Welcome to mirror list, hosted at ThFree Co, Russian Federation.

git.blender.org/blender.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorBrecht Van Lommel <brecht>2022-11-01 17:16:55 +0300
committerBrecht Van Lommel <brecht@blender.org>2022-11-08 14:28:40 +0300
commite1b3d9112730bc3b569ffff732a1558752ded146 (patch)
tree27caad945dcc9ce7313b4c84cd0efbab6f70503c /intern/cycles/util/half.h
parent32ec0521c542bb78a0080f8091856ec085030f09 (diff)
Refactor: replace Cycles sse/avx types by vectorized float4/int4/float8/int8
The distinction existed for legacy reasons, to easily port of Embree intersection code without affecting the main vector types. However we are now using SIMD for these types as well, so no good reason to keep the distinction. Also more consistently pass these vector types by value in inline functions. Previously it was partially changed for functions used by Metal to avoid having to add address space qualifiers, simple to do it everywhere. Also removes function declarations for vector math headers, serves no real purpose. Differential Revision: https://developer.blender.org/D16146
Diffstat (limited to 'intern/cycles/util/half.h')
-rw-r--r--intern/cycles/util/half.h16
1 files changed, 8 insertions, 8 deletions
diff --git a/intern/cycles/util/half.h b/intern/cycles/util/half.h
index c668638eb02..5665dd4c075 100644
--- a/intern/cycles/util/half.h
+++ b/intern/cycles/util/half.h
@@ -154,17 +154,17 @@ ccl_device_inline half float_to_half_display(const float f)
ccl_device_inline half4 float4_to_half4_display(const float4 f)
{
-#ifdef __KERNEL_SSE2__
+#ifdef __KERNEL_SSE__
/* CPU: SSE and AVX. */
- ssef x = min(max(load4f(f), 0.0f), 65504.0f);
+ float4 x = min(max(f, make_float4(0.0f)), make_float4(65504.0f));
# ifdef __KERNEL_AVX2__
- ssei rpack = _mm_cvtps_ph(x, 0);
+ int4 rpack = int4(_mm_cvtps_ph(x, 0));
# else
- ssei absolute = cast(x) & 0x7FFFFFFF;
- ssei Z = absolute + 0xC8000000;
- ssei result = andnot(absolute < 0x38800000, Z);
- ssei rshift = (result >> 13) & 0x7FFF;
- ssei rpack = _mm_packs_epi32(rshift, rshift);
+ int4 absolute = cast(x) & make_int4(0x7FFFFFFF);
+ int4 Z = absolute + make_int4(0xC8000000);
+ int4 result = andnot(absolute < make_int4(0x38800000), Z);
+ int4 rshift = (result >> 13) & make_int4(0x7FFF);
+ int4 rpack = int4(_mm_packs_epi32(rshift, rshift));
# endif
half4 h;
_mm_storel_pi((__m64 *)&h, _mm_castsi128_ps(rpack));