diff options
Diffstat (limited to 'test')
73 files changed, 5404 insertions, 156 deletions
diff --git a/test/Analysis/enum.cpp b/test/Analysis/enum.cpp index e26b8f00d98ca..b561e65ac86ea 100644 --- a/test/Analysis/enum.cpp +++ b/test/Analysis/enum.cpp @@ -24,3 +24,16 @@ void testCasting(int i) { clang_analyzer_eval(j == 0); // expected-warning{{FALSE}} } } + +enum class EnumBool : bool { + F = false, + T = true +}; + +bool testNoCrashOnSwitchEnumBool(EnumBool E) { + switch (E) { + case EnumBool::F: + return false; + } + return true; +} diff --git a/test/Analysis/gmalloc.c b/test/Analysis/gmalloc.c index 10c4fe2840545..50413e2e9b13c 100644 --- a/test/Analysis/gmalloc.c +++ b/test/Analysis/gmalloc.c @@ -13,6 +13,12 @@ gpointer g_realloc(gpointer mem, gsize n_bytes); gpointer g_try_malloc(gsize n_bytes); gpointer g_try_malloc0(gsize n_bytes); gpointer g_try_realloc(gpointer mem, gsize n_bytes); +gpointer g_malloc_n(gsize n_blocks, gsize n_block_bytes); +gpointer g_malloc0_n(gsize n_blocks, gsize n_block_bytes); +gpointer g_realloc_n(gpointer mem, gsize n_blocks, gsize n_block_bytes); +gpointer g_try_malloc_n(gsize n_blocks, gsize n_block_bytes); +gpointer g_try_malloc0_n(gsize n_blocks, gsize n_block_bytes); +gpointer g_try_realloc_n(gpointer mem, gsize n_blocks, gsize n_block_bytes); void g_free(gpointer mem); gpointer g_memdup(gconstpointer mem, guint byte_size); @@ -25,6 +31,12 @@ void f1() { gpointer g3 = g_try_malloc(n_bytes); gpointer g4 = g_try_malloc0(n_bytes); g3 = g_try_realloc(g3, n_bytes * 2); + gpointer g5 = g_malloc_n(n_bytes, sizeof(char)); + gpointer g6 = g_malloc0_n(n_bytes, sizeof(char)); + g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); + gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); + gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char)); + g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); g_free(g1); g_free(g2); @@ -38,6 +50,12 @@ void f2() { gpointer g3 = g_try_malloc(n_bytes); gpointer g4 = g_try_malloc0(n_bytes); g3 = g_try_realloc(g3, n_bytes * 2); + gpointer g5 = g_malloc_n(n_bytes, sizeof(char)); + gpointer g6 = g_malloc0_n(n_bytes, sizeof(char)); + g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); + gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); + gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char)); + g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); g_free(g1); g_free(g2); @@ -52,8 +70,100 @@ void f3() { gpointer g3 = g_try_malloc(n_bytes); gpointer g4 = g_try_malloc0(n_bytes); g3 = g_try_realloc(g3, n_bytes * 2); // expected-warning{{Potential leak of memory pointed to by 'g4'}} + gpointer g5 = g_malloc_n(n_bytes, sizeof(char)); + gpointer g6 = g_malloc0_n(n_bytes, sizeof(char)); + g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g6'}} + gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g5'}} + gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char)); + g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}} + + g_free(g1); // expected-warning{{Potential leak of memory pointed to by 'g7'}} + g_free(g2); + g_free(g3); +} + +void f4() { + gpointer g1 = g_malloc(n_bytes); + gpointer g2 = g_malloc0(n_bytes); + g1 = g_realloc(g1, n_bytes * 2); + gpointer g3 = g_try_malloc(n_bytes); + gpointer g4 = g_try_malloc0(n_bytes); + g3 = g_try_realloc(g3, n_bytes * 2); + gpointer g5 = g_malloc_n(n_bytes, sizeof(char)); + gpointer g6 = g_malloc0_n(n_bytes, sizeof(char)); + g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g6'}} + gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g5'}} + gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char)); + g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}} + + g_free(g1); // expected-warning{{Potential leak of memory pointed to by 'g7'}} + g_free(g2); + g_free(g3); + g_free(g4); +} + +void f5() { + gpointer g1 = g_malloc(n_bytes); + gpointer g2 = g_malloc0(n_bytes); + g1 = g_realloc(g1, n_bytes * 2); + gpointer g3 = g_try_malloc(n_bytes); + gpointer g4 = g_try_malloc0(n_bytes); + g3 = g_try_realloc(g3, n_bytes * 2); + gpointer g5 = g_malloc_n(n_bytes, sizeof(char)); + gpointer g6 = g_malloc0_n(n_bytes, sizeof(char)); + g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g6'}} + gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); + gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char)); + g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}} + + g_free(g1); // expected-warning{{Potential leak of memory pointed to by 'g7'}} + g_free(g2); + g_free(g3); + g_free(g4); + g_free(g5); +} + +void f6() { + gpointer g1 = g_malloc(n_bytes); + gpointer g2 = g_malloc0(n_bytes); + g1 = g_realloc(g1, n_bytes * 2); + gpointer g3 = g_try_malloc(n_bytes); + gpointer g4 = g_try_malloc0(n_bytes); + g3 = g_try_realloc(g3, n_bytes * 2); + gpointer g5 = g_malloc_n(n_bytes, sizeof(char)); + gpointer g6 = g_malloc0_n(n_bytes, sizeof(char)); + g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); + gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); + gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char)); + g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}} + + g_free(g1); // expected-warning{{Potential leak of memory pointed to by 'g7'}} + g_free(g2); + g_free(g3); + g_free(g4); + g_free(g5); + g_free(g6); +} + +void f7() { + gpointer g1 = g_malloc(n_bytes); + gpointer g2 = g_malloc0(n_bytes); + g1 = g_realloc(g1, n_bytes * 2); + gpointer g3 = g_try_malloc(n_bytes); + gpointer g4 = g_try_malloc0(n_bytes); + g3 = g_try_realloc(g3, n_bytes * 2); + gpointer g5 = g_malloc_n(n_bytes, sizeof(char)); + gpointer g6 = g_malloc0_n(n_bytes, sizeof(char)); + g5 = g_realloc_n(g5, n_bytes * 2, sizeof(char)); + gpointer g7 = g_try_malloc_n(n_bytes, sizeof(char)); + gpointer g8 = g_try_malloc0_n(n_bytes, sizeof(char)); + g7 = g_try_realloc_n(g7, n_bytes * 2, sizeof(char)); // expected-warning{{Potential leak of memory pointed to by 'g8'}} g_free(g1); g_free(g2); g_free(g3); + g_free(g4); + g_free(g5); + g_free(g6); + g_free(g7); } diff --git a/test/Analysis/inlining/inline-defensive-checks.c b/test/Analysis/inlining/inline-defensive-checks.c index 4029da651b65f..010d3a774756a 100644 --- a/test/Analysis/inlining/inline-defensive-checks.c +++ b/test/Analysis/inlining/inline-defensive-checks.c @@ -1,7 +1,7 @@ // RUN: %clang_analyze_cc1 -analyzer-checker=core -analyzer-config suppress-inlined-defensive-checks=true -verify %s // Perform inline defensive checks. -void idc(int *p) { +void idc(void *p) { if (p) ; } @@ -139,3 +139,42 @@ void idcTrackZeroThroughDoubleAssignemnt(int x) { int z = y; idcTriggerZeroValueThroughCall(z); } + +struct S { + int f1; + int f2; +}; + +void idcTrackZeroValueThroughUnaryPointerOperators(struct S *s) { + idc(s); + *(&(s->f1)) = 7; // no-warning +} + +void idcTrackZeroValueThroughUnaryPointerOperatorsWithOffset1(struct S *s) { + idc(s); + int *x = &(s->f2); + *x = 7; // no-warning +} + +void idcTrackZeroValueThroughUnaryPointerOperatorsWithOffset2(struct S *s) { + idc(s); + int *x = &(s->f2) - 1; + // FIXME: Should not warn. + *x = 7; // expected-warning{{Dereference of null pointer}} +} + +void idcTrackZeroValueThroughUnaryPointerOperatorsWithAssignment(struct S *s) { + idc(s); + int *x = &(s->f1); + *x = 7; // no-warning +} + + +struct S2 { + int a[1]; +}; + +void idcTrackZeroValueThroughUnaryPointerOperatorsWithArrayField(struct S2 *s) { + idc(s); + *(&(s->a[0])) = 7; // no-warning +} diff --git a/test/Analysis/inlining/inline-defensive-checks.cpp b/test/Analysis/inlining/inline-defensive-checks.cpp index 6a803fa695c6c..eaae8d2ae28f2 100644 --- a/test/Analysis/inlining/inline-defensive-checks.cpp +++ b/test/Analysis/inlining/inline-defensive-checks.cpp @@ -70,4 +70,17 @@ int *retNull() { void test(int *p1, int *p2) { idc(p1); Foo f(p1); -}
\ No newline at end of file +} + +struct Bar { + int x; +}; +void idcBar(Bar *b) { + if (b) + ; +} +void testRefToField(Bar *b) { + idcBar(b); + int &x = b->x; // no-warning + x = 5; +} diff --git a/test/Analysis/null-deref-offsets.c b/test/Analysis/null-deref-offsets.c new file mode 100644 index 0000000000000..988cec4985daa --- /dev/null +++ b/test/Analysis/null-deref-offsets.c @@ -0,0 +1,37 @@ +// RUN: %clang_analyze_cc1 -w -triple i386-apple-darwin10 -analyzer-checker=core,debug.ExprInspection -verify %s + +void clang_analyzer_eval(int); + +struct S { + int x, y; + int z[2]; +}; + +void testOffsets(struct S *s, int coin) { + if (s != 0) + return; + + // FIXME: Here we are testing the hack that computes offsets to null pointers + // as 0 in order to find null dereferences of not-exactly-null pointers, + // such as &(s->y) below, which is equal to 4 rather than 0 in run-time. + + // These are indeed null. + clang_analyzer_eval(s == 0); // expected-warning{{TRUE}} + clang_analyzer_eval(&(s->x) == 0); // expected-warning{{TRUE}} + + // FIXME: These should ideally be true. + clang_analyzer_eval(&(s->y) == 4); // expected-warning{{FALSE}} + clang_analyzer_eval(&(s->z[0]) == 8); // expected-warning{{FALSE}} + clang_analyzer_eval(&(s->z[1]) == 12); // expected-warning{{FALSE}} + + // FIXME: These should ideally be false. + clang_analyzer_eval(&(s->y) == 0); // expected-warning{{TRUE}} + clang_analyzer_eval(&(s->z[0]) == 0); // expected-warning{{TRUE}} + clang_analyzer_eval(&(s->z[1]) == 0); // expected-warning{{TRUE}} + + // But these should still be reported as null dereferences. + if (coin) + s->y = 5; // expected-warning{{Access to field 'y' results in a dereference of a null pointer (loaded from variable 's')}} + else + s->z[1] = 6; // expected-warning{{Array access (via field 'z') results in a null pointer dereference}} +} diff --git a/test/Analysis/uninit-const.cpp b/test/Analysis/uninit-const.cpp index 75e932a77cedb..db969bfb67d62 100644 --- a/test/Analysis/uninit-const.cpp +++ b/test/Analysis/uninit-const.cpp @@ -122,7 +122,7 @@ void f1(void) { } void f_uninit(void) { - int x; + int x; // expected-note {{'x' declared without an initial value}} doStuff_uninit(&x); // expected-warning {{1st function call argument is a pointer to uninitialized value}} // expected-note@-1 {{1st function call argument is a pointer to uninitialized value}} } diff --git a/test/CXX/modules-ts/basic/basic.link/module-declaration.cpp b/test/CXX/modules-ts/basic/basic.link/module-declaration.cpp new file mode 100644 index 0000000000000..ee696d14585ca --- /dev/null +++ b/test/CXX/modules-ts/basic/basic.link/module-declaration.cpp @@ -0,0 +1,55 @@ +// Tests for module-declaration syntax. +// +// RUN: rm -rf %t +// RUN: mkdir -p %t +// RUN: echo 'export module x; int a, b;' > %t/x.cppm +// RUN: echo 'export module x.y; int c;' > %t/x.y.cppm +// +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %t/x.cppm -o %t/x.pcm +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface -fmodule-file=%t/x.pcm %t/x.y.cppm -o %t/x.y.pcm +// +// Module implementation for unknown and known module. (The former is ill-formed.) +// FIXME: TEST=1 should fail because we don't have an interface for module z. +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=1 -DEXPORT= -DPARTITION= -DMODULE_NAME=z +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=2 -DEXPORT= -DPARTITION= -DMODULE_NAME=x +// +// Module interface for unknown and known module. (The latter is ill-formed due to +// redefinition.) +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=3 -DEXPORT=export -DPARTITION= -DMODULE_NAME=z +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=4 -DEXPORT=export -DPARTITION= -DMODULE_NAME=x +// +// Defining a module partition. +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=5 -DEXPORT=export -DPARTITION=partition -DMODULE_NAME=z +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=6 -DEXPORT= -DPARTITION=partition -DMODULE_NAME=z +// +// Miscellaneous syntax. +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=7 -DEXPORT= -DPARTITION=elderberry -DMODULE_NAME=z +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=8 -DEXPORT= -DPARTITION= -DMODULE_NAME='z [[]]' +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=9 -DEXPORT= -DPARTITION= -DMODULE_NAME='z [[fancy]]' +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DTEST=10 -DEXPORT= -DPARTITION= -DMODULE_NAME='z [[maybe_unused]]' + +EXPORT module PARTITION MODULE_NAME; +#if TEST == 4 +// expected-error@-2 {{redefinition of module 'x'}} +// expected-note-re@module-declaration.cpp:* {{loaded from '{{.*}}/x.pcm'}} +#elif TEST == 6 +// expected-error@-5 {{module partition must be declared 'export'}} +#elif TEST == 7 +// expected-error@-7 {{expected ';'}} expected-error@-7 {{requires a type specifier}} +#elif TEST == 9 +// expected-warning@-9 {{unknown attribute 'fancy' ignored}} +#elif TEST == 10 +// expected-error-re@-11 {{'maybe_unused' attribute cannot be applied to a module{{$}}}} +#else +// expected-no-diagnostics +#endif diff --git a/test/CodeGenCXX/modules-ts.cppm b/test/CXX/modules-ts/codegen-basics.cppm index 90f6b5319eb60..d1552d9edd263 100644 --- a/test/CodeGenCXX/modules-ts.cppm +++ b/test/CXX/modules-ts/codegen-basics.cppm @@ -1,7 +1,7 @@ // RUN: %clang_cc1 -fmodules-ts -std=c++1z -triple=x86_64-linux-gnu -fmodules-codegen -emit-module-interface %s -o %t.pcm // RUN: %clang_cc1 -fmodules-ts -std=c++1z -triple=x86_64-linux-gnu %t.pcm -emit-llvm -o - | FileCheck %s -module FooBar; +export module FooBar; export { // CHECK-LABEL: define i32 @_Z1fv( diff --git a/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.import/p1.cpp b/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.import/p1.cpp new file mode 100644 index 0000000000000..aaf43d6584a42 --- /dev/null +++ b/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.import/p1.cpp @@ -0,0 +1,41 @@ +// RUN: rm -rf %t +// RUN: mkdir -p %t +// RUN: echo 'export module x; int a, b;' > %t/x.cppm +// RUN: echo 'export module x.y; int c;' > %t/x.y.cppm +// +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %t/x.cppm -o %t/x.pcm +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface -fmodule-file=%t/x.pcm %t/x.y.cppm -o %t/x.y.pcm +// +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DMODULE_NAME=z +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ +// RUN: -DMODULE_X -DMODULE_NAME=x + +module MODULE_NAME; + +int use_1 = a; +#if !MODULE_X +// expected-error@-2 {{declaration of 'a' must be imported from module 'x' before it is required}} +// expected-note@x.cppm:1 {{here}} +#endif + +import x; + +int use_2 = b; // ok + +// There is no relation between module x and module x.y. +int use_3 = c; // expected-error {{declaration of 'c' must be imported from module 'x.y'}} + // expected-note@x.y.cppm:1 {{here}} + +import x [[]]; +import x [[foo]]; // expected-warning {{unknown attribute 'foo' ignored}} +import x [[noreturn]]; // expected-error {{'noreturn' attribute cannot be applied to a module import}} +import x [[blarg::noreturn]]; // expected-warning {{unknown attribute 'noreturn' ignored}} + +import x.y; +import x.; // expected-error {{expected a module name after 'import'}} +import .x; // expected-error {{expected a module name after 'import'}} + +int use_4 = c; // ok + +import blarg; // expected-error {{module 'blarg' not found}} diff --git a/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.interface/p1.cpp b/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.interface/p1.cpp new file mode 100644 index 0000000000000..afe3a7a48589c --- /dev/null +++ b/test/CXX/modules-ts/dcl.dcl/dcl.module/dcl.module.interface/p1.cpp @@ -0,0 +1,27 @@ +// RUN: %clang_cc1 -fmodules-ts %s -verify -o /dev/null +// RUN: %clang_cc1 -fmodules-ts %s -DINTERFACE -verify -o /dev/null +// RUN: %clang_cc1 -fmodules-ts %s -DIMPLEMENTATION -verify -o /dev/null +// +// RUN: %clang_cc1 -fmodules-ts %s -DBUILT_AS_INTERFACE -emit-module-interface -verify -o /dev/null +// RUN: %clang_cc1 -fmodules-ts %s -DINTERFACE -DBUILT_AS_INTERFACE -emit-module-interface -verify -o /dev/null +// RUN: %clang_cc1 -fmodules-ts %s -DIMPLEMENTATION -DBUILT_AS_INTERFACE -emit-module-interface -verify -o /dev/null + +#if INTERFACE +export module A; +#elif IMPLEMENTATION +module A; + #ifdef BUILT_AS_INTERFACE + // expected-error@-2 {{missing 'export' specifier in module declaration while building module interface}} + #endif +#else + #ifdef BUILT_AS_INTERFACE + // FIXME: Diagnose missing module declaration (at end of TU) + #endif +#endif + +export int a; +#ifndef INTERFACE +// expected-error@-2 {{export declaration can only be used within a module interface unit}} +#else +// expected-no-diagnostics +#endif diff --git a/test/CodeGen/asan-globals-gc.cpp b/test/CodeGen/asan-globals-gc.cpp new file mode 100644 index 0000000000000..6d64f41dda8dd --- /dev/null +++ b/test/CodeGen/asan-globals-gc.cpp @@ -0,0 +1,7 @@ +// RUN: %clang_cc1 -fsanitize=address -emit-llvm -o - -triple x86_64-windows-msvc %s | FileCheck %s --check-prefix=WITH-GC +// RUN: %clang_cc1 -fsanitize=address -emit-llvm -o - -triple x86_64-windows-msvc -fdata-sections %s | FileCheck %s --check-prefix=WITH-GC + +int global; + +// WITH-GC-NOT: call void @__asan_register_globals +// WITHOUT-GC: call void @__asan_register_globals diff --git a/test/CodeGen/catch-undef-behavior.c b/test/CodeGen/catch-undef-behavior.c index c5f3a79429a01..e67f0a1f8eec1 100644 --- a/test/CodeGen/catch-undef-behavior.c +++ b/test/CodeGen/catch-undef-behavior.c @@ -36,13 +36,7 @@ void foo() { // CHECK-COMMON: %[[I8PTR:.*]] = bitcast i32* %[[PTR:.*]] to i8* // CHECK-COMMON-NEXT: %[[SIZE:.*]] = call i64 @llvm.objectsize.i64.p0i8(i8* %[[I8PTR]], i1 false, i1 false) - // CHECK-COMMON-NEXT: %[[CHECK0:.*]] = icmp uge i64 %[[SIZE]], 4 - - // CHECK-COMMON: %[[PTRTOINT:.*]] = ptrtoint {{.*}}* %[[PTR]] to i64 - // CHECK-COMMON-NEXT: %[[MISALIGN:.*]] = and i64 %[[PTRTOINT]], 3 - // CHECK-COMMON-NEXT: %[[CHECK1:.*]] = icmp eq i64 %[[MISALIGN]], 0 - - // CHECK-COMMON: %[[OK:.*]] = and i1 %[[CHECK0]], %[[CHECK1]] + // CHECK-COMMON-NEXT: %[[OK:.*]] = icmp uge i64 %[[SIZE]], 4 // CHECK-UBSAN: br i1 %[[OK]], {{.*}} !prof ![[WEIGHT_MD:.*]], !nosanitize // CHECK-TRAP: br i1 %[[OK]], {{.*}} diff --git a/test/CodeGen/pgo-sample-thinlto-summary.c b/test/CodeGen/pgo-sample-thinlto-summary.c index a284af3c80877..7782aeb76616d 100644 --- a/test/CodeGen/pgo-sample-thinlto-summary.c +++ b/test/CodeGen/pgo-sample-thinlto-summary.c @@ -32,10 +32,10 @@ void unroll() { baz(i); } -// Checks if icp is invoked by normal compile, but not thinlto compile. +// Check that icp is not invoked (both -O2 and ThinLTO). // O2-LABEL: define void @icp // THINLTO-LABEL: define void @icp -// O2: if.true.direct_targ +// O2-NOT: if.true.direct_targ // ThinLTO-NOT: if.true.direct_targ void icp(void (*p)()) { p(); diff --git a/test/CodeGen/sanitize-recover.c b/test/CodeGen/sanitize-recover.c index 99eff8518ce84..d714d58c7f26c 100644 --- a/test/CodeGen/sanitize-recover.c +++ b/test/CodeGen/sanitize-recover.c @@ -22,15 +22,7 @@ void foo() { // PARTIAL: %[[SIZE:.*]] = call i64 @llvm.objectsize.i64.p0i8(i8* {{.*}}, i1 false, i1 false) // PARTIAL-NEXT: %[[CHECK0:.*]] = icmp uge i64 %[[SIZE]], 4 - // PARTIAL: %[[MISALIGN:.*]] = and i64 {{.*}}, 3 - // PARTIAL-NEXT: %[[CHECK1:.*]] = icmp eq i64 %[[MISALIGN]], 0 + // PARTIAL: br i1 %[[CHECK0]], {{.*}} !nosanitize - // PARTIAL: %[[CHECK01:.*]] = and i1 %[[CHECK1]], %[[CHECK0]] - - // PARTIAL: br i1 %[[CHECK01]], {{.*}} !nosanitize - // PARTIAL: br i1 %[[CHECK1]], {{.*}} !nosanitize - - // PARTIAL: call void @__ubsan_handle_type_mismatch_v1_abort( - // PARTIAL-NEXT: unreachable // PARTIAL: call void @__ubsan_handle_type_mismatch_v1( } diff --git a/test/CodeGen/split-debug-filename.c b/test/CodeGen/split-debug-filename.c index 43daeada96ab0..9ca7b0f3287a9 100644 --- a/test/CodeGen/split-debug-filename.c +++ b/test/CodeGen/split-debug-filename.c @@ -1,7 +1,12 @@ // RUN: %clang_cc1 -debug-info-kind=limited -split-dwarf-file foo.dwo -S -emit-llvm -o - %s | FileCheck %s +// RUN: %clang_cc1 -debug-info-kind=limited -enable-split-dwarf -split-dwarf-file foo.dwo -S -emit-llvm -o - %s | FileCheck --check-prefix=VANILLA %s int main (void) { return 0; } // Testing to ensure that the dwo name gets output into the compile unit. // CHECK: !DICompileUnit({{.