Thread View: gnu.gcc.bug
1 messages
1 total messages
Started by Walter Hunt
Wed, 02 Sep 1992 00:28
gcc 2.2.2 and abort()
Author: Walter Hunt
Date: Wed, 02 Sep 1992 00:28
Date: Wed, 02 Sep 1992 00:28
63 lines
2693 bytes
2693 bytes
@ I recently compiled up gcc 2.2.2 for my machine, with the
intent of doing some C++ programming. While compiling, I found one
of the files in my distribution to be corrupted. There may have
been further corruptions, but if there were, they resulted in
valid 'C' code since a "make bootstrap" completed fine. This may
or may not have anything to do with my results.
I then retrieved libg++-2.0, in order to have some basic C++
classes. While attempting to compile it, gcc (actually cc1plus)
bombed out with an abort at line 207 of file cp-tree.c. (Like so:
(gdb) run
Starting program: /usr/local/lib/gcc-lib/sun4/2.2.2/cc1plus /usr/tmp/cca28407.i -quiet -dumpbase BitSet.cc -version -o /usr/tmp/cca28407.s
GNU C++ version 2.2.2 (sparc) compiled by GNU C version 2.2.2.
Program received signal 6, Abort
0xf775c120 in kill ()
(gdb) where
#0 0xf775c120 in kill ()
#1 0xf777447c in abort ()
#2 0x3a7b8 in break_out_calls (exp=0x1770b8) at cp-tree.c:207
#3 0x3a7e0 in break_out_calls (exp=0x1770c8) at cp-tree.c:217
#4 0x3a848 in break_out_calls (exp=0x1770e8) at cp-tree.c:234
#5 0x3464c in build_modify_expr (lhs=0x1770e8, modifycode=NOP_EXPR,
rhs=0x1770a8) at cp-typeck.c:5157
#6 0xe894 in yyparse () at cp-parse.y:1211
#7 0x7b9c0 in compile_file (name=0xf7fff992 "/usr/tmp/cca28407.i")
at toplev.c:1638
#8 0x7e584 in main (argc=8, argv=0xf7fff8e4, envp=0xf7fff908) at toplev.c:3055
(gdb)
After some quick #if 0's, I determined that the line that was
making it croak was line 768 of BitSettoa.cc, which contained:
*fmt++ = (a & 1)? t : f;
Changing the line (and two more just like it in the same
routine) to either of:
if (a & 1) *fmt++ = t;
else *fmt++ = f;
or
*fmt = (a & 1)? t : f;
fmt++;
would compile fine.
Does this happen with the "official" 2.2.2 (i.e. do I have a
improperly built version)? Any chance of getting on a mailing list
for patches?
If it matters, I got all my sources from a uunet tape,
purchased a couple months ago.
Walter Hunt walter@optimage.com
OptImage Interactive Services Company, L.P. walter@aimla.com
11040 Santa Monica Blvd., Suite #300 uunet!{optimage,aimla}!walter
Los Angeles, Ca. 90025 (310) 445-5718
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads