[Bug middle-end/35142] MMX bad optimization with intrinsics

2008-02-09 Thread etjq78kl at free dot fr
--- Comment #2 from etjq78kl at free dot fr 2008-02-09 10:14 --- (In reply to comment #1) Glad to know that 4.3 solves this issue, but what about subl/addl $20, %esp at the begining and at the end. SSE code does not show any stack stack pointer manipulation (which is unneeded). --

[Bug middle-end/35142] MMX bad optimization with intrinsics

2008-02-08 Thread rguenth at gcc dot gnu dot org
--- Comment #1 from rguenth at gcc dot gnu dot org 2008-02-08 22:38 --- With gcc 4.3: _Z3fooR1XRKS_: .LFB124: subl$20, %esp .LCFI0: movl24(%esp), %eax movl28(%esp), %edx movq(%eax), %mm0 pxor(%edx), %mm0 movq%mm0, (