*}}, splitDebugFilename: "foo.dwo" + +// Testing to ensure that the dwo name is not output into the compile unit if +// it's for vanilla split-dwarf rather than split-dwarf for implicit modules. +// VANILLA-NOT: splitDebugFilename diff --git a/test/CodeGenCXX/invariant.group-for-vptrs.cpp b/test/CodeGenCXX/invariant.group-for-vptrs.cpp index 5c763538088b9..4e5d9b49d985d 100644 --- a/test/CodeGenCXX/invariant.group-for-vptrs.cpp +++ b/test/CodeGenCXX/invariant.group-for-vptrs.cpp @@ -12,15 +12,15 @@ struct D : A { void testExternallyVisible() { A *a = new A; - // CHECK: load {{.*}} !invariant.group ![[A_MD:[0-9]+]] + // CHECK: load {{.*}} !invariant.group ![[MD:[0-9]+]] a->foo(); D *d = new D; // CHECK: call void @_ZN1DC1Ev( - // CHECK: load {{.*}} !invariant.group ![[D_MD:[0-9]+]] + // CHECK: load {{.*}} !invariant.group ![[MD]] d->foo(); A *a2 = d; - // CHECK: load {{.*}} !invariant.group ![[A_MD]] + // CHECK: load {{.*}} !invariant.group ![[MD]] a2->foo(); } // CHECK-LABEL: {{^}}} @@ -40,35 +40,32 @@ struct C : B { // CHECK-LABEL: define void @_Z21testInternallyVisibleb( void testInternallyVisible(bool p) { B *b = new B; - // CHECK: = load {{.*}}, !invariant.group ![[B_MD:[0-9]+]] + // CHECK: = load {{.*}}, !invariant.group ![[MD]] b->bar(); // CHECK: call void @_ZN12_GLOBAL__N_11CC1Ev( C *c = new C; - // CHECK: = load {{.*}}, !invariant.group ![[C_MD:[0-9]+]] + // CHECK: = load {{.*}}, !invariant.group ![[MD]] c->bar(); } // Checking A::A() // CHECK-LABEL: define linkonce_odr void @_ZN1AC2Ev( -// CHECK: store {{.*}}, !invariant.group ![[A_MD]] +// CHECK: store {{.*}}, !invariant.group ![[MD]] // CHECK-LABEL: {{^}}} // Checking D::D() // CHECK-LABEL: define linkonce_odr void @_ZN1DC2Ev( // CHECK: = call i8* @llvm.invariant.group.barrier(i8* // CHECK: call void @_ZN1AC2Ev(%struct.A* -// CHECK: store {{.*}} !invariant.group ![[D_MD]] +// CHECK: store {{.*}} !invariant.group ![[MD]] // Checking B::B() // CHECK-LABEL: define internal void @_ZN12_GLOBAL__N_11BC2Ev( -// CHECK: store {{.*}}, !invariant.group ![[B_MD]] +// CHECK: store {{.*}}, !invariant.group ![[MD]] // Checking C::C() // CHECK-LABEL: define internal void @_ZN12_GLOBAL__N_11CC2Ev( -// CHECK: store {{.*}}, !invariant.group ![[C_MD]] +// CHECK: store {{.*}}, !invariant.group ![[MD]] -// CHECK: ![[A_MD]] = !{!"_ZTS1A"} -// CHECK: ![[D_MD]] = !{!"_ZTS1D"} -// CHECK: ![[B_MD]] = distinct !{} -// CHECK: ![[C_MD]] = distinct !{} +// CHECK: ![[MD]] = !{} diff --git a/test/CodeGenCXX/ubsan-suppress-checks.cpp b/test/CodeGenCXX/ubsan-suppress-checks.cpp index d0e7b32ef96c1..ae7c94b34f290 100644 --- a/test/CodeGenCXX/ubsan-suppress-checks.cpp +++ b/test/CodeGenCXX/ubsan-suppress-checks.cpp @@ -12,6 +12,7 @@ void load_non_null_pointers() { char c = "foo"[0]; + // CHECK-NOT: and i64 {{.*}}, !nosanitize // CHECK-NOT: icmp ne {{.*}}, null, !nosanitize // CHECK: ret void } @@ -43,10 +44,6 @@ struct A { }; f(); - // LAMBDA: %[[LAMBDAINT:[0-9]+]] = ptrtoint %class.anon* %[[FUNCVAR:.*]] to i64, !nosanitize - // LAMBDA: and i64 %[[LAMBDAINT]], 7, !nosanitize - // LAMBDA: call void @__ubsan_handle_type_mismatch - // LAMBDA-NOT: call void @__ubsan_handle_type_mismatch // LAMBDA: ret void } diff --git a/test/CoverageMapping/empty-destructor.cpp b/test/CoverageMapping/empty-destructor.cpp new file mode 100644 index 0000000000000..cfc29a75c6d74 --- /dev/null +++ b/test/CoverageMapping/empty-destructor.cpp @@ -0,0 +1,11 @@ +// RUN: %clang_cc1 -triple i686-windows -emit-llvm-only -fcoverage-mapping -dump-coverage-mapping -fprofile-instrument=clang %s | FileCheck %s + +struct A { + virtual ~A(); +}; + +// CHECK: ?PR32761@@YAXXZ: +// CHECK-NEXT: File 0, [[@LINE+1]]:16 -> [[@LINE+3]]:2 = #0 +void PR32761() { + A a; +} diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crt1.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crt1.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crt1.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crti.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crti.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crti.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crtn.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crtn.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/crtn.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtbegin.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtbegin.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtbegin.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtend.o b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtend.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv6hl_tree/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/crtend.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crt1.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crt1.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crt1.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crti.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crti.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crti.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crtn.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crtn.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/crtn.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtbegin.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtbegin.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtbegin.o diff --git a/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtend.o b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtend.o new file mode 100644 index 0000000000000..e69de29bb2d1d --- /dev/null +++ b/test/Driver/Inputs/opensuse_tumbleweed_armv7hl_tree/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/crtend.o diff --git a/test/Driver/fsanitize.c b/test/Driver/fsanitize.c index b6304b515b73b..05e239c74243c 100644 --- a/test/Driver/fsanitize.c +++ b/test/Driver/fsanitize.c @@ -358,6 +358,27 @@ // RUN: %clang -target i386-pc-openbsd -fsanitize=address %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ASAN-OPENBSD // CHECK-ASAN-OPENBSD: unsupported option '-fsanitize=address' for target 'i386-pc-openbsd' +// RUN: %clang -target x86_64-apple-darwin -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-X86-64-DARWIN +// CHECK-LSAN-X86-64-DARWIN-NOT: unsupported option + +// RUN: %clang -target x86_64-apple-iossimulator -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-X86-64-IOSSIMULATOR +// CHECK-LSAN-X86-64-IOSSIMULATOR-NOT: unsupported option + +// RUN: %clang -target x86_64-apple-tvossimulator -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-X86-64-TVOSSIMULATOR +// CHECK-LSAN-X86-64-TVOSSIMULATOR-NOT: unsupported option + +// RUN: %clang -target i386-apple-darwin -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-I386-DARWIN +// CHECK-LSAN-I386-DARWIN-NOT: unsupported option + +// RUN: %clang -target arm-apple-ios -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-ARM-IOS +// CHECK-LSAN-ARM-IOS-NOT: unsupported option + +// RUN: %clang -target i386-apple-iossimulator -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-I386-IOSSIMULATOR +// CHECK-LSAN-I386-IOSSIMULATOR-NOT: unsupported option + +// RUN: %clang -target i386-apple-tvossimulator -fsanitize=leak %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-LSAN-I386-TVOSSIMULATOR +// CHECK-LSAN-I386-TVOSSIMULATOR-NOT: unsupported option + // RUN: %clang -target i686-linux-gnu -fsanitize=efficiency-cache-frag %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ESAN-X86 // RUN: %clang -target i686-linux-gnu -fsanitize=efficiency-working-set %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ESAN-X86 // CHECK-ESAN-X86: error: unsupported option '-fsanitize=efficiency-{{.*}}' for target 'i686--linux-gnu' diff --git a/test/Driver/fuzzer.c b/test/Driver/fuzzer.c new file mode 100644 index 0000000000000..e534a73a647f9 --- /dev/null +++ b/test/Driver/fuzzer.c @@ -0,0 +1,20 @@ +// Test flags inserted by -fsanitize=fuzzer. + +// RUN: %clang -fsanitize=fuzzer %s -target x86_64-apple-darwin14 -### 2>&1 | FileCheck --check-prefixes=CHECK-FUZZER-LIB,CHECK-COVERAGE-FLAGS %s +// +// CHECK-FUZZER-LIB: libLLVMFuzzer.a +// CHECK-COVERAGE: -fsanitize-coverage-trace-pc-guard +// CHECK-COVERAGE-SAME: -fsanitize-coverage-indirect-calls +// CHECK-COVERAGE-SAME: -fsanitize-coverage-trace-cmp + +// RUN: %clang -fsanitize=fuzzer -target i386-unknown-linux -stdlib=platform %s -### 2>&1 | FileCheck --check-prefixes=CHECK-LIBCXX-LINUX %s +// +// CHECK-LIBCXX-LINUX: -lstdc++ + +// RUN: %clang -target x86_64-apple-darwin14 -fsanitize=fuzzer %s -### 2>&1 | FileCheck --check-prefixes=CHECK-LIBCXX-DARWIN %s +// +// CHECK-LIBCXX-DARWIN: -lc++ + +int LLVMFuzzerTestOneInput(const char *Data, long Size) { + return 0; +} diff --git a/test/Driver/hexagon-toolchain-elf.c b/test/Driver/hexagon-toolchain-elf.c index f6f2191b3eb64..98582450e3fd4 100644 --- a/test/Driver/hexagon-toolchain-elf.c +++ b/test/Driver/hexagon-toolchain-elf.c @@ -97,6 +97,22 @@ // CHECK024: "-cc1" {{.*}} "-target-cpu" "hexagonv62" // CHECK024: hexagon-link{{.*}}/Inputs/hexagon_tree/Tools/bin/../target/hexagon/lib/v62/crt0 +// RUN: %clang -### -target hexagon-unknown-elf \ +// RUN: -ccc-install-dir %S/Inputs/hexagon_tree/Tools/bin \ +// RUN: -O3 \ +// RUN: %s 2>&1 \ +// RUN: | FileCheck -check-prefix=CHECK025 %s +// CHECK025: "-ffp-contract=fast" +// CHECK025: hexagon-link + +// RUN: %clang -### -target hexagon-unknown-elf \ +// RUN: -ccc-install-dir %S/Inputs/hexagon_tree/Tools/bin \ +// RUN: -O3 -ffp-contract=off \ +// RUN: %s 2>&1 \ +// RUN: | FileCheck -check-prefix=CHECK026 %s +// CHECK026-NOT: "-ffp-contract=fast" +// CHECK026: hexagon-link + // ----------------------------------------------------------------------------- // Test Linker related args // ----------------------------------------------------------------------------- diff --git a/test/Driver/linux-ld.c b/test/Driver/linux-ld.c index e5aa870866d24..1c5f1a4556b1a 100644 --- a/test/Driver/linux-ld.c +++ b/test/Driver/linux-ld.c @@ -638,6 +638,46 @@ // CHECK-OPENSUSE-42-2-AARCH64: "{{.*}}/usr/lib64/gcc/aarch64-suse-linux/4.8{{/|\\\\}}crtend.o" // CHECK-OPENSUSE-42-2-AARCH64: "{{.*}}/usr/lib64/gcc/aarch64-suse-linux/4.8/../../../../lib64{{/|\\\\}}crtn.o" // +// Check openSUSE Tumbleweed on armv6hl +// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \ +// RUN: --target=armv6hl-suse-linux-gnueabi \ +// RUN: --gcc-toolchain="" \ +// RUN: --sysroot=%S/Inputs/opensuse_tumbleweed_armv6hl_tree \ +// RUN: | FileCheck --check-prefix=CHECK-OPENSUSE-TW-ARMV6HL %s +// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \ +// RUN: --target=armv6hl-suse-linux-gnueabi \ +// RUN: --gcc-toolchain="" \ +// RUN: --sysroot=%S/Inputs/opensuse_tumbleweed_armv6hl_tree \ +// RUN: | FileCheck --check-prefix=CHECK-OPENSUSE-TW-ARMV6HL %s +// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]" +// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crt1.o" +// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crti.o" +// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5{{/|\\\\}}crtbegin.o" +// CHECK-OPENSUSE-TW-ARMV6HL: "-L[[SYSROOT]]/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5" +// CHECK-OPENSUSE-TW-ARMV6HL: "-L[[SYSROOT]]/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/../../../../lib" +// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5{{/|\\\\}}crtend.o" +// CHECK-OPENSUSE-TW-ARMV6HL: "{{.*}}/usr/lib/gcc/armv6hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crtn.o" +// +// Check openSUSE Tumbleweed on armv7hl +// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \ +// RUN: --target=armv7hl-suse-linux-gnueabi \ +// RUN: --gcc-toolchain="" \ +// RUN: --sysroot=%S/Inputs/opensuse_tumbleweed_armv7hl_tree \ +// RUN: | FileCheck --check-prefix=CHECK-OPENSUSE-TW-ARMV7HL %s +// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \ +// RUN: --target=armv7hl-suse-linux-gnueabi \ +// RUN: --gcc-toolchain="" \ +// RUN: --sysroot=%S/Inputs/opensuse_tumbleweed_armv7hl_tree \ +// RUN: | FileCheck --check-prefix=CHECK-OPENSUSE-TW-ARMV7HL %s +// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]" +// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crt1.o" +// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crti.o" +// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5{{/|\\\\}}crtbegin.o" +// CHECK-OPENSUSE-TW-ARMV7HL: "-L[[SYSROOT]]/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5" +// CHECK-OPENSUSE-TW-ARMV7HL: "-L[[SYSROOT]]/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/../../../../lib" +// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5{{/|\\\\}}crtend.o" +// CHECK-OPENSUSE-TW-ARMV7HL: "{{.*}}/usr/lib/gcc/armv7hl-suse-linux-gnueabi/5/../../../../lib{{/|\\\\}}crtn.o" +// // Check dynamic-linker for different archs // RUN: %clang %s -### -o %t.o 2>&1 \ // RUN: --target=arm-linux-gnueabi \ diff --git a/test/Driver/modules-ts.cpp b/test/Driver/modules-ts.cpp index fd3391479eea8..3847b71f7b742 100644 --- a/test/Driver/modules-ts.cpp +++ b/test/Driver/modules-ts.cpp @@ -1,6 +1,6 @@ // Check compiling a module interface to a .pcm file. // -// RUN: %clang -fmodules-ts -x c++-module --precompile %s -Dimplementation= -o %t.pcm -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE +// RUN: %clang -fmodules-ts -x c++-module --precompile %s -o %t.pcm -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE // // CHECK-PRECOMPILE: -cc1 {{.*}} -emit-module-interface // CHECK-PRECOMPILE-SAME: -o {{.*}}.pcm @@ -18,7 +18,7 @@ // Check use of a .pcm file in another compilation. // -// RUN: %clang -fmodules-ts -fmodule-file=%t.pcm %s -c -o %t.o -v 2>&1 | FileCheck %s --check-prefix=CHECK-USE +// RUN: %clang -fmodules-ts -fmodule-file=%t.pcm -Dexport= %s -c -o %t.o -v 2>&1 | FileCheck %s --check-prefix=CHECK-USE // // CHECK-USE: -cc1 // CHECK-USE-SAME: -emit-obj @@ -28,11 +28,11 @@ // Check combining precompile and compile steps works. // -// RUN: %clang -fmodules-ts -x c++-module %s -Dimplementation= -c -o %t.pcm.o -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE --check-prefix=CHECK-COMPILE +// RUN: %clang -fmodules-ts -x c++-module %s -c -o %t.pcm.o -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE --check-prefix=CHECK-COMPILE // Check that .cppm is treated as a module implicitly. // RUN: cp %s %t.cppm -// RUN: %clang -fmodules-ts --precompile %t.cppm -Dimplementation= -o %t.pcm -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE +// RUN: %clang -fmodules-ts --precompile %t.cppm -o %t.pcm -v 2>&1 | FileCheck %s --check-prefix=CHECK-PRECOMPILE -// Note, we use -Dimplementation= to make this a valid module interface unit when building the interface. -module implementation foo; +// Note, we use -Dexport= to make this a module implementation unit when building the implementation. +export module foo; diff --git a/test/Driver/sanitizer-ld.c b/test/Driver/sanitizer-ld.c index c4a6f4377b896..efdd1245778d4 100644 --- a/test/Driver/sanitizer-ld.c +++ b/test/Driver/sanitizer-ld.c @@ -409,6 +409,15 @@ // CHECK-ASAN-DARWIN106-CXX: libclang_rt.asan_osx_dynamic.dylib // CHECK-ASAN-DARWIN106-CXX-NOT: -lc++abi +// RUN: %clangxx -fsanitize=leak %s -### -o %t.o 2>&1 \ +// RUN: -mmacosx-version-min=10.6 \ +// RUN: -target x86_64-apple-darwin13.4.0 -fuse-ld=ld -stdlib=platform \ +// RUN: --sysroot=%S/Inputs/basic_linux_tree \ +// RUN: | FileCheck --check-prefix=CHECK-LSAN-DARWIN106-CXX %s +// CHECK-LSAN-DARWIN106-CXX: "{{.*}}ld{{(.exe)?}}" +// CHECK-LSAN-DARWIN106-CXX: libclang_rt.lsan_osx_dynamic.dylib +// CHECK-LSAN-DARWIN106-CXX-NOT: -lc++abi + // RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \ // RUN: -target x86_64-unknown-linux -fuse-ld=ld -fsanitize=safe-stack \ // RUN: --sysroot=%S/Inputs/basic_linux_tree \ diff --git a/test/Driver/split-debug.c b/test/Driver/split-debug.c index eaa1e990c04cf..c12164b4e2f1b 100644 --- a/test/Driver/split-debug.c +++ b/test/Driver/split-debug.c @@ -36,53 +36,53 @@ // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -gmlt -fno-split-dwarf-inlining -S -### %s 2> %t // RUN: FileCheck -check-prefix=CHECK-GMLT-WITH-SPLIT < %t %s // -// CHECK-GMLT-WITH-SPLIT: "-split-dwarf=Enable" +// CHECK-GMLT-WITH-SPLIT: "-enable-split-dwarf" // CHECK-GMLT-WITH-SPLIT: "-debug-info-kind=line-tables-only" // CHECK-GMLT-WITH-SPLIT: "-split-dwarf-file" // RUN: %clang -target x86_64-unknown-linux-gnu -gmlt -gsplit-dwarf -fno-split-dwarf-inlining -S -### %s 2> %t // RUN: FileCheck -check-prefix=CHECK-SPLIT-WITH-GMLT < %t %s // -// CHECK-SPLIT-WITH-GMLT: "-split-dwarf=Enable" +// CHECK-SPLIT-WITH-GMLT: "-enable-split-dwarf" // CHECK-SPLIT-WITH-GMLT: "-debug-info-kind=line-tables-only" // CHECK-SPLIT-WITH-GMLT: "-split-dwarf-file" // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -fno-split-dwarf-inlining -S -### %s 2> %t // RUN: FileCheck -check-prefix=CHECK-SPLIT-WITH-NOINL < %t %s // -// CHECK-SPLIT-WITH-NOINL: "-split-dwarf=Enable" +// CHECK-SPLIT-WITH-NOINL: "-enable-split-dwarf" // CHECK-SPLIT-WITH-NOINL: "-debug-info-kind=limited" // CHECK-SPLIT-WITH-NOINL: "-split-dwarf-file" // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -gmlt -S -### %s 2> %t // RUN: FileCheck -check-prefix=CHECK-GMLT-OVER-SPLIT < %t %s // -// CHECK-GMLT-OVER-SPLIT-NOT: "-split-dwarf=Enable" +// CHECK-GMLT-OVER-SPLIT-NOT: "-enable-split-dwarf" // CHECK-GMLT-OVER-SPLIT: "-debug-info-kind=line-tables-only" // CHECK-GMLT-OVER-SPLIT-NOT: "-split-dwarf-file" // RUN: %clang -target x86_64-unknown-linux-gnu -gmlt -gsplit-dwarf -S -### %s 2> %t // RUN: FileCheck -check-prefix=CHECK-SPLIT-OVER-GMLT < %t %s // -// CHECK-SPLIT-OVER-GMLT: "-split-dwarf=Enable" "-debug-info-kind=limited" +// CHECK-SPLIT-OVER-GMLT: "-enable-split-dwarf" "-debug-info-kind=limited" // CHECK-SPLIT-OVER-GMLT: "-split-dwarf-file" // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -g0 -fno-split-dwarf-inlining -S -### %s 2> %t // RUN: FileCheck -check-prefix=CHECK-G0-OVER-SPLIT < %t %s // -// CHECK-G0-OVER-SPLIT-NOT: "-split-dwarf=Enable" +// CHECK-G0-OVER-SPLIT-NOT: "-enable-split-dwarf" // CHECK-G0-OVER-SPLIT-NOT: "-debug-info-kind // CHECK-G0-OVER-SPLIT-NOT: "-split-dwarf-file" // RUN: %clang -target x86_64-unknown-linux-gnu -gsplit-dwarf -g0 -S -### %s 2> %t // RUN: FileCheck -check-prefix=CHECK-G0-OVER-SPLIT < %t %s // -// CHECK-G0-OVER-SPLIT-NOT: "-split-dwarf=Enable" +// CHECK-G0-OVER-SPLIT-NOT: "-enable-split-dwarf" // CHECK-G0-OVER-SPLIT-NOT: "-debug-info-kind // CHECK-G0-OVER-SPLIT-NOT: "-split-dwarf-file" // RUN: %clang -target x86_64-unknown-linux-gnu -g0 -gsplit-dwarf -S -### %s 2> %t // RUN: FileCheck -check-prefix=CHECK-SPLIT-OVER-G0 < %t %s // -// CHECK-SPLIT-OVER-G0: "-split-dwarf=Enable" "-debug-info-kind=limited" +// CHECK-SPLIT-OVER-G0: "-enable-split-dwarf" "-debug-info-kind=limited" // CHECK-SPLIT-OVER-G0: "-split-dwarf-file" diff --git a/test/Format/incomplete.cpp b/test/Format/incomplete.cpp index 8a65fad657cd7..f92d576193730 100644 --- a/test/Format/incomplete.cpp +++ b/test/Format/incomplete.cpp @@ -1,6 +1,6 @@ // RUN: grep -Ev "// *[A-Z-]+:" %s | clang-format -style=LLVM -cursor=0 \ // RUN: | FileCheck -strict-whitespace %s -// CHECK: {{"IncompleteFormat": true}} +// CHECK: {{"IncompleteFormat": true, "Line": 2}} // CHECK: {{^int\ \i;$}} int i; // CHECK: {{^f \( g \(;$}} diff --git a/test/Headers/stdatomic.c b/test/Headers/stdatomic.c index 76112cfe40ea0..3643fd4245b31 100644 --- a/test/Headers/stdatomic.c +++ b/test/Headers/stdatomic.c @@ -1,4 +1,5 @@ // RUN: %clang_cc1 -std=c11 -E %s | FileCheck %s +// RUN: %clang_cc1 -std=c11 -fms-compatibility -E %s | FileCheck %s #include <stdatomic.h> int bool_lock_free = ATOMIC_BOOL_LOCK_FREE; diff --git a/test/Index/Core/external-source-symbol-attr.m b/test/Index/Core/external-source-symbol-attr.m new file mode 100644 index 0000000000000..49a075fc74e02 --- /dev/null +++ b/test/Index/Core/external-source-symbol-attr.m @@ -0,0 +1,100 @@ +// RUN: c-index-test core -print-source-symbols -- %s -target x86_64-apple-macosx10.7 | FileCheck %s + +#define EXT_DECL(mod_name) __attribute__((external_source_symbol(language="Swift", defined_in=mod_name))) +#define GEN_DECL(mod_name) __attribute__((external_source_symbol(language="Swift", defined_in=mod_name, generated_declaration))) +#define PUSH_GEN_DECL(mod_name) push(GEN_DECL(mod_name), apply_to=any(enum, objc_interface, objc_category, objc_protocol)) + +// This should not be indexed. +GEN_DECL("some_module") +@interface I1 +// CHECK-NOT: [[@LINE-1]]:12 | +-(void)method; +// CHECK-NOT: [[@LINE-1]]:8 | +@end + +EXT_DECL("some_module") +@interface I2 +// CHECK: [[@LINE-1]]:12 | class/Swift | I2 | c:@M@some_module@objc(cs)I2 | {{.