[RFC] Stabs parsing regression from GDB 6.6 to GDB 6.6.90

[Pedro Alves]

Daniel Jacobowitz wrote:
> If no one objects, I'll approve such a reversion for the branch and
> your patch for trunk.  Give it a day or two in case someone else has
> a better idea.

Just finished testing the reversion on 6.7 branch.  No results changes,
and I've confirmed that the 'unsigned long long' problem is cured by it.

I've checked this in to the branch ...

2007-10-08  Pedro Alves  <pedro_alves@portugalmail.pt>

         Revert:
         2007-08-10  Michael Snyder  <msnyder@access-company.com>

         * stabsread.c (read_huge_number): Attempt to compute value before
         values that it depends on.

... and these to head:

2007-10-09  Pedro Alves  <pedro_alves@portugalmail.pt>

         * stabsread.c (read_huge_number): Fix handling of octal
         representation when the bit width is known.
         (read_range_type): Record unsigned integral types with their size,
         when the type size is known.

2007-10-09  Pedro Alves  <pedro_alves@portugalmail.pt>

	* gdb.base/whatis.c (v_long_long, v_signed_long_long)
	(v_unsigned_long_long, v_long_long_array)
	(v_signed_long_long_array, v_unsigned_long_long_array)
	(slong_long_addr, a_slong_long_addr, v_long_long_pointer)
	(v_signed_long_long_pointer, v_unsigned_long_long_pointer)
	[!NO_LONG_LONG]: New.
	(t_struct, v_struct2, t_union, v_union2) [!NO_LONG_LONG]: Add
	v_long_long_member.

	(v_long_long_func, v_signed_long_long_func)
	(v_unsigned_long_long_func) [!NO_LONG_LONG]: New.
	(main) [!NO_LONG_LONG]: Initialize long long variants.

	* gdb.base/whatis.exp: If board file requests no_long_long, build
	test with NO_LONG_LONG defined.  Test long long, signed long long,
	and unsigned long long variants but only if board file doesn't
	disable it.

Full patches attached.

---

Testing on cygwin looks much better on the 6.7 branch compared to the
version currently shipped with Cygwin.  Here are .sum results
with 'set_board_info gdb,nosignals 1' - cygwin signals aren't
supported yet.

==> gdb-20060706-2/testresults/gdb.sum <==
# of expected passes            8855
# of unexpected failures        343
# of unexpected successes       1
# of expected failures          45
# of known failures             52
# of unresolved testcases       17
# of untested testcases         17
# of unsupported tests          30
/cygdrive/d/gdb/gdb-20060706-2/build/gdb/testsuite/../../gdb/gdb version 
6.5.50.20060706-cvs -nx

==> gdb-6.7/testresults/reverted/gdb.sum <==
# of expected passes            9471
# of unexpected failures        170
# of unexpected successes       1
# of expected failures          47
# of known failures             30
# of unresolved testcases       1
# of untested testcases         13
# of unsupported tests          32
/cygdrive/d/gdb/gdb-6.7/build-cygwin/gdb/testsuite/../../gdb/gdb version 
6.6.90.20071007-cvs -nx

-- 

Cheers,
Pedro Alves
-------------- next part --------------
A non-text attachment was scrubbed...
Name: revert_read_huge_number.diff
Type: text/x-diff
Size: 1026 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/gdb-patches/attachments/20071008/5c615281/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: read_huge_number2.diff
Type: text/x-diff
Size: 5161 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/gdb-patches/attachments/20071008/5c615281/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: whatis_long_long.diff
Type: text/x-diff
Size: 8184 bytes
Desc: not available
URL: <http://sourceware.org/pipermail/gdb-patches/attachments/20071008/5c615281/attachment-0002.bin>