diff options
author | Chris Wilson <chris@chris-wilson.co.uk> | 2017-04-27 13:46:51 +0300 |
---|---|---|
committer | Chris Wilson <chris@chris-wilson.co.uk> | 2017-04-27 14:22:13 +0300 |
commit | 3204c343bb691c42a5e568cbd2a9ec9b2b5703c0 (patch) | |
tree | a1e3c562ab38e3b6a6a3b1bebfde791d0f20c9be /drivers/gpu/drm/i915/intel_ringbuffer.c | |
parent | e3895af8bb12612972efb1a07f0bb02b2853afda (diff) | |
download | linux-3204c343bb691c42a5e568cbd2a9ec9b2b5703c0.tar.xz |
drm/i915: Defer context state allocation for legacy ring submission
Almost from the outset for execlists, we used deferred allocation of the
logical context and rings. Then we ported the infrastructure for pinning
contexts back to legacy, and so now we are able to also implement
deferred allocation for context objects prior to first use on the legacy
submission.
v2: We still need to differentiate between legacy engines, Joonas is
fixing that but I want this first ;) (Joonas)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170427104651.22394-1-chris@chris-wilson.co.uk
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Diffstat (limited to 'drivers/gpu/drm/i915/intel_ringbuffer.c')
-rw-r--r-- | drivers/gpu/drm/i915/intel_ringbuffer.c | 50 |
1 files changed, 50 insertions, 0 deletions
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 6836efb7e3d2..61f612454ce7 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1437,6 +1437,44 @@ static int context_pin(struct i915_gem_context *ctx) PIN_GLOBAL | PIN_HIGH); } +static struct i915_vma * +alloc_context_vma(struct intel_engine_cs *engine) +{ + struct drm_i915_private *i915 = engine->i915; + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + + obj = i915_gem_object_create(i915, i915->hw_context_size); + if (IS_ERR(obj)) + return ERR_CAST(obj); + + /* + * Try to make the context utilize L3 as well as LLC. + * + * On VLV we don't have L3 controls in the PTEs so we + * shouldn't touch the cache level, especially as that + * would make the object snooped which might have a + * negative performance impact. + * + * Snooping is required on non-llc platforms in execlist + * mode, but since all GGTT accesses use PAT entry 0 we + * get snooping anyway regardless of cache_level. + * + * This is only applicable for Ivy Bridge devices since + * later platforms don't have L3 control bits in the PTE. + */ + if (IS_IVYBRIDGE(i915)) { + /* Ignore any error, regard it as a simple optimisation */ + i915_gem_object_set_cache_level(obj, I915_CACHE_L3_LLC); + } + + vma = i915_vma_instance(obj, &i915->ggtt.base, NULL); + if (IS_ERR(vma)) + i915_gem_object_put(obj); + + return vma; +} + static int intel_ring_context_pin(struct intel_engine_cs *engine, struct i915_gem_context *ctx) { @@ -1449,6 +1487,18 @@ static int intel_ring_context_pin(struct intel_engine_cs *engine, return 0; GEM_BUG_ON(!ce->pin_count); /* no overflow please! */ + if (engine->id == RCS && !ce->state && engine->i915->hw_context_size) { + struct i915_vma *vma; + + vma = alloc_context_vma(engine); + if (IS_ERR(vma)) { + ret = PTR_ERR(vma); + goto error; + } + + ce->state = vma; + } + if (ce->state) { ret = context_pin(ctx); if (ret) |