From: "Srivatsa, Anusha" <anusha.srivatsa@intel.com>
To: "Roper, Matthew D" <matthew.d.roper@intel.com>
Cc: "intel-gfx@lists.freedesktop.org" <intel-gfx@lists.freedesktop.org>
Subject: Re: [Intel-gfx] [PATCH] drm/i915/tgl: Implement Wa_1606931601
Date: Fri, 31 Jan 2020 00:16:23 +0000 [thread overview]
Message-ID: <83F5C7385F545743AD4FB2A62F75B0734822AB42@ORSMSX108.amr.corp.intel.com> (raw)
In-Reply-To: <20200130204250.GC116423@mdroper-desk1.amr.corp.intel.com>
> -----Original Message-----
> From: Roper, Matthew D <matthew.d.roper@intel.com>
> Sent: Thursday, January 30, 2020 12:43 PM
> To: Srivatsa, Anusha <anusha.srivatsa@intel.com>
> Cc: intel-gfx@lists.freedesktop.org; Ceraolo Spurio, Daniele
> <daniele.ceraolospurio@intel.com>
> Subject: Re: [PATCH] drm/i915/tgl: Implement Wa_1606931601
>
> On Wed, Jan 29, 2020 at 02:42:06PM -0800, Anusha Srivatsa wrote:
> > Disable Inter and intra Read Suppression (bit 15) and Early Read and
> > Src Swap (bit 14) by setting the chicken register.
> >
> > BSpec: 46045,52890
> >
> > v2: Follow the Bspec implementation for the WA.
> > v3: Have 2 separate defines for bit 14 and 15.
> > - Rename register definitions with TGL_ prefix
>
> The hardware guys changed their mind again and we're back to only needing
> bit 14 now. They updated the bspec and the underlying database yet again.
> :-/
☹
> >
> > Cc: Matt Roper <matthew.d.roper@intel.com>
> > Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com>
> > ---
> > drivers/gpu/drm/i915/gt/intel_workarounds.c | 6 ++++++
> > drivers/gpu/drm/i915/i915_reg.h | 2 ++
> > 2 files changed, 8 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c
> > b/drivers/gpu/drm/i915/gt/intel_workarounds.c
> > index 5a7db279f702..1f84cd595f88 100644
> > --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
> > +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
> > @@ -593,6 +593,12 @@ static void tgl_ctx_workarounds_init(struct
> intel_engine_cs *engine,
> > wa_add(wal, FF_MODE2, FF_MODE2_TDS_TIMER_MASK, val,
> > IS_TGL_REVID(engine->i915, TGL_REVID_A0, TGL_REVID_A0) ? 0 :
> > FF_MODE2_TDS_TIMER_MASK);
> > +
> > + /* Wa_1606931601:tgl */
> > + WA_SET_BIT_MASKED(GEN7_ROW_CHICKEN2,
> > + GEN12_EARLY_READ_SRC0_DISABLE |
> > +
> GEN12_INTER_INTRA_READ_SUPPRESSION_DISABLE);
>
> I think Daniele already mentioned this on the other discussion, but
> ROW_CHICKEN2 isn't part of the context image on gen12 (see bspec page
> 46255). This is a change from ICL where it *was* part of the context (see
> bspec page 18907), so even though we handled this register in the
> ctx_workarounds_init for ICL, that's not the appropriate place to put it for
> TGL.
Agreed.
> Since this isn't a context workaround, we need to determine whether it's a
> general GT workaround (which would be initialized in
> tgl_gt_workarounds_init) or an engine workaround (which would be
> initialized in {rcs,xcs}_engine_wa_init. In this case the register we're
> modifying is 0xe49c; according to bspec page 52078 this falls in one of the
> forcewake ranges listed under the render engine column (0E000-0E8FF).
The register we are setting is 0xe4f4. This comes under Render engine. Adding this change to rcs_engine_wa_init().
> So I
> believe in this case that means we want to update
> rcs_engine_wa_init() with this workaround --- that ensures that the
> workaround will be re-applied any time the engine is reset (even if it's not a
> full-GPU reset).
Yes. Adding this to rcs_engine_wa_init().
Anusha
>
> Matt
>
> > +
> > }
> >
> > static void
> > diff --git a/drivers/gpu/drm/i915/i915_reg.h
> > b/drivers/gpu/drm/i915/i915_reg.h index 4c72b8ac0f2e..70ead809c706
> > 100644
> > --- a/drivers/gpu/drm/i915/i915_reg.h
> > +++ b/drivers/gpu/drm/i915/i915_reg.h
> > @@ -9149,6 +9149,8 @@ enum {
> > #define DOP_CLOCK_GATING_DISABLE (1 << 0)
> > #define PUSH_CONSTANT_DEREF_DISABLE (1 << 8)
> > #define GEN11_TDL_CLOCK_GATING_FIX_DISABLE (1 << 1)
> > +#define GEN12_EARLY_READ_SRC0_DISABLE (1 << 14)
> > +#define GEN12_INTER_INTRA_READ_SUPPRESSION_DISABLE (1 <<
> 15)
> >
> > #define HSW_ROW_CHICKEN3 _MMIO(0xe49c)
> > #define HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE (1 << 6)
> > --
> > 2.25.0
> >
>
> --
> Matt Roper
> Graphics Software Engineer
> VTT-OSGC Platform Enablement
> Intel Corporation
> (916) 356-2795
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2020-01-31 0:16 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-29 22:42 [Intel-gfx] [PATCH] drm/i915/tgl: Implement Wa_1606931601 Anusha Srivatsa
2020-01-30 3:59 ` [Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915/tgl: Implement Wa_1606931601 (rev3) Patchwork
2020-01-30 20:42 ` [Intel-gfx] [PATCH] drm/i915/tgl: Implement Wa_1606931601 Matt Roper
2020-01-31 0:16 ` Srivatsa, Anusha [this message]
-- strict thread matches above, loose matches on Subject: below --
2020-02-11 23:44 Anusha Srivatsa
2020-01-31 23:17 Anusha Srivatsa
2020-02-08 7:00 ` Matt Atwood
2020-01-24 21:29 Anusha Srivatsa
2020-01-22 23:40 Anusha Srivatsa
2020-01-23 17:50 ` Matt Roper
2020-01-24 19:36 ` Srivatsa, Anusha
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=83F5C7385F545743AD4FB2A62F75B0734822AB42@ORSMSX108.amr.corp.intel.com \
--to=anusha.srivatsa@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=matthew.d.roper@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;
as well as URLs for NNTP newsgroup(s).