Re: opt-info message change for vectorizer

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

On Wed, Aug 28, 2013 at 8:14 PM, Xinliang David Li <davidxl@google.com> wrote:
> Fixed as requested. I don't like the extra newline either, but I will
> leave that to Teresa.
>
> basic3.c:8:foo: note: loop vectorized
>
> basic3.c:8:foo: note: loop versioned for vectorization because of
> possible aliasing
>
> basic3.c:8:foo: note: loop peeled for vectorization to enhance alignment
>
> basic3.c:8:foo: note: loop with 7 iterations completely unrolled
>
> basic3.c:5:foo: note: loop with 7 iterations completely unrolled
>
>
> Is this version ok after testing?

-     "Vectorized basic-block\n");
+     "Basic block is vectorized\n");

lower case

Index: tree-vect-data-refs.c
===================================================================
--- tree-vect-data-refs.c (revision 201751)
+++ tree-vect-data-refs.c (working copy)
@@ -37,7 +37,7 @@ along with GCC; see the file COPYING3.
 #include "tree-scalar-evolution.h"
 #include "tree-vectorizer.h"
 #include "diagnostic-core.h"
-
+#include <stdlib.h>
 /* Need to include rtl.h, expr.h, etc. for optabs.  */
 #include "expr.h"
 #include "optabs.h"
@@ -1393,6 +1393,8 @@ vect_enhance_data_refs_alignment (loop_v

       supportable_dr_alignment = vect_supportable_dr_alignment (dr, true);
       do_peeling = vector_alignment_reachable_p (dr);
+      if (getenv("NOPEEL"))
+        do_peeling = false;
       if (do_peeling)
         {
           if (known_alignment_for_access_p (dr))

unrelated change

@@ -261,12 +262,20 @@ dump_loc (int dump_kind, FILE *dfile, so
   if (dump_kind)
     {
       if (LOCATION_LOCUS (loc) > BUILTINS_LOCATION)
-        fprintf (dfile, "\n%s:%d: note: ", LOCATION_FILE (loc),
-                 LOCATION_LINE (loc));
+        {
+          if (current_function_decl)
+            fprintf (dfile, "\n%s:%d:%s: note: ", LOCATION_FILE (loc),
+                     LOCATION_LINE (loc),
+                     gimple_decl_printable_name (current_function_decl, 1));
+          else
+            fprintf (dfile, "\n%s:%d: note: ", LOCATION_FILE (loc),
+                     LOCATION_LINE (loc));
+        }
       else if (current_function_decl)
-        fprintf (dfile, "\n%s:%d: note: ",
+        fprintf (dfile, "\n%s:%d:%s: note: ",
                  DECL_SOURCE_FILE (current_function_decl),
-                 DECL_SOURCE_LINE (current_function_decl));
+                 DECL_SOURCE_LINE (current_function_decl),
+                 gimple_decl_printable_name (current_function_decl, 1));
     }
 }

please not with this change (I oppose to it).

Ok with the change and not committing the hunks above.

Thanks,
Richard.








> thanks,
>
> David
>
> On Wed, Aug 28, 2013 at 2:45 AM, Richard Biener
> <richard.guenther@gmail.com> wrote:
>> On Tue, Aug 27, 2013 at 10:30 PM, Xinliang David Li <davidxl@google.com> wrote:
>>> If this is the convention, we should probably have another patch to
>>> fix all the existing opt-info messages.
>>
>> Yes please.
>>
>> Also ...
>>
>>
>>>>>> b.c:16:A::foo: note: Loop is vectorized
>>
>> "loop vectorized"
>>
>>>>>>
>>>>>> b.c:16:A::foo: note: Loop is versioned to remove aliases for vectorization
>>
>> "loop versioned for vectorization because of possible aliasing"
>>
>>>>>> b.c:16:A::foo: note: Loop is peeled to enhance alignment for vectorization
>>
>> "loop peeled for vectorization to enhance alignment"
>>
>>>>>> b.c:16:A::foo: note: Completely unroll loop 6 times
>>
>> maybe "loop with 6 iterations completely unrolled"
>>
>>>>>>
>>>>>> b.c:12:A::foo: note: Completely unroll loop 6 times
>>>>>>
>>
>> I hate the excessive vertical spacing as well.
>>
>> Richard.
>>
>>>>>> Ok after testing?
>>>>>>
>>>>>> thanks,
>>>>>>
>>>>>> David
>>>>