| .. |
|
CacheControl.cpp
|
feat(shader): decode cache invalidations and performance hints as no-ops
|
2026-10-01 22:45:25 +02:00 |
|
Float16Misc.cpp
|
feat(shader): implement f16 ldexp/frexp/norm/class, v_sat_pk_u8_i16 and v_mul/mac_legacy_f32
|
2026-10-02 06:46:26 +02:00 |
|
Float16Rounding.cpp
|
fix(shader): round f32 to f16 to nearest even on every driver
|
2026-10-02 06:40:10 +02:00 |
|
FmaRounding.cpp
|
fix(shader): keep fused multiply-adds fused and v_mad/v_mac_f32 rounded twice
|
2026-10-02 06:41:03 +02:00 |
|
LdsAtomics.cpp
|
feat(shader): implement the missing 32-bit LDS atomics and fix ds_min/max_f32
|
2026-10-02 07:11:48 +02:00 |
|
MixPrecision.cpp
|
ci: run the shader execution tests on lavapipe
|
2026-10-01 22:45:25 +02:00 |
|
PixelHelperLanes.cpp
|
fix(recompiler): start pixel waves without their helper invocations in EXEC
|
2026-10-02 21:08:32 +13:00 |
|
ReadLane.cpp
|
fix(shader): wrap the v_readlane_b32 lane select at the wave size
|
2026-10-02 18:23:56 +13:00 |
|
ScalarBitCount.cpp
|
feat(shader): implement s_bcnt0_i32_b64, s_ff0_i32_b64, s_flbit_i32(_i64), s_quadmask_b32
|
2026-10-02 09:59:47 +02:00 |
|
ScalarRelativeIndexing.cpp
|
feat(shader): implement s_movrels/s_movreld (b32, b64) and s_movrelsd_2_b32
|
2026-10-02 09:59:59 +02:00 |
|
SdwaDestination.cpp
|
ci: run the shader execution tests on lavapipe
|
2026-10-01 22:45:25 +02:00 |
|
ShaderClock.cpp
|
feat(shader): implement s_memtime and s_memrealtime
|
2026-10-02 07:58:41 +02:00 |
|
Vop3Aliases.cpp
|
feat(shader): decode VOP3 v_max/min_f16 and v_cvt_f32_f16/ubyte, accept abs/neg on f16 binaries
|
2026-10-01 22:45:25 +02:00 |
|
Vop3CompareModifiers.cpp
|
fix(shader): accept neg on VOP3-encoded float compares
|
2026-10-01 22:45:25 +02:00 |
|
Vop3Integer.cpp
|
feat(shader): implement VOP3 v_add/sub_nc_i32 with clamp, v_mul_lo_u16, v_mad_u16/i16/u32_u16/i32_i16, v_ashrrev_i64
|
2026-10-01 22:45:25 +02:00 |
|
Vop3Lerp.cpp
|
feat(shader): implement v_lerp_u8
|
2026-10-02 18:57:06 +02:00 |
|
Vop3Sad.cpp
|
feat(shader): implement v_sad_u8, v_sad_hi_u8, v_sad_u16 and v_msad_u8
|
2026-10-02 18:25:22 +02:00 |
|
VopcCompare.cpp
|
fix(shader): lower f16 NLE comparisons through swapped unordered less-than
|
2026-10-02 21:17:18 +03:00 |
|
VulkanTestDevice.hpp
|
ci: run the shader execution tests on lavapipe
|
2026-10-01 22:45:25 +02:00 |
|
WaveUniform.cpp
|
perf(recompiler): emit wave-uniform values once for both lanes of a two-lane invocation
|
2026-10-03 20:00:38 +13:00 |
|
WriteLane.cpp
|
fix(shader): select the written lane in v_writelane_b32
|
2026-10-02 09:59:37 +02:00 |
b45716210c