Re: regression tester mail...

[Geoff Keating <geoffk at geoffk dot org>]

Eric Christopher <echristo@redhat.com> writes:

> I was just running some tests and found a bunch of things failing that
> hadn't previously for mips (and others), and not sure when this happened
> (other than the last couple of days). The unordered failure for mips is
> mine - anyone else want to claim responsibility or know what patch
> caused all of the regressions off hand?

powerpc-eabisim g77.sum g77.f-torture/execute/19981119-0.f
powerpc-eabisim gcc.sum gcc.c-torture/compile/20010903-1.c,
powerpc-eabisim gcc.sum gcc.c-torture/execute/loop-2.c
powerpc-eabisim gcc.sum gcc.c-torture/execute/loop-2b.c
mips-elf gcc.sum gcc.c-torture/compile/20010903-1.c,
mips-elf gcc.sum gcc.c-torture/execute/931004-10.c
mips-elf gcc.sum gcc.c-torture/execute/931004-11.c
mips-elf gcc.sum gcc.c-torture/execute/931004-12.c
mips-elf gcc.sum gcc.c-torture/execute/931004-9.c
mips-elf gcc.sum gcc.c-torture/unsorted/loop386.c,
native gcc.sum gcc.c-torture/compile/20010903-1.c,

seem to be one of these:

+2002-07-21  Richard Henderson  <rth@redhat.com>
+
+       * expr.c (expand_expr) [TRY_FINALLY_EXPR]: Don't use
+       GOTO_SUBROUTINE_EXPR when finally_block can be re-expanded.
+
+2002-07-21  Richard Henderson  <rth@redhat.com>
+
+       * unroll.c (find_splittable_givs): Do not split DEST_ADDR givs
+       that are not unrolled completely.
+
+2002-07-21  Richard Henderson  <rth@redhat.com>
+
+       * loop.h (LOOP_AUTO_UNROLL): Rename from LOOP_FIRST_PASS.
+       * loop.c (strength_reduce): Update.
+       * toplev.c (rest_of_compilation): Do unrolling in the first
+       loop pass, not the second.
+
+2002-07-21  Richard Henderson  <rth@redhat.com>
+
+       * emit-rtl.c (set_mem_attributes): Preserve indirection of PARM_DECL
+       when flag_argument_noalias == 2.
+       * alias.c (nonoverlapping_memrefs_p): Handle that.
+       * print-rtl.c (print_mem_expr): Likewise.
+

mips-elf gcc.sum gcc.c-torture/execute/921029-1.c
native gcc.sum gcc.dg/20010912-1.c

is this:

+Sun Jul 21 00:54:54 CEST 2002  Jan Hubicka  <jh@suse.cz>
+
+       * gcse.c: Include cselib.h
+       (constptop_register): Break out from ...
+       (cprop_insn): ... here; kill basic_block argument.
+       (do_local_cprop, local_cprop_pass): New functions.
+       (one_cprop_pass): Call local_cprop_pass.
+

powerpc-eabisim gcc.sum gcc.dg/20010822-1.c
native gcc.sum gcc.dg/20010822-1.c

is this:

+Thu Jul 18 09:38:59 CEST 2002  Jan Hubicka  <jh@suse.cz>
+
+     * gcse.c (hoist_expr_reaches_here_p):  Stop once expr_bb is reached.
+
+     * gcse.c (try_replace_reg): Do not return false positives.
+

and finally,

mips-elf gcc.sum gcc.dg/unordered-1.c

is this:

+2002-07-03  Eric Christopher  <echristo@redhat.com>
+
+       * config/mips/linux.h: Add #undef for SUBTARGET_CPP_SPEC.
+       * config/mips/mips.h: Remove deprecated -m<processor> options
+       and cc1_cpu_spec associated.
+       (CONSTANT_ADDRESS_P): Fix last patch.
+       (ASM_DECLARE_FUNCTION_NAME): Declare. Fix comment.
+       * config/mips/mips.md (bungt, bunge, sungt_df, sungt_sf, sunge_df,
+       sunge_sf): Remove.
+

-- 
- Geoffrey Keating <geoffk@geoffk.org> <geoffk@redhat.com>