https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120233
Bug ID: 120233
Summary: [16 Regression] FAIL: gcc.target/i386/pr108938-3.c
scan-assembler-times bswap[\t ]+ 3
Product: gcc
Version: 16.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: hjl.tools at gmail dot com
CC: hubicka at ucw dot cz, liuhongt at gcc dot gnu.org,
ubizjak at gmail dot com
Target Milestone: ---
Target: x86-64
On x86-64, -m32 gave me
FAIL: gcc.target/i386/pr108938-3.c scan-assembler-times bswap[\t ]+ 3
The new code looks aweful:
foo2:
.LFB1:
.cfi_startproc
movl 8(%esp), %edx
movzwl (%edx), %eax
movzwl 2(%edx), %edx
movl %eax, %ecx
sall $16, %eax
sarw $8, %cx
movzwl %cx, %ecx
orl %ecx, %eax
movd %eax, %xmm0
movl %edx, %eax
sall $16, %edx
sarw $8, %ax
movdqa %xmm0, %xmm2
movzwl %ax, %eax
orl %eax, %edx
movl 4(%esp), %eax
movd %edx, %xmm1
punpcklbw %xmm1, %xmm2
punpcklbw %xmm1, %xmm0
pshufd $65, %xmm2, %xmm2
punpcklbw %xmm2, %xmm0
movd %xmm0, (%eax)
ret
.cfi_endproc
vs the old one:
foo2:
.LFB1:
.cfi_startproc
movl 8(%esp), %ecx
movl 4(%esp), %edx
movzwl (%ecx), %eax
rolw $8, %ax
movw %ax, (%edx)
movzwl 2(%ecx), %eax
rolw $8, %ax
movw %ax, 2(%edx)
ret
.cfi_endproc
