happyz
/
gemma.cpp
mirror of https://github.com/google/gemma.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Remove Bf16ToF32EO and use PromoteEvenTo and PromoteOddTo.

This commit is contained in:
Sam Kaufman 2024-04-29 14:13:07 -07:00
parent aa0b113214
commit f608337fef
2 changed files with 9 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
compression/compress-inl.h
View File

@ -51,20 +51,6 @@ namespace gcpp {
namespace HWY_NAMESPACE { namespace HWY_NAMESPACE {
namespace hn = hwy::HWY_NAMESPACE; namespace hn = hwy::HWY_NAMESPACE;
template <class DF, HWY_IF_F32_D(DF)>
HWY_INLINE void Bf16ToF32EO(const DF df32,
const hwy::bfloat16_t* HWY_RESTRICT in,
hn::Vec<DF>& v_even,
hn::Vec<DF>& v_odd) {
const hn::Repartition<hwy::bfloat16_t, DF> dbf16;
const hn::RebindToUnsigned<decltype(df32)> du32;
const auto odd = Set(du32, 0xFFFF0000u);
const auto interleaved = BitCast(du32, LoadU(dbf16, in));
v_even = BitCast(df32, hn::ShiftLeft<16>(interleaved));
v_odd = BitCast(df32, And(interleaved, odd));
}
// Enables generic code independent of compression type. // Enables generic code independent of compression type.
template <typename T> // primary, must specialize template <typename T> // primary, must specialize
struct CompressTraits {}; struct CompressTraits {};
@ -263,19 +249,19 @@ struct CompressTraits<hwy::bfloat16_t> {
VF32 be0, bo0, be1, bo1; VF32 be0, bo0, be1, bo1;
for (size_t i = 0; i < num; /* i += 2 * N */) { for (size_t i = 0; i < num; /* i += 2 * N */) {
const auto interleaved0 = hn::LoadU(dbf16, in + in_ofs + i);
const VF32 ae0 = Load(df32, vec_aligned + i); const VF32 ae0 = Load(df32, vec_aligned + i);
const VF32 ao0 = Load(df32, vec_aligned + i + (N / 2)); const VF32 ao0 = Load(df32, vec_aligned + i + (N / 2));
Bf16ToF32EO(df32, in + in_ofs + i, be0, bo0); sum0 = hn::MulAdd(ae0, hn::PromoteEvenTo(df32, interleaved0), sum0);
sum1 = hn::MulAdd(ao0, hn::PromoteOddTo(df32, interleaved0), sum1);
i += N; i += N;
sum0 = hn::MulAdd(ae0, be0, sum0);
sum1 = hn::MulAdd(ao0, bo0, sum1);
const auto interleaved1 = hn::LoadU(dbf16, in + in_ofs + i);
const VF32 ae1 = Load(df32, vec_aligned + i); const VF32 ae1 = Load(df32, vec_aligned + i);
const VF32 ao1 = Load(df32, vec_aligned + i + (N / 2)); const VF32 ao1 = Load(df32, vec_aligned + i + (N / 2));
Bf16ToF32EO(df32, in + in_ofs + i, be1, bo1); sum2 = hn::MulAdd(ae1, hn::PromoteEvenTo(df32, interleaved1), sum2);
sum3 = hn::MulAdd(ao1, hn::PromoteOddTo(df32, interleaved1), sum3);
i += N; i += N;
sum2 = hn::MulAdd(ae1, be1, sum2);
sum3 = hn::MulAdd(ao1, bo1, sum3);
} }
sum0 = Add(sum0, sum1); sum0 = Add(sum0, sum1);

7
gemma/ops.h
View File

@ -104,11 +104,10 @@ HWY_INLINE void ToEvenOddF32(
HWY_DASSERT(size % hn::Lanes(dbf16) == 0); HWY_DASSERT(size % hn::Lanes(dbf16) == 0);
HWY_DASSERT(hn::IsAligned(df, vec_aligned)); HWY_DASSERT(hn::IsAligned(df, vec_aligned));
VF32 veven, vodd;
for (size_t i = 0; i < size; i += hn::Lanes(dbf16)) { for (size_t i = 0; i < size; i += hn::Lanes(dbf16)) {
Bf16ToF32EO(df, vec_aligned + i, veven, vodd); const auto interleaved = hn::LoadU(dbf16, vec_aligned + i);
hn::Store(veven, df, out + i); hn::Store(hn::PromoteEvenTo(df, interleaved), df, out + i);
hn::Store(vodd, df, out + i + hn::Lanes(df)); hn::Store(hn::PromoteOddTo(df, interleaved), df, out + i + hn::Lanes(df));
} }
} }