sse.md (pbroadcast_evex_isa): New mode attr.

	* config/i386/sse.md (pbroadcast_evex_isa): New mode attr.
	(avx2_pbroadcast<mode>): Add another alternative with v instead
	of x constraints in it, using <pbroadcast_evex_isa> isa.
	(avx2_pbroadcast<mode>_1): Similarly, add two such alternatives.

	* gcc.target/i386/avx512bw-vpbroadcast-1.c: New test.
	* gcc.target/i386/avx512bw-vpbroadcast-2.c: New test.
	* gcc.target/i386/avx512bw-vpbroadcast-3.c: New test.
	* gcc.target/i386/avx512vl-vpbroadcast-1.c: New test.
	* gcc.target/i386/avx512vl-vpbroadcast-2.c: New test.
	* gcc.target/i386/avx512vl-vpbroadcast-3.c: New test.

From-SVN: r236369

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 531ad4d6df15fbed9f60c76464e3f734d84320ba..f3f6d38b9cfcb94aad3726482473915c50ee4e6d 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,5 +1,10 @@
 2016-05-18  Jakub Jelinek  <jakub@redhat.com>
 
+       * config/i386/sse.md (pbroadcast_evex_isa): New mode attr.
+       (avx2_pbroadcast<mode>): Add another alternative with v instead
+       of x constraints in it, using <pbroadcast_evex_isa> isa.
+       (avx2_pbroadcast<mode>_1): Similarly, add two such alternatives.
+
        * config/i386/sse.md (<ssse3_avx2>_palignr<mode>): Use
        constraint x instead of v in second alternative, add avx512bw
        alternative.

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 27796f4a62a4dfaa501f58d9cbbc6026df053af2..0bb217dcb35cbecdfdf480a4f888655fe31b5bc5 100644 (file)
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -16737,30 +16737,40 @@
    (set_attr "btver2_decode" "vector")
    (set_attr "mode" "OI")])
 
+(define_mode_attr pbroadcast_evex_isa
+  [(V64QI "avx512bw") (V32QI "avx512bw") (V16QI "avx512bw")
+   (V32HI "avx512bw") (V16HI "avx512bw") (V8HI "avx512bw")
+   (V16SI "avx512f") (V8SI "avx512f") (V4SI "avx512f")
+   (V8DI "avx512f") (V4DI "avx512f") (V2DI "avx512f")])
+
 (define_insn "avx2_pbroadcast<mode>"
-  [(set (match_operand:VI 0 "register_operand" "=x")
+  [(set (match_operand:VI 0 "register_operand" "=x,v")
        (vec_duplicate:VI
          (vec_select:<ssescalarmode>
-           (match_operand:<ssexmmmode> 1 "nonimmediate_operand" "xm")
+           (match_operand:<ssexmmmode> 1 "nonimmediate_operand" "xm,vm")
            (parallel [(const_int 0)]))))]
   "TARGET_AVX2"
   "vpbroadcast<ssemodesuffix>\t{%1, %0|%0, %<iptr>1}"
-  [(set_attr "type" "ssemov")
+  [(set_attr "isa" "*,<pbroadcast_evex_isa>")
+   (set_attr "type" "ssemov")
    (set_attr "prefix_extra" "1")
-   (set_attr "prefix" "vex")
+   (set_attr "prefix" "vex,evex")
    (set_attr "mode" "<sseinsnmode>")])
 
 (define_insn "avx2_pbroadcast<mode>_1"
-  [(set (match_operand:VI_256 0 "register_operand" "=x,x")
+  [(set (match_operand:VI_256 0 "register_operand" "=x,x,v,v")
        (vec_duplicate:VI_256
          (vec_select:<ssescalarmode>
-           (match_operand:VI_256 1 "nonimmediate_operand" "m,x")
+           (match_operand:VI_256 1 "nonimmediate_operand" "m,x,m,v")
            (parallel [(const_int 0)]))))]
   "TARGET_AVX2"
   "@
+   vpbroadcast<ssemodesuffix>\t{%1, %0|%0, %<iptr>1}
+   vpbroadcast<ssemodesuffix>\t{%x1, %0|%0, %x1}
    vpbroadcast<ssemodesuffix>\t{%1, %0|%0, %<iptr>1}
    vpbroadcast<ssemodesuffix>\t{%x1, %0|%0, %x1}"
-  [(set_attr "type" "ssemov")
+  [(set_attr "isa" "*,*,<pbroadcast_evex_isa>,<pbroadcast_evex_isa>")
+   (set_attr "type" "ssemov")
    (set_attr "prefix_extra" "1")
    (set_attr "prefix" "vex")
    (set_attr "mode" "<sseinsnmode>")])

diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index 91c7ebcda1b2c33482203a3cab4d2643dde70b9e..3df77a060679ac7ea19847e02cd5c23f41ef7122 100644 (file)
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -1,5 +1,12 @@
 2016-05-18  Jakub Jelinek  <jakub@redhat.com>
 
