Hi!
On Wed, 10 May 2017 18:30:54 +0200, Jakub Jelinek <ja...@redhat.com> wrote:
> On Tue, May 09, 2017 at 10:57:34PM +0200, Thomas Schwinge wrote:
> > Use "oacc kernels parallelized" attribute for parallelized OpenACC
> > kernels
> Ok.
Thanks. Committed to trunk in r247957:
commit 5dd0c4e81e7a79afccfc936407affbdda2e3b737
Author: tschwinge <tschwinge@138bc75d-0d04-0410-961f-82ee72b054a4>
Date: Fri May 12 09:18:34 2017 +0000
[PR middle-end/69921] Use "oacc kernels parallelized" attribute for
parallelized OpenACC kernels
gcc/
PR middle-end/69921
* tree-parloops.c (create_parallel_loop): Set "oacc kernels
parallelized" attribute for parallelized OpenACC kernels.
* omp-offload.c (execute_oacc_device_lower): Use it.
gcc/testsuite/
* c-c++-common/goacc/classify-kernels-unparallelized.c: Adjust.
* c-c++-common/goacc/classify-kernels.c: Likewise.
* c-c++-common/goacc/kernels-counter-vars-function-scope.c:
Likewise.
* c-c++-common/goacc/kernels-double-reduction-n.c: Likewise.
* c-c++-common/goacc/kernels-double-reduction.c: Likewise.
* c-c++-common/goacc/kernels-loop-2.c: Likewise.
* c-c++-common/goacc/kernels-loop-3.c: Likewise.
* c-c++-common/goacc/kernels-loop-g.c: Likewise.
* c-c++-common/goacc/kernels-loop-mod-not-zero.c: Likewise.
* c-c++-common/goacc/kernels-loop-n.c: Likewise.
* c-c++-common/goacc/kernels-loop-nest.c: Likewise.
* c-c++-common/goacc/kernels-loop.c: Likewise.
* c-c++-common/goacc/kernels-one-counter-var.c: Likewise.
* c-c++-common/goacc/kernels-reduction.c: Likewise.
* gfortran.dg/goacc/classify-kernels-unparallelized.f95: Likewise.
* gfortran.dg/goacc/classify-kernels.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-2.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data-2.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data-enter-exit.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data-update.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-n.f95: Likewise.
* gfortran.dg/goacc/kernels-loop.f95: Likewise.
git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@247957
138bc75d-0d04-0410-961f-82ee72b054a4
---
gcc/ChangeLog | 5 +++++
gcc/omp-offload.c | 24 ++++++++++++++++----
gcc/testsuite/ChangeLog | 26 ++++++++++++++++++++++
.../goacc/classify-kernels-unparallelized.c | 2 +-
.../c-c++-common/goacc/classify-kernels.c | 6 ++---
.../goacc/kernels-counter-vars-function-scope.c | 3 +--
.../goacc/kernels-double-reduction-n.c | 3 +--
.../c-c++-common/goacc/kernels-double-reduction.c | 3 +--
gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c | 3 +--
gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c | 3 +--
gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c | 3 +--
.../c-c++-common/goacc/kernels-loop-mod-not-zero.c | 3 +--
gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c | 3 +--
.../c-c++-common/goacc/kernels-loop-nest.c | 3 +--
gcc/testsuite/c-c++-common/goacc/kernels-loop.c | 3 +--
.../c-c++-common/goacc/kernels-one-counter-var.c | 3 +--
.../c-c++-common/goacc/kernels-reduction.c | 3 +--
.../goacc/classify-kernels-unparallelized.f95 | 2 +-
.../gfortran.dg/goacc/classify-kernels.f95 | 6 ++---
gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95 | 3 +--
.../gfortran.dg/goacc/kernels-loop-data-2.f95 | 3 +--
.../goacc/kernels-loop-data-enter-exit-2.f95 | 3 +--
.../goacc/kernels-loop-data-enter-exit.f95 | 3 +--
.../gfortran.dg/goacc/kernels-loop-data-update.f95 | 3 +--
.../gfortran.dg/goacc/kernels-loop-data.f95 | 3 +--
gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95 | 5 ++---
gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95 | 3 +--
gcc/tree-parloops.c | 16 +++++++------
28 files changed, 89 insertions(+), 60 deletions(-)
diff --git gcc/ChangeLog gcc/ChangeLog
index aeb22df..580a3db 100644
--- gcc/ChangeLog
+++ gcc/ChangeLog
@@ -1,5 +1,10 @@
2017-05-12 Thomas Schwinge <tho...@codesourcery.com>
+ PR middle-end/69921
+ * tree-parloops.c (create_parallel_loop): Set "oacc kernels
+ parallelized" attribute for parallelized OpenACC kernels.
+ * omp-offload.c (execute_oacc_device_lower): Use it.
+
* omp-expand.c (expand_omp_target) <GF_OMP_TARGET_KIND_OACC_KERNELS>:
Set "oacc kernels" attribute.
