From c99042ba973be37d9a5bcbb258f83251bf36b673 Mon Sep 17 00:00:00 2001
From: Alexey Bataev <a.bataev@hotmail.com>
Date: Thu, 15 Mar 2018 18:10:54 +0000
Subject: [OPENMP, NVPTX] Improve globalization of the variables captured by
 value.

If the variable is captured by value and the corresponding parameter in
the outlined function escapes its declaration context, this parameter
must be globalized. To globalize it we need to get the address of the
original parameter, load the value, store it to the global address and
use this global address instead of the original.

Patch improves globalization for parallel|teams regions + functions in
declare target regions.

llvm-svn: 327654
---
 clang/test/OpenMP/nvptx_teams_codegen.cpp | 24 ++++++++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)

(limited to 'clang/test/OpenMP/nvptx_teams_codegen.cpp')

diff --git a/clang/test/OpenMP/nvptx_teams_codegen.cpp b/clang/test/OpenMP/nvptx_teams_codegen.cpp
index b26d47c706a..c530d19fd7b 100644
--- a/clang/test/OpenMP/nvptx_teams_codegen.cpp
+++ b/clang/test/OpenMP/nvptx_teams_codegen.cpp
@@ -36,8 +36,12 @@ int main (int argc, char **argv) {
 // CK1:  store {{.+}} 0, {{.+}},
 // CK1:  store i{{[0-9]+}} [[ARGC]], i{{[0-9]+}}* [[ARGCADDR]],
 // CK1-64:  [[CONV:%.+]] = bitcast i{{[0-9]+}}* [[ARGCADDR]] to i{{[0-9]+}}*
-// CK1-64:  store i{{[0-9]+}}* [[CONV]], i{{[0-9]+}}** [[ARGCADDR_PTR]],
-// CK1-32:  store i{{[0-9]+}}* [[ARGCADDR]], i{{[0-9]+}}** [[ARGCADDR_PTR]],
+// CK1:  call i8* @__kmpc_data_sharing_push_stack(i{{[0-9]+}} 4, i16 0)
+// CK1-64:  [[ARG:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[CONV]]
+// CK1-32:  [[ARG:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[ARGCADDR]]
+// CK1:  [[ARGCADDR:%.+]] = getelementptr inbounds %struct.{{.*}}, %struct.{{.*}}* %{{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
+// CK1:  store i{{[0-9]+}} [[ARG]], i{{[0-9]+}}* [[ARGCADDR]],
+// CK1:  store i{{[0-9]+}}* [[ARGCADDR]], i{{[0-9]+}}** [[ARGCADDR_PTR]],
 // CK1:  [[ARGCADDR_PTR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[ARGCADDR_PTR]],
 // CK1:  store i{{[0-9]+}} 0, i{{[0-9]+}}* [[ARGCADDR_PTR_REF]],
 // CK1-NOT: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(
@@ -49,6 +53,10 @@ int main (int argc, char **argv) {
 // CK1: [[ARGCADDR_PTR:%.+]] = alloca i{{.+}}***,
 // CK1: [[ARGCADDR:%.+]] = alloca i{{.+}}**,
 // CK1: store i{{.+}}** [[ARGC]], i{{.+}}*** [[ARGCADDR]]
+// CK1: call i8* @__kmpc_data_sharing_push_stack(i{{[0-9]+}} {{4|8}}, i16 0)
+// CK1: [[ARG:%.+]] = load i{{[0-9]+}}**, i{{[0-9]+}}*** [[ARGCADDR]]
+// CK1: [[ARGCADDR:%.+]] = getelementptr inbounds %struct.{{.*}}, %struct.{{.*}}* %{{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
+// CK1: store i{{[0-9]+}}** [[ARG]], i{{[0-9]+}}*** [[ARGCADDR]],
 // CK1: store i8*** [[ARGCADDR]], i8**** [[ARGCADDR_PTR]],
 // CK1: [[ARGCADDR_PTR_REF:%.+]] = load i{{.+}}**, i{{.+}}*** [[ARGCADDR_PTR]],
 // CK1: store i{{[0-9]+}}** null, i{{[0-9]+}}*** [[ARGCADDR_PTR_REF]],
@@ -104,8 +112,12 @@ int main (int argc, char **argv) {
 // CK2-64: [[ACONV:%.+]] = bitcast i64* [[AADDR]] to i32*
 // CK2-64: [[BCONV:%.+]] = bitcast i64* [[BADDR]] to i32*
 // CK2-64: [[CONV:%.+]] = bitcast i64* [[ARGCADDR]] to i32*
-// CK2-64:  store i{{[0-9]+}}* [[CONV]], i{{[0-9]+}}** [[ARGCADDR_PTR]],
-// CK2-32:  store i{{[0-9]+}}* [[ARGCADDR]], i{{[0-9]+}}** [[ARGCADDR_PTR]],
+// CK2:  call i8* @__kmpc_data_sharing_push_stack(i{{[0-9]+}} 4, i16 0)
+// CK2-64:  [[ARG:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[CONV]]
+// CK2-32:  [[ARG:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[ARGCADDR]]
+// CK2:  [[ARGCADDR:%.+]] = getelementptr inbounds %struct.{{.*}}, %struct.{{.*}}* %{{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
+// CK2:  store i{{[0-9]+}} [[ARG]], i{{[0-9]+}}* [[ARGCADDR]],
+// CK2:  store i{{[0-9]+}}* [[ARGCADDR]], i{{[0-9]+}}** [[ARGCADDR_PTR]],
 // CK2:  [[ARGCADDR_PTR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[ARGCADDR_PTR]],
 // CK2: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[ARGCADDR_PTR_REF]],
 // CK2-NOT:  {{.+}} = call i32 @__kmpc_push_num_teams(
@@ -121,6 +133,10 @@ int main (int argc, char **argv) {
 // CK2: store i{{[0-9]+}} [[A_IN]], i{{[0-9]+}}* [[AADDR]],
 // CK2: store i{{[0-9]+}} [[B_IN]], i{{[0-9]+}}* [[BADDR]],
 // CK2: store i{{[0-9]+}}** [[ARGC]], i{{[0-9]+}}*** [[ARGCADDR]],
+// CK2: call i8* @__kmpc_data_sharing_push_stack(i{{[0-9]+}} {{4|8}}, i16 0)
+// CK2: [[ARG:%.+]] = load i{{[0-9]+}}**, i{{[0-9]+}}*** [[ARGCADDR]]
+// CK2: [[ARGCADDR:%.+]] = getelementptr inbounds %struct.{{.*}}, %struct.{{.*}}* %{{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
+// CK2: store i{{[0-9]+}}** [[ARG]], i{{[0-9]+}}*** [[ARGCADDR]],
 // CK2: store i{{[0-9]+}}*** [[ARGCADDR]], i{{[0-9]+}}**** [[ARGCADDR_PTR]],
 // CK2: [[ARGCADDR_PTR_REF:%.+]] = load i{{[0-9]+}}***, i{{[0-9]+}}**** [[ARGCADDR_PTR]],
 // CK2: store i{{[0-9]+}}** null, i{{[0-9]+}}*** [[ARGCADDR_PTR_REF]],
-- 
cgit v1.2.3