*}} | Decl | rel: 0 +-(void)method; +// CHECK: [[@LINE-1]]:8 | instance-method/Swift | method | c:@M@some_module@objc(cs)I2(im)method | -[I2 method] | Decl,Dyn,RelChild | rel: 1 +@end + +void test1(I1 *o) { +// CHECK: [[@LINE-1]]:12 | class/Swift | I1 | c:@M@some_module@objc(cs)I1 | + [o method]; + // CHECK: [[@LINE-1]]:6 | instance-method/Swift | method | c:@M@some_module@objc(cs)I1(im)method | +} + +EXT_DECL("some_module") +@protocol ExtProt +// CHECK: [[@LINE-1]]:11 | protocol/Swift | ExtProt | c:@M@some_module@objc(pl)ExtProt | +@end + +@interface I1(cat) +// CHECK: [[@LINE-1]]:15 | extension/ObjC | cat | c:@M@some_module@objc(cy)I1@cat | +-(void)cat_method; +// CHECK: [[@LINE-1]]:8 | instance-method/ObjC | cat_method | c:@M@some_module@objc(cs)I1(im)cat_method +@end + +EXT_DECL("cat_module") +@interface I1(cat2) +// CHECK: [[@LINE-1]]:15 | extension/Swift | cat2 | c:@CM@cat_module@some_module@objc(cy)I1@cat2 | +-(void)cat_method2; +// CHECK: [[@LINE-1]]:8 | instance-method/Swift | cat_method2 | c:@CM@cat_module@some_module@objc(cs)I1(im)cat_method2 +@end + +#define NS_ENUM(_name, _type) enum _name:_type _name; enum _name : _type + +#pragma clang attribute PUSH_GEN_DECL("modname") + +@interface I3 +// CHECK-NOT: [[@LINE-1]]:12 | +-(void)meth; +// CHECK-NOT: [[@LINE-1]]:8 | +@end + +@interface I3(cat) +// CHECK-NOT: [[@LINE-1]]:12 | +// CHECK-NOT: [[@LINE-2]]:15 | +-(void)meth2; +// CHECK-NOT: [[@LINE-1]]:8 | +@end + +@protocol ExtProt2 +// CHECK-NOT: [[@LINE-1]]:11 | +-(void)meth; +// CHECK-NOT: [[@LINE-1]]:8 | +@end + +typedef NS_ENUM(SomeEnum, int) { +// CHECK-NOT: [[@LINE-1]]:17 | + SomeEnumFirst = 0, + // CHECK-NOT: [[@LINE-1]]:3 | +}; + +#pragma clang attribute pop + +void test2(I3 *i3, id<ExtProt2> prot2, SomeEnum some) { + // CHECK: [[@LINE-1]]:12 | class/Swift | I3 | c:@M@modname@objc(cs)I3 | + // CHECK: [[@LINE-2]]:23 | protocol/Swift | ExtProt2 | c:@M@modname@objc(pl)ExtProt2 | + // CHECK: [[@LINE-3]]:40 | enum/Swift | SomeEnum | c:@M@modname@E@SomeEnum | + [i3 meth]; + // CHECK: [[@LINE-1]]:7 | instance-method/Swift | meth | c:@M@modname@objc(cs)I3(im)meth | + [i3 meth2]; + // CHECK: [[@LINE-1]]:7 | instance-method/Swift | meth2 | c:@CM@modname@objc(cs)I3(im)meth2 | + [prot2 meth]; + // CHECK: [[@LINE-1]]:10 | instance-method/Swift | meth | c:@M@modname@objc(pl)ExtProt2(im)meth | + some = SomeEnumFirst; + // CHECK: [[@LINE-1]]:10 | enumerator/Swift | SomeEnumFirst | c:@M@modname@E@SomeEnum@SomeEnumFirst | +} + +#pragma clang attribute PUSH_GEN_DECL("other_mod_for_cat") +@interface I3(cat_other_mod) +-(void)meth_other_mod; +@end +#pragma clang attribute pop + +void test3(I3 *i3) { + [i3 meth_other_mod]; + // CHECK: [[@LINE-1]]:7 | instance-method/Swift | meth_other_mod | c:@CM@other_mod_for_cat@modname@objc(cs)I3(im)meth_other_mod | +} diff --git a/test/Index/Core/index-source.cpp b/test/Index/Core/index-source.cpp index 5d4a1be456832..7bb366c53611b 100644 --- a/test/Index/Core/index-source.cpp +++ b/test/Index/Core/index-source.cpp @@ -111,3 +111,180 @@ template<> class PartialSpecilizationClass<int, int> { }; // CHECK: [[@LINE-1]]:7 | class(Gen,TS)/C++ | PartialSpecilizationClass | c:@S@PartialSpecilizationClass>#I#I | <no-cgname> | Def,RelSpecialization | rel: 1 // CHECK-NEXT: RelSpecialization | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass + +template<typename T, typename S> +class PseudoOverridesInSpecializations { + void function() { } + void function(int) { } + + static void staticFunction() { } + + int field; + static int variable; + + typedef T TypeDef; + using TypeAlias = T; + + enum anEnum { }; + + struct Struct { }; + union Union { }; + + using TypealiasOrRecord = void; + + template<typename U> struct InnerTemplate { }; + template<typename U> struct InnerTemplate <U*> { }; +}; + +template<> +class PseudoOverridesInSpecializations<double, int> { + void function() { } +// CHECK: [[@LINE-1]]:8 | instance-method/C++ | function | c:@S@PseudoOverridesInSpecializations>#d#I@F@function# | __ZN32PseudoOverridesInSpecializationsIdiE8functionEv | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | function | c:@ST>2#T#T@PseudoOverridesInSpecializations@F@function# + + void staticFunction() { } +// CHECK: [[@LINE-1]]:8 | instance-method/C++ | staticFunction | c:@S@PseudoOverridesInSpecializations>#d#I@F@staticFunction# | __ZN32PseudoOverridesInSpecializationsIdiE14staticFunctionEv | Def,RelChild | rel: 1 +// CHECK-NOT: RelOver + + int notOverridingField = 0; + +// CHECK-LABEL: checLabelBreak + int checLabelBreak = 0; + + int field = 0; +// CHECK: [[@LINE-1]]:7 | field/C++ | field | c:@S@PseudoOverridesInSpecializations>#d#I@FI@field | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | field | c:@ST>2#T#T@PseudoOverridesInSpecializations@FI@field + + static double variable; +// CHECK: [[@LINE-1]]:17 | static-property/C++ | variable | c:@S@PseudoOverridesInSpecializations>#d#I@variable | __ZN32PseudoOverridesInSpecializationsIdiE8variableE | Decl,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | variable | c:@ST>2#T#T@PseudoOverridesInSpecializations@variable + + typedef double TypeDef; +// CHECK: [[@LINE-1]]:18 | type-alias/C | TypeDef | c:index-source.cpp@S@PseudoOverridesInSpecializations>#d#I@T@TypeDef | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | TypeDef | c:index-source.cpp@ST>2#T#T@PseudoOverridesInSpecializations@T@TypeDef + + using TypeAlias = int; +// CHECK: [[@LINE-1]]:9 | type-alias/C++ | TypeAlias | c:@S@PseudoOverridesInSpecializations>#d#I@TypeAlias | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | TypeAlias | c:@ST>2#T#T@PseudoOverridesInSpecializations@TypeAlias + + enum anEnum { }; +// CHECK: [[@LINE-1]]:8 | enum/C | anEnum | c:@S@PseudoOverridesInSpecializations>#d#I@E@anEnum | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | anEnum | c:@ST>2#T#T@PseudoOverridesInSpecializations@E@anEnum + class Struct { }; +// CHECK: [[@LINE-1]]:9 | class/C++ | Struct | c:@S@PseudoOverridesInSpecializations>#d#I@S@Struct | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | Struct | c:@ST>2#T#T@PseudoOverridesInSpecializations@S@Struct + union Union { }; +// CHECK: [[@LINE-1]]:9 | union/C | Union | c:@S@PseudoOverridesInSpecializations>#d#I@U@Union | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | Union | c:@ST>2#T#T@PseudoOverridesInSpecializations@U@Union + + struct TypealiasOrRecord { }; +// CHECK: [[@LINE-1]]:10 | struct/C | TypealiasOrRecord | c:@S@PseudoOverridesInSpecializations>#d#I@S@TypealiasOrRecord | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | TypealiasOrRecord | c:@ST>2#T#T@PseudoOverridesInSpecializations@TypealiasOrRecord + + template<typename U> struct InnerTemplate { }; +// CHECK: [[@LINE-1]]:31 | struct(Gen)/C++ | InnerTemplate | c:@S@PseudoOverridesInSpecializations>#d#I@ST>1#T@InnerTemplate | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | InnerTemplate | c:@ST>2#T#T@PseudoOverridesInSpecializations@ST>1#T@InnerTemplate + template<typename U> struct InnerTemplate <U*> { }; +// CHECK-NOT: RelOver +}; + +template<typename S> +class PseudoOverridesInSpecializations<float, S> { + typedef float TypealiasOrRecord; +// CHECK: [[@LINE-1]]:17 | type-alias/C | TypealiasOrRecord | c:index-source.cpp@SP>1#T@PseudoOverridesInSpecializations>#f#t0.0@T@TypealiasOrRecord | <no-cgname> | Def,RelChild,RelOver,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | TypealiasOrRecord | c:@ST>2#T#T@PseudoOverridesInSpecializations@TypealiasOrRecord +}; + +template<typename T, typename U> +class ConflictingPseudoOverridesInSpecialization { + void foo(T x); + void foo(U x); +}; + +template<typename T> +class ConflictingPseudoOverridesInSpecialization<int, T> { + void foo(T x); +// CHECK: [[@LINE-1]]:8 | instance-method/C++ | foo | c:@SP>1#T@ConflictingPseudoOverridesInSpecialization>#I#t0.0@F@foo#S0_# | <no-cgname> | Decl,RelChild,RelOver,RelSpecialization | rel: 3 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelOver,RelSpecialization | foo | c:@ST>2#T#T@ConflictingPseudoOverridesInSpecialization@F@foo#t0.0# +// CHECK-NEXT: RelOver,RelSpecialization | foo | c:@ST>2#T#T@ConflictingPseudoOverridesInSpecialization@F@foo#t0.1# +}; + +template<typename T, typename U, int x> +void functionSpecialization(); + +template<typename T, typename U, int x> +void functionSpecialization() { } +// CHECK: [[@LINE-1]]:6 | function/C | functionSpecialization | c:@FT@>3#T#T#NIfunctionSpecialization#v# | <no-cgname> | Def | rel: 0 + +template<> +void functionSpecialization<int, int, 0>(); +// CHECK: [[@LINE-1]]:6 | function(Gen,TS)/C++ | functionSpecialization | c:@F@functionSpecialization<#I#I#VI0># | __Z22functionSpecializationIiiLi0EEvv | Decl,RelSpecialization | rel: 1 +// CHECK-NEXT: RelSpecialization | functionSpecialization | c:@FT@>3#T#T#NIfunctionSpecialization#v# + +template<> +void functionSpecialization<int, int, 0>() { } +// CHECK: [[@LINE-1]]:6 | function(Gen,TS)/C++ | functionSpecialization | c:@F@functionSpecialization<#I#I#VI0># | __Z22functionSpecializationIiiLi0EEvv | Def,RelSpecialization | rel: 1 +// CHECK-NEXT: RelSpecialization | functionSpecialization | c:@FT@>3#T#T#NIfunctionSpecialization#v# + +struct ContainsSpecializedMemberFunction { + template<typename T> + void memberSpecialization(); +}; + +template<typename T> +void ContainsSpecializedMemberFunction::memberSpecialization() { +// CHECK: [[@LINE-1]]:41 | instance-method/C++ | memberSpecialization | c:@S@ContainsSpecializedMemberFunction@FT@>1#TmemberSpecialization#v# | <no-cgname> | Def,RelChild | rel: 1 +// CHECK-NEXT: RelChild +} + +template<> +void ContainsSpecializedMemberFunction::memberSpecialization<int>() { +// CHECK: [[@LINE-1]]:41 | instance-method(Gen,TS)/C++ | memberSpecialization | c:@S@ContainsSpecializedMemberFunction@F@memberSpecialization<#I># | __ZN33ContainsSpecializedMemberFunction20memberSpecializationIiEEvv | Def,RelChild,RelSpecialization | rel: 2 +// CHECK-NEXT: RelChild +// CHECK-NEXT: RelSpecialization | memberSpecialization | c:@S@ContainsSpecializedMemberFunction@FT@>1#TmemberSpecialization#v# +} + +template<typename T> +class SpecializationDecl; +// CHECK: [[@LINE-1]]:7 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | <no-cgname> | Ref | rel: 0 + +template<typename T> +class SpecializationDecl { }; +// CHECK: [[@LINE-1]]:7 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | <no-cgname> | Def | rel: 0 + +template<> +class SpecializationDecl<int>; +// CHECK: [[@LINE-1]]:7 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | <no-cgname> | Ref | rel: 0 + +template<> +class SpecializationDecl<int> { }; +// CHECK: [[@LINE-1]]:7 | class(Gen,TS)/C++ | SpecializationDecl | c:@S@SpecializationDecl>#I | <no-cgname> | Def,RelSpecialization | rel: 1 +// CHECK-NEXT: RelSpecialization | SpecializationDecl | c:@ST>1#T@SpecializationDecl +// CHECK-NEXT: [[@LINE-3]]:7 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | <no-cgname> | Ref | rel: 0 + +template<typename T> +class PartialSpecilizationClass<Cls, T>; +// CHECK: [[@LINE-1]]:7 | class(Gen)/C++ | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass | <no-cgname> | Ref | rel: 0 +// CHECK-NEXT: [[@LINE-2]]:33 | class/C++ | Cls | c:@S@Cls | <no-cgname> | Ref | rel: 0 + +template<> +class PartialSpecilizationClass<Cls, Cls> : Cls { }; +// CHECK: [[@LINE-1]]:7 | class(Gen,TS)/C++ | PartialSpecilizationClass | c:@S@PartialSpecilizationClass>#$@S@Cls#S0_ | <no-cgname> | Def,RelSpecialization | rel: 1 +// CHECK-NEXT: RelSpecialization | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass +// CHECK-NEXT: [[@LINE-3]]:45 | class/C++ | Cls | c:@S@Cls | <no-cgname> | Ref,RelBase,RelCont | rel: 1 +// CHECK-NEXT: RelBase,RelCont | PartialSpecilizationClass | c:@S@PartialSpecilizationClass>#$@S@Cls#S0_ +// CHECK-NEXT: [[@LINE-5]]:7 | class(Gen)/C++ | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass | <no-cgname> | Ref | rel: 0 +// CHECK-NEXT: [[@LINE-6]]:33 | class/C++ | Cls | c:@S@Cls | <no-cgname> | Ref | rel: 0 +// CHECK-NEXT: [[@LINE-7]]:38 | class/C++ | Cls | c:@S@Cls | <no-cgname> | Ref | rel: 0 diff --git a/test/Index/Core/index-source.m b/test/Index/Core/index-source.m index f48891417f5fa..a2eef89deea7e 100644 --- a/test/Index/Core/index-source.m +++ b/test/Index/Core/index-source.m @@ -353,7 +353,8 @@ typedef MyGenCls<Base *><MyEnumerating> MyEnumerator; typedef NS_ENUM(AnotherEnum, int) { // CHECK-NOT: [[@LINE-1]]:17 | type-alias/C | AnotherEnum | -// CHECK: [[@LINE-2]]:17 | enum/C | AnotherEnum | [[AnotherEnum_USR:.*]] | {{.*}} | Ref,RelCont | rel: 1 +// CHECK-NOT: [[@LINE-2]]:17 | {{.*}} | Ref +// CHECK: [[@LINE-3]]:17 | enum/C | AnotherEnum | [[AnotherEnum_USR:.*]] | {{.*}} | Def | rel: 0 AnotherEnumFirst = 0, AnotherEnumSecond = 1, AnotherEnumThird = 2, diff --git a/test/Index/index-refs.cpp b/test/Index/index-refs.cpp index 5a1ba1e8b43ae..d8bede0b47611 100644 --- a/test/Index/index-refs.cpp +++ b/test/Index/index-refs.cpp @@ -105,6 +105,7 @@ int ginitlist[] = {EnumVal}; // CHECK: [indexDeclaration]: kind: c++-class-template | name: TS | {{.*}} | loc: 47:8 // CHECK-NEXT: [indexDeclaration]: kind: struct-template-partial-spec | name: TS | USR: c:@SP>1#T@TS>#t0.0#I | {{.*}} | loc: 50:8 // CHECK-NEXT: [indexDeclaration]: kind: typedef | name: MyInt | USR: c:index-refs.cpp@SP>1#T@TS>#t0.0#I@T@MyInt | {{.*}} | loc: 51:15 | semantic-container: [TS:50:8] | lexical-container: [TS:50:8] +// CHECK-NEXT: [indexEntityReference]: kind: c++-class-template | name: TS | USR: c:@ST>2#T#T@TS | lang: C++ | cursor: TemplateRef=TS:47:8 | loc: 50:8 | <parent>:: <<NULL>> | container: [TU] | refkind: direct /* when indexing implicit instantiations [indexDeclaration]: kind: struct-template-spec | name: TS | USR: c:@S@TS>#I | {{.*}} | loc: 50:8 [indexDeclaration]: kind: typedef | name: MyInt | USR: c:index-refs.cpp@593@S@TS>#I@T@MyInt | {{.*}} | loc: 51:15 | semantic-container: [TS:50:8] | lexical-container: [TS:50:8] diff --git a/test/Index/print-type.cpp b/test/Index/print-type.cpp index 108ba53c80b0b..ff150f7dd6138 100644 --- a/test/Index/print-type.cpp +++ b/test/Index/print-type.cpp @@ -71,6 +71,11 @@ struct Specialization<int>; Specialization<Specialization<bool>& > templRefParam; auto autoTemplRefParam = templRefParam; +template<typename T, typename U = int> +struct DefaultedTypeExample {}; + +typedef DefaultedTypeExample<int> DefaultedTypeAlias; + // RUN: c-index-test -test-print-type %s -std=c++14 | FileCheck %s // CHECK: Namespace=outer:1:11 (Definition) [type=] [typekind=Invalid] [isPOD=0] // CHECK: ClassTemplate=Foo:4:8 (Definition) [type=] [typekind=Invalid] [isPOD=0] @@ -115,7 +120,7 @@ auto autoTemplRefParam = templRefParam; // CHECK: TemplateRef=Baz:9:8 [type=] [typekind=Invalid] [isPOD=0] // CHECK: IntegerLiteral= [type=int] [typekind=Int] [isPOD=1] // CHECK: TemplateRef=Foo:4:8 [type=] [typekind=Invalid] [isPOD=0] -// CHECK: FieldDecl=qux:29:38 (Definition) [type=Qux<int, char *, Foo<int>, outer::inner::Bar::FooType>] [typekind=Unexposed] [templateargs/4= [type=int] [typekind=Int] [type=char *] [typekind=Pointer] [type=Foo<int>] [typekind=Unexposed] [type=outer::inner::Bar::FooType] [typekind=Typedef]] [canonicaltype=outer::Qux<int, char *, outer::Foo<int>, int>] [canonicaltypekind=Record] [canonicaltemplateargs/4= [type=int] [typekind=Int] [type=char *] [typekind=Pointer] [type=outer::Foo<int>] [typekind=Record] [type=int] [typekind=Int]] [isPOD=1] +// CHECK: FieldDecl=qux:29:38 (Definition) [type=Qux<int, char *, Foo<int>, outer::inner::Bar::FooType>] [typekind=Unexposed] [templateargs/4= [type=int] [typekind=Int] [type=char *] [typekind=Pointer] [type=outer::Foo<int>] [typekind=Record] [type=int] [typekind=Int]] [canonicaltype=outer::Qux<int, char *, outer::Foo<int>, int>] [canonicaltypekind=Record] [canonicaltemplateargs/4= [type=int] [typekind=Int] [type=char *] [typekind=Pointer] [type=outer::Foo<int>] [typekind=Record] [type=int] [typekind=Int]] [isPOD=1] // CHECK: TemplateRef=Qux:12:8 [type=] [typekind=Invalid] [isPOD=0] // CHECK: TemplateRef=Foo:4:8 [type=] [typekind=Invalid] [isPOD=0] // CHECK: FunctionTemplate=tbar:36:3 [type=T (int)] [typekind=FunctionProto] [canonicaltype=type-parameter-0-0 (int)] [canonicaltypekind=FunctionProto] [resulttype=T] [resulttypekind=Unexposed] [isPOD=0] @@ -177,3 +182,4 @@ auto autoTemplRefParam = templRefParam; // CHECK: VarDecl=autoTemplRefParam:72:6 (Definition) [type=Specialization<Specialization<bool> &>] [typekind=Auto] [templateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [canonicaltype=Specialization<Specialization<bool> &>] [canonicaltypekind=Record] [canonicaltemplateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [isPOD=1] // CHECK: UnexposedExpr=templRefParam:71:40 [type=const Specialization<Specialization<bool> &>] [typekind=Unexposed] const [templateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [canonicaltype=const Specialization<Specialization<bool> &>] [canonicaltypekind=Record] [canonicaltemplateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [isPOD=1] // CHECK: DeclRefExpr=templRefParam:71:40 [type=Specialization<Specialization<bool> &>] [typekind=Unexposed] [templateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [canonicaltype=Specialization<Specialization<bool> &>] [canonicaltypekind=Record] [canonicaltemplateargs/1= [type=Specialization<bool> &] [typekind=LValueReference]] [isPOD=1] +// CHECK: TypedefDecl=DefaultedTypeAlias:77:35 (Definition) [type=DefaultedTypeAlias] [typekind=Typedef] [templateargs/2= [type=int] [typekind=Int] [type=int] [typekind=Int]] [canonicaltype=DefaultedTypeExample<int, int>] [canonicaltypekind=Record] [canonicaltemplateargs/2= [type=int] [typekind=Int] [type=int] [typekind=Int]] [isPOD=0] diff --git a/test/Modules/Inputs/objc-desig-init/A.h b/test/Modules/Inputs/objc-desig-init/A.h new file mode 100644 index 0000000000000..02e0105f3addf --- /dev/null +++ b/test/Modules/Inputs/objc-desig-init/A.h @@ -0,0 +1 @@ +#import "A2.h" diff --git a/test/Modules/Inputs/objc-desig-init/A2.h b/test/Modules/Inputs/objc-desig-init/A2.h new file mode 100644 index 0000000000000..5a4ceb495b322 --- /dev/null +++ b/test/Modules/Inputs/objc-desig-init/A2.h @@ -0,0 +1,4 @@ +#import "Base.h" + +@interface A2 : Base +@end diff --git a/test/Modules/Inputs/objc-desig-init/Base.h b/test/Modules/Inputs/objc-desig-init/Base.h new file mode 100644 index 0000000000000..47184256b139f --- /dev/null +++ b/test/Modules/Inputs/objc-desig-init/Base.h @@ -0,0 +1,4 @@ +@class NSString; +@interface Base +- (id)initWithNibName:(NSString *)nibNameOrNil __attribute__((objc_designated_initializer)); +@end diff --git a/test/Modules/Inputs/objc-desig-init/X.h