* omp-general.c (oacc_set_fn_attrib): Remove is_kernel formal
diff --git gcc/omp-offload.c gcc/omp-offload.c
index d24f131..9372f9e 100644
--- gcc/omp-offload.c
+++ gcc/omp-offload.c
@@ -1444,6 +1444,13 @@ execute_oacc_device_lower ()
flag_openacc_dims = (char *)&flag_openacc_dims;
}
+ bool is_oacc_kernels
+ = (lookup_attribute ("oacc kernels",
+ DECL_ATTRIBUTES (current_function_decl)) != NULL);
+ bool is_oacc_kernels_parallelized
+ = (lookup_attribute ("oacc kernels parallelized",
+ DECL_ATTRIBUTES (current_function_decl)) != NULL);
+
/* Discover, partition and process the loops. */
oacc_loop *loops = oacc_loop_discovery ();
int fn_level = oacc_fn_attrib_level (attrs);
@@ -1453,17 +1460,26 @@ execute_oacc_device_lower ()
if (fn_level >= 0)
fprintf (dump_file, "Function is OpenACC routine level %d\n",
fn_level);
- else if (lookup_attribute ("oacc kernels",
- DECL_ATTRIBUTES (current_function_decl)))
- fprintf (dump_file, "Function is OpenACC kernels offload\n");
+ else if (is_oacc_kernels)
+ fprintf (dump_file, "Function is %s OpenACC kernels offload\n",
+ (is_oacc_kernels_parallelized
+ ? "parallelized" : "unparallelized"));
else
fprintf (dump_file, "Function is OpenACC parallel offload\n");
}
unsigned outer_mask = fn_level >= 0 ? GOMP_DIM_MASK (fn_level) - 1 : 0;
unsigned used_mask = oacc_loop_partition (loops, outer_mask);
+ /* OpenACC kernels constructs are special: they currently don't use the
+ generic oacc_loop infrastructure and attribute/dimension processing. */
+ if (is_oacc_kernels && is_oacc_kernels_parallelized)
+ {
+ /* Parallelized OpenACC kernels constructs use gang parallelism. See
+ also tree-parloops.c:create_parallel_loop. */
+ used_mask |= GOMP_DIM_MASK (GOMP_DIM_GANG);
+ }
+
int dims[GOMP_DIM_MAX];
-
oacc_validate_dims (current_function_decl, attrs, dims, fn_level, used_mask);
if (dump_file)
diff --git gcc/testsuite/ChangeLog gcc/testsuite/ChangeLog
index 52865d3..e1e2641 100644
--- gcc/testsuite/ChangeLog
+++ gcc/testsuite/ChangeLog
@@ -2,6 +2,32 @@
* c-c++-common/goacc/classify-kernels-unparallelized.c: Adjust.
* c-c++-common/goacc/classify-kernels.c: Likewise.
+ * c-c++-common/goacc/kernels-counter-vars-function-scope.c:
+ Likewise.
+ * c-c++-common/goacc/kernels-double-reduction-n.c: Likewise.
+ * c-c++-common/goacc/kernels-double-reduction.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-2.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-3.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-g.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-mod-not-zero.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-n.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-nest.c: Likewise.
+ * c-c++-common/goacc/kernels-loop.c: Likewise.
+ * c-c++-common/goacc/kernels-one-counter-var.c: Likewise.
+ * c-c++-common/goacc/kernels-reduction.c: Likewise.
+ * gfortran.dg/goacc/classify-kernels-unparallelized.f95: Likewise.
+ * gfortran.dg/goacc/classify-kernels.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-2.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data-2.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data-enter-exit.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data-update.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-n.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop.f95: Likewise.
+
+ * c-c++-common/goacc/classify-kernels-unparallelized.c: Adjust.
+ * c-c++-common/goacc/classify-kernels.c: Likewise.
* c-c++-common/goacc/classify-parallel.c: Likewise.
* c-c++-common/goacc/classify-routine.c: Likewise.
* gfortran.dg/goacc/classify-kernels-unparallelized.f95: Likewise.
diff --git gcc/testsuite/c-c++-common/goacc/classify-kernels-unparallelized.c
gcc/testsuite/c-c++-common/goacc/classify-kernels-unparallelized.c
index 70ff428..626f6b4 100644
--- gcc/testsuite/c-c++-common/goacc/classify-kernels-unparallelized.c
+++ gcc/testsuite/c-c++-common/goacc/classify-kernels-unparallelized.c
@@ -34,6 +34,6 @@ void KERNELS ()
/* Check the offloaded function's classification and compute dimensions (will
always be 1 x 1 x 1 for non-offloading compilation).
- { dg-final { scan-tree-dump-times "(?n)Function is OpenACC kernels offload"
1 "oaccdevlow" } }
+ { dg-final { scan-tree-dump-times "(?n)Function is unparallelized OpenACC
kernels offload" 1 "oaccdevlow" } }
{ dg-final { scan-tree-dump-times "(?n)Compute dimensions \\\[1, 1, 1\\\]"
1 "oaccdevlow" } }
{ dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } } */
diff --git gcc/testsuite/c-c++-common/goacc/classify-kernels.c
gcc/testsuite/c-c++-common/goacc/classify-kernels.c
index c8b0fda..95037e6 100644
--- gcc/testsuite/c-c++-common/goacc/classify-kernels.c
+++ gcc/testsuite/c-c++-common/goacc/classify-kernels.c
@@ -25,11 +25,11 @@ void KERNELS ()
/* Check that exactly one OpenACC kernels construct is analyzed, and that it
can be parallelized.
{ dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } }
- { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(0, , \\), oacc kernels, omp target entrypoint\\)\\)" 1 "parloops1" } }
+ { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } }
{ dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check the offloaded function's classification and compute dimensions (will
always be 1 x 1 x 1 for non-offloading compilation).
- { dg-final { scan-tree-dump-times "(?n)Function is OpenACC kernels offload"
1 "oaccdevlow" } }
+ { dg-final { scan-tree-dump-times "(?n)Function is parallelized OpenACC
kernels offload" 1 "oaccdevlow" } }
{ dg-final { scan-tree-dump-times "(?n)Compute dimensions \\\[1, 1, 1\\\]"
1 "oaccdevlow" } }
- { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } } */
+ { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels parallelized, oacc function \\(, , \\), oacc
kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } } */
diff --git
gcc/testsuite/c-c++-common/goacc/kernels-counter-vars-function-scope.c
gcc/testsuite/c-c++-common/goacc/kernels-counter-vars-function-scope.c
index 17f240e..c475333 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-counter-vars-function-scope.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-counter-vars-function-scope.c
@@ -45,9 +45,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-double-reduction-n.c
gcc/testsuite/c-c++-common/goacc/kernels-double-reduction-n.c
index 750f576..27ea2e9 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-double-reduction-n.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-double-reduction-n.c
@@ -27,10 +27,9 @@ foo (unsigned int n)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* { dg-final { scan-tree-dump-times "parallelizing outer loop" 1 "parloops1"
} } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-double-reduction.c
gcc/testsuite/c-c++-common/goacc/kernels-double-reduction.c
index df60d6a..0841e90 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-double-reduction.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-double-reduction.c
@@ -27,10 +27,9 @@ foo (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* { dg-final { scan-tree-dump-times "parallelizing outer loop" 1 "parloops1"
} } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c
index 913d91f..acef6a1 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c
@@ -59,11 +59,10 @@ main (void)
/* Check that only three loops are analyzed, and that all can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.2" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c
index 1822d2a..75e2bb7 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c
@@ -39,9 +39,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c
index e946319..73b469d 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c
@@ -7,9 +7,8 @@
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-mod-not-zero.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-mod-not-zero.c
index 9b63b45..5592623 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-mod-not-zero.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-mod-not-zero.c
@@ -43,9 +43,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c
index 279f797..e86be1b 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c
@@ -46,9 +46,8 @@ foo (COUNTERTYPE n)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-nest.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-nest.c
index db1071f..2b0e186 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-nest.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-nest.c
@@ -30,9 +30,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop.c
gcc/testsuite/c-c++-common/goacc/kernels-loop.c
index abf7a3c..9619d53 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop.c
@@ -46,9 +46,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-one-counter-var.c
gcc/testsuite/c-c++-common/goacc/kernels-one-counter-var.c
index 95f4817..69539b2 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-one-counter-var.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-one-counter-var.c
@@ -44,9 +44,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-reduction.c
gcc/testsuite/c-c++-common/goacc/kernels-reduction.c
index 6f5a418..4a18272 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-reduction.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-reduction.c
@@ -26,9 +26,8 @@ foo (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/gfortran.dg/goacc/classify-kernels-unparallelized.f95
gcc/testsuite/gfortran.dg/goacc/classify-kernels-unparallelized.f95
index 9887d35..4b282ca 100644
--- gcc/testsuite/gfortran.dg/goacc/classify-kernels-unparallelized.f95
+++ gcc/testsuite/gfortran.dg/goacc/classify-kernels-unparallelized.f95
@@ -36,6 +36,6 @@ end program main
! Check the offloaded function's classification and compute dimensions (will
! always be 1 x 1 x 1 for non-offloading compilation).
