Re: [PATCH GCC]Allow cfgcleanup to remove forwarder loop preheaders and latches

["H.J. Lu" <hjl dot tools at gmail dot com>]

On Sat, Mar 1, 2014 at 2:19 AM, Richard Biener
<richard.guenther@gmail.com> wrote:
> On Fri, Feb 28, 2014 at 7:23 PM, H.J. Lu <hjl.tools@gmail.com> wrote:
>> On Fri, Feb 28, 2014 at 9:42 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
>>> On Fri, Feb 28, 2014 at 9:25 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
>>>> On Fri, Feb 28, 2014 at 8:11 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
>>>>> On Fri, Feb 28, 2014 at 2:09 AM, Richard Biener
>>>>> <richard.guenther@gmail.com> wrote:
>>>>>> On Fri, Feb 28, 2014 at 10:09 AM, Richard Biener
>>>>>> <richard.guenther@gmail.com> wrote:
>>>>>>> On Fri, Feb 28, 2014 at 1:52 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
>>>>>>>> On Mon, Feb 24, 2014 at 9:12 PM, bin.cheng <bin.cheng@arm.com> wrote:
>>>>>>>>> Hi,
>>>>>>>>> This patch is to fix regression reported in PR60280 by removing forward loop
>>>>>>>>> headers/latches in cfg cleanup if possible.  Several tests are broken by
>>>>>>>>> this change since cfg cleanup is shared by all optimizers.  Some tests has
>>>>>>>>> already been fixed by recent patches, I went through and fixed the others.
>>>>>>>>> One case needs to be clarified is "gcc.dg/tree-prof/update-loopch.c".  When
>>>>>>>>> GCC removing a basic block, it checks profile information by calling
>>>>>>>>> check_bb_profile after redirecting incoming edges of the bb.  This certainly
>>>>>>>>> results in warnings about invalid profile information and causes the case to
>>>>>>>>> fail.  I will send a patch to skip checking profile information for a
>>>>>>>>> removing basic block in stage 1 if it sounds reasonable.  For now I just
>>>>>>>>> twisted the case itself.
>>>>>>>>>
>>>>>>>>> Bootstrap and tested on x86_64 and arm_a15.
>>>>>>>>>
>>>>>>>>> Is it OK?
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> 2014-02-25  Bin Cheng  <bin.cheng@arm.com>
>>>>>>>>>
>>>>>>>>>         PR target/60280
>>>>>>>>>         * tree-cfgcleanup.c (tree_forwarder_block_p): Protect loop
>>>>>>>>>         preheaders and latches only if requested.  Fix latch if it
>>>>>>>>>         is removed.
>>>>>>>>>         * tree-ssa-dom.c (tree_ssa_dominator_optimize): Set
>>>>>>>>>         LOOPS_HAVE_PREHEADERS.
>>>>>>>>>
>>>>>>>>
>>>>>>>> This change:
>>>>>>>>
>>>>>>>>         if (dest->loop_father->header == dest)
>>>>>>>> -  return false;
>>>>>>>> +  {
>>>>>>>> +    if (loops_state_satisfies_p (LOOPS_HAVE_PREHEADERS)
>>>>>>>> +        && bb->loop_father->header != dest)
>>>>>>>> +      return false;
>>>>>>>> +
>>>>>>>> +    if (loops_state_satisfies_p (LOOPS_HAVE_SIMPLE_LATCHES)
>>>>>>>> +        && bb->loop_father->header == dest)
>>>>>>>> +      return false;
>>>>>>>> +  }
>>>>>>>>      }
>>>>>>>>
>>>>>>>> miscompiled 435.gromacs in SPEC CPU 2006 on x32 with
>>>>>>>>
>>>>>>>> -O3 -funroll-loops -ffast-math -fwhole-program -flto=jobserver
>>>>>>>> -fuse-linker-plugin
>>>>>>>>
>>>>>>>> This patch changes loops without LOOPS_HAVE_PREHEADERS
>>>>>>>> nor LOOPS_HAVE_SIMPLE_LATCHES from returning false to returning
>>>>>>>> true.  I don't have a small testcase.  But this patch:
>>>>>>>>
>>>>>>>> diff --git a/gcc/tree-cfgcleanup.c b/gcc/tree-cfgcleanup.c
>>>>>>>> index b5c384b..2ba673c 100644
>>>>>>>> --- a/gcc/tree-cfgcleanup.c
>>>>>>>> +++ b/gcc/tree-cfgcleanup.c
>>>>>>>> @@ -323,6 +323,10 @@ tree_forwarder_block_p (basic_block bb, bool phi_wanted)
>>>>>>>>      if (loops_state_satisfies_p (LOOPS_HAVE_SIMPLE_LATCHES)
>>>>>>>>          && bb->loop_father->header == dest)
>>>>>>>>        return false;
>>>>>>>> +
>>>>>>>> +    if (!loops_state_satisfies_p (LOOPS_HAVE_PREHEADERS)
>>>>>>>> +        && !loops_state_satisfies_p (LOOPS_HAVE_SIMPLE_LATCHES))
>>>>>>>> +      return false;
>>>>>>>>    }
>>>>>>>>      }
>>>>>>>>
>>>>>>>> fixes the regression.  Does it make any senses?
>>>>>>>
>>>>>>> I think the preheader test isn't fully correct (bb may be in an inner loop
>>>>>>> for example).  So a more conservative variant would be
>>>>>>>
>>>>>>> Index: gcc/tree-cfgcleanup.c
>>>>>>> ===================================================================
>>>>>>> --- gcc/tree-cfgcleanup.c       (revision 208169)
>>>>>>> +++ gcc/tree-cfgcleanup.c       (working copy)
>>>>>>> @@ -316,13 +316,13 @@ tree_forwarder_block_p (basic_block bb,
>>>>>>>        /* Protect loop preheaders and latches if requested.  */
>>>>>>>        if (dest->loop_father->header == dest)
>>>>>>>         {
>>>>>>> -         if (loops_state_satisfies_p (LOOPS_HAVE_PREHEADERS)
>>>>>>> -             && bb->loop_father->header != dest)
>>>>>>> -           return false;
>>>>>>> -
>>>>>>> -         if (loops_state_satisfies_p (LOOPS_HAVE_SIMPLE_LATCHES)
>>>>>>> -             && bb->loop_father->header == dest)
>>>>>>> -           return false;
>>>>>>> +         if (bb->loop_father == dest->loop_father)
>>>>>>> +           return !loops_state_satisfies_p (LOOPS_HAVE_SIMPLE_LATCHES);
>>>>>>> +         else if (bb->loop_father == loop_outer (dest->loop_father))
>>>>>>> +           return !loops_state_satisfies_p (LOOPS_HAVE_PREHEADERS);
>>>>>>> +         /* Always preserve other edges into loop headers that are
>>>>>>> +            not simple latches or preheaders.  */
>>>>>>> +         return false;
>>>>>>>         }
>>>>>>>      }
>>>>>>>
>>>>>>> that makes sure we can properly update loop information.  It's also
>>>>>>> a more conservative change at this point which should still successfully
>>>>>>> remove simple latches and preheaders created by loop discovery.
>>>>>>
>>>>>> I think the patch makes sense anyway and thus I'll install it once it
>>>>>> passed bootstrap / regtesting.
>>>>>>
>>>>>> Another fix that may make sense is to restrict it to
>>>>>> !loops_state_satisfies_p (LOOPS_NEED_FIXUP), though cfgcleanup
>>>>>> itself can end up setting that ... which we eventually should fix if it
>>>>>> still happens.  That is, check if
>>>>>>
>>>>>> Index: gcc/tree-cfgcleanup.c
>>>>>> ===================================================================
>>>>>> --- gcc/tree-cfgcleanup.c       (revision 208169)
>>>>>> +++ gcc/tree-cfgcleanup.c       (working copy)
>>>>>>
>>>>>> @@ -729,8 +729,9 @@ cleanup_tree_cfg_noloop (void)
>>>>>>
>>>>>>    timevar_pop (TV_TREE_CLEANUP_CFG);
>>>>>>
>>>>>> -  if (changed && current_loops)
>>>>>> -    loops_state_set (LOOPS_NEED_FIXUP);
>>>>>> +  if (changed && current_loops
>>>>>> +      && !loops_state_satisfies_p (LOOPS_NEED_FIXUP))
>>>>>> +    verify_loop_structure ();
>>>>>>
>>>>>>    return changed;
>>>>>>  }
>>>>>>
>>>>>> trips anywhere (and apply fixes).  That's of course not appropriate at
>>>>>> this stage.
>>>>>>
>>>>>>> Does it fix 435.gromacs?
>>>>>
>>>>> I tried revision 208222 and it doesn't fix 435.gromacs.
>>>>
>>>> Remove
>>>>
>>>>           else if (bb->loop_father == loop_outer (dest->loop_father))
>>>>             return !loops_state_satisfies_p (LOOPS_HAVE_PREHEADERS);
>>>
>>> Should we also check other loop state, like LOOPS_HAVE_SIMPLE_LATCHES
>>> or LOOPS_MAY_HAVE_MULTIPLE_LATCHES here?
>>>
>>
>> This patch:
>>
>> diff --git a/gcc/tree-cfgcleanup.c b/gcc/tree-cfgcleanup.c
>> index 926d300..fb1c63d 100644
>> --- a/gcc/tree-cfgcleanup.c
>> +++ b/gcc/tree-cfgcleanup.c
>> @@ -328,7 +328,9 @@ tree_forwarder_block_p (basic_block bb, bool phi_wanted)
>>                   (LOOPS_MAY_HAVE_MULTIPLE_LATCHES));
>>        }
>>      else if (bb->loop_father == loop_outer (dest->loop_father))
>> -      return !loops_state_satisfies_p (LOOPS_HAVE_PREHEADERS);
>> +      return (!loops_state_satisfies_p (LOOPS_HAVE_PREHEADERS)
>> +         && !loops_state_satisfies_p
>> +          (LOOPS_MAY_HAVE_MULTIPLE_LATCHES));
>>      /* Always preserve other edges into loop headers that are
>>         not simple latches or preheaders.  */
>>      return false;
>>
>> works.  Does it look right?
>
> No.  Latches and pre-headers are independent.  The above effectively
> disables removing preheaders.
>
> Unfortunately I don't have a x32 runtime environment to reproduce the
> issue.
>
> So, can you narrow it down by bisecting the actual preheader removal?
> Like, add a static counter for the number of times returned true above
> and return false if it is between getenv("from") and getenv("to") and
> then narrow down from and to?  Then look at the actual transform
> done if it hopefully narrows down to a single removal ... (should be
> enough to do that for the ltrans stage, so eventually you can narrow
> it down to a specific ltrans unit first).

I opened:

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60418

-- 
H.J.