Re: [PATCH PR69652, Regression]

[Richard Biener <richard dot guenther at gmail dot com>]

On Fri, Feb 5, 2016 at 3:54 PM, Yuri Rumyantsev <ysrumyan@gmail.com> wrote:
> Hi All,
>
> Here is updated patch - I came back to move call statements also since
>  masked loads are presented by internal call. I also assume that for
> the following simple loop
>   for (i = 0; i < n; i++)
>     if (b1[i])
>       a1[i] = sqrtf(a2[i] * a2[i] + a3[i] * a3[i]);
> motion must be done for all vector statements in semi-hammock including SQRT.
>
> Bootstrap and regression testing did not show any new failures.
> Is it OK for trunk?

The patch is incredibly hard to parse due to the re-indenting.  Please
consider sending
diffs with -b.

This issue exposes that you are moving (masked) stores across loads without
checking aliasing.  In the specific case those loads are dead and thus
this is safe
but in general I thought we were checking that we are using the same VUSE
during the sinking operation.

Thus, I'd rather have

+             /* Check that LHS does not have uses outside of STORE_BB.  */
+             res = true;
+             FOR_EACH_IMM_USE_FAST (use_p, imm_iter, lhs)
+               {
+                 gimple *use_stmt;
+                 use_stmt = USE_STMT (use_p);
+                 if (is_gimple_debug (use_stmt))
+                   continue;
+                 if (gimple_bb (use_stmt) != store_bb)
+                   {
+                     res = false;
+                     break;
+                   }
+               }

also check for the dead code case and DCE those stmts here.  Like so:

   if (has_zero_uses (lhs))
    {
      gsi_remove (&gsi_from, true);
      continue;
    }

before the above loop.

Richard.

> ChangeLog:
>
> 2016-02-05  Yuri Rumyantsev  <ysrumyan@gmail.com>
>
> PR tree-optimization/69652
> * tree-vect-loop.c (optimize_mask_stores): Move declaration of STMT1
> to nested loop, introduce new SCALAR_VUSE vector to keep vuse of all
> skipped scalar statements, introduce variable LAST_VUSE to keep
> vuse of LAST_STORE, add assertion that SCALAR_VUSE is empty in the
> begining of current masked store processing, did source re-formatting,
> skip parsing of debug gimples, stop processing if a gimple with
> volatile operand has been encountered, save scalar statement
> with vuse in SCALAR_VUSE, skip processing debug statements in IMM_USE
> iterator, change vuse of all saved scalar statements to LAST_VUSE if
> it makes sence.
>
> gcc/testsuite/ChangeLog:
> * gcc.dg/torture/pr69652.c: New test.
>
> 2016-02-04 19:40 GMT+03:00 Jakub Jelinek <jakub@redhat.com>:
>> On Thu, Feb 04, 2016 at 05:46:27PM +0300, Yuri Rumyantsev wrote:
>>> Here is a patch that cures the issues with non-correct vuse for scalar
>>> statements during code motion, i.e. if vuse of scalar statement is
>>> vdef of masked store which has been sunk to new basic block, we must
>>> fix it up.  The patch also fixed almost all remarks pointed out by
>>> Jacub.
>>>
>>> Bootstrapping and regression testing on v86-64 did not show any new failures.
>>> Is it OK for trunk?
>>>
>>> ChangeLog:
>>> 2016-02-04  Yuri Rumyantsev  <ysrumyan@gmail.com>
>>>
>>> PR tree-optimization/69652
>>> * tree-vect-loop.c (optimize_mask_stores): Move declaration of STMT1
>>> to nested loop, introduce new SCALAR_VUSE vector to keep vuse of all
>>> skipped scalar statements, introduce variable LAST_VUSE that has
>>> vuse of LAST_STORE, add assertion that SCALAR_VUSE is empty in the
>>> begining of current masked store processing, did source re-formatting,
>>> skip parsing of debug gimples, stop processing when call or gimple
>>> with volatile operand habe been encountered, save scalar statement
>>> with vuse in SCALAR_VUSE, skip processing debug statements in IMM_USE
>>> iterator, change vuse of all saved scalar statements to LAST_VUSE if
>>> it makes sence.
>>>
>>> gcc/testsuite/ChangeLog:
>>> * gcc.dg/torture/pr69652.c: New test.
>>
>> Your mailer breaks ChangeLog formatting, so it is hard to check the
>> formatting of the ChangeLog entry.
>>
>> diff --git a/gcc/testsuite/gcc.dg/torture/pr69652.c b/gcc/testsuite/gcc.dg/torture/pr69652.c
>> new file mode 100644
>> index 0000000..91f30cf
>> --- /dev/null
>> +++ b/gcc/testsuite/gcc.dg/torture/pr69652.c
>> @@ -0,0 +1,14 @@
>> +/* { dg-do compile } */
>> +/* { dg-options "-O2 -ffast-math -ftree-vectorize " } */
>> +/* { dg-additional-options "-mavx" { target { i?86-*-* x86_64-*-* } } } */
>> +
>> +void fn1(double **matrix, int column, int row, int n)
>> +{
>> +  int k;
>> +  for (k = 0; k < n; k++)
>> +    if (matrix[row][k] != matrix[column][k])
>> +      {
>> +       matrix[column][k] = -matrix[column][k];
>> +       matrix[row][k] = matrix[row][k] - matrix[column][k];
>> +      }
>> +}
>> \ No newline at end of file
>>
>> Please make sure the last line of the test is a new-line.
>>
>> @@ -6971,6 +6972,8 @@ optimize_mask_stores (struct loop *loop)
>>            gsi_next (&gsi))
>>         {
>>           stmt = gsi_stmt (gsi);
>> +         if (is_gimple_debug (stmt))
>> +           continue;
>>           if (is_gimple_call (stmt)
>>               && gimple_call_internal_p (stmt)
>>               && gimple_call_internal_fn (stmt) == IFN_MASK_STORE)
>>
>> This is not needed, you do something only for is_gimple_call,
>> which is never true if is_gimple_debug, so the code used to be fine as is.
>>
>> +             /* Skip debug sstatements.  */
>>
>> s/ss/s/
>>
>> +             if (is_gimple_debug (gsi_stmt (gsi)))
>> +               continue;
>> +             stmt1 = gsi_stmt (gsi);
>> +             /* Do not consider writing to memory,volatile and call
>>
>> Missing space after ,
>>
>> +             /* Skip scalar statements.  */
>> +             if (!VECTOR_TYPE_P (TREE_TYPE (lhs)))
>> +               {
>> +                 /* If scalar statement has vuse we need to modify it
>> +                    when another masked store will be sunk.  */
>> +                 if (gimple_vuse (stmt1))
>> +                   scalar_vuse.safe_push (stmt1);
>>                   continue;
>> +               }
>>
>> I don't think it is safe to take for granted that the scalar stmts are all
>> going to be DCEd, but I could be wrong.
>>
>> +             /* Check that LHS does not have uses outside of STORE_BB.  */
>> +             res = true;
>> +             FOR_EACH_IMM_USE_FAST (use_p, imm_iter, lhs)
>> +               {
>> +                 gimple *use_stmt;
>> +                 use_stmt = USE_STMT (use_p);
>> +                 if (is_gimple_debug (use_stmt))
>> +                   continue;
>>
>> Ignoring debug stmts to make decision whether you move or not is
>> of course the right thing to do.  But IMHO you should remember if
>> you saw any is_gimple_debug stmts in some bool var.
>>
>> +                 if (gimple_bb (use_stmt) != store_bb)
>> +                   {
>> +                     res = false;
>> +                     break;
>> +                   }
>> +               }
>> +             if (!res)
>> +               break;
>>
>> -               if (gimple_vuse (stmt1)
>> -                   && gimple_vuse (stmt1) != gimple_vuse (last_store))
>> -                 break;
>> +             if (gimple_vuse (stmt1)
>> +                 && gimple_vuse (stmt1) != gimple_vuse (last_store))
>> +               break;
>>
>> +             /* Can move STMT1 to STORE_BB.  */
>> +             if (dump_enabled_p ())
>> +               {
>> +                 dump_printf_loc (MSG_NOTE, vect_location,
>> +                                  "Move stmt to created bb\n");
>> +                 dump_gimple_stmt (MSG_NOTE, TDF_SLIM, stmt1, 0);
>> +               }
>>
>> And if yes, invalidate them here, because the move would otherwise
>> generate invalid IL.
>>
>> +             gsi_move_before (&gsi_from, &gsi_to);
>> +             /* Shift GSI_TO for further insertion.  */
>> +             gsi_prev (&gsi_to);
>> +           }
>> +         /* Put other masked stores with the same mask to STORE_BB.  */
>> +         if (worklist.is_empty ()
>> +             || gimple_call_arg (worklist.last (), 2) != mask
>> +             || worklist.last () != stmt1)
>> +           break;
>> +         last = worklist.pop ();
>>         }
>>        add_phi_arg (phi, gimple_vuse (last_store), e, UNKNOWN_LOCATION);
>> +      /* Mask stores motion could crossing scalar statements with vuse
>> +        which should be corrected.  */
>>
>> s/crossing/cross/
>> That said, I'm not really sure if without some verification if such
>> reads are really dead it is safe to skip them and update this way.
>> Richard?
>>
>> +      last_vuse = gimple_vuse (last_store);
>> +      while (!scalar_vuse.is_empty ())
>> +       {
>> +          stmt = scalar_vuse.pop ();
>> +          if (gimple_vuse (stmt) != last_vuse)
>> +             {
>> +               gimple_set_vuse (stmt, last_vuse);
>> +               update_stmt (stmt);
>> +             }
>> +       }
>>      }
>>  }
>>
>>         Jakub