-! { dg-final { scan-tree-dump-times "(?n)Function is OpenACC kernels offload"
1 "oaccdevlow" } }
+! { dg-final { scan-tree-dump-times "(?n)Function is unparallelized OpenACC
kernels offload" 1 "oaccdevlow" } }
! { dg-final { scan-tree-dump-times "(?n)Compute dimensions \\\[1, 1, 1\\\]" 1
"oaccdevlow" } }
! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } }
diff --git gcc/testsuite/gfortran.dg/goacc/classify-kernels.f95
gcc/testsuite/gfortran.dg/goacc/classify-kernels.f95
index 69c89a9..da025c1 100644
--- gcc/testsuite/gfortran.dg/goacc/classify-kernels.f95
+++ gcc/testsuite/gfortran.dg/goacc/classify-kernels.f95
@@ -27,11 +27,11 @@ end program main
! Check that exactly one OpenACC kernels construct is analyzed, and that it
! can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } }
-! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(0, , \\), oacc kernels, omp target entrypoint\\)\\)" 1 "parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check the offloaded function's classification and compute dimensions (will
! always be 1 x 1 x 1 for non-offloading compilation).
-! { dg-final { scan-tree-dump-times "(?n)Function is OpenACC kernels offload"
1 "oaccdevlow" } }
+! { dg-final { scan-tree-dump-times "(?n)Function is parallelized OpenACC
kernels offload" 1 "oaccdevlow" } }
! { dg-final { scan-tree-dump-times "(?n)Compute dimensions \\\[1, 1, 1\\\]" 1
"oaccdevlow" } }
-! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels parallelized, oacc function \\(, , \\), oacc
kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95
index 865f7a6..516aede 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95
@@ -34,11 +34,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-2.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-2.f95
index c9f3a62..ff3788a 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-2.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-2.f95
@@ -40,11 +40,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95
index 3361607..60a5c96 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95
@@ -40,11 +40,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit.f95
index 5ba56fb..ce04749 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit.f95
@@ -38,11 +38,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-update.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-update.f95
index a622a96..d2de138 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-update.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-update.f95
@@ -38,10 +38,9 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 2
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 2 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 2 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data.f95
index 4ec2ac3..92872b2 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data.f95
@@ -38,11 +38,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95
index 409fe6f..079712f2 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95
@@ -32,10 +32,9 @@ end module test
! Check that only one loop is analyzed, and that it can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } }
+! TODO, PR70545.
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" { xfail *-*-* } } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function __test_MOD_foo._omp_fn.0
" 1 "optimized" } }
-
-! TODO, PR70545.
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" {
xfail *-*-* } } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95
index ae2cac6..cc9a3a9 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95
@@ -30,9 +30,8 @@ end program main
! Check that only one loop is analyzed, and that it can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" } }
diff --git gcc/tree-parloops.c gcc/tree-parloops.c
index 6ce9d84..f826154 100644
--- gcc/tree-parloops.c
+++ gcc/tree-parloops.c
@@ -2040,19 +2040,20 @@ create_parallel_loop (struct loop *loop, tree loop_fn,
tree data,
tree cvar, cvar_init, initvar, cvar_next, cvar_base, type;
edge exit, nexit, guard, end, e;
- /* Prepare the GIMPLE_OMP_PARALLEL statement. */
if (oacc_kernels_p)
{
gcc_checking_assert (lookup_attribute ("oacc kernels",
DECL_ATTRIBUTES (cfun->decl)));
-
- tree clause = build_omp_clause (loc, OMP_CLAUSE_NUM_GANGS);
- OMP_CLAUSE_NUM_GANGS_EXPR (clause)
- = build_int_cst (integer_type_node, n_threads);
- oacc_set_fn_attrib (cfun->decl, clause, NULL);
+ /* Indicate to later processing that this is a parallelized OpenACC
+ kernels construct. */
+ DECL_ATTRIBUTES (cfun->decl)
+ = tree_cons (get_identifier ("oacc kernels parallelized"),
+ NULL_TREE, DECL_ATTRIBUTES (cfun->decl));
}
else
{
+ /* Prepare the GIMPLE_OMP_PARALLEL statement. */
+
basic_block bb = loop_preheader_edge (loop)->src;
basic_block paral_bb = single_pred (bb);
gsi = gsi_last_bb (paral_bb);
@@ -2154,7 +2155,8 @@ create_parallel_loop (struct loop *loop, tree loop_fn,
tree data,
/* Emit GIMPLE_OMP_FOR. */
if (oacc_kernels_p)
- /* In combination with the NUM_GANGS on the parallel. */
+ /* Parallelized OpenACC kernels constructs use gang parallelism. See also
+ omp-offload.c:execute_oacc_device_lower. */
t = build_omp_clause (loc, OMP_CLAUSE_GANG);
else
{
Committed to gomp-4_0-branch in r247958:
commit 2630763958847ab6841dc1164d29ead4ac90fe00
Author: tschwinge <tschwinge@138bc75d-0d04-0410-961f-82ee72b054a4>
Date: Fri May 12 09:20:35 2017 +0000
[PR middle-end/69921] Use "oacc kernels parallelized" attribute for
parallelized OpenACC kernels
gcc/
PR middle-end/69921
* tree-parloops.c (create_parallel_loop): Set "oacc kernels
parallelized" attribute for parallelized OpenACC kernels.