b/test/Modules/Inputs/objc-desig-init/X.h new file mode 100644 index 0000000000000..cb46ba70a3746 --- /dev/null +++ b/test/Modules/Inputs/objc-desig-init/X.h @@ -0,0 +1,4 @@ +#import "A2.h" + +@interface X : A2 +@end diff --git a/test/Modules/Inputs/objc-desig-init/module.modulemap b/test/Modules/Inputs/objc-desig-init/module.modulemap new file mode 100644 index 0000000000000..0150efb8eb181 --- /dev/null +++ b/test/Modules/Inputs/objc-desig-init/module.modulemap @@ -0,0 +1,9 @@ +module Base { + header "Base.h" + export * +} + +module A { + header "A.h" + export * +} diff --git a/test/Modules/Inputs/template-default-args/a.h b/test/Modules/Inputs/template-default-args/a.h index 532cbc8be0353..1541d5f6a1b51 100644 --- a/test/Modules/Inputs/template-default-args/a.h +++ b/test/Modules/Inputs/template-default-args/a.h @@ -14,3 +14,16 @@ struct FriendL { template<typename T> friend struct L; }; END + +namespace DeferredLookup { + template<typename T, typename U = T> using X = U; + template<typename T> void f() { (void) X<T>(); } + template<typename T> int n = X<T>(); + template<typename T> struct S { X<T> xt; enum E : int; }; + template<typename T> enum S<T>::E : int { a = X<T>() }; + + namespace Indirect { + template<typename, bool = true> struct A {}; + template<typename> struct B { template<typename T> using C = A<T>; }; + } +} diff --git a/test/Modules/Inputs/template-default-args/d.h b/test/Modules/Inputs/template-default-args/d.h index 5961c91a82ec9..07ececc1f41d4 100644 --- a/test/Modules/Inputs/template-default-args/d.h +++ b/test/Modules/Inputs/template-default-args/d.h @@ -4,3 +4,10 @@ struct FriendL { template<typename T> friend struct L; }; END + +namespace DeferredLookup { + namespace Indirect { + template<typename, bool = true> struct A {}; + template<typename> struct B { template<typename T> using C = A<T>; }; + } +} diff --git a/test/Modules/localsubmodulevis.m b/test/Modules/localsubmodulevis.m index f8c4b3c755eb3..bf317651df968 100644 --- a/test/Modules/localsubmodulevis.m +++ b/test/Modules/localsubmodulevis.m @@ -1,4 +1,4 @@ -// RUN: rm -rf %t +// RUN: rm -rf %t.a %t.b // RUN: %clang_cc1 -fmodules -fmodules-local-submodule-visibility -fimplicit-module-maps -fmodules-cache-path=%t.a -I %S/Inputs/preprocess -verify %s // RUN: %clang_cc1 -fmodules -fmodules-local-submodule-visibility -fimplicit-module-maps -fmodules-cache-path=%t.b -I %S/Inputs/preprocess -x c -verify -x c %s diff --git a/test/Modules/objc-designated-init-mod.m b/test/Modules/objc-designated-init-mod.m new file mode 100644 index 0000000000000..15c25a3ec2ef0 --- /dev/null +++ b/test/Modules/objc-designated-init-mod.m @@ -0,0 +1,17 @@ +// RUN: rm -rf %t +// RUN: %clang_cc1 -fmodules-cache-path=%t -fmodules -fimplicit-module-maps -I %S/Inputs/objc-desig-init %s -verify +// expected-no-diagnostics + +#import "X.h" +#import "Base.h" +#import "A.h" + +@implementation X + +- (instancetype)initWithNibName:(NSString *)nibName { + if ((self = [super initWithNibName:nibName])) { + return self; + } + return self; +} +@end diff --git a/test/Modules/template-default-args.cpp b/test/Modules/template-default-args.cpp index 9d16cbf43420c..c51cb28408844 100644 --- a/test/Modules/template-default-args.cpp +++ b/test/Modules/template-default-args.cpp @@ -44,3 +44,20 @@ H<> h; // expected-error {{default argument of 'H' must be imported from module I<> i; L<> *l; END + +namespace DeferredLookup { + template<typename T, typename U = T> using X = U; + template<typename T> void f() { (void) X<T>(); } + template<typename T> int n = X<T>(); // expected-warning {{extension}} + template<typename T> struct S { X<T> xt; enum E : int; }; + template<typename T> enum S<T>::E : int { a = X<T>() }; + + void test() { + f<int>(); + n<int> = 1; + S<int> s; + S<int>::E e = S<int>::E::a; + + Indirect::B<int>::C<int> indirect; + } +} diff --git a/test/OpenMP/capturing_in_templates.cpp b/test/OpenMP/capturing_in_templates.cpp new file mode 100644 index 0000000000000..a4dcbee8fab97 --- /dev/null +++ b/test/OpenMP/capturing_in_templates.cpp @@ -0,0 +1,30 @@ +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-ibm-linux-gnu -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s +// expected-no-diagnostics + +template <typename T1, typename T2> +struct pair { + T1 t1; + T2 t2; + pair(T1 t1, T2 t2) : t1(t1), t2(t2) {} +}; + +template <typename T1, typename T2> +pair<T1, T2> make_pair(T1 &&t1, T2 &&t2) { + return {t1, t2}; +} + +// CHECK-LABEL: @main +int main(int argc, char **argv) { +// CHECK: call i32 @__tgt_target(i32 -1, i8* @{{.+}}.region_id, i32 0, i8** null, i8** null, i64* null, i32* null) +#pragma omp target + { + for (int i = 0; i < 64; ++i) { + for (int j = 0; j < 64; ++j) { + auto foo = make_pair(i * i, j * j); + } + } + } + return 0; +} + +// CHECK: call {{.+}} @{{.*}}make_pair diff --git a/test/OpenMP/distribute_parallel_for_codegen.cpp b/test/OpenMP/distribute_parallel_for_codegen.cpp new file mode 100644 index 0000000000000..62c38f1888517 --- /dev/null +++ b/test/OpenMP/distribute_parallel_for_codegen.cpp @@ -0,0 +1,2260 @@ +// Test host code gen +// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64 +// RUN: %clang_cc1 -DLAMBDA -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -DLAMBDA -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64 +// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32 +// RUN: %clang_cc1 -DLAMBDA -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -DLAMBDA -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32 + +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64 +// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64 +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32 +// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32 +// expected-no-diagnostics +#ifndef HEADER +#define HEADER + + +template <typename T> +T tmain() { + T *a, *b, *c; + int n = 10000; + int ch = 100; + + // no schedule clauses + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + } + + // dist_schedule: static no chunk + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for dist_schedule(static) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + } + + // dist_schedule: static chunk + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for dist_schedule(static, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + } + + // schedule: static no chunk + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for schedule(static) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + } + + // schedule: static chunk + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for schedule(static, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + } + + // schedule: dynamic no chunk + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for schedule(dynamic) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + } + + // schedule: dynamic chunk + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for schedule(dynamic, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + } + + return T(); +} + +int main() { + double *a, *b, *c; + int n = 10000; + int ch = 100; + +#ifdef LAMBDA + // LAMBDA-LABEL: @main + // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]]( + [&]() { + // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]]( + + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN_1:@.+]]( + + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN_2:@.+]]( + + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN_3:@.+]]( + + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN_4:@.+]]( + + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN_5:@.+]]( + + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN_6:@.+]]( + + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN_7:@.+]]( + + // no schedule clauses + #pragma omp target + #pragma omp teams + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_1]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}}) + + #pragma omp distribute parallel for + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // LAMBDA: define{{.+}} void [[OMP_OUTLINED_1]]( + // LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca + // LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca + // LAMBDA-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca + // LAMBDA-DAG: [[OMP_ST:%.omp.stride]] = alloca + + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + + // check EUB for distribute + // LAMBDA-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], + // LAMBDA: [[NUM_IT_1:%.+]] = load{{.+}}, + // LAMBDA-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] + // LAMBDA: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] + // LAMBDA-DAG: [[EUB_TRUE]]: + // LAMBDA: [[NUM_IT_2:%.+]] = load{{.+}}, + // LAMBDA: br label %[[EUB_END:.+]] + // LAMBDA-DAG: [[EUB_FALSE]]: + // LAMBDA: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], + // LAMBDA: br label %[[EUB_END]] + // LAMBDA-DAG: [[EUB_END]]: + // LAMBDA-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] + // LAMBDA: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + + // initialize omp.iv + // LAMBDA: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], + // LAMBDA: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], + // LAMBDA: br label %[[OMP_JUMP_BACK:.+]] + + // check exit condition + // LAMBDA: [[OMP_JUMP_BACK]]: + // LAMBDA-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], + // LAMBDA-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], + // LAMBDA: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] + // LAMBDA: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // LAMBDA: [[DIST_BODY]]: + // LAMBDA-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to + // LAMBDA-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to + // check that distlb and distub are properly passed to fork_call + // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) + // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) + // LAMBDA: br label %[[DIST_INC:.+]] + + // increment by stride (distInc - 'parallel for' executes the whole chunk) and latch + // LAMBDA: [[DIST_INC]]: + // LAMBDA-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], + // LAMBDA-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], + // LAMBDA: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]] + // LAMBDA: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], + // LAMBDA: br label %[[OMP_JUMP_BACK]] + + // LAMBDA-DAG: call void @__kmpc_for_static_fini( + // LAMBDA: ret + + // implementation of 'parallel for' + // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_1]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + + // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + + // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used + // In this case we use EUB + // LAMBDA-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], + // LAMBDA: [[PF_NUM_IT_1:%.+]] = load{{.+}}, + // LAMBDA-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] + // LAMBDA: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] + // LAMBDA: [[PF_EUB_TRUE]]: + // LAMBDA: [[PF_NUM_IT_2:%.+]] = load{{.+}}, + // LAMBDA: br label %[[PF_EUB_END:.+]] + // LAMBDA-DAG: [[PF_EUB_FALSE]]: + // LAMBDA: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], + // LAMBDA: br label %[[PF_EUB_END]] + // LAMBDA-DAG: [[PF_EUB_END]]: + // LAMBDA-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] + // LAMBDA: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + + // initialize omp.iv + // LAMBDA: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label %[[OMP_PF_JUMP_BACK:.+]] + + // check exit condition + // LAMBDA: [[OMP_PF_JUMP_BACK]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], + // LAMBDA: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] + // LAMBDA: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // LAMBDA: [[PF_BODY]]: + // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label {{.+}} + + // check stride 1 for 'for' in 'distribute parallel for' + // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 + // LAMBDA: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label %[[OMP_PF_JUMP_BACK]] + + // LAMBDA-DAG: call void @__kmpc_for_static_fini( + // LAMBDA: ret + + [&]() { + a[i] = b[i] + c[i]; + }(); + } + + // dist_schedule: static no chunk (same sa default - no dist_schedule) + #pragma omp target + #pragma omp teams + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_2]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_2:@.+]] to {{.+}}) + + #pragma omp distribute parallel for dist_schedule(static) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // LAMBDA: define{{.+}} void [[OMP_OUTLINED_2]]( + // LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca + // LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca + // LAMBDA-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca + // LAMBDA-DAG: [[OMP_ST:%.omp.stride]] = alloca + + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + + // check EUB for distribute + // LAMBDA-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], + // LAMBDA: [[NUM_IT_1:%.+]] = load{{.+}}, + // LAMBDA-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] + // LAMBDA: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] + // LAMBDA-DAG: [[EUB_TRUE]]: + // LAMBDA: [[NUM_IT_2:%.+]] = load{{.+}}, + // LAMBDA: br label %[[EUB_END:.+]] + // LAMBDA-DAG: [[EUB_FALSE]]: + // LAMBDA: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], + // LAMBDA: br label %[[EUB_END]] + // LAMBDA-DAG: [[EUB_END]]: + // LAMBDA-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] + // LAMBDA: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + + // initialize omp.iv + // LAMBDA: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], + // LAMBDA: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], + // LAMBDA: br label %[[OMP_JUMP_BACK:.+]] + + // check exit condition + // LAMBDA: [[OMP_JUMP_BACK]]: + // LAMBDA-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], + // LAMBDA-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], + // LAMBDA: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] + // LAMBDA: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // LAMBDA: [[DIST_BODY]]: + // LAMBDA-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to + // LAMBDA-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to + // check that distlb and distub are properly passed to fork_call + // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) + // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) + // LAMBDA: br label %[[DIST_INC:.+]] + + // increment by stride (distInc - 'parallel for' executes the whole chunk) and latch + // LAMBDA: [[DIST_INC]]: + // LAMBDA-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], + // LAMBDA-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], + // LAMBDA: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]] + // LAMBDA: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], + // LAMBDA: br label %[[OMP_JUMP_BACK]] + + // LAMBDA-DAG: call void @__kmpc_for_static_fini( + // LAMBDA: ret + + // implementation of 'parallel for' + // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_2]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + + // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + + // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used + // In this case we use EUB + // LAMBDA-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], + // LAMBDA: [[PF_NUM_IT_1:%.+]] = load{{.+}}, + // LAMBDA-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] + // LAMBDA: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] + // LAMBDA: [[PF_EUB_TRUE]]: + // LAMBDA: [[PF_NUM_IT_2:%.+]] = load{{.+}}, + // LAMBDA: br label %[[PF_EUB_END:.+]] + // LAMBDA-DAG: [[PF_EUB_FALSE]]: + // LAMBDA: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], + // LAMBDA: br label %[[PF_EUB_END]] + // LAMBDA-DAG: [[PF_EUB_END]]: + // LAMBDA-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] + // LAMBDA: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + + // initialize omp.iv + // LAMBDA: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label %[[OMP_PF_JUMP_BACK:.+]] + + // check exit condition + // LAMBDA: [[OMP_PF_JUMP_BACK]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], + // LAMBDA: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] + // LAMBDA: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // LAMBDA: [[PF_BODY]]: + // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label {{.+}} + + // check stride 1 for 'for' in 'distribute parallel for' + // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 + // LAMBDA: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label %[[OMP_PF_JUMP_BACK]] + + // LAMBDA-DAG: call void @__kmpc_for_static_fini( + // LAMBDA: ret + [&]() { + a[i] = b[i] + c[i]; + }(); + } + + // dist_schedule: static chunk + #pragma omp target + #pragma omp teams + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_3]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_3:@.+]] to {{.+}}) + + #pragma omp distribute parallel for dist_schedule(static, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // LAMBDA: define{{.+}} void [[OMP_OUTLINED_3]]( + // LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca + // LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca + // LAMBDA-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca + // LAMBDA-DAG: [[OMP_ST:%.omp.stride]] = alloca + + // unlike the previous tests, in this one we have a outer and inner loop for 'distribute' + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 91, + // LAMBDA: br label %[[DIST_OUTER_LOOP_HEADER:.+]] + + // LAMBDA: [[DIST_OUTER_LOOP_HEADER]]: + // check EUB for distribute + // LAMBDA-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], + // LAMBDA: [[NUM_IT_1:%.+]] = load{{.+}}, + // LAMBDA-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] + // LAMBDA: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] + // LAMBDA-DAG: [[EUB_TRUE]]: + // LAMBDA: [[NUM_IT_2:%.+]] = load{{.+}}, + // LAMBDA: br label %[[EUB_END:.+]] + // LAMBDA-DAG: [[EUB_FALSE]]: + // LAMBDA: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], + // LAMBDA: br label %[[EUB_END]] + // LAMBDA-DAG: [[EUB_END]]: + // LAMBDA-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] + // LAMBDA: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + + // initialize omp.iv + // LAMBDA: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], + // LAMBDA: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], + + // check exit condition + // LAMBDA-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], + // LAMBDA-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], + // LAMBDA: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] + // LAMBDA: br {{.+}} [[CMP_IV_UB]], label %[[DIST_OUTER_LOOP_BODY:.+]], label %[[DIST_OUTER_LOOP_END:.+]] + + // LAMBDA: [[DIST_OUTER_LOOP_BODY]]: + // LAMBDA: br label %[[DIST_INNER_LOOP_HEADER:.+]] + + // LAMBDA: [[DIST_INNER_LOOP_HEADER]]: + // LAMBDA-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}} [[OMP_IV]], + // LAMBDA-DAG: [[OMP_UB_VAL_4:%.+]] = load {{.+}} [[OMP_UB]], + // LAMBDA: [[CMP_IV_UB_2:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_2]], [[OMP_UB_VAL_4]] + // LAMBDA: br{{.+}} [[CMP_IV_UB_2]], label %[[DIST_INNER_LOOP_BODY:.+]], label %[[DIST_INNER_LOOP_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // LAMBDA: [[DIST_INNER_LOOP_BODY]]: + // LAMBDA-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}} + // LAMBDA-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}} + // check that distlb and distub are properly passed to fork_call + // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) + // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) + // LAMBDA: br label %[[DIST_INNER_LOOP_INC:.