh.dixit@intel.com>,
<igt-dev@lists.freedesktop.org>
Subject: Re: [igt-dev] [PATCH i-g-t 3/4] tests/gem_ctx_freq: Disable SLPC efficient freq for the test
Date: Wed, 7 Jun 2023 23:01:57 -0700 [thread overview]
Message-ID: <b52f62a7-2ced-f835-9609-b7d51b30a491@intel.com> (raw)
In-Reply-To: <20230606012150.2119065-4-ashutosh.dixit@intel.com>
On 6/5/2023 6:21 PM, Ashutosh Dixit wrote:
> When SLPC efficient freq is enabled, GPU frequencies can be outside the
> min/max range set from sysfs. Any test which compares actual GPU frequency
> against set frequencies is therefore meaninful only when SLPC efficient
> freq is disabled.
>
> Bug: https://gitlab.freedesktop.org/drm/intel/-/issues/6786
> Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
> ---
> tests/i915/gem_ctx_freq.c | 15 ++++++++++++++-
> 1 file changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/tests/i915/gem_ctx_freq.c b/tests/i915/gem_ctx_freq.c
> index b2f6c2af55e1..6dfcf4187d4b 100644
> --- a/tests/i915/gem_ctx_freq.c
> +++ b/tests/i915/gem_ctx_freq.c
> @@ -131,6 +131,15 @@ static bool get_sysfs_freq(int dirfd, uint32_t *min, uint32_t *max)
> igt_sysfs_rps_scanf(dirfd, RPS_MAX_FREQ_MHZ, "%u", max) == 1);
> }
>
> +static void slpc_ignore_efficient_freq(int dirfd, bool val)
> +{
> + if (!(gem_using_guc_submission(i915) && i915_is_slpc_enabled(i915)))
> + return;
There was a reason why I had both, not remembering now.
> +
> + igt_require(igt_sysfs_has_attr(dirfd, "slpc_ignore_eff_freq"));
> + igt_assert(igt_sysfs_set_u32(dirfd, "slpc_ignore_eff_freq", val));
> +}
Can we add this as a helper in lib/igt_pm.c? I suspect there might be
other present/future tests that may need to use this.
Thanks,
Vinay.
> +
> static void sysfs_range(int dirfd, int gt)
> {
> #define N_STEPS 10
> @@ -147,9 +156,11 @@ static void sysfs_range(int dirfd, int gt)
> * The sysfs interface sets the global limits and overrides the
> * user's request. So we can to check that if the user requests
> * a range outside of the sysfs, the requests are only run at the
> - * constriained sysfs range.
> + * constrained sysfs range. With GuC SLPC this requires disabling
> + * efficient freq.
> */
>
> + slpc_ignore_efficient_freq(dirfd, true);
> igt_require(get_sysfs_freq(dirfd, &sys_min, &sys_max));
> igt_info("System min freq: %dMHz; max freq: %dMHz\n", sys_min, sys_max);
>
> @@ -199,6 +210,8 @@ static void __restore_sysfs_freq(int dirfd)
> {
> char buf[256];
>
> + slpc_ignore_efficient_freq(dirfd, false);
> +
> if (igt_sysfs_read(sysfs, "gt_RPn_freq_mhz", buf, sizeof(buf)) > 0)
> igt_sysfs_rps_set(dirfd, RPS_MIN_FREQ_MHZ, buf);
>
next prev parent reply other threads:[~2023-06-08 6:02 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-06-06 1:21 [igt-dev] [PATCH i-g-t 0/4] gem_ctx_freq: multi-gt support and disable efficient freq Ashutosh Dixit
2023-06-06 1:21 ` [igt-dev] [PATCH i-g-t 1/4] tests/gem_ctx_freq: Add support for multi-gts Ashutosh Dixit
2023-06-07 23:47 ` Dixit, Ashutosh
2023-06-06 1:21 ` [igt-dev] [PATCH i-g-t 2/4] lib/igt_sysfs: Don't assert in igt_sysfs_dir_id_to_name Ashutosh Dixit
2023-06-06 1:27 ` Dixit, Ashutosh
2023-06-07 20:53 ` Umesh Nerlige Ramappa
2023-06-06 1:21 ` [igt-dev] [PATCH i-g-t 3/4] tests/gem_ctx_freq: Disable SLPC efficient freq for the test Ashutosh Dixit
2023-06-07 21:35 ` Umesh Nerlige Ramappa
2023-06-08 2:28 ` Dixit, Ashutosh
2023-06-08 6:01 ` Belgaumkar, Vinay [this message]
2023-06-08 15:50 ` Dixit, Ashutosh
2023-06-06 1:21 ` [igt-dev] [PATCH i-g-t 4/4] HAX: Add gem_ctx_freq to BAT testlist Ashutosh Dixit
2023-06-06 2:27 ` [igt-dev] ✓ Fi.CI.BAT: success for gem_ctx_freq: multi-gt support and disable efficient freq Patchwork
2023-06-06 23:36 ` [igt-dev] ✓ Fi.CI.IGT: " 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=b52f62a7-2ced-f835-9609-b7d51b30a491@intel.com \
--to=vinay.belgaumkar@intel.com \
--cc=ashutosh.dixit@intel.com \
--cc=igt-dev@lists.freedesktop.org \
/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