* omp-low.c (execute_oacc_device_lower): Use it.
* config/nvptx/nvptx.c (nvptx_goacc_validate_dims): Likewise.
* omp-low.c (set_oacc_fn_attrib): Make it "static".
* omp-low.h (set_oacc_fn_attrib): Remove prototype.
gcc/testsuite/
* c-c++-common/goacc/classify-kernels-unparallelized.c: Adjust.
* c-c++-common/goacc/classify-kernels.c: Likewise.
* c-c++-common/goacc/kernels-acc-loop-reduction.c: Likewise.
* c-c++-common/goacc/kernels-acc-loop-smaller-equal.c: Likewise.
* c-c++-common/goacc/kernels-counter-vars-function-scope.c:
Likewise.
* c-c++-common/goacc/kernels-double-reduction-n.c: Likewise.
* c-c++-common/goacc/kernels-double-reduction.c: Likewise.
* c-c++-common/goacc/kernels-loop-2-acc-loop.c: Likewise.
* c-c++-common/goacc/kernels-loop-2.c: Likewise.
* c-c++-common/goacc/kernels-loop-3-acc-loop.c: Likewise.
* c-c++-common/goacc/kernels-loop-3.c: Likewise.
* c-c++-common/goacc/kernels-loop-acc-loop.c: Likewise.
* c-c++-common/goacc/kernels-loop-data-2.c: Likewise.
* c-c++-common/goacc/kernels-loop-data-enter-exit-2.c: Likewise.
* c-c++-common/goacc/kernels-loop-data-enter-exit.c: Likewise.
* c-c++-common/goacc/kernels-loop-data-update.c: Likewise.
* c-c++-common/goacc/kernels-loop-data.c: Likewise.
* c-c++-common/goacc/kernels-loop-g.c: Likewise.
* c-c++-common/goacc/kernels-loop-mod-not-zero.c: Likewise.
* c-c++-common/goacc/kernels-loop-n-acc-loop.c: Likewise.
* c-c++-common/goacc/kernels-loop-n.c: Likewise.
* c-c++-common/goacc/kernels-loop-nest.c: Likewise.
* c-c++-common/goacc/kernels-loop.c: Likewise.
* c-c++-common/goacc/kernels-one-counter-var.c: Likewise.
* c-c++-common/goacc/kernels-parallel-loop-data-enter-exit.c:
Likewise.
* c-c++-common/goacc/kernels-reduction.c: Likewise.
* gfortran.dg/goacc/classify-kernels-unparallelized.f95: Likewise.
* gfortran.dg/goacc/classify-kernels.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-2.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data-2.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data-enter-exit.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data-update.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-data.f95: Likewise.
* gfortran.dg/goacc/kernels-loop-n.f95: Likewise.
* gfortran.dg/goacc/kernels-loop.f95: Likewise.
* gfortran.dg/goacc/kernels-parallel-loop-data-enter-exit.f95:
Likewise.
trunk r247957
git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/branches/gomp-4_0-branch@247958
138bc75d-0d04-0410-961f-82ee72b054a4
---
gcc/ChangeLog.gomp | 8 +++++
gcc/config/nvptx/nvptx.c | 31 ++++++----------
gcc/omp-low.c | 26 +++++++++++---
gcc/omp-low.h | 1 -
gcc/testsuite/ChangeLog.gomp | 41 ++++++++++++++++++++++
.../goacc/classify-kernels-unparallelized.c | 2 +-
.../c-c++-common/goacc/classify-kernels.c | 6 ++--
.../goacc/kernels-acc-loop-reduction.c | 3 +-
.../goacc/kernels-acc-loop-smaller-equal.c | 3 +-
.../goacc/kernels-counter-vars-function-scope.c | 3 +-
.../goacc/kernels-double-reduction-n.c | 3 +-
.../c-c++-common/goacc/kernels-double-reduction.c | 3 +-
.../c-c++-common/goacc/kernels-loop-2-acc-loop.c | 3 +-
gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c | 3 +-
.../c-c++-common/goacc/kernels-loop-3-acc-loop.c | 3 +-
gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c | 3 +-
.../c-c++-common/goacc/kernels-loop-acc-loop.c | 3 +-
.../c-c++-common/goacc/kernels-loop-data-2.c | 3 +-
.../goacc/kernels-loop-data-enter-exit-2.c | 3 +-
.../goacc/kernels-loop-data-enter-exit.c | 3 +-
.../c-c++-common/goacc/kernels-loop-data-update.c | 3 +-
.../c-c++-common/goacc/kernels-loop-data.c | 3 +-
gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c | 3 +-
.../c-c++-common/goacc/kernels-loop-mod-not-zero.c | 3 +-
.../c-c++-common/goacc/kernels-loop-n-acc-loop.c | 3 +-
gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c | 3 +-
.../c-c++-common/goacc/kernels-loop-nest.c | 3 +-
gcc/testsuite/c-c++-common/goacc/kernels-loop.c | 3 +-
.../c-c++-common/goacc/kernels-one-counter-var.c | 3 +-
.../goacc/kernels-parallel-loop-data-enter-exit.c | 3 +-
.../c-c++-common/goacc/kernels-reduction.c | 3 +-
.../goacc/classify-kernels-unparallelized.f95 | 2 +-
.../gfortran.dg/goacc/classify-kernels.f95 | 6 ++--
gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95 | 3 +-
.../gfortran.dg/goacc/kernels-loop-data-2.f95 | 3 +-
.../goacc/kernels-loop-data-enter-exit-2.f95 | 3 +-
.../goacc/kernels-loop-data-enter-exit.f95 | 3 +-
.../gfortran.dg/goacc/kernels-loop-data-update.f95 | 3 +-
.../gfortran.dg/goacc/kernels-loop-data.f95 | 3 +-
gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95 | 7 ++--
gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95 | 3 +-
.../kernels-parallel-loop-data-enter-exit.f95 | 3 +-
gcc/tree-parloops.c | 16 +++++----
43 files changed, 132 insertions(+), 110 deletions(-)
diff --git gcc/ChangeLog.gomp gcc/ChangeLog.gomp
index 8cd46c3..b34e987 100644
--- gcc/ChangeLog.gomp
+++ gcc/ChangeLog.gomp
@@ -1,5 +1,13 @@
2017-05-12 Thomas Schwinge <tho...@codesourcery.com>
+ PR middle-end/69921
+ * tree-parloops.c (create_parallel_loop): Set "oacc kernels
+ parallelized" attribute for parallelized OpenACC kernels.
+ * omp-low.c (execute_oacc_device_lower): Use it.
+ * config/nvptx/nvptx.c (nvptx_goacc_validate_dims): Likewise.
+ * omp-low.c (set_oacc_fn_attrib): Make it "static".
+ * omp-low.h (set_oacc_fn_attrib): Remove prototype.
+
* omp-low.c (expand_omp_target) <GF_OMP_TARGET_KIND_OACC_KERNELS>:
Set "oacc kernels" attribute.
(set_oacc_fn_attrib): Remove is_kernel formal parameter. Adjust
diff --git gcc/config/nvptx/nvptx.c gcc/config/nvptx/nvptx.c
index f3c1525..fd8da8d 100644
--- gcc/config/nvptx/nvptx.c
+++ gcc/config/nvptx/nvptx.c
@@ -4285,28 +4285,17 @@ nvptx_goacc_validate_dims (tree decl, int dims[], int
fn_level)
higher, as otherwise we're not expecting any parallelization to
happen. */
if (optimize >= 2
- && lookup_attribute ("oacc kernels", DECL_ATTRIBUTES (decl)))
+ && lookup_attribute ("oacc kernels", DECL_ATTRIBUTES (decl))
+ && !lookup_attribute ("oacc kernels parallelized",
+ DECL_ATTRIBUTES (decl)))
{
- bool avoid_offloading_p = true;
- for (unsigned ix = 0; ix != GOMP_DIM_MAX; ix++)
- {
- if (dims[ix] == 0 || dims[ix] > 1)
- {
- avoid_offloading_p = false;
- break;
- }
- }
- if (avoid_offloading_p)
- {
- warning_at (DECL_SOURCE_LOCATION (decl), 0,
- "OpenACC kernels construct will be executed"
- " sequentially; will by default avoid offloading to"
- " prevent data copy penalty");
- DECL_ATTRIBUTES (decl)
- = tree_cons (get_identifier ("omp avoid offloading"),
- NULL_TREE, DECL_ATTRIBUTES (decl));
-
- }
+ warning_at (DECL_SOURCE_LOCATION (decl), 0,
+ "OpenACC kernels construct will be executed"
+ " sequentially; will by default avoid offloading to"
+ " prevent data copy penalty");
+ DECL_ATTRIBUTES (decl)
+ = tree_cons (get_identifier ("omp avoid offloading"),
+ NULL_TREE, DECL_ATTRIBUTES (decl));
}
}
diff --git gcc/omp-low.c gcc/omp-low.c
index 18872f9..0fbc3ff 100644
--- gcc/omp-low.c
+++ gcc/omp-low.c
@@ -13138,7 +13138,7 @@ replace_oacc_fn_attrib (tree fn, tree dims)
function attribute. Push any that are non-constant onto the ARGS
list, along with an appropriate GOMP_LAUNCH_DIM tag. */
-void
+static void
set_oacc_fn_attrib (tree fn, tree clauses, vec<tree> *args)
{
/* Must match GOMP_DIM ordering. */
@@ -21171,6 +21171,13 @@ execute_oacc_device_lower ()
flag_openacc_dims = (char *)&flag_openacc_dims;
}
+ bool is_oacc_kernels
+ = (lookup_attribute ("oacc kernels",
+ DECL_ATTRIBUTES (current_function_decl)) != NULL);
+ bool is_oacc_kernels_parallelized
+ = (lookup_attribute ("oacc kernels parallelized",
+ DECL_ATTRIBUTES (current_function_decl)) != NULL);
+
/* Discover, partition and process the loops. */
oacc_loop *loops = oacc_loop_discovery ();
int fn_level = oacc_fn_attrib_level (attrs);
@@ -21180,17 +21187,26 @@ execute_oacc_device_lower ()
if (fn_level >= 0)
fprintf (dump_file, "Function is OpenACC routine level %d\n",
fn_level);
- else if (lookup_attribute ("oacc kernels",
- DECL_ATTRIBUTES (current_function_decl)))
- fprintf (dump_file, "Function is OpenACC kernels offload\n");
+ else if (is_oacc_kernels)
+ fprintf (dump_file, "Function is %s OpenACC kernels offload\n",
+ (is_oacc_kernels_parallelized
+ ? "parallelized" : "unparallelized"));
else
fprintf (dump_file, "Function is OpenACC parallel offload\n");
}
unsigned outer_mask = fn_level >= 0 ? GOMP_DIM_MASK (fn_level) - 1 : 0;
unsigned used_mask = oacc_loop_partition (loops, outer_mask);
+ /* OpenACC kernels constructs are special: they currently don't use the
+ generic oacc_loop infrastructure and attribute/dimension processing. */
+ if (is_oacc_kernels && is_oacc_kernels_parallelized)
+ {
+ /* Parallelized OpenACC kernels constructs use gang parallelism. See
+ also tree-parloops.c:create_parallel_loop. */
+ used_mask |= GOMP_DIM_MASK (GOMP_DIM_GANG);
+ }
+
int dims[GOMP_DIM_MAX];
-
oacc_validate_dims (current_function_decl, attrs, dims, fn_level, used_mask);
if (dump_file)
diff --git gcc/omp-low.h gcc/omp-low.h
index 0ea5308..65b2433 100644
--- gcc/omp-low.h
+++ gcc/omp-low.h
@@ -34,7 +34,6 @@ extern void replace_oacc_fn_attrib (tree, tree);
extern int verify_oacc_routine_clauses (tree, tree *, location_t, const char
*);
extern tree build_oacc_routine_dims (tree);
extern tree get_oacc_fn_attrib (tree);
-extern void set_oacc_fn_attrib (tree, tree, vec<tree> *);
extern int get_oacc_ifn_dim_arg (const gimple *);
extern int get_oacc_fn_dim_size (tree, int);
diff --git gcc/testsuite/ChangeLog.gomp gcc/testsuite/ChangeLog.gomp
index 9f2b7be..dadff1a 100644
--- gcc/testsuite/ChangeLog.gomp
+++ gcc/testsuite/ChangeLog.gomp
@@ -1,5 +1,46 @@
2017-05-12 Thomas Schwinge <tho...@codesourcery.com>
+ * c-c++-common/goacc/classify-kernels-unparallelized.c: Adjust.
