Re: LuaJIT2 performance for number crunching

[Miles Bader <miles@...>]

KHMan <keinhong@gmail.com> writes:
>>> Sorry to barge in, it is a worrying difference. One thing is bugging me:
>>> Is the C code running SSE2? IIRC gcc -O2 does not normally enable SSE2.
>>
>> It will enable it by default when it thinks it's possible.
>
> But this is not standard behaviour. I haven't heard of such a change
> in gcc default behaviour from the past few release notices. Would be
> much obliged if you can correct me on this.

It is on x86_64 systems, using the x86_64 abi; e.g.:

   $ gcc -o - -S -O2 -fverbose-asm -x c /dev/null | grep sse
   # options passed:  /dev/null -mtune=generic -march=x86-64 -auxbase-strip -
   # -mieee-fp -mmmx -mno-sse4 -mpush-args -mred-zone -msse -msse2

   $ gcc -o - -m32 -S -O2 -fverbose-asm -x c /dev/null | grep sse
   # options passed:  -imultilib 32 /dev/null -m32 -mtune=generic -march=i586
   # -mfp-ret-in-387 -mfused-madd -mglibc -mieee-fp -mno-red-zone -mno-sse4

   $ gcc --version
   gcc (Debian 4.5.2-2) 4.5.2
   Copyright (C) 2010 Free Software Foundation, Inc.
   This is free software; see the source for copying conditions.  There is NO
   warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

-miles

-- 
Alliance, n. In international politics, the union of two thieves who have
their hands so deeply inserted in each other's pockets that they cannot
separately plunder a third.