arm.c (arm_preferred_simd_mode): Check TARGET_NEON_VECTORIZE_DOUBLE instead of…

arm.c (arm_preferred_simd_mode): Check TARGET_NEON_VECTORIZE_DOUBLE instead of TARGET_NEON_VECTORIZE_QUAD.


	* config/arm/arm.c (arm_preferred_simd_mode): Check
	TARGET_NEON_VECTORIZE_DOUBLE instead of
	TARGET_NEON_VECTORIZE_QUAD.
	(arm_autovectorize_vector_sizes): Likewise.
	* config/arm/arm.opt (mvectorize-with-neon-quad): Make inverse
	mask of mvectorize-with-neon-double.  Add RejectNegative.
	(mvectorize-with-neon-double): New.

From-SVN: r178588

gcc/ChangeLog

+2011-09-06  Ira Rosen  <ira.rosen@linaro.org>
+
+	 * config/arm/arm.c (arm_preferred_simd_mode): Check
+	TARGET_NEON_VECTORIZE_DOUBLE instead of
+	TARGET_NEON_VECTORIZE_QUAD.
+	(arm_autovectorize_vector_sizes): Likewise.
+	* config/arm/arm.opt (mvectorize-with-neon-quad): Make inverse
+	mask of mvectorize-with-neon-double.  Add RejectNegative.
+	(mvectorize-with-neon-double): New.
+
 2011-09-06  Richard Guenther  <rguenther@suse.de>
 
 	* tree-ssa-sccvn.c (visit_use): CSE stmt pieces and simplify

gcc/config/arm/arm.c

@@ -23026,7 +23026,7 @@ arm_array_mode_supported_p (enum machine_mode mode,
   return false;
 }
 
-/* Use the option -mvectorize-with-neon-quad to override the use of doubleword
+/* Use the option -mvectorize-with-neon-double to override the use of quardword
    registers when autovectorizing for Neon, at least until multiple vector
    widths are supported properly by the middle-end.  */
 
@@ -23037,15 +23037,15 @@ arm_preferred_simd_mode (enum machine_mode mode)
     switch (mode)
       {
       case SFmode:
-	return TARGET_NEON_VECTORIZE_QUAD ? V4SFmode : V2SFmode;
+	return TARGET_NEON_VECTORIZE_DOUBLE ? V2SFmode : V4SFmode;
       case SImode:
-	return TARGET_NEON_VECTORIZE_QUAD ? V4SImode : V2SImode;
+	return TARGET_NEON_VECTORIZE_DOUBLE ? V2SImode : V4SImode;
       case HImode:
-	return TARGET_NEON_VECTORIZE_QUAD ? V8HImode : V4HImode;
+	return TARGET_NEON_VECTORIZE_DOUBLE ? V4HImode : V8HImode;
       case QImode:
-	return TARGET_NEON_VECTORIZE_QUAD ? V16QImode : V8QImode;
+	return TARGET_NEON_VECTORIZE_DOUBLE ? V8QImode : V16QImode;
       case DImode:
-	if (TARGET_NEON_VECTORIZE_QUAD)
+	if (!TARGET_NEON_VECTORIZE_DOUBLE)
 	  return V2DImode;
 	break;
 
@@ -24268,7 +24268,7 @@ arm_expand_sync (enum machine_mode mode,
 static unsigned int
 arm_autovectorize_vector_sizes (void)
 {
-  return TARGET_NEON_VECTORIZE_QUAD ? 16 | 8 : 0;
+  return TARGET_NEON_VECTORIZE_DOUBLE ? 0 : (16 | 8);
 }
 
 static bool

gcc/config/arm/arm.opt

@@ -238,9 +238,13 @@ Target Report RejectNegative Mask(LITTLE_WORDS)
 Assume big endian bytes, little endian words.  This option is deprecated.
 
 mvectorize-with-neon-quad
-Target Report Mask(NEON_VECTORIZE_QUAD)
+Target Report RejectNegative InverseMask(NEON_VECTORIZE_DOUBLE)
 Use Neon quad-word (rather than double-word) registers for vectorization
 
