Not only is i915_vma_pin() called for every single object on every single execbuf, it is usually a simple increment as the VMA is already bound for execution by the GPU. Rearrange the tests for unbound and pin_count overflow so that we can do the increment and test very cheaply and compact enough to inline the operation into execbuf. The trick used is to note that we can check for an overflow bit (keeping space available for it inside the flags) at the same time as checking the binding bits. Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> --- drivers/gpu/drm/i915/i915_gem.c | 18 +++++-------- drivers/gpu/drm/i915/i915_gem_gtt.h | 53 ++++++++++++++++++++++++++----------- 2 files changed, 44 insertions(+), 27 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index eeea7dab70f1..a666f72121eb 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -3696,23 +3696,19 @@ void __i915_vma_set_map_and_fenceable(struct i915_vma *vma) obj->map_and_fenceable = mappable && fenceable; } -int -i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags) +int __i915_vma_do_pin(struct i915_vma *vma, + u64 size, u64 alignment, u64 flags) { - unsigned int bound; + unsigned int bound = vma->flags; int ret; GEM_BUG_ON((flags & (PIN_GLOBAL | PIN_USER)) == 0); GEM_BUG_ON((flags & PIN_GLOBAL) && !i915_vma_is_ggtt(vma)); - bound = vma->flags; - if (WARN_ON((bound & I915_VMA_PIN_MASK) == I915_VMA_PIN_MASK)) - return -EBUSY; - - /* Pin early to prevent the shrinker/eviction logic from destroying - * our vma as we insert and bind. - */ - __i915_vma_pin(vma); + if (WARN_ON(bound & I915_VMA_PIN_OVERFLOW)) { + ret = -EBUSY; + goto err; + } if ((bound & (I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND)) == 0) { ret = i915_vma_insert(vma, size, alignment, flags); diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h index 0308c7401405..a7e05b4d7d2b 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.h +++ b/drivers/gpu/drm/i915/i915_gem_gtt.h @@ -193,13 +193,15 @@ struct i915_vma { * In the worst case this is 1 + 1 + 1 + 2*2 = 7. That would fit into 3 * bits with absolutely no headroom. So use 4 bits. */ #define I915_VMA_PIN_MASK 0xf +#define I915_VMA_PIN_OVERFLOW BIT(5) /** Flags and address space this VMA is bound to */ -#define I915_VMA_GLOBAL_BIND BIT(5) -#define I915_VMA_LOCAL_BIND BIT(6) +#define I915_VMA_GLOBAL_BIND BIT(6) +#define I915_VMA_LOCAL_BIND BIT(7) +#define I915_VMA_BIND_MASK (I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND | I915_VMA_PIN_OVERFLOW) -#define I915_VMA_GGTT BIT(7) -#define I915_VMA_CLOSED BIT(8) +#define I915_VMA_GGTT BIT(8) +#define I915_VMA_CLOSED BIT(9) unsigned int active; struct i915_gem_active last_read[I915_NUM_ENGINES]; @@ -616,20 +618,39 @@ i915_ggtt_view_equal(const struct i915_ggtt_view *a, return true; } -int __must_check -i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags); /* Flags used by pin/bind&friends. */ -#define PIN_MAPPABLE BIT(0) -#define PIN_NONBLOCK BIT(1) -#define PIN_GLOBAL BIT(2) -#define PIN_OFFSET_BIAS BIT(3) -#define PIN_USER BIT(4) -#define PIN_UPDATE BIT(5) -#define PIN_ZONE_4G BIT(6) -#define PIN_HIGH BIT(7) -#define PIN_OFFSET_FIXED BIT(8) +#define PIN_NONBLOCK BIT(0) +#define PIN_MAPPABLE BIT(1) +#define PIN_ZONE_4G BIT(2) + +#define PIN_MBZ BIT(5) /* I915_VMA_PIN_OVERFLOW */ +#define PIN_GLOBAL BIT(6) /* I915_VMA_GLOBAL_BIND */ +#define PIN_USER BIT(7) /* I915_VMA_LOCAL_BIND */ +#define PIN_UPDATE BIT(8) + +#define PIN_HIGH BIT(9) +#define PIN_OFFSET_BIAS BIT(10) +#define PIN_OFFSET_FIXED BIT(11) #define PIN_OFFSET_MASK (~4095) +int __i915_vma_do_pin(struct i915_vma *vma, + u64 size, u64 alignment, u64 flags); +static inline int __must_check +i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags) +{ + BUILD_BUG_ON(PIN_MBZ != I915_VMA_PIN_OVERFLOW); + BUILD_BUG_ON(PIN_GLOBAL != I915_VMA_GLOBAL_BIND); + BUILD_BUG_ON(PIN_USER != I915_VMA_LOCAL_BIND); + + /* Pin early to prevent the shrinker/eviction logic from destroying + * our vma as we insert and bind. + */ + if (likely(((++vma->flags ^ flags) & I915_VMA_BIND_MASK) == 0)) + return 0; + + return __i915_vma_do_pin(vma, size, alignment, flags); +} + static inline int i915_vma_pin_count(const struct i915_vma *vma) { return vma->flags & I915_VMA_PIN_MASK; @@ -643,7 +664,7 @@ static inline bool i915_vma_is_pinned(const struct i915_vma *vma) static inline void __i915_vma_pin(struct i915_vma *vma) { vma->flags++; - GEM_BUG_ON(!i915_vma_is_pinned(vma)); + GEM_BUG_ON(vma->flags & I915_VMA_PIN_OVERFLOW); } static inline void __i915_vma_unpin(struct i915_vma *vma) -- 2.8.1 _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx