зеркало из https://github.com/mozilla/gecko-dev.git
Bug 1635649 - optimize texture(sampler2DArray) for same layer in SWGL. r=jrmuizel
Differential Revision: https://phabricator.services.mozilla.com/D73993
This commit is contained in:
Lee Salzman
Родитель
f4e2895339
Коммит
c864b9fc4f
|
|
@ -10,6 +10,18 @@
|
|||
|
||||
namespace glsl {
|
||||
|
||||
#if USE_SSE2
|
||||
bool test_all(Bool cond) { return _mm_movemask_ps(cond) == 0xF; }
|
||||
bool test_any(Bool cond) { return _mm_movemask_ps(cond) != 0; }
|
||||
bool test_none(Bool cond) { return _mm_movemask_ps(cond) == 0; }
|
||||
#else
|
||||
bool test_all(Bool cond) {
|
||||
return bit_cast<uint32_t>(CONVERT(cond, U8)) == 0xFFFFFFFFU;
|
||||
}
|
||||
bool test_any(Bool cond) { return bit_cast<uint32_t>(CONVERT(cond, U8)) != 0; }
|
||||
bool test_none(Bool cond) { return bit_cast<uint32_t>(CONVERT(cond, U8)) == 0; }
|
||||
#endif
|
||||
|
||||
float make_float(float n) { return n; }
|
||||
|
||||
float make_float(int32_t n) { return float(n); }
|
||||
|
|
@ -2235,7 +2247,8 @@ vec4 texelFetchRGBA8(sampler2D sampler, ivec2 P) {
|
|||
}
|
||||
|
||||
vec4 texelFetchRGBA8(sampler2DArray sampler, ivec3 P) {
|
||||
I32 offset = P.x + P.y * sampler->stride + P.z * sampler->height_stride;
|
||||
assert(test_all(P.z == P.z.x));
|
||||
I32 offset = P.x + P.y * sampler->stride + P.z.x * sampler->height_stride;
|
||||
return fetchOffsetsRGBA8(sampler, offset);
|
||||
}
|
||||
|
||||
|
|
@ -2253,7 +2266,8 @@ vec4 texelFetchR8(sampler2D sampler, ivec2 P) {
|
|||
}
|
||||
|
||||
vec4 texelFetchR8(sampler2DArray sampler, ivec3 P) {
|
||||
I32 offset = P.x + P.y * sampler->stride + P.z * sampler->height_stride;
|
||||
assert(test_all(P.z == P.z.x));
|
||||
I32 offset = P.x + P.y * sampler->stride + P.z.x * sampler->height_stride;
|
||||
return vec4(fetchOffsetsR8(sampler, offset), 0.0f, 0.0f, 1.0f);
|
||||
}
|
||||
|
||||
|
|
@ -2269,8 +2283,9 @@ vec4 texelFetchFloat(sampler2D sampler, ivec2 P) {
|
|||
return fetchOffsetsFloat(sampler, offset);
|
||||
}
|
||||
|
||||
vec4 texelFetchFloat(sampler2DArray sampler, ivec3 P) {
|
||||
I32 offset = P.x * 4 + P.y * sampler->stride + P.z * sampler->height_stride;
|
||||
SI vec4 texelFetchFloat(sampler2DArray sampler, ivec3 P) {
|
||||
assert(test_all(P.z == P.z.x));
|
||||
I32 offset = P.x * 4 + P.y * sampler->stride + P.z.x * sampler->height_stride;
|
||||
return fetchOffsetsFloat(sampler, offset);
|
||||
}
|
||||
|
||||
|
|
@ -2349,7 +2364,7 @@ vec4 texelFetch(sampler2DRect sampler, ivec2 P) {
|
|||
return fetchOffsetsRGBA8(sampler, offset);
|
||||
}
|
||||
|
||||
vec4 texelFetch(sampler2DArray sampler, ivec3 P, int lod) {
|
||||
SI vec4 texelFetch(sampler2DArray sampler, ivec3 P, int lod) {
|
||||
assert(lod == 0);
|
||||
P = clamp2DArray(P, sampler);
|
||||
if (sampler->format == TextureFormat::RGBA32F) {
|
||||
|
|
@ -2460,7 +2475,7 @@ SI T linearQuantize(T P, float scale, S sampler) {
|
|||
}
|
||||
|
||||
template <typename S>
|
||||
vec4 textureLinearRGBA8(S sampler, vec2 P, I32 zoffset = 0) {
|
||||
vec4 textureLinearRGBA8(S sampler, vec2 P, int32_t zoffset = 0) {
|
||||
assert(sampler->format == TextureFormat::RGBA8);
|
||||
|
||||
#if USE_SSE2
|
||||
|
|
@ -2481,7 +2496,7 @@ vec4 textureLinearRGBA8(S sampler, vec2 P, I32 zoffset = 0) {
|
|||
_mm_set1_epi16(sampler->stride));
|
||||
row0 = _mm_add_epi32(row0, _mm_unpackhi_epi16(clampyx, _mm_setzero_si128()));
|
||||
// Add in layer offset if available
|
||||
row0 = _mm_add_epi32(row0, zoffset);
|
||||
row0 = _mm_add_epi32(row0, _mm_set1_epi32(zoffset));
|
||||
|
||||
// Check if fractional coords are all zero, in which case skip filtering.
|
||||
__m128i fracyx = _mm_packs_epi32(frac.y, frac.x);
|
||||
|
|
@ -2622,7 +2637,7 @@ vec4 textureLinearRGBA8(S sampler, vec2 P, I32 zoffset = 0) {
|
|||
}
|
||||
|
||||
template <typename S>
|
||||
static U16 textureLinearPackedR8(S sampler, ivec2 i, I32 zoffset) {
|
||||
static U16 textureLinearPackedR8(S sampler, ivec2 i, int32_t zoffset) {
|
||||
assert(sampler->format == TextureFormat::R8);
|
||||
ivec2 frac = i & (I32)0x7F;
|
||||
i >>= 7;
|
||||
|
|
@ -2659,7 +2674,7 @@ static U16 textureLinearPackedR8(S sampler, ivec2 i, I32 zoffset) {
|
|||
}
|
||||
|
||||
template <typename S>
|
||||
vec4 textureLinearR8(S sampler, vec2 P, I32 zoffset = 0) {
|
||||
vec4 textureLinearR8(S sampler, vec2 P, int32_t zoffset = 0) {
|
||||
assert(sampler->format == TextureFormat::R8);
|
||||
|
||||
#if USE_SSE2
|
||||
|
|
@ -2680,7 +2695,7 @@ vec4 textureLinearR8(S sampler, vec2 P, I32 zoffset = 0) {
|
|||
_mm_set1_epi16(sampler->stride));
|
||||
row0 = _mm_add_epi32(row0, _mm_unpackhi_epi16(clampyx, _mm_setzero_si128()));
|
||||
// Add in layer offset if available
|
||||
row0 = _mm_add_epi32(row0, zoffset);
|
||||
row0 = _mm_add_epi32(row0, _mm_set1_epi32(zoffset));
|
||||
|
||||
__m128i fracyx = _mm_packs_epi32(frac.y, frac.x);
|
||||
|
||||
|
|
@ -2740,7 +2755,7 @@ vec4 textureLinearR8(S sampler, vec2 P, I32 zoffset = 0) {
|
|||
}
|
||||
|
||||
template <typename S>
|
||||
vec4 textureLinearRGBA32F(S sampler, vec2 P, I32 zoffset = 0) {
|
||||
vec4 textureLinearRGBA32F(S sampler, vec2 P, int32_t zoffset = 0) {
|
||||
assert(sampler->format == TextureFormat::RGBA32F);
|
||||
P.x *= sampler->width;
|
||||
P.y *= sampler->height;
|
||||
|
|
@ -2777,15 +2792,15 @@ vec4 textureLinearRGBA32F(S sampler, vec2 P, I32 zoffset = 0) {
|
|||
return pixel_float_to_vec4(c0, c1, c2, c3);
|
||||
}
|
||||
|
||||
vec4 texture(sampler2D sampler, vec2 P) {
|
||||
SI vec4 texture(sampler2D sampler, vec2 P) {
|
||||
if (sampler->filter == TextureFilter::LINEAR) {
|
||||
if (sampler->format == TextureFormat::RGBA32F) {
|
||||
return textureLinearRGBA32F(sampler, P);
|
||||
} else if (sampler->format == TextureFormat::RGBA8) {
|
||||
if (sampler->format == TextureFormat::RGBA8) {
|
||||
return textureLinearRGBA8(sampler, P);
|
||||
} else {
|
||||
assert(sampler->format == TextureFormat::R8);
|
||||
} else if (sampler->format == TextureFormat::R8) {
|
||||
return textureLinearR8(sampler, P);
|
||||
} else {
|
||||
assert(sampler->format == TextureFormat::RGBA32F);
|
||||
return textureLinearRGBA32F(sampler, P);
|
||||
}
|
||||
} else {
|
||||
ivec2 coord(roundzero(P.x, sampler->width), roundzero(P.y, sampler->height));
|
||||
|
|
@ -2804,17 +2819,20 @@ vec4 texture(sampler2DRect sampler, vec2 P) {
|
|||
}
|
||||
}
|
||||
|
||||
vec4 texture(sampler2DArray sampler, vec3 P) {
|
||||
SI vec4 texture(sampler2DArray sampler, vec3 P) {
|
||||
if (sampler->filter == TextureFilter::LINEAR) {
|
||||
I32 zoffset =
|
||||
clampCoord(roundeven(P.z, 1.0f), sampler->depth) * sampler->height_stride;
|
||||
if (sampler->format == TextureFormat::RGBA32F) {
|
||||
return textureLinearRGBA32F(sampler, vec2(P.x, P.y), zoffset);
|
||||
} else if (sampler->format == TextureFormat::RGBA8) {
|
||||
// SSE2 can generate slow code for 32-bit multiply, and we never actually sample
|
||||
// from different layers in one chunk, so do cheaper scalar multiplication instead.
|
||||
assert(test_all(P.z == P.z.x));
|
||||
int32_t zoffset =
|
||||
clampCoord(roundeven(P.z.x, 1.0f), sampler->depth) * sampler->height_stride;
|
||||
if (sampler->format == TextureFormat::RGBA8) {
|
||||
return textureLinearRGBA8(sampler, vec2(P.x, P.y), zoffset);
|
||||
} else {
|
||||
assert(sampler->format == TextureFormat::R8);
|
||||
} else if (sampler->format == TextureFormat::R8) {
|
||||
return textureLinearR8(sampler, vec2(P.x, P.y), zoffset);
|
||||
} else {
|
||||
assert(sampler->format == TextureFormat::RGBA32F);
|
||||
return textureLinearRGBA32F(sampler, vec2(P.x, P.y), zoffset);
|
||||
}
|
||||
} else {
|
||||
// just do nearest for now
|
||||
|
|
@ -3213,16 +3231,4 @@ Array<vec2, SIZE> if_then_else(I32 c, Array<vec2, SIZE> t,
|
|||
return r;
|
||||
}
|
||||
|
||||
#if USE_SSE2
|
||||
bool test_all(Bool cond) { return _mm_movemask_ps(cond) == 0xF; }
|
||||
bool test_any(Bool cond) { return _mm_movemask_ps(cond) != 0; }
|
||||
bool test_none(Bool cond) { return _mm_movemask_ps(cond) == 0; }
|
||||
#else
|
||||
bool test_all(Bool cond) {
|
||||
return bit_cast<uint32_t>(CONVERT(cond, U8)) == 0xFFFFFFFFU;
|
||||
}
|
||||
bool test_any(Bool cond) { return bit_cast<uint32_t>(CONVERT(cond, U8)) != 0; }
|
||||
bool test_none(Bool cond) { return bit_cast<uint32_t>(CONVERT(cond, U8)) == 0; }
|
||||
#endif
|
||||
|
||||
} // namespace glsl
|
||||
|
|
|
|||
Загрузка…
Ссылка в новой задаче