+]] + + // check DistInc + // LAMBDA: [[DIST_INNER_LOOP_INC]]: + // LAMBDA-DAG: [[OMP_IV_VAL_3:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], + // LAMBDA-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], + // LAMBDA: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_3]], [[OMP_ST_VAL_1]] + // LAMBDA: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], + // LAMBDA: br label %[[DIST_INNER_LOOP_HEADER]] + + // LAMBDA: [[DIST_INNER_LOOP_END]]: + // LAMBDA: br label %[[DIST_OUTER_LOOP_INC:.+]] + + // LAMBDA: [[DIST_OUTER_LOOP_INC]]: + // check NextLB and NextUB + // LAMBDA-DAG: [[OMP_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_LB]], + // LAMBDA-DAG: [[OMP_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]], + // LAMBDA-DAG: [[OMP_LB_NEXT:%.+]] = add{{.+}} [[OMP_LB_VAL_2]], [[OMP_ST_VAL_2]] + // LAMBDA: store{{.+}} [[OMP_LB_NEXT]], {{.+}}* [[OMP_LB]], + // LAMBDA-DAG: [[OMP_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_UB]], + // LAMBDA-DAG: [[OMP_ST_VAL_3:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]], + // LAMBDA-DAG: [[OMP_UB_NEXT:%.+]] = add{{.+}} [[OMP_UB_VAL_5]], [[OMP_ST_VAL_3]] + // LAMBDA: store{{.+}} [[OMP_UB_NEXT]], {{.+}}* [[OMP_UB]], + // LAMBDA: br label %[[DIST_OUTER_LOOP_HEADER]] + + // outer loop exit + // LAMBDA: [[DIST_OUTER_LOOP_END]]: + // LAMBDA-DAG: call void @__kmpc_for_static_fini( + // LAMBDA: ret + + // skip implementation of 'parallel for': using default scheduling and was tested above + [&]() { + a[i] = b[i] + c[i]; + }(); + } + + // schedule: static no chunk + #pragma omp target + #pragma omp teams + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_4]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_4:@.+]] to {{.+}}) + + #pragma omp distribute parallel for schedule(static) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // LAMBDA: define{{.+}} void [[OMP_OUTLINED_4]]( + // LAMBDA-DAG: [[OMP_IV:%.omp.iv]] = alloca + // LAMBDA-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca + // LAMBDA-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca + // LAMBDA-DAG: [[OMP_ST:%.omp.stride]] = alloca + + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_4:@.+]] to {{.+}}, + // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case + // LAMBDA: ret + + // 'parallel for' implementation is the same as the case without schedule clase (static no chunk is the default) + // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_4]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + + // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + + // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used + // In this case we use EUB + // LAMBDA-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], + // LAMBDA: [[PF_NUM_IT_1:%.+]] = load{{.+}}, + // LAMBDA-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] + // LAMBDA: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] + // LAMBDA: [[PF_EUB_TRUE]]: + // LAMBDA: [[PF_NUM_IT_2:%.+]] = load{{.+}}, + // LAMBDA: br label %[[PF_EUB_END:.+]] + // LAMBDA-DAG: [[PF_EUB_FALSE]]: + // LAMBDA: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], + // LAMBDA: br label %[[PF_EUB_END]] + // LAMBDA-DAG: [[PF_EUB_END]]: + // LAMBDA-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] + // LAMBDA: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + + // initialize omp.iv + // LAMBDA: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label %[[OMP_PF_JUMP_BACK:.+]] + + // check exit condition + // LAMBDA: [[OMP_PF_JUMP_BACK]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], + // LAMBDA: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] + // LAMBDA: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // LAMBDA: [[PF_BODY]]: + // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label {{.+}} + + // check stride 1 for 'for' in 'distribute parallel for' + // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 + // LAMBDA: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label %[[OMP_PF_JUMP_BACK]] + + // LAMBDA-DAG: call void @__kmpc_for_static_fini( + // LAMBDA: ret + + [&]() { + a[i] = b[i] + c[i]; + }(); + } + + // schedule: static chunk + #pragma omp target + #pragma omp teams + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_5]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_5:@.+]] to {{.+}}) + + #pragma omp distribute parallel for schedule(static, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // LAMBDA: define{{.+}} void [[OMP_OUTLINED_5]]( + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_5:@.+]] to {{.+}}, + // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case + // LAMBDA: ret + + // 'parallel for' implementation using outer and inner loops and PrevEUB + // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_5]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}}) + // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 33, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + + // check PrevEUB (using PrevUB instead of NumIt as upper bound) + // LAMBDA: [[OMP_PF_OUTER_LOOP_HEADER]]: + // LAMBDA-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], + // LAMBDA-64-DAG: [[OMP_PF_UB_VAL_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_1]] to + // LAMBDA: [[PF_PREV_UB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // LAMBDA-64-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_CONV]], [[PF_PREV_UB_VAL_1]] + // LAMBDA-32-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_PREV_UB_VAL_1]] + // LAMBDA: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] + // LAMBDA: [[PF_EUB_TRUE]]: + // LAMBDA: [[PF_PREV_UB_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // LAMBDA: br label %[[PF_EUB_END:.+]] + // LAMBDA-DAG: [[PF_EUB_FALSE]]: + // LAMBDA: [[OMP_PF_UB_VAL_2:%.+]] = load{{.+}} [[OMP_PF_UB]], + // LAMBDA-64: [[OMP_PF_UB_VAL_2_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_2]] to + // LAMBDA: br label %[[PF_EUB_END]] + // LAMBDA-DAG: [[PF_EUB_END]]: + // LAMBDA-64-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2_CONV]], %[[PF_EUB_FALSE]] ] + // LAMBDA-32-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2]], %[[PF_EUB_FALSE]] ] + // LAMBDA-64-DAG: [[PF_EUB_RES_CONV:%.+]] = trunc{{.+}} [[PF_EUB_RES]] to + // LAMBDA-64: store{{.+}} [[PF_EUB_RES_CONV]],{{.+}} [[OMP_PF_UB]], + // LAMBDA-32: store{{.+}} [[PF_EUB_RES]], {{.+}} [[OMP_PF_UB]], + + // initialize omp.iv (IV = LB) + // LAMBDA: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + + // outer loop: while (IV < UB) { + // LAMBDA-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], + // LAMBDA: [[PF_CMP_IV_UB_1:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] + // LAMBDA: br{{.+}} [[PF_CMP_IV_UB_1]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_BODY]]: + // LAMBDA: br label %[[OMP_PF_INNER_FOR_HEADER:.+]] + + // LAMBDA: [[OMP_PF_INNER_FOR_HEADER]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], + // LAMBDA: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] + // LAMBDA: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + + // LAMBDA: [[OMP_PF_INNER_LOOP_BODY]]: + // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // skip body branch + // LAMBDA: br{{.+}} + // LAMBDA: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + + // IV = IV + 1 and inner loop latch + // LAMBDA: [[OMP_PF_INNER_LOOP_INC]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], + // LAMBDA-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 + // LAMBDA-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], + // LAMBDA: br label %[[OMP_PF_INNER_FOR_HEADER]] + + // check NextLB and NextUB + // LAMBDA: [[OMP_PF_INNER_LOOP_END]]: + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_INC]]: + // LAMBDA-DAG: [[OMP_PF_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA-DAG: [[OMP_PF_ST_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]], + // LAMBDA-DAG: [[OMP_PF_LB_NEXT:%.+]] = add{{.+}} [[OMP_PF_LB_VAL_2]], [[OMP_PF_ST_VAL_1]] + // LAMBDA: store{{.+}} [[OMP_PF_LB_NEXT]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], + // LAMBDA-DAG: [[OMP_PF_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]], + // LAMBDA-DAG: [[OMP_PF_UB_NEXT:%.+]] = add{{.+}} [[OMP_PF_UB_VAL_5]], [[OMP_PF_ST_VAL_2]] + // LAMBDA: store{{.+}} [[OMP_PF_UB_NEXT]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_END]]: + // LAMBDA-DAG: call void @__kmpc_for_static_fini( + // LAMBDA: ret + [&]() { + a[i] = b[i] + c[i]; + }(); + } + + // schedule: dynamic no chunk + #pragma omp target + #pragma omp teams + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_6]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_6:@.+]] to {{.+}}) + + #pragma omp distribute parallel for schedule(dynamic) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // LAMBDA: define{{.+}} void [[OMP_OUTLINED_6]]( + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_6:@.+]] to {{.+}}, + // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case + // LAMBDA: ret + + // 'parallel for' implementation using outer and inner loops and PrevEUB + // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_6]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], + // LAMBDA: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}}) + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_HEADER]]: + // LAMBDA: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]]) + // LAMBDA: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0 + // LAMBDA: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + + // initialize omp.iv (IV = LB) + // LAMBDA: [[OMP_PF_OUTER_LOOP_BODY]]: + // LAMBDA-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]] + + // LAMBDA: [[OMP_PF_INNER_LOOP_HEADER]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], + // LAMBDA: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] + // LAMBDA: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + + // LAMBDA: [[OMP_PF_INNER_LOOP_BODY]]: + // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // skip body branch + // LAMBDA: br{{.+}} + // LAMBDA: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + + // IV = IV + 1 and inner loop latch + // LAMBDA: [[OMP_PF_INNER_LOOP_INC]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], + // LAMBDA-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 + // LAMBDA-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], + // LAMBDA: br label %[[OMP_PF_INNER_LOOP_HEADER]] + + // check NextLB and NextUB + // LAMBDA: [[OMP_PF_INNER_LOOP_END]]: + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_INC]]: + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_END]]: + // LAMBDA: ret + [&]() { + a[i] = b[i] + c[i]; + }(); + } + + // schedule: dynamic chunk + #pragma omp target + #pragma omp teams + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN_7]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_7:@.+]] to {{.+}}) + + #pragma omp distribute parallel for schedule(dynamic, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // LAMBDA: define{{.+}} void [[OMP_OUTLINED_7]]( + // LAMBDA: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_7:@.+]] to {{.+}}, + // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case + // LAMBDA: ret + + // 'parallel for' implementation using outer and inner loops and PrevEUB + // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED_7]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}}) + // LAMBDA-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + // LAMBDA-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // LAMBDA-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // LAMBDA-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // LAMBDA-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // LAMBDA-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // LAMBDA-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // LAMBDA-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], + // LAMBDA: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}}) + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_HEADER]]: + // LAMBDA: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]]) + // LAMBDA: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0 + // LAMBDA: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + + // initialize omp.iv (IV = LB) + // LAMBDA: [[OMP_PF_OUTER_LOOP_BODY]]: + // LAMBDA-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // LAMBDA-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // LAMBDA: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]] + + // LAMBDA: [[OMP_PF_INNER_LOOP_HEADER]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // LAMBDA-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], + // LAMBDA: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] + // LAMBDA: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + + // LAMBDA: [[OMP_PF_INNER_LOOP_BODY]]: + // LAMBDA-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // skip body branch + // LAMBDA: br{{.+}} + // LAMBDA: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + + // IV = IV + 1 and inner loop latch + // LAMBDA: [[OMP_PF_INNER_LOOP_INC]]: + // LAMBDA-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], + // LAMBDA-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 + // LAMBDA-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], + // LAMBDA: br label %[[OMP_PF_INNER_LOOP_HEADER]] + + // check NextLB and NextUB + // LAMBDA: [[OMP_PF_INNER_LOOP_END]]: + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_INC]]: + // LAMBDA: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + + // LAMBDA: [[OMP_PF_OUTER_LOOP_END]]: + // LAMBDA: ret + [&]() { + a[i] = b[i] + c[i]; + }(); + } + }(); + return 0; +#else + // CHECK-LABEL: @main + + // CHECK: call i{{[0-9]+}} @__tgt_target_teams( + // CHECK: call void [[OFFLOADING_FUN_1:@.+]]( + + // CHECK: call i{{[0-9]+}} @__tgt_target_teams( + // CHECK: call void [[OFFLOADING_FUN_2:@.+]]( + + // CHECK: call i{{[0-9]+}} @__tgt_target_teams( + // CHECK: call void [[OFFLOADING_FUN_3:@.+]]( + + // CHECK: call i{{[0-9]+}} @__tgt_target_teams( + // CHECK: call void [[OFFLOADING_FUN_4:@.+]]( + + // CHECK: call i{{[0-9]+}} @__tgt_target_teams( + // CHECK: call void [[OFFLOADING_FUN_5:@.+]]( + + // CHECK: call i{{[0-9]+}} @__tgt_target_teams( + // CHECK: call void [[OFFLOADING_FUN_6:@.+]]( + + // CHECK: call i{{[0-9]+}} @__tgt_target_teams( + // CHECK: call void [[OFFLOADING_FUN_7:@.+]]( + + // CHECK: call{{.+}} [[TMAIN:@.+]]() + + // no schedule clauses + #pragma omp target + #pragma omp teams + // CHECK: define internal void [[OFFLOADING_FUN_1]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}}) + + #pragma omp distribute parallel for + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // CHECK: define{{.+}} void [[OMP_OUTLINED_1]]( + // CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca + // CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca + // CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca + // CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca + + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + + // check EUB for distribute + // CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], + // CHECK: [[NUM_IT_1:%.+]] = load{{.+}}, + // CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] + // CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] + // CHECK-DAG: [[EUB_TRUE]]: + // CHECK: [[NUM_IT_2:%.+]] = load{{.+}}, + // CHECK: br label %[[EUB_END:.+]] + // CHECK-DAG: [[EUB_FALSE]]: + // CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], + // CHECK: br label %[[EUB_END]] + // CHECK-DAG: [[EUB_END]]: + // CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] + // CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + + // initialize omp.iv + // CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], + // CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], + // CHECK: br label %[[OMP_JUMP_BACK:.+]] + + // check exit condition + // CHECK: [[OMP_JUMP_BACK]]: + // CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], + // CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], + // CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] + // CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // CHECK: [[DIST_BODY]]: + // CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], + // CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], + // CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}} + // check that distlb and distub are properly passed to fork_call + // CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) + // CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) + // CHECK: br label %[[DIST_INC:.+]] + + // increment by stride (distInc - 'parallel for' executes the whole chunk) and latch + // CHECK: [[DIST_INC]]: + // CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], + // CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], + // CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]] + // CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], + // CHECK: br label %[[OMP_JUMP_BACK]] + + // CHECK-DAG: call void @__kmpc_for_static_fini( + // CHECK: ret + + // implementation of 'parallel for' + // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_1]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + + // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + + // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used + // In this case we use EUB + // CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], + // CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}}, + // CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] + // CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] + // CHECK: [[PF_EUB_TRUE]]: + // CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}}, + // CHECK: br label %[[PF_EUB_END:.+]] + // CHECK-DAG: [[PF_EUB_FALSE]]: + // CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], + // CHECK: br label %[[PF_EUB_END]] + // CHECK-DAG: [[PF_EUB_END]]: + // CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] + // CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + + // initialize omp.iv + // CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // CHECK: br label %[[OMP_PF_JUMP_BACK:.+]] + + // check exit condition + // CHECK: [[OMP_PF_JUMP_BACK]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], + // CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], + // CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] + // CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // CHECK: [[PF_BODY]]: + // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK: br label {{.+}} + + // check stride 1 for 'for' in 'distribute parallel for' + // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 + // CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], + // CHECK: br label %[[OMP_PF_JUMP_BACK]] + + // CHECK-DAG: call void @__kmpc_for_static_fini( + // CHECK: ret + } + + // dist_schedule: static no chunk + #pragma omp target + #pragma omp teams + // CHECK: define{{.+}} void [[OFFLOADING_FUN_2]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_2:@.+]] to {{.+}}) + + #pragma omp distribute parallel for dist_schedule(static) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // CHECK: define{{.+}} void [[OMP_OUTLINED_2]]( + // CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca + // CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca + // CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca + // CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca + + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + + // check EUB for distribute + // CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], + // CHECK: [[NUM_IT_1:%.+]] = load{{.+}}, + // CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] + // CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] + // CHECK-DAG: [[EUB_TRUE]]: + // CHECK: [[NUM_IT_2:%.+]] = load{{.+}}, + // CHECK: br label %[[EUB_END:.+]] + // CHECK-DAG: [[EUB_FALSE]]: + // CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], + // CHECK: br label %[[EUB_END]] + // CHECK-DAG: [[EUB_END]]: + // CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] + // CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + + // initialize omp.iv + // CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], + // CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], + // CHECK: br label %[[OMP_JUMP_BACK:.+]] + + // check exit condition + // CHECK: [[OMP_JUMP_BACK]]: + // CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], + // CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], + // CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] + // CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // CHECK: [[DIST_BODY]]: + // CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], + // CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], + // CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}} + // check that distlb and distub are properly passed to fork_call + // CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) + // CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) + // CHECK: br label %[[DIST_INC:.+]] + + // increment by stride (distInc - 'parallel for' executes the whole chunk) and latch + // CHECK: [[DIST_INC]]: + // CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], + // CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], + // CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]] + // CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], + // CHECK: br label %[[OMP_JUMP_BACK]] + + // CHECK-DAG: call void @__kmpc_for_static_fini( + // CHECK: ret + + // implementation of 'parallel for' + // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_2]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + + // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + + // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used + // In this case we use EUB + // CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], + // CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}}, + // CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] + // CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] + // CHECK: [[PF_EUB_TRUE]]: + // CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}}, + // CHECK: br label %[[PF_EUB_END:.+]] + // CHECK-DAG: [[PF_EUB_FALSE]]: + // CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], + // CHECK: br label %[[PF_EUB_END]] + // CHECK-DAG: [[PF_EUB_END]]: + // CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] + // CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + + // initialize omp.iv + // CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // CHECK: br label %[[OMP_PF_JUMP_BACK:.+]] + + // check exit condition + // CHECK: [[OMP_PF_JUMP_BACK]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], + // CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], + // CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] + // CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // CHECK: [[PF_BODY]]: + // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK: br label {{.+}} + + // check stride 1 for 'for' in 'distribute parallel for' + // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 + // CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], + // CHECK: br label %[[OMP_PF_JUMP_BACK]] + + // CHECK-DAG: call void @__kmpc_for_static_fini( + // CHECK: ret + } + + // dist_schedule: static chunk + #pragma omp target + #pragma omp teams + // CHECK: define{{.+}} void [[OFFLOADING_FUN_3]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_3:@.+]] to {{.+}}) + + #pragma omp distribute parallel for dist_schedule(static, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // CHECK: define{{.+}} void [[OMP_OUTLINED_3]]( + // CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca + // CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca + // CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca + // CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca + + // unlike the previous tests, in this one we have a outer and inner loop for 'distribute' + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 91, + // CHECK: br label %[[DIST_OUTER_LOOP_HEADER:.+]] + + // CHECK: [[DIST_OUTER_LOOP_HEADER]]: + // check EUB for distribute + // CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], + // CHECK: [[NUM_IT_1:%.+]] = load{{.+}}, + // CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] + // CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] + // CHECK-DAG: [[EUB_TRUE]]: + // CHECK: [[NUM_IT_2:%.+]] = load{{.+}}, + // CHECK: br label %[[EUB_END:.+]] + // CHECK-DAG: [[EUB_FALSE]]: + // CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], + // CHECK: br label %[[EUB_END]] + // CHECK-DAG: [[EUB_END]]: + // CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] + // CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + + // initialize omp.iv + // CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], + // CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], + + // check exit condition + // CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], + // CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], + // CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] + // CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_OUTER_LOOP_BODY:.+]], label %[[DIST_OUTER_LOOP_END:.+]] + + // CHECK: [[DIST_OUTER_LOOP_BODY]]: + // CHECK: br label %[[DIST_INNER_LOOP_HEADER:.+]] + + // CHECK: [[DIST_INNER_LOOP_HEADER]]: + // CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}} [[OMP_IV]], + // CHECK-DAG: [[OMP_UB_VAL_4:%.+]] = load {{.+}} [[OMP_UB]], + // CHECK: [[CMP_IV_UB_2:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_2]], [[OMP_UB_VAL_4]] + // CHECK: br{{.+}} [[CMP_IV_UB_2]], label %[[DIST_INNER_LOOP_BODY:.+]], label %[[DIST_INNER_LOOP_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // CHECK: [[DIST_INNER_LOOP_BODY]]: + // CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], + // CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], + // CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}} + // check that distlb and distub are properly passed to fork_call + // CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) + // CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) + // CHECK: br label %[[DIST_INNER_LOOP_INC:.+]] + + // check DistInc + // CHECK: [[DIST_INNER_LOOP_INC]]: + // CHECK-DAG: [[OMP_IV_VAL_3:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], + // CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], + // CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_3]], [[OMP_ST_VAL_1]] + // CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], + // CHECK: br label %[[DIST_INNER_LOOP_HEADER]] + + // CHECK: [[DIST_INNER_LOOP_END]]: + // CHECK: br label %[[DIST_OUTER_LOOP_INC:.+]] + + // CHECK: [[DIST_OUTER_LOOP_INC]]: + // check NextLB and NextUB + // CHECK-DAG: [[OMP_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_LB]], + // CHECK-DAG: [[OMP_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]], + // CHECK-DAG: [[OMP_LB_NEXT:%.+]] = add{{.+}} [[OMP_LB_VAL_2]], [[OMP_ST_VAL_2]] + // CHECK: store{{.+}} [[OMP_LB_NEXT]], {{.+}}* [[OMP_LB]], + // CHECK-DAG: [[OMP_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_UB]], + // CHECK-DAG: [[OMP_ST_VAL_3:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]], + // CHECK-DAG: [[OMP_UB_NEXT:%.+]] = add{{.+}} [[OMP_UB_VAL_5]], [[OMP_ST_VAL_3]] + // CHECK: store{{.+}} [[OMP_UB_NEXT]], {{.+}}* [[OMP_UB]], + // CHECK: br label %[[DIST_OUTER_LOOP_HEADER]] + + // outer loop exit + // CHECK: [[DIST_OUTER_LOOP_END]]: + // CHECK-DAG: call void @__kmpc_for_static_fini( + // CHECK: ret + + // skip implementation of 'parallel for': using default scheduling and was tested above + } + + // schedule: static no chunk + #pragma omp target + #pragma omp teams + // CHECK: define{{.+}} void [[OFFLOADING_FUN_4]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_4:@.+]] to {{.+}}) + + #pragma omp distribute parallel for schedule(static) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // CHECK: define{{.+}} void [[OMP_OUTLINED_4]]( + // CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca + // CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca + // CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca + // CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca + + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + // CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_4:@.+]] to {{.+}}, + // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case + // CHECK: ret + + // 'parallel for' implementation is the same as the case without schedule clase (static no chunk is the default) + // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_4]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + + // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + + // PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used + // In this case we use EUB + // CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], + // CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}}, + // CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] + // CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] + // CHECK: [[PF_EUB_TRUE]]: + // CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}}, + // CHECK: br label %[[PF_EUB_END:.+]] + // CHECK-DAG: [[PF_EUB_FALSE]]: + // CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], + // CHECK: br label %[[PF_EUB_END]] + // CHECK-DAG: [[PF_EUB_END]]: + // CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] + // CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + + // initialize omp.iv + // CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // CHECK: br label %[[OMP_PF_JUMP_BACK:.+]] + + // check exit condition + // CHECK: [[OMP_PF_JUMP_BACK]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], + // CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], + // CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] + // CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + + // check that PrevLB and PrevUB are passed to the 'for' + // CHECK: [[PF_BODY]]: + // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK: br label {{.+}} + + // check stride 1 for 'for' in 'distribute parallel for' + // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 + // CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], + // CHECK: br label %[[OMP_PF_JUMP_BACK]] + + // CHECK-DAG: call void @__kmpc_for_static_fini( + // CHECK: ret + } + + // schedule: static chunk + #pragma omp target + #pragma omp teams + // CHECK: define{{.+}} void [[OFFLOADING_FUN_5]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_5:@.+]] to {{.+}}) + + #pragma omp distribute parallel for schedule(static, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // CHECK: define{{.+}} void [[OMP_OUTLINED_5]]( + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + // CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_5:@.+]] to {{.+}}, + // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case + // CHECK: ret + + // 'parallel for' implementation using outer and inner loops and PrevEUB + // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_5]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}}) + // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 33, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + + // check PrevEUB (using PrevUB instead of NumIt as upper bound) + // CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]: + // CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], + // CHECK-64-DAG: [[OMP_PF_UB_VAL_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_1]] to + // CHECK: [[PF_PREV_UB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // CHECK-64-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_CONV]], [[PF_PREV_UB_VAL_1]] + // CHECK-32-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_PREV_UB_VAL_1]] + // CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] + // CHECK: [[PF_EUB_TRUE]]: + // CHECK: [[PF_PREV_UB_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // CHECK: br label %[[PF_EUB_END:.+]] + // CHECK-DAG: [[PF_EUB_FALSE]]: + // CHECK: [[OMP_PF_UB_VAL_2:%.+]] = load{{.+}} [[OMP_PF_UB]], + // CHECK-64: [[OMP_PF_UB_VAL_2_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_2]] to + // CHECK: br label %[[PF_EUB_END]] + // CHECK-DAG: [[PF_EUB_END]]: + // CHECK-64-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2_CONV]], %[[PF_EUB_FALSE]] ] + // CHECK-32-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2]], %[[PF_EUB_FALSE]] ] + // CHECK-64-DAG: [[PF_EUB_RES_CONV:%.+]] = trunc{{.+}} [[PF_EUB_RES]] to + // CHECK-64: store{{.+}} [[PF_EUB_RES_CONV]],{{.+}} [[OMP_PF_UB]], + // CHECK-32: store{{.+}} [[PF_EUB_RES]], {{.+}} [[OMP_PF_UB]], + + // initialize omp.iv (IV = LB) + // CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + + // outer loop: while (IV < UB) { + // CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], + // CHECK: [[PF_CMP_IV_UB_1:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] + // CHECK: br{{.+}} [[PF_CMP_IV_UB_1]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + + // CHECK: [[OMP_PF_OUTER_LOOP_BODY]]: + // CHECK: br label %[[OMP_PF_INNER_FOR_HEADER:.+]] + + // CHECK: [[OMP_PF_INNER_FOR_HEADER]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], + // CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] + // CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + + // CHECK: [[OMP_PF_INNER_LOOP_BODY]]: + // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // skip body branch + // CHECK: br{{.+}} + // CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + + // IV = IV + 1 and inner loop latch + // CHECK: [[OMP_PF_INNER_LOOP_INC]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], + // CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 + // CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], + // CHECK: br label %[[OMP_PF_INNER_FOR_HEADER]] + + // check NextLB and NextUB + // CHECK: [[OMP_PF_INNER_LOOP_END]]: + // CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + + // CHECK: [[OMP_PF_OUTER_LOOP_INC]]: + // CHECK-DAG: [[OMP_PF_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK-DAG: [[OMP_PF_ST_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]], + // CHECK-DAG: [[OMP_PF_LB_NEXT:%.+]] = add{{.+}} [[OMP_PF_LB_VAL_2]], [[OMP_PF_ST_VAL_1]] + // CHECK: store{{.+}} [[OMP_PF_LB_NEXT]], {{.+}}* [[OMP_PF_LB]], + // CHECK-DAG: [[OMP_PF_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], + // CHECK-DAG: [[OMP_PF_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]], + // CHECK-DAG: [[OMP_PF_UB_NEXT:%.+]] = add{{.+}} [[OMP_PF_UB_VAL_5]], [[OMP_PF_ST_VAL_2]] + // CHECK: store{{.+}} [[OMP_PF_UB_NEXT]], {{.+}}* [[OMP_PF_UB]], + // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + + // CHECK: [[OMP_PF_OUTER_LOOP_END]]: + // CHECK-DAG: call void @__kmpc_for_static_fini( + // CHECK: ret + } + + // schedule: dynamic no chunk + #pragma omp target + #pragma omp teams + // CHECK: define{{.+}} void [[OFFLOADING_FUN_6]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_6:@.+]] to {{.+}}) + + #pragma omp distribute parallel for schedule(dynamic) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // CHECK: define{{.+}} void [[OMP_OUTLINED_6]]( + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + // CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_6:@.+]] to {{.+}}, + // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case + // CHECK: ret + + // 'parallel for' implementation using outer and inner loops and PrevEUB + // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_6]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // CHECK-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], + // CHECK: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}}) + // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + + // CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]: + // CHECK: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]]) + // CHECK: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0 + // CHECK: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + + // initialize omp.iv (IV = LB) + // CHECK: [[OMP_PF_OUTER_LOOP_BODY]]: + // CHECK-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]] + + // CHECK: [[OMP_PF_INNER_LOOP_HEADER]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], + // CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] + // CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + + // CHECK: [[OMP_PF_INNER_LOOP_BODY]]: + // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // skip body branch + // CHECK: br{{.+}} + // CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + + // IV = IV + 1 and inner loop latch + // CHECK: [[OMP_PF_INNER_LOOP_INC]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], + // CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 + // CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], + // CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER]] + + // check NextLB and NextUB + // CHECK: [[OMP_PF_INNER_LOOP_END]]: + // CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + + // CHECK: [[OMP_PF_OUTER_LOOP_INC]]: + // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + + // CHECK: [[OMP_PF_OUTER_LOOP_END]]: + // CHECK: ret + } + + // schedule: dynamic chunk + #pragma omp target + #pragma omp teams + // CHECK: define{{.+}} void [[OFFLOADING_FUN_7]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_7:@.+]] to {{.+}}) + + #pragma omp distribute parallel for schedule(dynamic, ch) + for (int i = 0; i < n; ++i) { + a[i] = b[i] + c[i]; + // CHECK: define{{.+}} void [[OMP_OUTLINED_7]]( + // CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + // CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_7:@.+]] to {{.+}}, + // skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case + // CHECK: ret + + // 'parallel for' implementation using outer and inner loops and PrevEUB + // CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_7]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}}) + // CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + // CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + + // initialize lb and ub to PrevLB and PrevUB + // CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], + // CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], + // CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} + // CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], + // CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], + // CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], + // CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], + // CHECK-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], + // CHECK: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}}) + // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + + // CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]: + // CHECK: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]]) + // CHECK: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0 + // CHECK: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + + // initialize omp.iv (IV = LB) + // CHECK: [[OMP_PF_OUTER_LOOP_BODY]]: + // CHECK-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], + // CHECK-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + // CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]] + + // CHECK: [[OMP_PF_INNER_LOOP_HEADER]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], + // CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] + // CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + + // CHECK: [[OMP_PF_INNER_LOOP_BODY]]: + // CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], + // skip body branch + // CHECK: br{{.