omp-low.c (lower_rec_input_clauses): For lastprivate clauses in ctx->for_simd_scan_ph...

	* omp-low.c (lower_rec_input_clauses): For lastprivate clauses in
	ctx->for_simd_scan_phase simd copy the outer var to the privatized
	variable(s).  For conditional lastprivate look through outer
	GIMPLE_OMP_SCAN context.
	(lower_omp_1): For conditional lastprivate look through outer
	GIMPLE_OMP_SCAN context.

	* testsuite/libgomp.c/scan-19.c: New test.
	* testsuite/libgomp.c/scan-20.c: New test.

From-SVN: r273169

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index ae3fbe4..edb8df5 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,5 +1,12 @@
 2019-07-06  Jakub Jelinek  <jakub@redhat.com>
 
+       * omp-low.c (lower_rec_input_clauses): For lastprivate clauses in
+       ctx->for_simd_scan_phase simd copy the outer var to the privatized
+       variable(s).  For conditional lastprivate look through outer
+       GIMPLE_OMP_SCAN context.
+       (lower_omp_1): For conditional lastprivate look through outer
+       GIMPLE_OMP_SCAN context.
+
        * omp-low.c (struct omp_context): Rename combined_into_simd_safelen0
        member to combined_into_simd_safelen1.
        (lower_rec_input_clauses, lower_omp_1): Adjust uses.

diff --git a/gcc/omp-low.c b/gcc/omp-low.c
index b687fe1..3fe466f 100644 (file)
--- a/gcc/omp-low.c
+++ b/gcc/omp-low.c
@@ -5006,6 +5006,17 @@ lower_rec_input_clauses (tree clauses, gimple_seq *ilist, gimple_seq *dlist,
                            lower_omp (&tseq, ctx->outer);
                          gimple_seq_add_seq (&llist[1], tseq);
                        }
+                     if (OMP_CLAUSE_CODE (c) == OMP_CLAUSE_LASTPRIVATE
+                         && ctx->for_simd_scan_phase)
+                       {
+                         x = unshare_expr (ivar);
+                         tree orig_v
+                           = build_outer_var_ref (var, ctx,
+                                                  OMP_CLAUSE_LASTPRIVATE);
+                         x = lang_hooks.decls.omp_clause_assign_op (c, x,
+                                                                    orig_v);
+                         gimplify_and_add (x, &llist[0]);
+                       }
                      if (y)
                        {
                          y = lang_hooks.decls.omp_clause_dtor (c, ivar);
@@ -5035,6 +5046,16 @@ lower_rec_input_clauses (tree clauses, gimple_seq *ilist, gimple_seq *dlist,
                }
              if (nx)
                gimplify_and_add (nx, ilist);
+             if (OMP_CLAUSE_CODE (c) == OMP_CLAUSE_LASTPRIVATE
+                 && is_simd
+                 && ctx->for_simd_scan_phase)
+               {
+                 tree orig_v = build_outer_var_ref (var, ctx,
+                                                    OMP_CLAUSE_LASTPRIVATE);
+                 x = lang_hooks.decls.omp_clause_assign_op (c, new_var,
+                                                            orig_v);
+                 gimplify_and_add (x, ilist);
+               }
              /* FALLTHRU */
 
            do_dtor:
@@ -5709,11 +5730,12 @@ lower_rec_input_clauses (tree clauses, gimple_seq *ilist, gimple_seq *dlist,
                    && OMP_CLAUSE_LASTPRIVATE_CONDITIONAL (c))
                  {
                    tree o = lookup_decl (OMP_CLAUSE_DECL (c), ctx);
-                   tree *v
-                     = ctx->lastprivate_conditional_map->get (o);
-                   tree po = lookup_decl (OMP_CLAUSE_DECL (c), ctx->outer);
-                   tree *pv
-                     = ctx->outer->lastprivate_conditional_map->get (po);
+                   omp_context *outer = ctx->outer;
+                   if (gimple_code (outer->stmt) == GIMPLE_OMP_SCAN)
+                     outer = outer->outer;
+                   tree *v = ctx->lastprivate_conditional_map->get (o);
+                   tree po = lookup_decl (OMP_CLAUSE_DECL (c), outer);
+                   tree *pv = outer->lastprivate_conditional_map->get (po);
                    *v = *pv;
                  }
            }
@@ -12421,7 +12443,11 @@ lower_omp_1 (gimple_stmt_iterator *gsi_p, omp_context *ctx)
              {
                tree clauses;
                if (up->combined_into_simd_safelen1)
-                 up = up->outer;
+                 {
+                   up = up->outer;
+                   if (gimple_code (up->stmt) == GIMPLE_OMP_SCAN)
+                     up = up->outer;
+                 }
                if (gimple_code (up->stmt) == GIMPLE_OMP_FOR)
                  clauses = gimple_omp_for_clauses (up->stmt);
                else

diff --git a/libgomp/ChangeLog b/libgomp/ChangeLog
index 8ffe379..107c5c9 100644 (file)
--- a/libgomp/ChangeLog
+++ b/libgomp/ChangeLog
@@ -1,5 +1,8 @@
 2019-07-06  Jakub Jelinek  <jakub@redhat.com>
 
