+Wed Mar 9 15:23:19 1994 Jim Kingdon (kingdon@lioth.cygnus.com)
+
+ * stabsread.c (common_block_end, fix_common_block): Stash the
+ struct pending * in the SYMBOL_TYPE, not the SYMBOL_NAMESPACE, so
+ as to not assume that a pointer fits in an enum.
+
Wed Mar 9 18:56:36 1994 Kung Hsu (kung@mexican.cygnus.com)
* os9kread.c (fill_sym): check compiler verion number for pre-
* remote-os9k.c: fix bug in 'set remotebaud' function.
* remote-os9k.c (rombug_link): minimize checking so to improve
speed.
- * symfile.c (symfile_command): check if failed to link, also make
+ * symfile.c (symbol_file_command): check if failed to link, also make
the command be able to accept more than one filenames.
* target.c (target_link): check if failed to link with rombug.
* config/i386/tm-i386os9k.h : add #define DECR_PC_AFTER_BREAK 0.
for (j = common_block_i; j < common_block->nsyms; j++)
add_symbol_to_list (common_block->symbol[j], &new);
- SYMBOL_NAMESPACE (sym) = (enum namespace)((long) new);
+ SYMBOL_TYPE (sym) = (struct type *) new;
/* Should we be putting local_symbols back to what it was?
Does it matter? */
struct symbol *sym;
int valu;
{
- struct pending *next = (struct pending *) SYMBOL_NAMESPACE (sym);
+ struct pending *next = (struct pending *) SYMBOL_TYPE (sym);
for ( ; next; next = next->next)
{
register int j;
have the correct data for that slot yet.
The use of the LOC_BLOCK code in this chain is nonstandard--
- it refers to a FORTRAN common block rather than the usual meaning. */
+ it refers to a FORTRAN common block rather than the usual meaning, and
+ the such LOC_BLOCK symbols use their fields in nonstandard ways. */
EXTERN struct symbol *global_sym_chain[HASHSIZE];