aditya/llama.cpp
1
0
Fork 0
mirror of https://git.adityakumar.xyz/llama.cpp.git synced 2024-11-09 15:29:43 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

inline -> static inline for "bytesFromNibbles" (#161)

Browse source 
Without "static" prefix, it fails to compile in clang
This commit is contained in:
hoangmit 2023-03-15 15:05:14 -04:00 committed by GitHub
parent 47857e564c
commit 113e685d18
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
ggml.c
View file

@ -364,7 +364,7 @@ static const size_t CACHE_LINE_SIZE_F32 = CACHE_LINE_SIZE/sizeof(float);
#if __AVX2__ #if __AVX2__
// Unpack 32 4-bit fields into 32 bytes // Unpack 32 4-bit fields into 32 bytes
// The output vector contains 32 bytes, each one in [ 0 .. 15 ] interval // The output vector contains 32 bytes, each one in [ 0 .. 15 ] interval
inline __m256i bytesFromNibbles( const uint8_t* rsi ) static inline __m256i bytesFromNibbles( const uint8_t* rsi )
{ {
// Load 16 bytes from memory // Load 16 bytes from memory
__m128i tmp = _mm_loadu_si128( ( const __m128i* )rsi ); __m128i tmp = _mm_loadu_si128( ( const __m128i* )rsi );
@ -381,7 +381,7 @@ inline __m256i bytesFromNibbles( const uint8_t* rsi )
return bytes; return bytes;
} }
inline __m128i packNibbles( __m256i bytes ) static inline __m128i packNibbles( __m256i bytes )
{ {
// Move bits within 16-bit lanes from 0000_abcd_0000_efgh into 0000_0000_abcd_efgh // Move bits within 16-bit lanes from 0000_abcd_0000_efgh into 0000_0000_abcd_efgh
const __m256i lowByte = _mm256_set1_epi16( 0xFF ); const __m256i lowByte = _mm256_set1_epi16( 0xFF );