+       * gcc.target/i386/avx512bw-vpbroadcast-1.c: New test.
+       * gcc.target/i386/avx512bw-vpbroadcast-2.c: New test.
+       * gcc.target/i386/avx512bw-vpbroadcast-3.c: New test.
+       * gcc.target/i386/avx512vl-vpbroadcast-1.c: New test.
+       * gcc.target/i386/avx512vl-vpbroadcast-2.c: New test.
+       * gcc.target/i386/avx512vl-vpbroadcast-3.c: New test.
+
        * gcc.target/i386/avx512vl-vpalignr-3.c: New test.
        * gcc.target/i386/avx512bw-vpalignr-3.c: New test.
 

diff --git a/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-1.c b/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-1.c
new file mode 100644 (file)
index 0000000..da6dba5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-1.c
@@ -0,0 +1,104 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mavx512vl -mavx512bw" } */
+
+#include <x86intrin.h>
+
+void
+f1 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = _mm_broadcastb_epi8 (a);
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */
+
+void
+f2 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = _mm_broadcastw_epi16 (a);
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */
+
+void
+f3 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = _mm_broadcastd_epi32 (a);
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */
+
+void
+f4 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = _mm_broadcastq_epi64 (a);
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastq\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */
+
+void
+f5 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  register __m256i b __asm ("xmm17");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  b = _mm256_broadcastb_epi8 (a);
+  asm volatile ("" : "+v" (b));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */
+
+void
+f6 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  register __m256i b __asm ("xmm17");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  b = _mm256_broadcastw_epi16 (a);
+  asm volatile ("" : "+v" (b));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */
+
+void
+f7 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  register __m256i b __asm ("xmm17");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  b = _mm256_broadcastd_epi32 (a);
+  asm volatile ("" : "+v" (b));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */
+
+void
+f8 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  register __m256i b __asm ("xmm17");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  b = _mm256_broadcastq_epi64 (a);
+  asm volatile ("" : "+v" (b));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastq\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */

diff --git a/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-2.c b/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-2.c
new file mode 100644 (file)
index 0000000..2d623c6
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-2.c
@@ -0,0 +1,68 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mavx512vl -mavx512bw" } */
+
+typedef char V1 __attribute__((vector_size (16)));
+typedef short V2 __attribute__((vector_size (16)));
+typedef char V5 __attribute__((vector_size (32)));
+typedef short V6 __attribute__((vector_size (32)));
+typedef int V7 __attribute__((vector_size (32)));
+
+void
+f1 (V1 x)
+{
+  register V1 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V1) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */
+
+void
+f2 (V2 x)
+{
+  register V2 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V2) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */
+
+void
+f5 (V5 x)
+{
+  register V5 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V5) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*(xmm16\[^\n\r]*ymm16|ymm16\[^\n\r]*xmm16)" } } */
+
+void
+f6 (V6 x)
+{
+  register V6 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V6) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*(xmm16\[^\n\r]*ymm16|ymm16\[^\n\r]*xmm16)" } } */
+
+void
+f7 (V7 x)
+{
+  register V7 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V7) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*(xmm16\[^\n\r]*ymm16|ymm16\[^\n\r]*xmm16)" } } */

diff --git a/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-3.c b/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-3.c
new file mode 100644 (file)
index 0000000..ff3d275
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-3.c
@@ -0,0 +1,58 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mavx512vl -mavx512bw" } */
+
+typedef char V1 __attribute__((vector_size (16)));
+typedef short V2 __attribute__((vector_size (16)));
+typedef char V5 __attribute__((vector_size (32)));
+typedef short V6 __attribute__((vector_size (32)));
+typedef int V7 __attribute__((vector_size (32)));
+
+void
+f1 (V1 *x)
+{
+  register V1 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V1) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*xmm16" } } */
+
+void
+f2 (V2 *x)
+{
+  register V2 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V2) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*xmm16" } } */
+
+void
+f5 (V5 *x)
+{
+  register V5 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V5) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*ymm16" } } */
+
+void
+f6 (V6 *x)
+{
+  register V6 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V6) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*ymm16" } } */
+
+void
+f7 (V7 *x)
+{
+  register V7 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V7) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*ymm16" } } */