+mvectorize-with-neon-double
+Target Report RejectNegative Mask(NEON_VECTORIZE_DOUBLE)
+Use Neon double-word (rather than quad-word) registers for vectorization
+
 mword-relocations
 Target Report Var(target_word_relocations) Init(TARGET_DEFAULT_WORD_RELOCATIONS)
 Only generate absolute relocations on word sized values.

gcc/testsuite/ChangeLog

+2011-09-06  Ira Rosen  <ira.rosen@linaro.org>
+
+	* lib/target-supports.exp (check_effective_target_vect_multiple_sizes):
+	New procedure.
+	(add_options_for_quad_vectors): Replace with ...
+	(add_options_for_double_vectors): ... this.
+	* gfortran.dg/vect/pr19049.f90: Expect more printings on targets that
+	support multiple vector sizes since the vectorizer attempts to
+	vectorize with both vector sizes.
+	* gcc.dg/vect/no-vfa-vect-79.c,
+	gcc.dg/vect/no-vfa-vect-102a.c, gcc.dg/vect/vect-outer-1a.c,
+	gcc.dg/vect/vect-outer-1b.c, gcc.dg/vect/vect-outer-2b.c,
+	gcc.dg/vect/vect-outer-3a.c, gcc.dg/vect/no-vfa-vect-37.c,
+	gcc.dg/vect/vect-outer-3b.c, gcc.dg/vect/no-vfa-vect-101.c,
+	gcc.dg/vect/no-vfa-vect-102.c, gcc.dg/vect/vect-reduc-dot-s8b.c,
+	gcc.dg/vect/vect-outer-1.c, gcc.dg/vect/vect-104.c: Likewise.
+	* gcc.dg/vect/vect-42.c: Run with 64 bit vectors if applicable.
+	* gcc.dg/vect/vect-multitypes-6.c, gcc.dg/vect/vect-52.c,
+	gcc.dg/vect/vect-54.c, gcc.dg/vect/vect-46.c, gcc.dg/vect/vect-48.c,
+	gcc.dg/vect/vect-96.c, gcc.dg/vect/vect-multitypes-3.c,
+	gcc.dg/vect/vect-40.c: Likewise.
+	* gcc.dg/vect/vect-outer-5.c: Remove quad-vectors option as
+	redundant.
+	* gcc.dg/vect/vect-109.c, gcc.dg/vect/vect-peel-1.c,
+	gcc.dg/vect/vect-peel-2.c, gcc.dg/vect/slp-25.c,
+	gcc.dg/vect/vect-multitypes-1.c, gcc.dg/vect/slp-3.c,
+	gcc.dg/vect/no-vfa-pr29145.c, gcc.dg/vect/vect-multitypes-4.c:
+	Likewise.
+	* gcc.dg/vect/vect-peel-4.c: Make ia global.
+
 2011-09-05  Richard Sandiford  <rdsandiford@googlemail.com>
 
 	PR target/49606

gcc/testsuite/gcc.dg/vect/no-vfa-pr29145.c

 /* { dg-require-effective-target vect_int } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/no-vfa-vect-101.c

@@ -45,6 +45,7 @@ int main (void)
 }
 
 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 0 "vect" } } */
-/* { dg-final { scan-tree-dump-times "can't determine dependence" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "can't determine dependence" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "can't determine dependence" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */
 

gcc/testsuite/gcc.dg/vect/no-vfa-vect-102.c

@@ -53,6 +53,7 @@ int main (void)
 }
 
 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 0 "vect" } } */
-/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */
 

gcc/testsuite/gcc.dg/vect/no-vfa-vect-102a.c

@@ -53,6 +53,7 @@ int main (void)
 }
 
 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 0 "vect" } } */
-/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */
 

gcc/testsuite/gcc.dg/vect/no-vfa-vect-37.c

@@ -58,5 +58,6 @@ int main (void)
    If/when the aliasing problems are resolved, unalignment may
    prevent vectorization on some targets.  */
 /* { dg-final { scan-tree-dump-times "vectorized 2 loops" 1 "vect" { xfail *-*-* } } } */
-/* { dg-final { scan-tree-dump-times "can't determine dependence between" 2 "vect" } } */
+/* { dg-final { scan-tree-dump-times "can't determine dependence" 2 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "can't determine dependence" 4 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

