target/8659: trying to delete frame instruction with -fpic

[Franz Sirl]

At 18:39 27.11.2002, Marcus Meissner wrote:
>options passed:  -fpreprocessed -O2 -fpic
>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
>  -fkeep-static-consts -fcaller-saves -freg-struct-return -fgcse -fgcse-lm
>  -fgcse-sm -frerun-cse-after-loop -frerun-loop-opt
>  -fdelete-null-pointer-checks -fschedule-insns -fschedule-insns2
>  -fsched-interblock -fsched-spec -fbranch-count-reg -fpic -freorder-blocks
>  -fcprop-registers -fcommon -fgnu-linker -fregmove -foptimize-register-move
>  -fargument-alias -fstrict-aliasing -fmerge-constants -fident -fpeephole2
>  -fguess-branch-probability -fmath-errno -ftrapping-math
>  -fnew-ra-pre-reload -mpowerpc -mnew-mnemonics -msched-prolog
   ^^^^^^^^^^^^^^^^^^^^

Wow, that's the new Register-Allocator present as an experimental feature 
in mainline (3.3exp), it probably won't be officially supported and fully 
tested before 3.4. So I guess Suse backported it to 3.2.1, which is nice 
for developers, but I strongly hope this didn't make it into an Suse 
distribution release...

Franz.