From: Mika Kuoppala <mika.kuoppala@linux.intel.com>
To: intel-gfx@lists.freedesktop.org
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Subject: Re: [PATCH] drm/i915/bdw: WaProgramL3SqcReg1Default
Date: Wed, 01 Oct 2014 16:46:50 +0300 [thread overview]
Message-ID: <874mvnucl1.fsf@gaia.fi.intel.com> (raw)
In-Reply-To: <1412089877-3998-1-git-send-email-rodrigo.vivi@intel.com>
Rodrigo Vivi <rodrigo.vivi@intel.com> writes:
> Program the default initial value of the L3SqcReg1 on BDW for performance
>
> v2: Default confirmed and using intel_ring_emit_wa as Mika pointed out.
>
> Cc: Mika Kuoppala <mika.kuoppala@intel.com>
> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> ---
> drivers/gpu/drm/i915/i915_reg.h | 3 +++
> drivers/gpu/drm/i915/intel_ringbuffer.c | 5 ++++-
> 2 files changed, 7 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
> index 36a847a..33143cc 100644
> --- a/drivers/gpu/drm/i915/i915_reg.h
> +++ b/drivers/gpu/drm/i915/i915_reg.h
> @@ -4972,6 +4972,9 @@ enum punit_power_well {
> #define GEN7_L3SQCREG1 0xB010
> #define VLV_B0_WA_L3SQCREG1_VALUE 0x00D30000
>
> +#define GEN8_L3SQCREG1 0xB100
> +#define BDW_WA_L3SQCREG1_DEFAULT 0x00610000
> +
This is the default value after reset. I have experimented with other
values and nothing improves above noise level. Further, my suggestion to
using 0x00810000 will cause a gpu hang...so there is dragons here.
As you are writing the same default value that is already in register,
and there is no indication that we should refresh anything, I would
say this patch is not needed until someone comes along and shows
something better than what the default is.
Thanks,
-Mika
> #define GEN7_L3CNTLREG1 0xB01C
> #define GEN7_WA_FOR_GEN7_L3_CONTROL 0x3C47FF8C
> #define GEN7_L3AGDIS (1<<19)
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index 816a692..a37675d 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -707,7 +707,7 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
> * update the number of dwords required based on the
> * actual number of workarounds applied
> */
> - ret = intel_ring_begin(ring, 18);
> + ret = intel_ring_begin(ring, 21);
> if (ret)
> return ret;
>
> @@ -751,6 +751,9 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
> intel_ring_emit_wa(ring, GEN7_GT_MODE,
> GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4);
>
> + /* WaProgramL3SqcReg1Default:bdw */
> + intel_ring_emit_wa(ring, GEN8_L3SQCREG1, BDW_WA_L3SQCREG1_DEFAULT);
> +
> intel_ring_advance(ring);
>
> DRM_DEBUG_DRIVER("Number of Workarounds applied: %d\n",
> --
> 1.9.3
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2014-10-01 13:47 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-20 0:16 [PATCH 1/5] drm/i915: Add IS_BDW_GT3 macro Rodrigo Vivi
2014-09-20 0:16 ` [PATCH 2/5] drm/i915/bdw: WaDisableFenceDestinationToSLM Rodrigo Vivi
2014-09-25 12:37 ` Mika Kuoppala
2014-09-29 12:32 ` Daniel Vetter
2014-09-29 12:49 ` Daniel Vetter
2014-09-30 15:02 ` Mika Kuoppala
2014-09-30 16:27 ` Daniel Vetter
2014-09-20 0:16 ` [PATCH 3/5] drm/i915/bdw: WaProgramL3SqcReg1Default Rodrigo Vivi
2014-09-26 12:03 ` Mika Kuoppala
2014-09-26 18:32 ` Rodrigo Vivi
2014-09-26 19:06 ` Ville Syrjälä
2014-09-30 15:11 ` [PATCH] " Rodrigo Vivi
2014-10-01 13:46 ` Mika Kuoppala [this message]
2014-09-20 0:16 ` [PATCH 4/5] drm/i915: Let number of workarounds more clear Rodrigo Vivi
2014-09-24 22:44 ` [PATCH] " Rodrigo Vivi
2014-09-26 12:22 ` [PATCH 4/5] " Mika Kuoppala
2014-09-26 12:56 ` Jani Nikula
2014-09-26 12:58 ` Chris Wilson
2014-09-26 13:02 ` Damien Lespiau
2014-09-26 14:16 ` Mika Kuoppala
2014-09-26 14:56 ` Chris Wilson
2014-09-30 22:13 ` Rodrigo Vivi
2014-09-20 0:16 ` [PATCH 5/5] drm/i915/bdw: Remove BDW preproduction W/As until C stepping Rodrigo Vivi
2014-09-24 22:46 ` [PATCH] " Rodrigo Vivi
2014-09-30 22:14 ` Rodrigo Vivi
2014-10-01 14:11 ` Mika Kuoppala
2014-10-09 14:11 ` Rodrigo Vivi
2014-10-19 12:14 ` Daniel Vetter
2014-09-29 19:58 ` [PATCH 1/5] drm/i915: Add IS_BDW_GT3 macro Jesse Barnes
2015-03-31 23:03 [PATCH] drm/i915/bdw: WaProgramL3SqcReg1Default Rodrigo Vivi
2015-04-01 5:52 ` shuang.he
2015-04-01 7:49 ` Jani Nikula
2015-04-01 8:31 ` Ville Syrjälä
2015-04-01 14:49 ` Vivi, Rodrigo
2015-04-01 15:41 ` Ville Syrjälä
2015-04-07 8:26 ` Daniel Vetter
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=874mvnucl1.fsf@gaia.fi.intel.com \
--to=mika.kuoppala@linux.intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=rodrigo.vivi@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.