+       * testsuite/libgomp.c/scan-19.c: New test.
+       * testsuite/libgomp.c/scan-20.c: New test.
+
        * testsuite/libgomp.c/scan-11.c: New test.
        * testsuite/libgomp.c/scan-12.c: New test.
        * testsuite/libgomp.c/scan-13.c: New test.

diff --git a/libgomp/testsuite/libgomp.c/scan-19.c b/libgomp/testsuite/libgomp.c/scan-19.c
new file mode 100644 (file)
index 0000000..19d31d9
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/scan-19.c
@@ -0,0 +1,119 @@
+/* { dg-require-effective-target size32plus } */
+/* { dg-additional-options "-O2 -fopenmp -fdump-tree-vect-details" } */
+/* { dg-additional-options "-mavx" { target avx_runtime } } */
+/* { dg-final { scan-tree-dump-times "vectorized \[2-6] loops" 2 "vect" { target sse2_runtime } } } */
+
+extern void abort (void);
+int r, a[1024], b[1024], x, y, z;
+
+__attribute__((noipa)) void
+foo (int *a, int *b)
+{
+  #pragma omp for simd reduction (inscan, +:r) lastprivate (conditional: z) firstprivate (x) private (y)
+  for (int i = 0; i < 1024; i++)
+    {
+      { y = a[i]; r += y + x + 12; }
+      #pragma omp scan inclusive(r)
+      { b[i] = r; if ((i & 1) == 0 && i < 937) z = r; }
+    }
+}
+
+__attribute__((noipa)) int
+bar (void)
+{
+  int s = 0;
+  #pragma omp parallel
+  #pragma omp for simd reduction (inscan, +:s) firstprivate (x) private (y) lastprivate (z)
+  for (int i = 0; i < 1024; i++)
+    {
+      { y = 2 * a[i]; s += y; z = y; }
+      #pragma omp scan inclusive(s)
+      { y = s; b[i] = y + x + 12; }
+    }
+  return s;
+}
+
+__attribute__((noipa)) void
+baz (int *a, int *b)
+{
+  #pragma omp parallel for simd reduction (inscan, +:r) firstprivate (x) lastprivate (x) if (simd: 0)
+  for (int i = 0; i < 1024; i++)
+    {
+      { r += a[i]; if (i == 1023) x = 29; }
+      #pragma omp scan inclusive(r)
+      b[i] = r;
+    }
+}
+
+__attribute__((noipa)) int
+qux (void)
+{
+  int s = 0;
+  #pragma omp parallel for simd simdlen (1) reduction (inscan, +:s) lastprivate (conditional: x, y)
+  for (int i = 0; i < 1024; i++)
+    {
+      { s += 2 * a[i]; if ((a[i] & 1) == 1 && i < 825) x = a[i]; }
+      #pragma omp scan inclusive(s)
+      { b[i] = s; if ((a[i] & 1) == 0 && i < 829) y = a[i]; }
+    }
+  return s;
+}
+
+int
+main ()
+{
+  int s = 0;
+  x = -12;
+  for (int i = 0; i < 1024; ++i)
+    {
+      a[i] = i;
+      b[i] = -1;
+      asm ("" : "+g" (i));
+    }
+  #pragma omp parallel
+  foo (a, b);
+  if (r != 1024 * 1023 / 2 || x != -12 || z != b[936])
+    abort ();
+  for (int i = 0; i < 1024; ++i)
+    {
+      s += i;
+      if (b[i] != s)
+       abort ();
+      else
+       b[i] = 25;
+    }
+  if (bar () != 1024 * 1023 || x != -12 || z != 2 * 1023)
+    abort ();
+  s = 0;
+  for (int i = 0; i < 1024; ++i)
+    {
+      s += 2 * i;
+      if (b[i] != s)
+       abort ();
+      else
+       b[i] = -1;
+    }
+  r = 0;
+  baz (a, b);
+  if (r != 1024 * 1023 / 2 || x != 29)
+    abort ();
+  s = 0;
+  for (int i = 0; i < 1024; ++i)
+    {
+      s += i;
+      if (b[i] != s)
+       abort ();
+      else
+       b[i] = -25;
+    }
+  if (qux () != 1024 * 1023 || x != 823 || y != 828)
+    abort ();
+  s = 0;
+  for (int i = 0; i < 1024; ++i)
+    {
+      s += 2 * i;
+      if (b[i] != s)
+       abort ();
+    }
+  return 0;
+}

