OpenE2K
/
binutils-gdb
12
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

* buildsym.c (finish_block): If we pop the context stack and it is 

not empty, complain () instead of abort ().
This commit is contained in:
Jim Kingdon 1993-10-18 15:40:04 +00:00
parent f843c95fc3
commit f91b837dd0
2 changed files with 24 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
gdb/ChangeLog
View File

@ -1,3 +1,19 @@
Mon Oct 18 10:28:08 1993 Jim Kingdon (kingdon@lioth.cygnus.com)
* buildsym.c (finish_block): If we pop the context stack and it is
not empty, complain () instead of abort ().
Sun Oct 17 19:42:31 1993 Jim Kingdon (kingdon@lioth.cygnus.com)
* parse.c, parser-defs.h (follow_types): New function.
* c-exp.y (ptype : typebase abs_decl): Use it.
* c-exp.y (ptype): Add support for type qualifiers after the
typebase. The typebase rule already has support for them before
the typebase.
* Makefile.in: Change the expected number of shift/reduce
conflicts to 6. This is OK--the 2 new conflicts are basically the
same as one of the old ones.
Sun Oct 17 13:04:49 1993 Fred Fish (fnf@cygnus.com)
* Makefile.in (VERSION): Bump to 4.10.3.

11
gdb/buildsym.c
View File

@ -667,11 +667,16 @@ end_symtab (end_addr, sort_pending, sort_linevec, objfile, section)
finish_block (cstk->name, &local_symbols, cstk->old_blocks,
cstk->start_addr, end_addr, objfile);
/* Debug: if context stack still has something in it,
we are in trouble. */
if (context_stack_depth > 0)
{
abort ();
/* This is said to happen with SCO. The old coffread.c code
simply emptied the context stack, so we do the same. FIXME:
Find out why it is happening. This is not believed to happen
in most cases (even for coffread.c); it used to be an abort(). */
static struct complaint msg =
{"Context stack not empty in end_symtab", 0, 0};
complain (&msg);
context_stack_depth = 0;
}
}