+}} + // CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + + // IV = IV + 1 and inner loop latch + // CHECK: [[OMP_PF_INNER_LOOP_INC]]: + // CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], + // CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 + // CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], + // CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER]] + + // check NextLB and NextUB + // CHECK: [[OMP_PF_INNER_LOOP_END]]: + // CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + + // CHECK: [[OMP_PF_OUTER_LOOP_INC]]: + // CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + + // CHECK: [[OMP_PF_OUTER_LOOP_END]]: + // CHECK: ret + } + + return tmain<int>(); +#endif +} + +// check code +// CHECK: define{{.+}} [[TMAIN]]() + +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_1:@.+]]( + +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_2:@.+]]( + +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_3:@.+]]( + +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_4:@.+]]( + +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_5:@.+]]( + +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_6:@.+]]( + +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_7:@.+]]( + +// CHECK: define{{.+}} void [[OFFLOADING_FUN_1]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[OMP_OUTLINED_1]]( +// CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca +// CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca +// CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca +// CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca + +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + +// check EUB for distribute +// CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], +// CHECK: [[NUM_IT_1:%.+]] = load{{.+}}, +// CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] +// CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] +// CHECK-DAG: [[EUB_TRUE]]: +// CHECK: [[NUM_IT_2:%.+]] = load{{.+}}, +// CHECK: br label %[[EUB_END:.+]] +// CHECK-DAG: [[EUB_FALSE]]: +// CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], +// CHECK: br label %[[EUB_END]] +// CHECK-DAG: [[EUB_END]]: +// CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] +// CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + +// initialize omp.iv +// CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], +// CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], +// CHECK: br label %[[OMP_JUMP_BACK:.+]] + +// check exit condition +// CHECK: [[OMP_JUMP_BACK]]: +// CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], +// CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], +// CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] +// CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]] + +// check that PrevLB and PrevUB are passed to the 'for' +// CHECK: [[DIST_BODY]]: +// CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], +// CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], +// CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}} +// check that distlb and distub are properly passed to fork_call +// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) +// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) +// CHECK: br label %[[DIST_INC:.+]] + +// increment by stride (distInc - 'parallel for' executes the whole chunk) and latch +// CHECK: [[DIST_INC]]: +// CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], +// CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], +// CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]] +// CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], +// CHECK: br label %[[OMP_JUMP_BACK]] + +// CHECK-DAG: call void @__kmpc_for_static_fini( +// CHECK: ret + +// implementation of 'parallel for' +// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_1]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + +// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + +// initialize lb and ub to PrevLB and PrevUB +// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], +// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], +// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + +// PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used +// In this case we use EUB +// CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], +// CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}}, +// CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] +// CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] +// CHECK: [[PF_EUB_TRUE]]: +// CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}}, +// CHECK: br label %[[PF_EUB_END:.+]] +// CHECK-DAG: [[PF_EUB_FALSE]]: +// CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], +// CHECK: br label %[[PF_EUB_END]] +// CHECK-DAG: [[PF_EUB_END]]: +// CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] +// CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + +// initialize omp.iv +// CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], +// CHECK: br label %[[OMP_PF_JUMP_BACK:.+]] + +// check exit condition +// CHECK: [[OMP_PF_JUMP_BACK]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], +// CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], +// CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] +// CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + +// check that PrevLB and PrevUB are passed to the 'for' +// CHECK: [[PF_BODY]]: +// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK: br label {{.+}} + +// check stride 1 for 'for' in 'distribute parallel for' +// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 +// CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], +// CHECK: br label %[[OMP_PF_JUMP_BACK]] + +// CHECK-DAG: call void @__kmpc_for_static_fini( +// CHECK: ret + +// CHECK: define{{.+}} void [[OFFLOADING_FUN_2]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_2:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[OMP_OUTLINED_2]]( +// CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca +// CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca +// CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca +// CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca + +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, + +// check EUB for distribute +// CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], +// CHECK: [[NUM_IT_1:%.+]] = load{{.+}}, +// CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] +// CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] +// CHECK-DAG: [[EUB_TRUE]]: +// CHECK: [[NUM_IT_2:%.+]] = load{{.+}}, +// CHECK: br label %[[EUB_END:.+]] +// CHECK-DAG: [[EUB_FALSE]]: +// CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], +// CHECK: br label %[[EUB_END]] +// CHECK-DAG: [[EUB_END]]: +// CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] +// CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + +// initialize omp.iv +// CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], +// CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], +// CHECK: br label %[[OMP_JUMP_BACK:.+]] + +// check exit condition +// CHECK: [[OMP_JUMP_BACK]]: +// CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], +// CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], +// CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] +// CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_BODY:.+]], label %[[DIST_END:.+]] + +// check that PrevLB and PrevUB are passed to the 'for' +// CHECK: [[DIST_BODY]]: +// CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], +// CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], +// CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}} +// check that distlb and distub are properly passed to fork_call +// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) +// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_2:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) +// CHECK: br label %[[DIST_INC:.+]] + +// increment by stride (distInc - 'parallel for' executes the whole chunk) and latch +// CHECK: [[DIST_INC]]: +// CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], +// CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], +// CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_2]], [[OMP_ST_VAL_1]] +// CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], +// CHECK: br label %[[OMP_JUMP_BACK]] + +// CHECK-DAG: call void @__kmpc_for_static_fini( +// CHECK: ret + +// implementation of 'parallel for' +// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_2]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + +// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + +// initialize lb and ub to PrevLB and PrevUB +// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], +// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], +// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + +// PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used +// In this case we use EUB +// CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], +// CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}}, +// CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] +// CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] +// CHECK: [[PF_EUB_TRUE]]: +// CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}}, +// CHECK: br label %[[PF_EUB_END:.+]] +// CHECK-DAG: [[PF_EUB_FALSE]]: +// CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], +// CHECK: br label %[[PF_EUB_END]] +// CHECK-DAG: [[PF_EUB_END]]: +// CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] +// CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + +// initialize omp.iv +// CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], +// CHECK: br label %[[OMP_PF_JUMP_BACK:.+]] + +// check exit condition +// CHECK: [[OMP_PF_JUMP_BACK]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], +// CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], +// CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] +// CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + +// check that PrevLB and PrevUB are passed to the 'for' +// CHECK: [[PF_BODY]]: +// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK: br label {{.+}} + +// check stride 1 for 'for' in 'distribute parallel for' +// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 +// CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], +// CHECK: br label %[[OMP_PF_JUMP_BACK]] + +// CHECK-DAG: call void @__kmpc_for_static_fini( +// CHECK: ret + +// CHECK: define{{.+}} void [[OFFLOADING_FUN_3]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_3:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[OMP_OUTLINED_3]]( +// CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca +// CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca +// CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca +// CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca + +// unlike the previous tests, in this one we have a outer and inner loop for 'distribute' +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 91, +// CHECK: br label %[[DIST_OUTER_LOOP_HEADER:.+]] + +// CHECK: [[DIST_OUTER_LOOP_HEADER]]: +// check EUB for distribute +// CHECK-DAG: [[OMP_UB_VAL_1:%.+]] = load{{.+}} [[OMP_UB]], +// CHECK: [[NUM_IT_1:%.+]] = load{{.+}}, +// CHECK-DAG: [[CMP_UB_NUM_IT:%.+]] = icmp sgt {{.+}} [[OMP_UB_VAL_1]], [[NUM_IT_1]] +// CHECK: br {{.+}} [[CMP_UB_NUM_IT]], label %[[EUB_TRUE:.+]], label %[[EUB_FALSE:.+]] +// CHECK-DAG: [[EUB_TRUE]]: +// CHECK: [[NUM_IT_2:%.+]] = load{{.+}}, +// CHECK: br label %[[EUB_END:.+]] +// CHECK-DAG: [[EUB_FALSE]]: +// CHECK: [[OMP_UB_VAL2:%.+]] = load{{.+}} [[OMP_UB]], +// CHECK: br label %[[EUB_END]] +// CHECK-DAG: [[EUB_END]]: +// CHECK-DAG: [[EUB_RES:%.+]] = phi{{.+}} [ [[NUM_IT_2]], %[[EUB_TRUE]] ], [ [[OMP_UB_VAL2]], %[[EUB_FALSE]] ] +// CHECK: store{{.+}} [[EUB_RES]], {{.+}}* [[OMP_UB]], + +// initialize omp.iv +// CHECK: [[OMP_LB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_LB]], +// CHECK: store {{.+}} [[OMP_LB_VAL_1]], {{.+}}* [[OMP_IV]], + +// check exit condition +// CHECK-DAG: [[OMP_IV_VAL_1:%.+]] = load {{.+}} [[OMP_IV]], +// CHECK-DAG: [[OMP_UB_VAL_3:%.+]] = load {{.+}} [[OMP_UB]], +// CHECK: [[CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_1]], [[OMP_UB_VAL_3]] +// CHECK: br {{.+}} [[CMP_IV_UB]], label %[[DIST_OUTER_LOOP_BODY:.+]], label %[[DIST_OUTER_LOOP_END:.+]] + +// CHECK: [[DIST_OUTER_LOOP_BODY]]: +// CHECK: br label %[[DIST_INNER_LOOP_HEADER:.+]] + +// CHECK: [[DIST_INNER_LOOP_HEADER]]: +// CHECK-DAG: [[OMP_IV_VAL_2:%.+]] = load {{.+}} [[OMP_IV]], +// CHECK-DAG: [[OMP_UB_VAL_4:%.+]] = load {{.+}} [[OMP_UB]], +// CHECK: [[CMP_IV_UB_2:%.+]] = icmp sle {{.+}} [[OMP_IV_VAL_2]], [[OMP_UB_VAL_4]] +// CHECK: br{{.+}} [[CMP_IV_UB_2]], label %[[DIST_INNER_LOOP_BODY:.+]], label %[[DIST_INNER_LOOP_END:.+]] + +// check that PrevLB and PrevUB are passed to the 'for' +// CHECK: [[DIST_INNER_LOOP_BODY]]: +// CHECK-DAG: [[OMP_PREV_LB:%.+]] = load {{.+}}, {{.+}} [[OMP_LB]], +// CHECK-64-DAG: [[OMP_PREV_LB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_LB]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB:%.+]] = load {{.+}}, {{.+}} [[OMP_UB]], +// CHECK-64-DAG: [[OMP_PREV_UB_EXT:%.+]] = zext {{.+}} [[OMP_PREV_UB]] to {{.+}} +// check that distlb and distub are properly passed to fork_call +// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_EXT]], i{{[0-9]+}} [[OMP_PREV_UB_EXT]], {{.+}}) +// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_3:@.+]] to {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB]], i{{[0-9]+}} [[OMP_PREV_UB]], {{.+}}) +// CHECK: br label %[[DIST_INNER_LOOP_INC:.+]] + +// check DistInc +// CHECK: [[DIST_INNER_LOOP_INC]]: +// CHECK-DAG: [[OMP_IV_VAL_3:%.+]] = load {{.+}}, {{.+}}* [[OMP_IV]], +// CHECK-DAG: [[OMP_ST_VAL_1:%.+]] = load {{.+}}, {{.+}}* [[OMP_ST]], +// CHECK: [[OMP_IV_INC:%.+]] = add{{.+}} [[OMP_IV_VAL_3]], [[OMP_ST_VAL_1]] +// CHECK: store{{.+}} [[OMP_IV_INC]], {{.+}}* [[OMP_IV]], +// CHECK: br label %[[DIST_INNER_LOOP_HEADER]] + +// CHECK: [[DIST_INNER_LOOP_END]]: +// CHECK: br label %[[DIST_OUTER_LOOP_INC:.+]] + +// CHECK: [[DIST_OUTER_LOOP_INC]]: +// check NextLB and NextUB +// CHECK-DAG: [[OMP_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_LB]], +// CHECK-DAG: [[OMP_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]], +// CHECK-DAG: [[OMP_LB_NEXT:%.+]] = add{{.+}} [[OMP_LB_VAL_2]], [[OMP_ST_VAL_2]] +// CHECK: store{{.+}} [[OMP_LB_NEXT]], {{.+}}* [[OMP_LB]], +// CHECK-DAG: [[OMP_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_UB]], +// CHECK-DAG: [[OMP_ST_VAL_3:%.+]] = load{{.+}}, {{.+}} [[OMP_ST]], +// CHECK-DAG: [[OMP_UB_NEXT:%.+]] = add{{.+}} [[OMP_UB_VAL_5]], [[OMP_ST_VAL_3]] +// CHECK: store{{.+}} [[OMP_UB_NEXT]], {{.+}}* [[OMP_UB]], +// CHECK: br label %[[DIST_OUTER_LOOP_HEADER]] + +// outer loop exit +// CHECK: [[DIST_OUTER_LOOP_END]]: +// CHECK-DAG: call void @__kmpc_for_static_fini( +// CHECK: ret + +// skip implementation of 'parallel for': using default scheduling and was tested above + +// CHECK: define{{.+}} void [[OFFLOADING_FUN_4]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_4:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[OMP_OUTLINED_4]]( +// CHECK-DAG: [[OMP_IV:%.omp.iv]] = alloca +// CHECK-DAG: [[OMP_LB:%.omp.comb.lb]] = alloca +// CHECK-DAG: [[OMP_UB:%.omp.comb.ub]] = alloca +// CHECK-DAG: [[OMP_ST:%.omp.stride]] = alloca + +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, +// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_4:@.+]] to {{.+}}, +// skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case +// CHECK: ret + +// 'parallel for' implementation is the same as the case without schedule clase (static no chunk is the default) +// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_4]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) + +// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, + +// initialize lb and ub to PrevLB and PrevUB +// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], +// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], +// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 34, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) + +// PrevEUB is only used when 'for' has a chunked schedule, otherwise EUB is used +// In this case we use EUB +// CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], +// CHECK: [[PF_NUM_IT_1:%.+]] = load{{.+}}, +// CHECK-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_NUM_IT_1]] +// CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] +// CHECK: [[PF_EUB_TRUE]]: +// CHECK: [[PF_NUM_IT_2:%.+]] = load{{.+}}, +// CHECK: br label %[[PF_EUB_END:.+]] +// CHECK-DAG: [[PF_EUB_FALSE]]: +// CHECK: [[OMP_PF_UB_VAL2:%.+]] = load{{.+}} [[OMP_PF_UB]], +// CHECK: br label %[[PF_EUB_END]] +// CHECK-DAG: [[PF_EUB_END]]: +// CHECK-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_NUM_IT_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL2]], %[[PF_EUB_FALSE]] ] +// CHECK: store{{.+}} [[PF_EUB_RES]],{{.+}} [[OMP_PF_UB]], + +// initialize omp.iv +// CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], +// CHECK: br label %[[OMP_PF_JUMP_BACK:.+]] + +// check exit condition +// CHECK: [[OMP_PF_JUMP_BACK]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load {{.+}} [[OMP_PF_IV]], +// CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load {{.+}} [[OMP_PF_UB]], +// CHECK: [[PF_CMP_IV_UB:%.+]] = icmp sle {{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] +// CHECK: br {{.+}} [[PF_CMP_IV_UB]], label %[[PF_BODY:.+]], label %[[PF_END:.+]] + +// check that PrevLB and PrevUB are passed to the 'for' +// CHECK: [[PF_BODY]]: +// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK: br label {{.+}} + +// check stride 1 for 'for' in 'distribute parallel for' +// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load {{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK: [[OMP_PF_IV_INC:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_2]], 1 +// CHECK: store{{.+}} [[OMP_PF_IV_INC]], {{.+}}* [[OMP_PF_IV]], +// CHECK: br label %[[OMP_PF_JUMP_BACK]] + +// CHECK-DAG: call void @__kmpc_for_static_fini( +// CHECK: ret + +// CHECK: define{{.+}} void [[OFFLOADING_FUN_5]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_5:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[OMP_OUTLINED_5]]( +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, +// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_5:@.+]] to {{.+}}, +// skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case +// CHECK: ret + +// 'parallel for' implementation using outer and inner loops and PrevEUB +// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_5]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}}) +// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + +// initialize lb and ub to PrevLB and PrevUB +// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], +// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], +// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, {{.+}} 33, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]],{{.+}}) +// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + +// check PrevEUB (using PrevUB instead of NumIt as upper bound) +// CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]: +// CHECK-DAG: [[OMP_PF_UB_VAL_1:%.+]] = load{{.+}} [[OMP_PF_UB]], +// CHECK-64-DAG: [[OMP_PF_UB_VAL_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_1]] to +// CHECK: [[PF_PREV_UB_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], +// CHECK-64-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_CONV]], [[PF_PREV_UB_VAL_1]] +// CHECK-32-DAG: [[PF_CMP_UB_NUM_IT:%.