All of lore.kernel.org
 help / color / mirror / Atom feed
From: Imre Deak <imre.deak@intel.com>
To: intel-gfx@lists.freedesktop.org
Cc: Eero Tamminen <eero.t.tamminen@intel.com>
Subject: [PATCH v3 2/3] drm/i915/bxt: Fix inadvertent CPU snooping due to incorrect MOCS config
Date: Fri,  1 Jul 2016 16:40:05 +0300	[thread overview]
Message-ID: <1467380406-11954-3-git-send-email-imre.deak@intel.com> (raw)
In-Reply-To: <1467380406-11954-1-git-send-email-imre.deak@intel.com>

Setting a write-back cache policy in the MOCS entry definition also
implies snooping, which has a considerable overhead. This is
unexpected for a few reasons:
- From user-space's point of view since it didn't want a coherent
  surface (it didn't set the buffer as such via the set caching IOCTL).
- There is a separate MOCS entry field for snooping (which we never
  set).
- This MOCS table is about caching in (e)LLC and there is no (e)LLC on
  BXT. There is a separate table for L3 cache control.

Considering the above the current behavior of snooping looks like an
unintentional side-effect of the WB setting. Changing it to be LLC-UC
gets rid of the snooping without any ill-effects. For a coherent
surface the application would use a separate MOCS entry at index 1 and
call the set caching IOCTL to setup the PTE entries for the
corresponding buffer to be snooped. In the future we could also add a
new MOCS entry for coherent surfaces.

This resulted in 70% improvement in synthetic texturing benchmarks.

Kudos to Valtteri Rantala, Eero Tamminen and Michael T Frederick and
Ville who helped to narrow the source of problem to the kernel and to
the snooping behaviour in particular.

With a follow-up change to adjust the 3rd entry value
igt/gem_mocs_settings is passing after this change.

v2:
- Rebase on v2 of patch 1/2.
v3:
- Set the entry as LLC uncached instead of PTE-passthrough. This way
  we also keep snooping disabled, but we also make the cacheability/
  coherency setting indepent of the PTE which is managed by the
  kernel. (Chris)

CC: Rong R Yang <rong.r.yang@intel.com>
CC: Yakui Zhao <yakui.zhao@intel.com>
CC: Valtteri Rantala <valtteri.rantala@intel.com>
CC: Eero Tamminen <eero.t.tamminen@intel.com>
CC: Michael T Frederick <michael.t.frederick@intel.com>
CC: Ville Syrjälä <ville.syrjala@linux.intel.com>
CC: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/intel_mocs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_mocs.c b/drivers/gpu/drm/i915/intel_mocs.c
index d36e609..927825f 100644
--- a/drivers/gpu/drm/i915/intel_mocs.c
+++ b/drivers/gpu/drm/i915/intel_mocs.c
@@ -149,8 +149,8 @@ static const struct drm_i915_mocs_entry broxton_mocs_table[] = {
 	  .l3cc_value =    L3_ESC(0) | L3_SCC(0) | L3_CACHEABILITY(L3_WB),
 	},
 	{
-	  /* 0x0000003b */
-	  .control_value = LE_CACHEABILITY(LE_WB) |
+	  /* 0x00000039 */
+	  .control_value = LE_CACHEABILITY(LE_UC) |
 			   LE_TGT_CACHE(LE_TC_LLC_ELLC) |
 			   LE_LRUM(3) | LE_AOM(0) | LE_RSC(0) | LE_SCC(0) |
 			   LE_PFM(0) | LE_SCF(0),
-- 
2.5.0

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2016-07-01 13:40 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-07-01 13:40 [PATCH v3 0/3] drm/i915/bxt: Fix performance due to bogus MOCS entry Imre Deak
2016-07-01 13:40 ` [PATCH v3 1/3] drm/i915/gen9: Clean up MOCS table definitions Imre Deak
2016-07-01 21:23   ` Bish, Jim
2016-07-01 21:47   ` Francisco Jerez
2016-07-12 11:08     ` Imre Deak
2016-07-13  2:10   ` Zhao Yakui
2016-07-01 13:40 ` Imre Deak [this message]
2016-07-13  2:32   ` [PATCH v3 2/3] drm/i915/bxt: Fix inadvertent CPU snooping due to incorrect MOCS config Zhao Yakui
2016-07-14  8:33   ` Yang, Rong R
2016-07-01 13:40 ` [PATCH v3 3/3] drm/i915: Give proper names to MOCS entries Imre Deak
2016-07-01 13:49   ` Chris Wilson
2016-07-01 13:56     ` Imre Deak
2016-07-01 14:32   ` [PATCH v4 " Imre Deak
2016-07-13  2:22   ` [PATCH v3 " Zhao Yakui
2016-07-13 10:04     ` Imre Deak
2016-07-14  1:38       ` Zhao Yakui
2016-07-01 14:54 ` ✗ Ro.CI.BAT: warning for drm/i915/bxt: Fix performance due to bogus MOCS entry Patchwork
2016-07-19 18:15   ` Imre Deak
2016-07-01 15:15 ` ✗ Ro.CI.BAT: warning for drm/i915/bxt: Fix performance due to bogus MOCS entry (rev2) Patchwork
2016-07-18 14:28 ` [PATCH v3 0/3] drm/i915/bxt: Fix performance due to bogus MOCS entry Ville Syrjälä

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=1467380406-11954-3-git-send-email-imre.deak@intel.com \
    --to=imre.deak@intel.com \
    --cc=eero.t.tamminen@intel.com \
    --cc=intel-gfx@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 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.