gcc/testsuite/gcc.dg/vect/no-vfa-vect-79.c

@@ -46,5 +46,6 @@ int main (void)
   If/when the aliasing problems are resolved, unalignment may
   prevent vectorization on some targets.  */
 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect"  { xfail *-*-* } } } */
-/* { dg-final { scan-tree-dump-times "can't determine dependence between" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "can't determine dependence" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "can't determine dependence" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

gcc/testsuite/gcc.dg/vect/slp-25.c

 /* { dg-require-effective-target vect_int } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/slp-3.c

 /* { dg-require-effective-target vect_int } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-104.c

@@ -64,6 +64,7 @@ int main (void)
 }
 
 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 0 "vect" } } */
-/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "possible dependence between data-refs" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */
 

gcc/testsuite/gcc.dg/vect/vect-109.c

 /* { dg-require-effective-target vect_int } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-40.c

 /* { dg-require-effective-target vect_float } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-42.c

 /* { dg-require-effective-target vect_float } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-46.c

 /* { dg-require-effective-target vect_float } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-48.c

 /* { dg-require-effective-target vect_float } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-52.c

 /* { dg-require-effective-target vect_float } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-54.c

 /* { dg-require-effective-target vect_float } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-96.c

 /* { dg-require-effective-target vect_int } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-multitypes-1.c

 /* { dg-require-effective-target vect_int } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-multitypes-3.c

 /* { dg-require-effective-target vect_int } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-multitypes-4.c

 /* { dg-require-effective-target vect_int } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-multitypes-6.c

 /* { dg-require-effective-target vect_int } */
+/* { dg-add-options double_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-outer-1.c

@@ -22,5 +22,6 @@ foo (){
 }
 
 /* { dg-final { scan-tree-dump-times "OUTER LOOP VECTORIZED" 1 "vect" { xfail *-*-* } } } */
-/* { dg-final { scan-tree-dump-times "strided access in outer loop" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

gcc/testsuite/gcc.dg/vect/vect-outer-1a.c

@@ -20,5 +20,6 @@ foo (){
 }
 
 /* { dg-final { scan-tree-dump-times "OUTER LOOP VECTORIZED" 1 "vect" { xfail *-*-* } } } */
-/* { dg-final { scan-tree-dump-times "strided access in outer loop" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

gcc/testsuite/gcc.dg/vect/vect-outer-1b.c

@@ -22,5 +22,6 @@ foo (){
 }
 
 /* { dg-final { scan-tree-dump-times "OUTER LOOP VECTORIZED" 1 "vect" { xfail *-*-* } } } */
-/* { dg-final { scan-tree-dump-times "strided access in outer loop" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

gcc/testsuite/gcc.dg/vect/vect-outer-2b.c

@@ -37,5 +37,6 @@ int main (void)
   return 0;
 }
 
-/* { dg-final { scan-tree-dump-times "strided access in outer loop." 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 2 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

gcc/testsuite/gcc.dg/vect/vect-outer-3a.c

@@ -49,5 +49,6 @@ int main (void)
 }
 
 /* { dg-final { scan-tree-dump-times "OUTER LOOP VECTORIZED" 1 "vect" { xfail vect_no_align } } } */
-/* { dg-final { scan-tree-dump-times "step doesn't divide the vector-size" 2 "vect" } } */
+/* { dg-final { scan-tree-dump-times "step doesn't divide the vector-size" 2 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "step doesn't divide the vector-size" 3 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

gcc/testsuite/gcc.dg/vect/vect-outer-3b.c

@@ -49,5 +49,6 @@ int main (void)
 }
 
 /* { dg-final { scan-tree-dump-times "OUTER LOOP VECTORIZED" 1 "vect" { xfail *-*-* } } } */
-/* { dg-final { scan-tree-dump-times "strided access in outer loop" 2 "vect" } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 2 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "strided access in outer loop" 4 "vect" { target vect_multiple_sizes } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

gcc/testsuite/gcc.dg/vect/vect-outer-5.c

 /* { dg-require-effective-target vect_float } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include <signal.h>

gcc/testsuite/gcc.dg/vect/vect-peel-1.c

 /* { dg-require-effective-target vect_int } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-peel-2.c

