On Sat, Dec 29, 2001 at 06:44:26PM -0700, Lawrence Gold wrote:
> With "-O2 -march=athlon -fstrict-aliasing -finline-limit=5000
> -fschedule-insns2", it runs just fine.
Why do you specify -fschedule-insns2 ? It should be already active with -O2.
Probably you already know it, but to check the optimization flags you can compile
a simple source file with the additional options :
-S -fverbose-asm
The resulting asm file will show all the passed and enabled options :
.file "flags.c"
# GNU C version 3.0.3 (i686-pc-linux-gnu)
# compiled by GNU C version 3.0.3.
# options passed: -lang-c -D__GNUC__=3 -D__GNUC_MINOR__=0
# -D__GNUC_PATCHLEVEL__=3 -D__ELF__ -Dunix -Dlinux -D__ELF__ -D__unix__
# -D__linux__ -D__unix -D__linux -Asystem=posix -D__OPTIMIZE__
# -D__STDC_HOSTED__=1 -Wall -Wno-sign-compare -Wno-unused -Acpu=i386
# -Amachine=i386 -Di386 -D__i386 -D__i386__ -D__i686 -D__i686__
# -D__pentiumpro -D__pentiumpro__ -D__tune_i686__ -D__tune_pentiumpro__
# -DNDEBUG -DUSE_ARCH_i686 -march=i686 -O3 -Wall -Wno-sign-compare
# -Wno-unused -fomit-frame-pointer -fstrict-aliasing -fverbose-asm
# options enabled: -fdefer-pop -fomit-frame-pointer
# -foptimize-sibling-calls -fcse-follow-jumps -fcse-skip-blocks
# -fexpensive-optimizations -fthread-jumps -fstrength-reduce -fpeephole
# -fforce-mem -ffunction-cse -finline-functions -finline
# -fkeep-static-consts -fcaller-saves -fpcc-struct-return -fgcse
# -frerun-cse-after-loop -frerun-loop-opt -fdelete-null-pointer-checks
# -fschedule-insns2 -fsched-interblock -fsched-spec -fbranch-count-reg
# -freorder-blocks -frename-registers -fcommon -fverbose-asm -fgnu-linker
# -fregmove -foptimize-register-move -fargument-alias -fstrict-aliasing
# -fident -fpeephole2 -fguess-branch-probability -fmath-errno -m80387
# -mhard-float -mno-soft-float -mieee-fp -mfp-ret-in-387 -march=i686
Anyway, now I'm compiling AdvanceMAME for Linux with these options with gcc 3.0.3:
-O3 -fomit-frame-pointer -fstrict-aliasing -march=? (i686 for me)
The only exception is the z80.c file which requires :
-O2 -fomit-frame-pointer -fstrict-aliasing -march=?
otherwise it doesn't compile.
--
Andrea Mazzoleni
935A 2D3C 5C70 BCD6 CB0C ED89 7C19 4321 6340 3F6D
_______________________________________________
Xmame mailing list
[EMAIL PROTECTED]
http://toybox.twisted.org.uk/mailman/listinfo/xmame