diff --git a/libgomp/testsuite/libgomp.c/scan-20.c b/libgomp/testsuite/libgomp.c/scan-20.c
new file mode 100644 (file)
index 0000000..1840164
--- /dev/null
+++ b/libgomp/testsuite/libgomp.c/scan-20.c
@@ -0,0 +1,119 @@
+/* { dg-require-effective-target size32plus } */
+/* { dg-additional-options "-O2 -fopenmp -fdump-tree-vect-details" } */
+/* { dg-additional-options "-mavx" { target avx_runtime } } */
+/* { dg-final { scan-tree-dump-times "vectorized \[2-6] loops" 2 "vect" { target sse2_runtime } } } */
+
+extern void abort (void);
+int r, a[1024], b[1024], x, y, z;
+
+__attribute__((noipa)) void
+foo (int *a, int *b)
+{
+  #pragma omp for simd reduction (inscan, +:r) lastprivate (conditional: z) firstprivate (x) private (y) simdlen(1)
+  for (int i = 0; i < 1024; i++)
+    {
+      { b[i] = r; if ((i & 1) == 0 && i < 937) z = r; }
+      #pragma omp scan exclusive(r)
+      { y = a[i]; r += y + x + 12; }
+    }
+}
+
+__attribute__((noipa)) int
+bar (void)
+{
+  int s = 0;
+  #pragma omp parallel
+  #pragma omp for simd reduction (inscan, +:s) firstprivate (x) private (y) lastprivate (z) if (0)
+  for (int i = 0; i < 1024; i++)
+    {
+      { y = s; b[i] = y + x + 12; }
+      #pragma omp scan exclusive(s)
+      { y = 2 * a[i]; s += y; z = y; }
+    }
+  return s;
+}
+
+__attribute__((noipa)) void
+baz (int *a, int *b)
+{
+  #pragma omp parallel for simd reduction (inscan, +:r) firstprivate (x) lastprivate (x)
+  for (int i = 0; i < 1024; i++)
+    {
+      b[i] = r;
+      #pragma omp scan exclusive(r)
+      { r += a[i]; if (i == 1023) x = 29; }
+    }
+}
+
+__attribute__((noipa)) int
+qux (void)
+{
+  int s = 0;
+  #pragma omp parallel for simd reduction (inscan, +:s) lastprivate (conditional: x, y)
+  for (int i = 0; i < 1024; i++)
+    {
+      { b[i] = s; if ((a[i] & 1) == 0 && i < 829) y = a[i]; }
+      #pragma omp scan exclusive(s)
+      { s += 2 * a[i]; if ((a[i] & 1) == 1 && i < 825) x = a[i]; }
+    }
+  return s;
+}
+
+int
+main ()
+{
+  int s = 0;
+  x = -12;
+  for (int i = 0; i < 1024; ++i)
+    {
+      a[i] = i;
+      b[i] = -1;
+      asm ("" : "+g" (i));
+    }
+  #pragma omp parallel
+  foo (a, b);
+  if (r != 1024 * 1023 / 2 || x != -12 || z != b[936])
+    abort ();
+  for (int i = 0; i < 1024; ++i)
+    {
+      if (b[i] != s)
+       abort ();
+      else
+       b[i] = 25;
+      s += i;
+    }
+  if (bar () != 1024 * 1023 || x != -12 || z != 2 * 1023)
+    abort ();
+  s = 0;
+  for (int i = 0; i < 1024; ++i)
+    {
+      if (b[i] != s)
+       abort ();
+      else
+       b[i] = -1;
+      s += 2 * i;
+    }
+  r = 0;
+  baz (a, b);
+  if (r != 1024 * 1023 / 2 || x != 29)
+    abort ();
+  s = 0;
+  for (int i = 0; i < 1024; ++i)
+    {
+      if (b[i] != s)
+       abort ();
+      else
+       b[i] = -25;
+      s += i;
+    }
+  if (qux () != 1024 * 1023 || x != 823 || y != 828)
+    abort ();
+  s = 0;
+  for (int i = 0; i < 1024; ++i)
+    {
+      if (b[i] != s)
+       abort ();
+      s += 2 * i;
+    }
+  return 0;
+}