Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
To: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [Intel-gfx] [PATCH v7 11/11] drm/i915/perf: Wa_14017512683: Disable OAM if media C6 is enabled in BIOS
Date: Sun, 19 Mar 2023 20:56:00 -0700	[thread overview]
Message-ID: <87v8iwt6sv.wl-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <20230317231641.2815418-12-umesh.nerlige.ramappa@intel.com>

On Fri, 17 Mar 2023 16:16:41 -0700, Umesh Nerlige Ramappa wrote:
>

Hi Umesh,

Please read from bottom to top. Mostly nits but let's see what you think.

> OAM does not work with media C6 enabled on some steppings of MTL.
> Disable OAM if we detect that media C6 was enabled in bios.
>
> v2: (Ashutosh)
> - Remove drm_notice from the driver load path
> - Log a drm_err when opening an OAM stream on affected steppings
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
> ---
>  drivers/gpu/drm/i915/i915_perf.c | 41 ++++++++++++++++++++++++++++++++
>  1 file changed, 41 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
> index 18afa76653b7..823379d63caf 100644
> --- a/drivers/gpu/drm/i915/i915_perf.c
> +++ b/drivers/gpu/drm/i915/i915_perf.c
> @@ -209,6 +209,7 @@
>  #include "gt/intel_gt_regs.h"
>  #include "gt/intel_lrc.h"
>  #include "gt/intel_lrc_reg.h"
> +#include "gt/intel_rc6.h"
>  #include "gt/intel_ring.h"
>  #include "gt/uc/intel_guc_slpc.h"
>
> @@ -4216,6 +4217,20 @@ static int read_properties_unlocked(struct i915_perf *perf,
>		return -EINVAL;
>	}
>
> +	/*
> +	 * Wa_14017512683: mtl[a0..c0): Use of OAM must be preceded with Media
> +	 * C6 disable in BIOS. Fail if Media C6 is enabled on steppings where OAM
> +	 * does not work as expected.
> +	 */
> +	if (IS_MTL_MEDIA_STEP(props->engine->i915, STEP_A0, STEP_C0) &&
> +	    props->engine->gt->type == GT_MEDIA &&

Instead of gt type I think it's better to check perf_group->type. That is
why as I said below maybe better to have a valid perf_group even in this
case.

> +	    intel_check_bios_c6_setup(&props->engine->gt->rc6)) {
> +		drm_dbg(&perf->i915->drm,
> +			"OAM requires media C6 to be disabled in BIOS\n");
> +		return -EINVAL;
> +	}

So now we can change this check to something like:

	if (engine->perf_group->type == OAM && i915_perf->mtl_bios_mc6_enabled)

> +
> +
>	if (!engine_supports_oa(props->engine)) {
>		drm_dbg(&perf->i915->drm,
>			"Engine not supported by OA %d:%d\n",
> @@ -4897,6 +4912,15 @@ static u32 num_perf_groups_per_gt(struct intel_gt *gt)
>
>  static u32 __oam_engine_group(struct intel_engine_cs *engine)
>  {
> +	/*
> +	 * Wa_14017512683: mtl[a0..c0): Use of OAM must be preceded with Media
> +	 * C6 disable in BIOS. To disable use of OAM with media engines, set the
> +	 * oa_group to PERF_GROUP_INVALID.
> +	 */
> +	if (IS_MTL_MEDIA_STEP(engine->i915, STEP_A0, STEP_C0) &&
> +	    intel_check_bios_c6_setup(&engine->gt->rc6))
> +		return PERF_GROUP_INVALID;

I think we should just remove this. Let the perf group be valid in this
case too since we cannot support OA for a different reason. Then we can use
the OAM perf group above.

Though if we drop this we have only 2 instances of the checks instead of 3
so maybe ok to not have i915_perf->mtl_bios_mc6_enabled?

> +
>	if (GRAPHICS_VER_FULL(engine->i915) >= IP_VER(12, 70)) {
>		/*
>		 * There's 1 SAMEDIA gt and 1 OAM per SAMEDIA gt. All media slices
> @@ -5316,6 +5340,23 @@ int i915_perf_ioctl_version(struct drm_i915_private *i915)
>	 *
>	 * 7: Add support for video decode and enhancement classes.
>	 */
> +
> +	/*
> +	 * Wa_14017512683: mtl[a0..c0): Use of OAM must be preceded with Media
> +	 * C6 disable in BIOS. If Media C6 is enabled in BIOS, return version 6
> +	 * to indicate that OA media is not supported.
> +	 */
> +	if (IS_MTL_MEDIA_STEP(i915, STEP_A0, STEP_C0)) {
> +		struct intel_gt *gt;
> +		int i;
> +
> +		for_each_gt(gt, i915, i) {
> +			if (gt->type == GT_MEDIA &&
> +			    intel_check_bios_c6_setup(&gt->rc6))

Rather than repeat this comment and these checks (MTL stepping and
intel_check_bios_c6_setup) 3 times, I think it will be simpler to set a
variable (say i915_perf->mtl_bios_mc6_enabled) in i915_perf_init() and
reuse that in the 3 places.


> +				return 6;
> +		}
> +	}
> +
>	return 7;
>  }

Thanks.
--
Ashutosh

  reply	other threads:[~2023-03-20  4:01 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-17 23:16 [Intel-gfx] [PATCH v7 00/11] Add OAM support for MTL Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 01/11] drm/i915/perf: Drop wakeref on GuC RC error Umesh Nerlige Ramappa
2023-03-20 10:16   ` Jani Nikula
2023-03-20 19:24     ` Umesh Nerlige Ramappa
2023-03-21  9:59       ` Jani Nikula
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 02/11] drm/i915/mtl: Synchronize i915/BIOS on C6 enabling Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 03/11] drm/i915/perf: Validate OA sseu config outside switch Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 04/11] drm/i915/perf: Group engines into respective OA groups Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 05/11] drm/i915/perf: Fail modprobe if i915_perf_init fails on OOM Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 06/11] drm/i915/perf: Parse 64bit report header formats correctly Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 07/11] drm/i915/perf: Handle non-power-of-2 reports Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 08/11] drm/i915/perf: Add engine class instance parameters to perf Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 09/11] drm/i915/perf: Add support for OA media units Umesh Nerlige Ramappa
2023-03-20  2:06   ` Dixit, Ashutosh
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 10/11] drm/i915/perf: Pass i915 object to perf revision helper Umesh Nerlige Ramappa
2023-03-17 23:16 ` [Intel-gfx] [PATCH v7 11/11] drm/i915/perf: Wa_14017512683: Disable OAM if media C6 is enabled in BIOS Umesh Nerlige Ramappa
2023-03-20  3:56   ` Dixit, Ashutosh [this message]
2023-03-21  0:27     ` Umesh Nerlige Ramappa
2023-03-21  0:30       ` Dixit, Ashutosh
2023-03-18  0:10 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for Add OAM support for MTL Patchwork
2023-03-18  0:10 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2023-03-18  0:10 ` [Intel-gfx] ✗ Fi.CI.DOCS: " Patchwork
2023-03-18  0:27 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2023-03-18  1:35 ` [Intel-gfx] ✓ 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=87v8iwt6sv.wl-ashutosh.dixit@intel.com \
    --to=ashutosh.dixit@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=umesh.nerlige.ramappa@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