diff options
Diffstat (limited to 'absl/crc')
| -rw-r--r-- | absl/crc/internal/crc32_x86_arm_combined_simd.h | 14 |
1 files changed, 9 insertions, 5 deletions
diff --git a/absl/crc/internal/crc32_x86_arm_combined_simd.h b/absl/crc/internal/crc32_x86_arm_combined_simd.h index d2fc6211..d3eedd58 100644 --- a/absl/crc/internal/crc32_x86_arm_combined_simd.h +++ b/absl/crc/internal/crc32_x86_arm_combined_simd.h @@ -225,8 +225,8 @@ inline void V128_Store(V128* dst, V128 data) { // Using inline assembly as clang does not generate the pmull2 instruction and // performance drops by 15-20%. -// TODO(b/193678732): Investigate why the compiler decides not to generate -// such instructions and why it becomes so much worse. +// TODO(b/193678732): Investigate why there is a slight performance hit when +// using intrinsics instead of inline assembly. inline V128 V128_PMulHi(const V128 l, const V128 r) { uint64x2_t res; __asm__ __volatile__("pmull2 %0.1q, %1.2d, %2.2d \n\t" @@ -235,10 +235,14 @@ inline V128 V128_PMulHi(const V128 l, const V128 r) { return res; } +// TODO(b/193678732): Investigate why the compiler decides to move the constant +// loop multiplicands from GPR to Neon registers every loop iteration. inline V128 V128_PMulLow(const V128 l, const V128 r) { - return reinterpret_cast<V128>(vmull_p64( - reinterpret_cast<poly64_t>(vget_low_p64(vreinterpretq_p64_u64(l))), - reinterpret_cast<poly64_t>(vget_low_p64(vreinterpretq_p64_u64(r))))); + uint64x2_t res; + __asm__ __volatile__("pmull %0.1q, %1.1d, %2.1d \n\t" + : "=w"(res) + : "w"(l), "w"(r)); + return res; } inline V128 V128_PMul01(const V128 l, const V128 r) { |