https://bugs.llvm.org/show_bug.cgi?id=49931
Bug ID: 49931
Summary: Codegen regression between 10 and 11 with
-mprefer-vector-width=512
Product: libraries
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: Backend: X86
Assignee: [email protected]
Reporter: [email protected]
CC: [email protected], [email protected],
[email protected], [email protected],
[email protected]
typedef unsigned char uint8_t;
void foo( uint8_t *__restrict dst, uint8_t *__restrict src )
{
for( int x = 0; x < 16; x++ )
dst[x] = (dst[x]-src[x])>>2;
}
Clang flags: -O3 -mavx512vnni -mprefer-vector-width=512
See the change in the use of ZMM registers.
LLVM 11+:
foo(unsigned char*, unsigned char*): #
@foo(unsigned char*, unsigned char*)
vpmovzxbw ymm0, xmmword ptr [rdi] # ymm0 =
mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero,mem[8],zero,mem[9],zero,mem[10],zero,mem[11],zero,mem[12],zero,mem[13],zero,mem[14],zero,mem[15],zero
vpmovzxbw ymm1, xmmword ptr [rsi] # ymm1 =
mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero,mem[8],zero,mem[9],zero,mem[10],zero,mem[11],zero,mem[12],zero,mem[13],zero,mem[14],zero,mem[15],zero
vpsubw ymm0, ymm0, ymm1
vpsrlw ymm0, ymm0, 2
vpmovzxwd zmm0, ymm0 # zmm0 =
ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
vpmovdb xmmword ptr [rdi], zmm0
vzeroupper
ret
LLVM 10:
foo(unsigned char*, unsigned char*): #
@foo(unsigned char*, unsigned char*)
vpmovzxbd zmm0, xmmword ptr [rdi] # zmm0 =
mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
vpmovzxbd zmm1, xmmword ptr [rsi] # zmm1 =
mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero,mem[8],zero,zero,zero,mem[9],zero,zero,zero,mem[10],zero,zero,zero,mem[11],zero,zero,zero,mem[12],zero,zero,zero,mem[13],zero,zero,zero,mem[14],zero,zero,zero,mem[15],zero,zero,zero
vpsubd zmm0, zmm0, zmm1
vpsrld zmm0, zmm0, 2
vpmovdb xmmword ptr [rdi], zmm0
vzeroupper
ret
Godbolt:
https://godbolt.org/z/4zPv8rMvP
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
