From: tim.gore@intel.com
To: intel-gfx@lists.freedesktop.org
Cc: mika.kuoppala@intel.com
Subject: [PATCH] drm/i915: implement WaIncreaseDefaultTLBEntries
Date: Thu, 4 Feb 2016 11:49:34 +0000 [thread overview]
Message-ID: <1454586574-2343-1-git-send-email-tim.gore@intel.com> (raw)
From: Tim Gore <tim.gore@intel.com>
WaIncreaseDefaultTLBEntries increases the number of TLB
entries available for GPGPU workloads and gives significant
( > 10% ) performance gain for some OCL benchmarks.
Put this in a new function that can be a place for
workarounds that are GT related but not required per ring.
This function is called on driver load and also after a
reset and on resume, so it is safe for workarounds that get
clobbered in these situations. This function currently has
just this one workaround.
v2: This was originally split into 3 patches but following
review feedback was squashed into 1.
I have not incorporated some style comments from Chris
Wilson as I felt that after defining and intialising a
temporary variable and then adding an additional if block
to only write the register if the temporary variable had
been set, this didn't really give a net gain.
v3: Resending in the hope that BAT will run
v4: Change subject line to trigger BAT (please!)
Signed-off-by: Tim Gore <tim.gore@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 21 +++++++++++++++++++++
drivers/gpu/drm/i915/i915_reg.h | 7 +++++++
2 files changed, 28 insertions(+)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 7377b67..5cd5f8b 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -2132,6 +2132,25 @@ static void i915_address_space_init(struct i915_address_space *vm,
list_add_tail(&vm->global_link, &dev_priv->vm_list);
}
+static void gtt_write_workarounds(struct drm_device *dev)
+{
+ struct drm_i915_private *dev_priv = dev->dev_private;
+
+ /* This function is for gtt related workarounds. This function is
+ * called on driver load and after a GPU reset, so you can place
+ * workarounds here even if they get overwritten by GPU reset.
+ */
+ /* WaIncreaseDefaultTLBEntries:chv,bdw,skl,bxt */
+ if (IS_BROADWELL(dev))
+ I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_BDW);
+ else if (IS_CHERRYVIEW(dev))
+ I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_CHV);
+ else if (IS_SKYLAKE(dev))
+ I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_SKL);
+ else if (IS_BROXTON(dev))
+ I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_BXT);
+}
+
int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
{
struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2148,6 +2167,8 @@ int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
int i915_ppgtt_init_hw(struct drm_device *dev)
{
+ gtt_write_workarounds(dev);
+
/* In the case of execlists, PPGTT is enabled by the context descriptor
* and the PDPs are contained within the context itself. We don't
* need to do anything here. */
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 65e32a3..0ab6312 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -8170,4 +8170,11 @@ enum skl_disp_power_wells {
#define GEN9_VEBOX_MOCS(i) _MMIO(0xcb00 + (i) * 4) /* Video MOCS registers */
#define GEN9_BLT_MOCS(i) _MMIO(0xcc00 + (i) * 4) /* Blitter MOCS registers */
+/* gamt regs */
+#define GEN8_L3_LRA_1_GPGPU _MMIO(0x4dd4)
+#define GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_BDW 0x67F1427F /* max/min for LRA1/2 */
+#define GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_CHV 0x5FF101FF /* max/min for LRA1/2 */
+#define GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_SKL 0x67F1427F /* " " */
+#define GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_BXT 0x5FF101FF /* " " */
+
#endif /* _I915_REG_H_ */
--
1.9.1
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next reply other threads:[~2016-02-04 11:49 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-04 11:49 tim.gore [this message]
2016-02-04 13:09 ` ✓ Fi.CI.BAT: success for drm/i915: implement WaIncreaseDefaultTLBEntries Patchwork
2016-02-04 14:25 ` Tvrtko Ursulin
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=1454586574-2343-1-git-send-email-tim.gore@intel.com \
--to=tim.gore@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=mika.kuoppala@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).