+ * c-c++-common/goacc/classify-kernels.c: Likewise.
+ * c-c++-common/goacc/kernels-acc-loop-reduction.c: Likewise.
+ * c-c++-common/goacc/kernels-acc-loop-smaller-equal.c: Likewise.
+ * c-c++-common/goacc/kernels-counter-vars-function-scope.c:
+ Likewise.
+ * c-c++-common/goacc/kernels-double-reduction-n.c: Likewise.
+ * c-c++-common/goacc/kernels-double-reduction.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-2-acc-loop.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-2.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-3-acc-loop.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-3.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-acc-loop.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-data-2.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-data-enter-exit-2.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-data-enter-exit.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-data-update.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-data.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-g.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-mod-not-zero.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-n-acc-loop.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-n.c: Likewise.
+ * c-c++-common/goacc/kernels-loop-nest.c: Likewise.
+ * c-c++-common/goacc/kernels-loop.c: Likewise.
+ * c-c++-common/goacc/kernels-one-counter-var.c: Likewise.
+ * c-c++-common/goacc/kernels-parallel-loop-data-enter-exit.c:
+ Likewise.
+ * c-c++-common/goacc/kernels-reduction.c: Likewise.
+ * gfortran.dg/goacc/classify-kernels-unparallelized.f95: Likewise.
+ * gfortran.dg/goacc/classify-kernels.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-2.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data-2.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data-enter-exit.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data-update.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-data.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop-n.f95: Likewise.
+ * gfortran.dg/goacc/kernels-loop.f95: Likewise.
+ * gfortran.dg/goacc/kernels-parallel-loop-data-enter-exit.f95:
+ Likewise.
+
* c-c++-common/goacc/oaccdevlow-kernels.c: Adjust.
* c-c++-common/goacc/oaccdevlow-parallel.c: Likewise.
* c-c++-common/goacc/oaccdevlow-routine.c: Likewise.
diff --git gcc/testsuite/c-c++-common/goacc/classify-kernels-unparallelized.c
gcc/testsuite/c-c++-common/goacc/classify-kernels-unparallelized.c
index 70ff428..626f6b4 100644
--- gcc/testsuite/c-c++-common/goacc/classify-kernels-unparallelized.c
+++ gcc/testsuite/c-c++-common/goacc/classify-kernels-unparallelized.c
@@ -34,6 +34,6 @@ void KERNELS ()
/* Check the offloaded function's classification and compute dimensions (will
always be 1 x 1 x 1 for non-offloading compilation).
- { dg-final { scan-tree-dump-times "(?n)Function is OpenACC kernels offload"
1 "oaccdevlow" } }
+ { dg-final { scan-tree-dump-times "(?n)Function is unparallelized OpenACC
kernels offload" 1 "oaccdevlow" } }
{ dg-final { scan-tree-dump-times "(?n)Compute dimensions \\\[1, 1, 1\\\]"
1 "oaccdevlow" } }
{ dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } } */
diff --git gcc/testsuite/c-c++-common/goacc/classify-kernels.c
gcc/testsuite/c-c++-common/goacc/classify-kernels.c
index c8b0fda..95037e6 100644
--- gcc/testsuite/c-c++-common/goacc/classify-kernels.c
+++ gcc/testsuite/c-c++-common/goacc/classify-kernels.c
@@ -25,11 +25,11 @@ void KERNELS ()
/* Check that exactly one OpenACC kernels construct is analyzed, and that it
can be parallelized.
{ dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } }
- { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(0, , \\), oacc kernels, omp target entrypoint\\)\\)" 1 "parloops1" } }
+ { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } }
{ dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check the offloaded function's classification and compute dimensions (will
always be 1 x 1 x 1 for non-offloading compilation).
