Skip to contentSkip to conversation

drm/i915/hsw: Change default LLC age to 3

PATCH Applied Ben Widawsky 2 participants 2 messages Aug 5, 2013

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v4 1/1 drm/i915/hsw: Change default LLC age to 3

The default LLC age was changed:
commit 0d8ff15e9a15f2b393e53337a107b7a1e5919b6d

Author: Ben Widawsky <[email protected]>
Date:   Thu Jul 4 11:02:03 2013 -0700

drm/i915/hsw: Set correct Haswell PTE encodings.

On the surface it would seem setting a default age wouldn't matter because all GEM BOs are aged similarly, so the order in which objects are evicted would not be subject to aging. The current working theory as to why this caused a regression though is that LLC is a bit special in that it is shared with the CPU. Presumably (not verified) the CPU fetches cachelines with age 3, and therefore recently cached GPU objects would be evicted before similar CPU object first when the LLC is full. It stands to reason therefore that this would negatively impact CPU bound benchmarks - but those seem to be low on the priority list.

eLLC OTOH does not have this same property as LLC. It should be used entirely for the GPU, and so the age really shouldn't matter. Furthermore, we have no evidence to suggest one is better than another on eLLC. Since we've never properly supported eLLC before no, there should be no regression. If the GPU client really wants "younger" objects, they should use MOCS.

v2: Drop the extra #define (Chad)

v3: Actually git add

v4: Pimped commit message

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=67062

---
 drivers/gpu/drm/i915/i915_gem_gtt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +2 -1
index e7b4204..3e7f124 100644
@@ -52,6 +52,7 @@
  */
 #define HSW_CACHEABILITY_CONTROL(bits)	((((bits) & 0x7) << 1) | \
 					 (((bits) & 0x8) << (11 - 3)))
+#define HSW_WB_LLC_AGE3			HSW_CACHEABILITY_CONTROL(0x2)
 #define HSW_WB_LLC_AGE0			HSW_CACHEABILITY_CONTROL(0x3)
 #define HSW_WB_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0xb)
 
@@ -105,7 +106,7 @@ static gen6_gtt_pte_t hsw_pte_encode(dma_addr_t addr,
 	pte |= HSW_PTE_ADDR_ENCODE(addr);
 
 	if (level != I915_CACHE_NONE)
-		pte |= HSW_WB_LLC_AGE0;
+		pte |= HSW_WB_LLC_AGE3;
 
 	return pte;
 }
Signature · 1 line
1.8.3.4
#1 lore raw
Quoted 26 lines · Ben Widawsky

The default LLC age was changed:
commit 0d8ff15e9a15f2b393e53337a107b7a1e5919b6d

Author: Ben Widawsky <[email protected]>
Date:   Thu Jul 4 11:02:03 2013 -0700

drm/i915/hsw: Set correct Haswell PTE encodings.

On the surface it would seem setting a default age wouldn't matter because all GEM BOs are aged similarly, so the order in which objects are evicted would not be subject to aging. The current working theory as to why this caused a regression though is that LLC is a bit special in that it is shared with the CPU. Presumably (not verified) the CPU fetches cachelines with age 3, and therefore recently cached GPU objects would be evicted before similar CPU object first when the LLC is full. It stands to reason therefore that this would negatively impact CPU bound benchmarks - but those seem to be low on the priority list.

eLLC OTOH does not have this same property as LLC. It should be used entirely for the GPU, and so the age really shouldn't matter. Furthermore, we have no evidence to suggest one is better than another on eLLC. Since we've never properly supported eLLC before no, there should be no regression. If the GPU client really wants "younger" objects, they should use MOCS.

v2: Drop the extra #define (Chad)

Quoted 7 lines

v3: Actually git add

v4: Pimped commit message

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=67062

Queued for -next, thanks for the patch.
-Daniel

Signature · 3 lines
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch