This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug rtl-optimization/50567] Reload pass generates sub-optimal spill code for registers in presence of a vec_concat insn
- From: "siddhesh.poyarekar at gmail dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Thu, 29 Sep 2011 15:24:52 +0000
- Subject: [Bug rtl-optimization/50567] Reload pass generates sub-optimal spill code for registers in presence of a vec_concat insn
- Auto-submitted: auto-generated
- References: <bug-50567-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50567
--- Comment #2 from Siddhesh Poyarekar <siddhesh.poyarekar at gmail dot com> 2011-09-29 15:24:52 UTC ---
Thanks, that eliminated the spill to stack. The extra xmm1 to xmm0 move still
remains:
process:
.LFB0:
.cfi_startproc
movq (%rdi), %rax
cmpq %rsi, %rdi
movq %rax, %rdx
jae .L2
movq (%rsi), %rdx
.L2:
movd %rdx, %xmm1
pinsrq $1, %rax, %xmm1
movdqa %xmm1, %xmm0
ret