[PATCH] Reduce cost of aligned sse register store.

[liuhongt]

Make them be equal to cost of unaligned ones to avoid odd alignment
peeling.

Impact for SPEC2017 on CLX:
fprate:
  503.bwaves_r    BuildSame
  507.cactuBSSN_r     -0.22
  508.namd_r          -0.02
  510.parest_r        -0.28
  511.povray_r        -0.20
  519.lbm_r       BuildSame
  521.wrf_r           -0.58
  526.blender_r       -0.30
  527.cam4_r           1.07
  538.imagick_r        0.01
  544.nab_r           -0.09
  549.fotonik3d_r BuildSame
  554.roms_r      BuildSame
intrate:
  500.perlbench_r     -0.25
  502.gcc_r           -0.15
  505.mcf_r       BuildSame
  520.omnetpp_r        1.03
  523.xalancbmk_r     -0.13
  525.x264_r          -0.05
  531.deepsjeng_r     -0.27
  541.leela_r         -0.24
  548.exchange2_r     -0.06
  557.xz_r            -0.10
  999.specrand_ir      2.69

Bootstrapped and regtested on x86_64-linux-gnu{-m32,}.
Ready to push to trunk.

gcc/ChangeLog:

	PR target/102543
	* config/i386/x86-tune-costs.h (skylake_cost): Reduce cost of
	storing 256/512-bit SSE register to be equal to cost of
	unaligned store to avoid odd alignment peeling.
	(icelake_cost): Ditto.

gcc/testsuite/ChangeLog:

	* gcc.target/i386/pr102543.c: New test.
---
 gcc/config/i386/x86-tune-costs.h         |  4 +--
 gcc/testsuite/gcc.target/i386/pr102543.c | 35 ++++++++++++++++++++++++
 2 files changed, 37 insertions(+), 2 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/i386/pr102543.c

diff --git a/gcc/config/i386/x86-tune-costs.h b/gcc/config/i386/x86-tune-costs.h
index dd5563d2e64..60d50c97fca 100644
--- a/gcc/config/i386/x86-tune-costs.h
+++ b/gcc/config/i386/x86-tune-costs.h
@@ -1903,7 +1903,7 @@ struct processor_costs skylake_cost = {
   {6, 6, 6},				/* cost of storing integer registers */
   {6, 6, 6, 10, 20},			/* cost of loading SSE register
 					   in 32bit, 64bit, 128bit, 256bit and 512bit */
-  {8, 8, 8, 12, 24},			/* cost of storing SSE register
+  {8, 8, 8, 8, 16},			/* cost of storing SSE register
 					   in 32bit, 64bit, 128bit, 256bit and 512bit */
   {6, 6, 6, 10, 20},			/* cost of unaligned loads.  */
   {8, 8, 8, 8, 16},			/* cost of unaligned stores.  */
@@ -2029,7 +2029,7 @@ struct processor_costs icelake_cost = {
   {6, 6, 6},				/* cost of storing integer registers */
   {6, 6, 6, 10, 20},			/* cost of loading SSE register
 					   in 32bit, 64bit, 128bit, 256bit and 512bit */
-  {8, 8, 8, 12, 24},			/* cost of storing SSE register
+  {8, 8, 8, 8, 16},			/* cost of storing SSE register
 					   in 32bit, 64bit, 128bit, 256bit and 512bit */
   {6, 6, 6, 10, 20},			/* cost of unaligned loads.  */
   {8, 8, 8, 8, 16},			/* cost of unaligned stores.  */
diff --git a/gcc/testsuite/gcc.target/i386/pr102543.c b/gcc/testsuite/gcc.target/i386/pr102543.c
new file mode 100644
index 00000000000..893eb9a5902
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr102543.c
@@ -0,0 +1,35 @@
+/* PR target/102543 */
+/* { dg-do compile } */
+/* { dg-options "-Ofast -march=skylake-avx512 -fdump-tree-optimized" } */
+/* { dg-final { scan-tree-dump-not "MEM\\\[" "optimized" } } */
+
+struct a
+{
+  int a[100];
+};
+typedef struct a misaligned_t __attribute__ ((aligned (8)));
+typedef struct a aligned_t __attribute__ ((aligned (32)));
+
+__attribute__ ((used))
+__attribute__ ((noinline))
+void
+t(void *a, int misaligned, aligned_t *d)
+{
+  int i,v;
+  for (i=0;i<100;i++)
+    {
+      if (misaligned)
+	v=((misaligned_t *)a)->a[i];
+      else
+	v=((aligned_t *)a)->a[i];
+      d->a[i]+=v;
+    }
+}
+struct b {int v; misaligned_t m;aligned_t aa;} b;
+aligned_t d;
+int
+main()
+{
+  t(&b.m, 1, &d);
+  return 0;
+}
-- 
2.18.2