kate/BadFish
1
0
Fork 0
mirror of https://github.com/sockspls/badfish synced 2025-04-30 16:53:09 +00:00
Code Issues Releases Wiki Activity

Use packed 32-bit MMX operations for updating the PSQT accumulator

Browse source 
This improves the speed of NNUE by a bit on old hardware that code path
is intended for, like a Pentium III 1.13 GHz:

10 repeats of "./stockfish bench 16 1 13 default depth NNUE":

Before:
54 642 504 897 cycles (± 0.12%)
62 301 937 829 instructions (± 0.03%)

After:
54 320 821 928 cycles (± 0.13%)
62 084 742 699 instructions (± 0.02%)

Speed of go depth 20 from startpos:

Before: 53103 nps
After: 53856 nps

closes https://github.com/official-stockfish/Stockfish/pull/3476

No functional change.
This commit is contained in:
Fanael Linithien 2021-05-18 19:17:59 +02:00 • committed by Joost VandeVondele
parent 0faf81d1f6
commit 038487f954
1 changed files with 5 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
src/nnue/nnue_feature_transformer.h
View file

@ -84,18 +84,18 @@ namespace Stockfish::Eval::NNUE {
#elif USE_MMX
typedef __m64 vec_t;
typedef std::int32_t psqt_vec_t;
typedef __m64 psqt_vec_t;
#define vec_load(a) (*(a))
#define vec_store(a,b) *(a)=(b)
#define vec_add_16(a,b) _mm_add_pi16(a,b)
#define vec_sub_16(a,b) _mm_sub_pi16(a,b)
#define vec_load_psqt(a) (*(a))
#define vec_store_psqt(a,b) *(a)=(b)
#define vec_add_psqt_32(a,b) a+b
#define vec_sub_psqt_32(a,b) a-b
#define vec_zero_psqt() 0
#define vec_add_psqt_32(a,b) _mm_add_pi32(a,b)
#define vec_sub_psqt_32(a,b) _mm_sub_pi32(a,b)
#define vec_zero_psqt() _mm_setzero_si64()
static constexpr IndexType NumRegs = 8;
static constexpr IndexType NumPsqtRegs = 8;
static constexpr IndexType NumPsqtRegs = 4;
#elif USE_NEON
typedef int16x8_t vec_t;