diff --git a/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-1.c b/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-1.c
new file mode 100644 (file)
index 0000000..4545809
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-1.c
@@ -0,0 +1,104 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mavx512vl -mno-avx512bw" } */
+
+#include <x86intrin.h>
+
+void
+f1 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = _mm_broadcastb_epi8 (a);
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*xmm16" } } */
+
+void
+f2 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = _mm_broadcastw_epi16 (a);
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*xmm16" } } */
+
+void
+f3 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = _mm_broadcastd_epi32 (a);
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */
+
+void
+f4 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = _mm_broadcastq_epi64 (a);
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastq\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */
+
+void
+f5 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  register __m256i b __asm ("xmm17");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  b = _mm256_broadcastb_epi8 (a);
+  asm volatile ("" : "+v" (b));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*\[xy]mm1\[67]" } } */
+
+void
+f6 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  register __m256i b __asm ("xmm17");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  b = _mm256_broadcastw_epi16 (a);
+  asm volatile ("" : "+v" (b));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*\[xy]mm1\[67]" } } */
+
+void
+f7 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  register __m256i b __asm ("xmm17");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  b = _mm256_broadcastd_epi32 (a);
+  asm volatile ("" : "+v" (b));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */
+
+void
+f8 (__m128i x)
+{
+  register __m128i a __asm ("xmm16");
+  register __m256i b __asm ("xmm17");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  b = _mm256_broadcastq_epi64 (a);
+  asm volatile ("" : "+v" (b));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastq\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */

diff --git a/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-2.c b/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-2.c
new file mode 100644 (file)
index 0000000..aed9914
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-2.c
@@ -0,0 +1,68 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mavx512vl -mno-avx512bw" } */
+
+typedef char V1 __attribute__((vector_size (16)));
+typedef short V2 __attribute__((vector_size (16)));
+typedef char V5 __attribute__((vector_size (32)));
+typedef short V6 __attribute__((vector_size (32)));
+typedef int V7 __attribute__((vector_size (32)));
+
+void
+f1 (V1 x)
+{
+  register V1 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V1) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*xmm16" } } */
+
+void
+f2 (V2 x)
+{
+  register V2 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V2) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*xmm16" } } */
+
+void
+f5 (V5 x)
+{
+  register V5 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V5) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*\[xy]mm16" } } */
+
+void
+f6 (V6 x)
+{
+  register V6 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V6) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*\[xy]mm16" } } */
+
+void
+f7 (V7 x)
+{
+  register V7 a __asm ("xmm16");
+  a = x;
+  asm volatile ("" : "+v" (a));
+  a = __builtin_shuffle (a, (V7) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*(xmm16\[^\n\r]*ymm16|ymm16\[^\n\r]*xmm16)" } } */

diff --git a/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-3.c b/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-3.c
new file mode 100644 (file)
index 0000000..b93bd36
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-3.c
@@ -0,0 +1,58 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mavx512vl -mno-avx512bw" } */
+
+typedef char V1 __attribute__((vector_size (16)));
+typedef short V2 __attribute__((vector_size (16)));
+typedef char V5 __attribute__((vector_size (32)));
+typedef short V6 __attribute__((vector_size (32)));
+typedef int V7 __attribute__((vector_size (32)));
+
+void
+f1 (V1 *x)
+{
+  register V1 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V1) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*xmm16" } } */
+
+void
+f2 (V2 *x)
+{
+  register V2 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V2) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*xmm16" } } */
+
+void
+f5 (V5 *x)
+{
+  register V5 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V5) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*ymm16" } } */
+
+void
+f6 (V6 *x)
+{
+  register V6 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V6) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*ymm16" } } */
+
+void
+f7 (V7 *x)
+{
+  register V7 a __asm ("xmm16");
+  a = __builtin_shuffle (*x, (V7) { 0 });
+  asm volatile ("" : "+v" (a));
+}
+
+/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*ymm16" } } */