- { dg-final { scan-tree-dump-times "(?n)Function is OpenACC kernels offload"
1 "oaccdevlow" } }
+ { dg-final { scan-tree-dump-times "(?n)Function is parallelized OpenACC
kernels offload" 1 "oaccdevlow" } }
{ dg-final { scan-tree-dump-times "(?n)Compute dimensions \\\[1, 1, 1\\\]"
1 "oaccdevlow" } }
- { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } } */
+ { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels parallelized, oacc function \\(, , \\), oacc
kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } } */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-acc-loop-reduction.c
gcc/testsuite/c-c++-common/goacc/kernels-acc-loop-reduction.c
index b52f280..4824e53 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-acc-loop-reduction.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-acc-loop-reduction.c
@@ -16,9 +16,8 @@ foo (int n, unsigned int *a)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*\\._omp_fn\\.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-acc-loop-smaller-equal.c
gcc/testsuite/c-c++-common/goacc/kernels-acc-loop-smaller-equal.c
index dd10c46..d70afb0 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-acc-loop-smaller-equal.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-acc-loop-smaller-equal.c
@@ -16,9 +16,8 @@ foo (int n)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*\\._omp_fn\\.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git
gcc/testsuite/c-c++-common/goacc/kernels-counter-vars-function-scope.c
gcc/testsuite/c-c++-common/goacc/kernels-counter-vars-function-scope.c
index 17f240e..c475333 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-counter-vars-function-scope.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-counter-vars-function-scope.c
@@ -45,9 +45,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-double-reduction-n.c
gcc/testsuite/c-c++-common/goacc/kernels-double-reduction-n.c
index 750f576..27ea2e9 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-double-reduction-n.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-double-reduction-n.c
@@ -27,10 +27,9 @@ foo (unsigned int n)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* { dg-final { scan-tree-dump-times "parallelizing outer loop" 1 "parloops1"
} } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-double-reduction.c
gcc/testsuite/c-c++-common/goacc/kernels-double-reduction.c
index df60d6a..0841e90 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-double-reduction.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-double-reduction.c
@@ -27,10 +27,9 @@ foo (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* { dg-final { scan-tree-dump-times "parallelizing outer loop" 1 "parloops1"
} } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-2-acc-loop.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-2-acc-loop.c
index 21b2a70..7b127cb 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-2-acc-loop.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-2-acc-loop.c
@@ -9,11 +9,10 @@
/* Check that only three loops are analyzed, and that all can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.2" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c
index 913d91f..acef6a1 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-2.c
@@ -59,11 +59,10 @@ main (void)
/* Check that only three loops are analyzed, and that all can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.2" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-3-acc-loop.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-3-acc-loop.c
index d82debc..a040e09 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-3-acc-loop.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-3-acc-loop.c
@@ -8,9 +8,8 @@
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c
index 1822d2a..75e2bb7 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-3.c
@@ -39,9 +39,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-acc-loop.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-acc-loop.c
index dc3bb43..070a5b5 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-acc-loop.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-acc-loop.c
@@ -8,9 +8,8 @@
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-data-2.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-data-2.c
index 8046ae9..7180021 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-data-2.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-data-2.c
@@ -59,11 +59,10 @@ main (void)
/* Check that only three loops are analyzed, and that all can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.2" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-data-enter-exit-2.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-data-enter-exit-2.c
index ac977d2..0c9f833 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-data-enter-exit-2.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-data-enter-exit-2.c
@@ -57,11 +57,10 @@ main (void)
/* Check that only three loops are analyzed, and that all can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.2" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-data-enter-exit.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-data-enter-exit.c
index 72e18a7..0bd21b6 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-data-enter-exit.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-data-enter-exit.c
@@ -54,11 +54,10 @@ main (void)
/* Check that only three loops are analyzed, and that all can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.2" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-data-update.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-data-update.c
index 94d5702..dd5a841 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-data-update.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-data-update.c
@@ -55,10 +55,9 @@ main (void)
/* Check that only two loops are analyzed, and that both can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 2
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 2 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 2 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-data.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-data.c
index cc49699..a658182 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-data.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-data.c
@@ -53,11 +53,10 @@ main (void)
/* Check that only three loops are analyzed, and that all can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.2" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c
index e946319..73b469d 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-g.c
@@ -7,9 +7,8 @@
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-mod-not-zero.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-mod-not-zero.c
index 9b63b45..5592623 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-mod-not-zero.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-mod-not-zero.c
@@ -43,9 +43,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-n-acc-loop.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-n-acc-loop.c
index 685cb7f..1f25e63 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-n-acc-loop.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-n-acc-loop.c
@@ -8,9 +8,8 @@
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c
index 279f797..e86be1b 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-n.c
@@ -46,9 +46,8 @@ foo (COUNTERTYPE n)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop-nest.c
gcc/testsuite/c-c++-common/goacc/kernels-loop-nest.c
index db1071f..2b0e186 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop-nest.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop-nest.c
@@ -30,9 +30,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-loop.c
gcc/testsuite/c-c++-common/goacc/kernels-loop.c
index abf7a3c..9619d53 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-loop.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-loop.c
@@ -46,9 +46,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-one-counter-var.c
gcc/testsuite/c-c++-common/goacc/kernels-one-counter-var.c
index 95f4817..69539b2 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-one-counter-var.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-one-counter-var.c
@@ -44,9 +44,8 @@ main (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git
gcc/testsuite/c-c++-common/goacc/kernels-parallel-loop-data-enter-exit.c
gcc/testsuite/c-c++-common/goacc/kernels-parallel-loop-data-enter-exit.c
index 70c5469..58c9416 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-parallel-loop-data-enter-exit.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-parallel-loop-data-enter-exit.c
@@ -55,11 +55,10 @@ main (void)
/* Check that only two loops are analyzed, and that both can be
parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 2
"parloops1" { xfail *-*-* } } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 2 "parloops1" { xfail *-*-* } } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" { xfail *-*-* } } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.0" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.1" 1
"optimized" } } */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*main._omp_fn.2" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 2 "parloops1" {
xfail *-*-* } } } */
diff --git gcc/testsuite/c-c++-common/goacc/kernels-reduction.c
gcc/testsuite/c-c++-common/goacc/kernels-reduction.c
index 6f5a418..4a18272 100644
--- gcc/testsuite/c-c++-common/goacc/kernels-reduction.c
+++ gcc/testsuite/c-c++-common/goacc/kernels-reduction.c
@@ -26,9 +26,8 @@ foo (void)
/* Check that only one loop is analyzed, and that it can be parallelized. */
/* { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } } */
+/* { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } } */
/* { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } } */
/* Check that the loop has been split off into a function. */
/* { dg-final { scan-tree-dump-times "(?n);; Function .*foo.*._omp_fn.0" 1
"optimized" } } */
-
-/* { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" }
} */
diff --git gcc/testsuite/gfortran.dg/goacc/classify-kernels-unparallelized.f95
gcc/testsuite/gfortran.dg/goacc/classify-kernels-unparallelized.f95
index 9887d35..4b282ca 100644
--- gcc/testsuite/gfortran.dg/goacc/classify-kernels-unparallelized.f95
+++ gcc/testsuite/gfortran.dg/goacc/classify-kernels-unparallelized.f95
@@ -36,6 +36,6 @@ end program main
! Check the offloaded function's classification and compute dimensions (will
! always be 1 x 1 x 1 for non-offloading compilation).
