Since you’re here...

We hope you will consider supporting us today. We need your support to continue to exist, because good entries are more and more work time. Every reader contribution, however big or small, is so valuable. Support "Chess Engines Diary" even a small amount– and it only takes a minute. Thank you.
============================== My email: jotes@go2.pl



Stockfish 16120309 - new version

on

Stockfish, UCI chess engine
Leader of the ranking list JCER = 3368

🔬 Author: Andrey Neporada 
More:
Timestamp: 1480754227 

 Help GCC to optimize msb() to single instruction 

 GCC compiles builtin_clzll to “63 ^ BSR”. BSR is processor instruction "Bit Scan Reverse". 
So old msb() function is basically 63 - 63 ^ BSR. 
Unfortunately, GCC fails to simplify this expression. 

 Old function compiles to 
bsrq %rdi, %rdi 
movl $63, %eax 
xorq $63, %rdi 
subl %edi, %eax 
ret 

 New function compiles to 
bsrq %rdi, %rax 
ret 

 BTW, Clang compiles both function to the same (optimal) code. 

 No functional change.


Stockfish 16120309 - download

Comments

Post a Comment