 /* { dg-require-effective-target vect_int } */
-/* { dg-add-options quad_vectors } */
 
 #include <stdarg.h>
 #include "tree-vect.h"

gcc/testsuite/gcc.dg/vect/vect-peel-4.c

@@ -6,12 +6,12 @@
 #define N 128
 
 int ib[N+7];
+int ia[N+1];
 
 __attribute__ ((noinline))
 int main1 ()
 {
   int i;
-  int ia[N+1];
 
   /* Don't peel keeping one load and the store aligned.  */
   for (i = 0; i <= N; i++)

gcc/testsuite/gcc.dg/vect/vect-reduc-dot-s8b.c

@@ -58,7 +58,8 @@ int main (void)
 }
 
 /* { dg-final { scan-tree-dump-times "vect_recog_dot_prod_pattern: detected" 1 "vect" { xfail *-*-* } } } */
-/* { dg-final { scan-tree-dump-times "vect_recog_widen_mult_pattern: detected" 1 "vect" } } */
+/* { dg-final { scan-tree-dump-times "vect_recog_widen_mult_pattern: detected" 1 "vect" { xfail vect_multiple_sizes } } } */
+/* { dg-final { scan-tree-dump-times "vect_recog_widen_mult_pattern: detected" 2 "vect" { target vect_multiple_sizes } } } */
 
 /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { xfail *-*-* } } } */
 

gcc/testsuite/gfortran.dg/vect/pr19049.f90

@@ -19,6 +19,7 @@ subroutine s111 (ntimes,ld,n,ctime,dtime,a,b,c,d,e,aa,bb,cc)
       end
 
 ! { dg-final { scan-tree-dump-times "vectorized 1 loops" 0 "vect" } }
-! { dg-final { scan-tree-dump-times "complicated access pattern" 1 "vect" } }
+! { dg-final { scan-tree-dump-times "complicated access pattern" 1 "vect" { xfail vect_multiple_sizes } } }
+! { dg-final { scan-tree-dump-times "complicated access pattern" 2 "vect" { target vect_multiple_sizes } } }
 ! { dg-final { cleanup-tree-dump "vect" } }
 

gcc/testsuite/lib/target-supports.exp

@@ -3375,6 +3375,24 @@ foreach N {2 3 4 8} {
     }]
 }
 
+# Return 1 if the target supports multiple vector sizes
+
+proc check_effective_target_vect_multiple_sizes { } {
+    global et_vect_multiple_sizes
+
+    if [info exists et_vect_multiple_sizes_saved] {
+        verbose "check_effective_target_vect_multiple_sizes: using cached result" 2
+    } else {
+        set et_vect_multiple_sizes_saved 0
+        if { ([istarget arm*-*-*] && [check_effective_target_arm_neon]) } {
+           set et_vect_multiple_sizes_saved 1
+        }
+    }
+
+    verbose "check_effective_target_vect_multiple_sizes: returning $et_vect_multiple_sizes_saved" 2
+    return $et_vect_multiple_sizes_saved
+}
+
 # Return 1 if the target supports section-anchors
 
 proc check_effective_target_section_anchors { } {
@@ -3758,11 +3776,11 @@ proc add_options_for_bind_pic_locally { flags } {
     return $flags
 }
 
-# Add to FLAGS the flags needed to enable 128-bit vectors.
+# Add to FLAGS the flags needed to enable 64-bit vectors.
 
-proc add_options_for_quad_vectors { flags } {
+proc add_options_for_double_vectors { flags } {
     if [is-effective-target arm_neon_ok] {
-	return "$flags -mvectorize-with-neon-quad"
+	return "$flags -mvectorize-with-neon-double"
     }
 
     return $flags