This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

3.1 regrename.c doesn't handle which_alternative==-1


I just came across a 3.1 ICE in regrename.c, see

http://gcc.gnu.org/ml/gcc-prs/2002-04/msg01616.html

which seems to happen because nothing in regrename.c handles the
case which_alternative==-1 (which seems to indicate a failure in
some other part of the program); here is some backtrace info that
I neglected to put into the original PR:

(gdb) backtrace
#0  copyprop_hardreg_forward_1 (bb=0x13bd468, vd=0x1df46bc)
    at ../../gcc/regrename.c:1564
#1  0x00244810 in copyprop_hardreg_forward () at ../../gcc/regrename.c:1750
#2  0x001afcc8 in rest_of_compilation (decl=0x7dc980)
    at ../../gcc/toplev.c:3328
#3  0x00044b7c in c_expand_body (fndecl=0x7dc980, nested_p=0, can_defer_p=1)
    at ../../gcc/c-decl.c:7110
#4  0x00044914 in finish_function (nested=0, can_defer_p=1)
    at ../../gcc/c-decl.c:6977
#5  0x00032378 in yyparse_1 () at c-parse.y:386
#6  0x001ae3fc in compile_file () at ../../gcc/toplev.c:2120
#7  0x001b3384 in do_compile () at ../../gcc/toplev.c:5201
#8  0x001b3414 in toplev_main (argc=10, argv=0xffbef7a4)
    at ../../gcc/toplev.c:5234

Is copyprop_hardreg_forward supposed to do anything at all when
which_alternative is -1?

Brad


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]