Re: [new-regalloc-branch] bootstrap failure on SPARC

[Daniel Berlin <dan at cgsoftware dot com>]

Dan Nicolaescu <dann@godzilla.ICS.UCI.EDU> writes:

> On sparc-sun-solaris2.8
> 
> configured with:
> 
> configure --disable-nls --disable-multilib --enable-languages=c
> 
> I get:
> 
> stage1/xgcc -Bstage1/ -B/home/dann/gcc-newRA/sparc-sun-solaris2.8/bin/ -c  -DIN_GCC    -g -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wtraditional -pedantic -Wno-long-long  -DHAVE_CONFIG_H    -I. -I. -I../../gcc-CVS/gcc -I../../gcc-CVS/gcc/. -I../../gcc-CVS/gcc/config -I../../gcc-CVS/gcc/../include ../../gcc-CVS/gcc/cpplex.c -o cpplex.o
> ../../gcc-CVS/gcc/cpplex.c: In function `cpp_parse_escape':
> ../../gcc-CVS/gcc/cpplex.c:1883: Internal compiler error in subreg_hard_regno, at emit-rtl.c:713
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
> 
> about 20 or so other files fail at this stage with the same ICE in subreg_hard_regno, at emit-rtl.c:713
> all other files are compiled without an error. 
Whee.
This is because denis removed a supposedly incorrect test in
get_free_reg.
I don't think it was really incorrect.
Reverting his patch makes these failures go away.
Denis, what exactly about the get_free_reg test was incorrect?

> 
> If needed I can send the preprocessed file. 
> 
>         --dan

-- 
"Years ago, I worked in a natural, organic health food store in
Seattle, Washington.  One day a man walked in and asked, "If I
can melt dry ice, can I swim without getting wet?"  Two days
later I was fired for eating cotton candy and drinking straight
Bosco on the job.
"-Steven Wright