-! { dg-final { scan-tree-dump-times "(?n)Function is OpenACC kernels offload"
1 "oaccdevlow" } }
+! { dg-final { scan-tree-dump-times "(?n)Function is unparallelized OpenACC
kernels offload" 1 "oaccdevlow" } }
! { dg-final { scan-tree-dump-times "(?n)Compute dimensions \\\[1, 1, 1\\\]" 1
"oaccdevlow" } }
! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } }
diff --git gcc/testsuite/gfortran.dg/goacc/classify-kernels.f95
gcc/testsuite/gfortran.dg/goacc/classify-kernels.f95
index 69c89a9..da025c1 100644
--- gcc/testsuite/gfortran.dg/goacc/classify-kernels.f95
+++ gcc/testsuite/gfortran.dg/goacc/classify-kernels.f95
@@ -27,11 +27,11 @@ end program main
! Check that exactly one OpenACC kernels construct is analyzed, and that it
! can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } }
-! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(0, , \\), oacc kernels, omp target entrypoint\\)\\)" 1 "parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check the offloaded function's classification and compute dimensions (will
! always be 1 x 1 x 1 for non-offloading compilation).
-! { dg-final { scan-tree-dump-times "(?n)Function is OpenACC kernels offload"
1 "oaccdevlow" } }
+! { dg-final { scan-tree-dump-times "(?n)Function is parallelized OpenACC
kernels offload" 1 "oaccdevlow" } }
! { dg-final { scan-tree-dump-times "(?n)Compute dimensions \\\[1, 1, 1\\\]" 1
"oaccdevlow" } }
-! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc function
\\(1, 1, 1\\), oacc kernels parallelized, oacc function \\(, , \\), oacc
kernels, omp target entrypoint\\)\\)" 1 "oaccdevlow" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95
index 865f7a6..516aede 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-2.f95
@@ -34,11 +34,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-2.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-2.f95
index c9f3a62..ff3788a 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-2.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-2.f95
@@ -40,11 +40,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95
index 3361607..60a5c96 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit-2.f95
@@ -40,11 +40,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit.f95
index 5ba56fb..ce04749 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-enter-exit.f95
@@ -38,11 +38,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-update.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-update.f95
index a622a96..d2de138 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-update.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data-update.f95
@@ -38,10 +38,9 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 2
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 2 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 2 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-data.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-data.f95
index 4ec2ac3..92872b2 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-data.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-data.f95
@@ -38,11 +38,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 3
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 3 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 3 "parloops1" } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95
index bdfebde..7c56897 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop-n.f95
@@ -32,11 +32,10 @@ end module test
! Check that only one loop is analyzed, and that it can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } }
+! This failure was introduced with the GOMP_MAP_POINTER ->
+! GOMP_MAP_FIRSTPRIVATE_POINTER conversion.
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" { xfail *-*-* } } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function __test_MOD_foo._omp_fn.0
" 1 "optimized" } }
-
-! This failure was introduced with the GOMP_MAP_POINTER ->
-! GOMP_MAP_FIRSTPRIVATE_POINTER conversion.
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" {
xfail *-*-* } } }
diff --git gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95
gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95
index ae2cac6..cc9a3a9 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-loop.f95
@@ -30,9 +30,8 @@ end program main
! Check that only one loop is analyzed, and that it can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 1
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 1 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 1 "parloops1" } }
diff --git
gcc/testsuite/gfortran.dg/goacc/kernels-parallel-loop-data-enter-exit.f95
gcc/testsuite/gfortran.dg/goacc/kernels-parallel-loop-data-enter-exit.f95
index 49b767c..16c9b80 100644
--- gcc/testsuite/gfortran.dg/goacc/kernels-parallel-loop-data-enter-exit.f95
+++ gcc/testsuite/gfortran.dg/goacc/kernels-parallel-loop-data-enter-exit.f95
@@ -39,11 +39,10 @@ end program main
! Check that only three loops are analyzed, and that all can be parallelized.
! { dg-final { scan-tree-dump-times "SUCCESS: may be parallelized" 2
"parloops1" } }
+! { dg-final { scan-tree-dump-times "(?n)__attribute__\\(\\(oacc kernels
parallelized, oacc function \\(, , \\), oacc kernels, omp target
entrypoint\\)\\)" 2 "parloops1" } }
! { dg-final { scan-tree-dump-not "FAILED:" "parloops1" } }
! Check that the loop has been split off into a function.
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.0 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.1 " 1
"optimized" } }
! { dg-final { scan-tree-dump-times "(?n);; Function MAIN__._omp_fn.2 " 1
"optimized" } }
-
-! { dg-final { scan-tree-dump-times "(?n)oacc function \\(0," 2 "parloops1" } }
diff --git gcc/tree-parloops.c gcc/tree-parloops.c
index a94f7a2..02736e0 100644
--- gcc/tree-parloops.c
+++ gcc/tree-parloops.c
@@ -2038,19 +2038,20 @@ create_parallel_loop (struct loop *loop, tree loop_fn,
tree data,
tree cvar, cvar_init, initvar, cvar_next, cvar_base, type;
edge exit, nexit, guard, end, e;
- /* Prepare the GIMPLE_OMP_PARALLEL statement. */
if (oacc_kernels_p)
{
gcc_checking_assert (lookup_attribute ("oacc kernels",
DECL_ATTRIBUTES (cfun->decl)));
-
- tree clause = build_omp_clause (loc, OMP_CLAUSE_NUM_GANGS);
- OMP_CLAUSE_NUM_GANGS_EXPR (clause)
- = build_int_cst (integer_type_node, n_threads);
- set_oacc_fn_attrib (cfun->decl, clause, NULL);
+ /* Indicate to later processing that this is a parallelized OpenACC
+ kernels construct. */
+ DECL_ATTRIBUTES (cfun->decl)
+ = tree_cons (get_identifier ("oacc kernels parallelized"),
+ NULL_TREE, DECL_ATTRIBUTES (cfun->decl));
}
else
{
+ /* Prepare the GIMPLE_OMP_PARALLEL statement. */
+
basic_block bb = loop_preheader_edge (loop)->src;
basic_block paral_bb = single_pred (bb);
gsi = gsi_last_bb (paral_bb);
@@ -2152,7 +2153,8 @@ create_parallel_loop (struct loop *loop, tree loop_fn,
tree data,
/* Emit GIMPLE_OMP_FOR. */
if (oacc_kernels_p)
- /* In combination with the NUM_GANGS on the parallel. */
+ /* Parallelized OpenACC kernels constructs use gang parallelism. See also
+ omp-low.c:execute_oacc_device_lower. */
t = build_omp_clause (loc, OMP_CLAUSE_GANG);
else
{
Grüße
Thomas