+]] = icmp{{.+}} [[OMP_PF_UB_VAL_1]], [[PF_PREV_UB_VAL_1]] +// CHECK: br i1 [[PF_CMP_UB_NUM_IT]], label %[[PF_EUB_TRUE:.+]], label %[[PF_EUB_FALSE:.+]] +// CHECK: [[PF_EUB_TRUE]]: +// CHECK: [[PF_PREV_UB_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], +// CHECK: br label %[[PF_EUB_END:.+]] +// CHECK-DAG: [[PF_EUB_FALSE]]: +// CHECK: [[OMP_PF_UB_VAL_2:%.+]] = load{{.+}} [[OMP_PF_UB]], +// CHECK-64: [[OMP_PF_UB_VAL_2_CONV:%.+]] = sext{{.+}} [[OMP_PF_UB_VAL_2]] to +// CHECK: br label %[[PF_EUB_END]] +// CHECK-DAG: [[PF_EUB_END]]: +// CHECK-64-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2_CONV]], %[[PF_EUB_FALSE]] ] +// CHECK-32-DAG: [[PF_EUB_RES:%.+]] = phi{{.+}} [ [[PF_PREV_UB_VAL_2]], %[[PF_EUB_TRUE]] ], [ [[OMP_PF_UB_VAL_2]], %[[PF_EUB_FALSE]] ] +// CHECK-64-DAG: [[PF_EUB_RES_CONV:%.+]] = trunc{{.+}} [[PF_EUB_RES]] to +// CHECK-64: store{{.+}} [[PF_EUB_RES_CONV]],{{.+}} [[OMP_PF_UB]], +// CHECK-32: store{{.+}} [[PF_EUB_RES]], {{.+}} [[OMP_PF_UB]], + +// initialize omp.iv (IV = LB) +// CHECK: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], + +// outer loop: while (IV < UB) { +// CHECK-DAG: [[OMP_PF_IV_VAL_1:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK-DAG: [[OMP_PF_UB_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], +// CHECK: [[PF_CMP_IV_UB_1:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_1]], [[OMP_PF_UB_VAL_3]] +// CHECK: br{{.+}} [[PF_CMP_IV_UB_1]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + +// CHECK: [[OMP_PF_OUTER_LOOP_BODY]]: +// CHECK: br label %[[OMP_PF_INNER_FOR_HEADER:.+]] + +// CHECK: [[OMP_PF_INNER_FOR_HEADER]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], +// CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] +// CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + +// CHECK: [[OMP_PF_INNER_LOOP_BODY]]: +// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// skip body branch +// CHECK: br{{.+}} +// CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + +// IV = IV + 1 and inner loop latch +// CHECK: [[OMP_PF_INNER_LOOP_INC]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], +// CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 +// CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], +// CHECK: br label %[[OMP_PF_INNER_FOR_HEADER]] + +// check NextLB and NextUB +// CHECK: [[OMP_PF_INNER_LOOP_END]]: +// CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + +// CHECK: [[OMP_PF_OUTER_LOOP_INC]]: +// CHECK-DAG: [[OMP_PF_LB_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK-DAG: [[OMP_PF_ST_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]], +// CHECK-DAG: [[OMP_PF_LB_NEXT:%.+]] = add{{.+}} [[OMP_PF_LB_VAL_2]], [[OMP_PF_ST_VAL_1]] +// CHECK: store{{.+}} [[OMP_PF_LB_NEXT]], {{.+}}* [[OMP_PF_LB]], +// CHECK-DAG: [[OMP_PF_UB_VAL_5:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], +// CHECK-DAG: [[OMP_PF_ST_VAL_2:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_ST]], +// CHECK-DAG: [[OMP_PF_UB_NEXT:%.+]] = add{{.+}} [[OMP_PF_UB_VAL_5]], [[OMP_PF_ST_VAL_2]] +// CHECK: store{{.+}} [[OMP_PF_UB_NEXT]], {{.+}}* [[OMP_PF_UB]], +// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + +// CHECK: [[OMP_PF_OUTER_LOOP_END]]: +// CHECK-DAG: call void @__kmpc_for_static_fini( +// CHECK: ret + +// CHECK: define{{.+}} void [[OFFLOADING_FUN_6]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED_6:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[OMP_OUTLINED_6]]( +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, +// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_6:@.+]] to {{.+}}, +// skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case +// CHECK: ret + +// 'parallel for' implementation using outer and inner loops and PrevEUB +// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_6]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}) +// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + +// initialize lb and ub to PrevLB and PrevUB +// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], +// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], +// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], +// CHECK-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], +// CHECK: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}}) +// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + +// CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]: +// CHECK: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]]) +// CHECK: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0 +// CHECK: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + +// initialize omp.iv (IV = LB) +// CHECK: [[OMP_PF_OUTER_LOOP_BODY]]: +// CHECK-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], +// CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]] + +// CHECK: [[OMP_PF_INNER_LOOP_HEADER]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], +// CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] +// CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + +// CHECK: [[OMP_PF_INNER_LOOP_BODY]]: +// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// skip body branch +// CHECK: br{{.+}} +// CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + +// IV = IV + 1 and inner loop latch +// CHECK: [[OMP_PF_INNER_LOOP_INC]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], +// CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 +// CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], +// CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER]] + +// check NextLB and NextUB +// CHECK: [[OMP_PF_INNER_LOOP_END]]: +// CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + +// CHECK: [[OMP_PF_OUTER_LOOP_INC]]: +// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + +// CHECK: [[OMP_PF_OUTER_LOOP_END]]: +// CHECK: ret + +// CHECK: define{{.+}} void [[OFFLOADING_FUN_7]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 5, {{.+}}* [[OMP_OUTLINED_7:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[OMP_OUTLINED_7]]( +// CHECK: call void @__kmpc_for_static_init_4({{.+}}, {{.+}}, i32 92, +// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_7:@.+]] to {{.+}}, +// skip rest of implementation of 'distribute' as it is tested above for default dist_schedule case +// CHECK: ret + +// 'parallel for' implementation using outer and inner loops and PrevEUB +// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_7]]({{.+}}, {{.+}}, i{{[0-9]+}} [[OMP_PREV_LB_IN:%.+]], i{{[0-9]+}} [[OMP_PREV_UB_IN:%.+]], {{.+}}, {{.+}}, {{.+}}, {{.+}}, {{.+}}) +// CHECK-DAG: [[OMP_PF_LB:%.omp.lb]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_UB:%.omp.ub]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_IV:%.omp.iv]] = alloca{{.+}}, +// CHECK-DAG: [[OMP_PF_ST:%.omp.stride]] = alloca{{.+}}, + +// initialize lb and ub to PrevLB and PrevUB +// CHECK-DAG: store{{.+}} [[OMP_PREV_LB_IN]], {{.+}}* [[PREV_LB_ADDR:%.+]], +// CHECK-DAG: store{{.+}} [[OMP_PREV_UB_IN]], {{.+}}* [[PREV_UB_ADDR:%.+]], +// CHECK-DAG: [[OMP_PREV_LB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_LB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_LB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_LB_VAL]] to {{.+}} +// CHECK-DAG: [[OMP_PREV_UB_VAL:%.+]] = load{{.+}}, {{.+}}* [[PREV_UB_ADDR]], +// CHECK-64-DAG: [[OMP_PREV_UB_TRC:%.+]] = trunc{{.+}} [[OMP_PREV_UB_VAL]] to {{.+}} +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_LB_TRC]], {{.+}}* [[OMP_PF_LB]], +// CHECK-64-DAG: store{{.+}} [[OMP_PREV_UB_TRC]], {{.+}}* [[OMP_PF_UB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_LB_VAL]], {{.+}}* [[OMP_PF_LB]], +// CHECK-32-DAG: store{{.+}} [[OMP_PREV_UB_VAL]], {{.+}}* [[OMP_PF_UB]], +// CHECK-DAG: [[OMP_PF_LB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK-DAG: [[OMP_PF_UB_VAL:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_UB]], +// CHECK: call void @__kmpc_dispatch_init_4({{.+}}, {{.+}}, {{.+}} 35, {{.+}} [[OMP_PF_LB_VAL]], {{.+}} [[OMP_PF_UB_VAL]], {{.+}}, {{.+}}) +// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER:.+]] + +// CHECK: [[OMP_PF_OUTER_LOOP_HEADER]]: +// CHECK: [[IS_FIN:%.+]] = call{{.+}} @__kmpc_dispatch_next_4({{.+}}, {{.+}}, {{.+}}, {{.+}}* [[OMP_PF_LB]], {{.+}}* [[OMP_PF_UB]], {{.+}}* [[OMP_PF_ST]]) +// CHECK: [[IS_FIN_CMP:%.+]] = icmp{{.+}} [[IS_FIN]], 0 +// CHECK: br{{.+}} [[IS_FIN_CMP]], label %[[OMP_PF_OUTER_LOOP_BODY:.+]], label %[[OMP_PF_OUTER_LOOP_END:.+]] + +// initialize omp.iv (IV = LB) +// CHECK: [[OMP_PF_OUTER_LOOP_BODY]]: +// CHECK-DAG: [[OMP_PF_LB_VAL_1:%.+]] = load{{.+}}, {{.+}} [[OMP_PF_LB]], +// CHECK-DAG: store {{.+}} [[OMP_PF_LB_VAL_1]], {{.+}}* [[OMP_PF_IV]], +// CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER:.+]] + +// CHECK: [[OMP_PF_INNER_LOOP_HEADER]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_2:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// CHECK-DAG: [[OMP_PF_UB_VAL_4:%.+]] = load{{.+}}, {{.+}}* [[OMP_PF_UB]], +// CHECK: [[PF_CMP_IV_UB_2:%.+]] = icmp{{.+}} [[OMP_PF_IV_VAL_2]], [[OMP_PF_UB_VAL_4]] +// CHECK: br{{.+}} [[PF_CMP_IV_UB_2]], label %[[OMP_PF_INNER_LOOP_BODY:.+]], label %[[OMP_PF_INNER_LOOP_END:.+]] + +// CHECK: [[OMP_PF_INNER_LOOP_BODY]]: +// CHECK-DAG: {{.+}} = load{{.+}}, {{.+}}* [[OMP_PF_IV]], +// skip body branch +// CHECK: br{{.+}} +// CHECK: br label %[[OMP_PF_INNER_LOOP_INC:.+]] + +// IV = IV + 1 and inner loop latch +// CHECK: [[OMP_PF_INNER_LOOP_INC]]: +// CHECK-DAG: [[OMP_PF_IV_VAL_3:%.+]] = load{{.+}}, {{.+}}* [[OMP_IV]], +// CHECK-DAG: [[OMP_PF_NEXT_IV:%.+]] = add{{.+}} [[OMP_PF_IV_VAL_3]], 1 +// CHECK-DAG: store{{.+}} [[OMP_PF_NEXT_IV]], {{.+}}* [[OMP_IV]], +// CHECK: br label %[[OMP_PF_INNER_LOOP_HEADER]] + +// check NextLB and NextUB +// CHECK: [[OMP_PF_INNER_LOOP_END]]: +// CHECK: br label %[[OMP_PF_OUTER_LOOP_INC:.+]] + +// CHECK: [[OMP_PF_OUTER_LOOP_INC]]: +// CHECK: br label %[[OMP_PF_OUTER_LOOP_HEADER]] + +// CHECK: [[OMP_PF_OUTER_LOOP_END]]: +// CHECK: ret +#endif diff --git a/test/OpenMP/distribute_parallel_for_firstprivate_codegen.cpp b/test/OpenMP/distribute_parallel_for_firstprivate_codegen.cpp new file mode 100644 index 0000000000000..d12c0aa11cf26 --- /dev/null +++ b/test/OpenMP/distribute_parallel_for_firstprivate_codegen.cpp @@ -0,0 +1,619 @@ +// RxUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64 +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64 +// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32 +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32 + +// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64 +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64 +// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32 +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32 +// expected-no-diagnostics +#ifndef HEADER +#define HEADER + +template <class T> +struct S { + T f; + S(T a) : f(a) {} + S() : f() {} + operator T() { return T(); } + ~S() {} +}; + +// CHECK: [[S_FLOAT_TY:%.+]] = type { float } +// CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} } +template <typename T> +T tmain() { + S<T> test; + T t_var = T(); + T vec[] = {1, 2}; + S<T> s_arr[] = {1, 2}; + S<T> &var = test; + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for firstprivate(t_var, vec, s_arr, s_arr, var, var) + for (int i = 0; i < 2; ++i) { + vec[i] = t_var; + s_arr[i] = var; + } + return T(); +} + +int main() { + static int svar; + volatile double g; + volatile double &g1 = g; + + #ifdef LAMBDA + // LAMBDA-LABEL: @main + // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]]( + [&]() { + static float sfvar; + // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]]( + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN:@.+]]( + + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED:@.+]] to {{.+}}) + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for firstprivate(g, g1, svar, sfvar) + for (int i = 0; i < 2; ++i) { + // LAMBDA-64: define{{.*}} internal{{.*}} void [[OMP_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i{{[0-9]+}} [[G_IN:%.+]], i{{[0-9]+}} [[G1_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]], i{{[0-9]+}} [[SFVAR_IN:%.+]]) + // LAMBDA-32: define{{.*}} internal{{.*}} void [[OMP_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, double* {{.+}} [[G_IN:%.+]], i{{[0-9]+}} [[G1_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]], i{{[0-9]+}} [[SFVAR_IN:%.+]]) + + // addr alloca's + // LAMBDA-64: [[G_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA-32: [[G_ADDR:%.+]] = alloca double*, + // LAMBDA: [[G1_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SFVAR_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[G1_REF:%.+]] = alloca double*, + // LAMBDA: [[TMP:%.+]] = alloca double*, + + // private alloca's + // LAMBDA: [[G_PRIV:%.+]] = alloca double, + // LAMBDA: [[G1_PRIV:%.+]] = alloca double, + // LAMBDA: [[TMP_PRIV:%.+]] = alloca double*, + // LAMBDA: [[SVAR_PRIV:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SFVAR_PRIV:%.+]] = alloca float, + + // transfer input parameters into addr alloca's + // LAMBDA-DAG: store {{.+}} [[G_IN]], {{.+}} [[G_ADDR]], + // LAMBDA-DAG: store {{.+}} [[G1_IN]], {{.+}} [[G1_ADDR]], + // LAMBDA-DAG: store {{.+}} [[SVAR_IN]], {{.+}} [[SVAR_ADDR]], + // LAMBDA-DAG: store {{.+}} [[SFVAR_IN]], {{.+}} [[SFVAR_ADDR]], + + // init private alloca's with addr alloca's + // g + // LAMBDA-64-DAG: [[G_CONV:%.+]] = bitcast {{.+}}* [[G_ADDR]] to + // LAMBDA-32-DAG: [[G_CONV:%.+]] = load {{.+}}*, {{.+}}** [[G_ADDR]] + // LAMBDA-DAG: [[G_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[G_CONV]], + // LAMBDA-DAG: store {{.+}} [[G_ADDR_VAL]], {{.+}}* [[G_PRIV]], + + // g1 + // LAMBDA-DAG: [[G1_CONV:%.+]] = bitcast {{.+}}* [[G1_ADDR]] to + // LAMBDA-DAG: store {{.+}}* [[G1_CONV]], {{.+}}** [[G1_REF]], + // LAMBDA-DAG: [[G1_REF_VAL:%.+]] = load {{.+}}*, {{.+}}** [[G1_REF]], + // LAMBDA-DAG: store {{.+}}* [[G1_REF_VAL]], {{.+}}** [[TMP]], + // LAMBDA-DAG: [[TMP_REF:%.+]] = load {{.+}}*, {{.+}}** [[TMP]], + // LAMBDA-DAG: [[TMP_VAL:%.+]] = load {{.+}}, {{.+}}* [[TMP_REF]], + // LAMBDA-DAG: store {{.+}} [[TMP_VAL]], {{.+}}* [[G1_PRIV]] + // LAMBDA-DAG: store {{.+}}* [[G1_PRIV]], {{.+}}** [[TMP_PRIV]], + + // svar + // LAMBDA-64-DAG: [[SVAR_CONV:%.+]] = bitcast {{.+}}* [[SVAR_ADDR]] to + // LAMBDA-64-DAG: [[SVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_CONV]], + // LAMBDA-32-DAG: [[SVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_ADDR]], + // LAMBDA-DAG: store {{.+}} [[SVAR_VAL]], {{.+}}* [[SVAR_PRIV]], + + // sfvar + // LAMBDA-DAG: [[SFVAR_CONV:%.+]] = bitcast {{.+}}* [[SFVAR_ADDR]] to + // LAMBDA-DAG: [[SFVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SFVAR_CONV]], + // LAMBDA-DAG: store {{.+}} [[SFVAR_VAL]], {{.+}}* [[SFVAR_PRIV]], + + // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4( + // pass firstprivate parameters to parallel outlined function + // g + // LAMBDA-64-DAG: [[G_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[G_PRIV]], + // LAMBDA-64: [[G_CAST_CONV:%.+]] = bitcast {{.+}}* [[G_CAST:%.+]] to + // LAMBDA-64-DAG: store {{.+}} [[G_PRIV_VAL]], {{.+}}* [[G_CAST_CONV]], + // LAMBDA-64-DAG: [[G_PAR:%.+]] = load {{.+}}, {{.+}}* [[G_CAST]], + + // g1 + // LAMBDA-DAG: [[TMP_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[TMP_PRIV]], + // LAMBDA-DAG: [[G1_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[TMP_PRIV_VAL]], + // LAMBDA: [[G1_CAST_CONV:%.+]] = bitcast {{.+}}* [[G1_CAST:%.+]] to + // LAMBDA-DAG: store {{.+}} [[G1_PRIV_VAL]], {{.+}}* [[G1_CAST_CONV]], + // LAMBDA-DAG: [[G1_PAR:%.+]] = load {{.+}}, {{.+}}* [[G1_CAST]], + + // svar + // LAMBDA: [[SVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_PRIV]], + // LAMBDA-64-DAG: [[SVAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[SVAR_CAST:%.+]] to + // LAMBDA-64-DAG: store {{.+}} [[SVAR_VAL]], {{.+}}* [[SVAR_CAST_CONV]], + // LAMBDA-32-DAG: store {{.+}} [[SVAR_VAL]], {{.+}}* [[SVAR_CAST:%.+]], + // LAMBDA-DAG: [[SVAR_PAR:%.+]] = load {{.+}}, {{.+}}* [[SVAR_CAST]], + + // sfvar + // LAMBDA: [[SFVAR_VAL:%.+]] = load {{.+}}, {{.+}}* [[SFVAR_PRIV]], + // LAMBDA-DAG: [[SFVAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[SFVAR_CAST:%.+]] to + // LAMBDA-DAG: store {{.+}} [[SFVAR_VAL]], {{.+}}* [[SFVAR_CAST_CONV]], + // LAMBDA-DAG: [[SFVAR_PAR:%.+]] = load {{.+}}, {{.+}}* [[SFVAR_CAST]], + + // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[G_PAR]], {{.+}} [[G1_PAR]], {{.+}} [[SVAR_PAR]], {{.+}} [[SFVAR_PAR]]) + // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[G_PRIV]], {{.+}} [[G1_PAR]], {{.+}} [[SVAR_PAR]], {{.+}} [[SFVAR_PAR]]) + // LAMBDA: call {{.*}}void @__kmpc_for_static_fini( + // LAMBDA: ret void + + + // LAMBDA-64: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, i{{[0-9]+}} [[G_IN:%.+]], i{{[0-9]+}} [[G1_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]], i{{[0-9]+}} [[SFVAR_IN:%.+]]) + // LAMBDA-32: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, double* {{.+}} [[G_IN:%.+]], i{{[0-9]+}} [[G1_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]], i{{[0-9]+}} [[SFVAR_IN:%.+]]) + // skip initial params + // LAMBDA: {{.+}} = alloca{{.+}}, + // LAMBDA: {{.+}} = alloca{{.+}}, + // LAMBDA: {{.+}} = alloca{{.+}}, + // LAMBDA: {{.+}} = alloca{{.+}}, + + // addr alloca's + // LAMBDA-64: [[G_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA-32: [[G_ADDR:%.+]] = alloca double*, + // LAMBDA: [[G1_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SFVAR_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[G1_REF:%.+]] = alloca double*, + + // private alloca's (only for 32-bit) + // LAMBDA-32: [[G_PRIV:%.+]] = alloca double, + + // transfer input parameters into addr alloca's + // LAMBDA-DAG: store {{.+}} [[G_IN]], {{.+}} [[G_ADDR]], + // LAMBDA-DAG: store {{.+}} [[G1_IN]], {{.+}} [[G1_ADDR]], + // LAMBDA-DAG: store {{.+}} [[SVAR_IN]], {{.+}} [[SVAR_ADDR]], + // LAMBDA-DAG: store {{.+}} [[SFVAR_IN]], {{.+}} [[SFVAR_ADDR]], + + // prepare parameters for lambda + // g + // LAMBDA-64-DAG: [[G_CONV:%.+]] = bitcast {{.+}}* [[G_ADDR]] to + // LAMBDA-32-DAG: [[G_ADDR_REF:%.+]] = load {{.+}}*, {{.+}}** [[G_ADDR]] + // LAMBDA-32-DAG: [[G_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[G_ADDR_REF]], + // LAMBDA-32-DAG: store {{.+}} [[G_ADDR_VAL]], {{.+}}* [[G_PRIV]], + + // g1 + // LAMBDA-DAG: [[G1_CONV:%.+]] = bitcast {{.+}}* [[G1_ADDR]] to + // LAMBDA-DAG: store {{.+}}* [[G1_CONV]], {{.+}}* [[G1_REF]], + + // svar + // LAMBDA-64-DAG: [[SVAR_CONV:%.+]] = bitcast {{.+}}* [[SVAR_ADDR]] to + + // sfvar + // LAMBDA-DAG: [[SFVAR_CONV:%.+]] = bitcast {{.+}}* [[SFVAR_ADDR]] to + + // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4( + g = 1; + g1 = 1; + svar = 3; + sfvar = 4.0; + // LAMBDA-64: store double 1.0{{.+}}, double* [[G_CONV]], + // LAMBDA-32: store double 1.0{{.+}}, double* [[G_PRIV]], + // LAMBDA: [[G1_REF_REF:%.+]] = load {{.+}}*, {{.+}}** [[G1_REF]], + // LAMBDA: store {{.+}} 1.0{{.+}}, {{.+}}* [[G1_REF_REF]], + // LAMBDA-64: store {{.+}} 3, {{.+}}* [[SVAR_CONV]], + // LAMBDA-32: store {{.+}} 3, {{.+}}* [[SVAR_ADDR]], + // LAMBDA: store {{.+}} 4.0{{.+}}, {{.+}}* [[SFVAR_CONV]], + + // pass params to inner lambda + // LAMBDA: [[G_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 + // LAMBDA-64: store double* [[G_CONV]], double** [[G_PRIVATE_ADDR_REF]], + // LAMBDA-32: store double* [[G_PRIV]], double** [[G_PRIVATE_ADDR_REF]], + // LAMBDA: [[G1_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1 + // LAMBDA: [[G1_REF_REF:%.+]] = load double*, double** [[G1_REF]], + // LAMBDA: store double* [[G1_REF_REF]], double** [[G1_PRIVATE_ADDR_REF]], + // LAMBDA: [[SVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 2 + // LAMBDA-64: store i{{[0-9]+}}* [[SVAR_CONV]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR_REF]] + // LAMBDA-32: store i{{[0-9]+}}* [[SVAR_ADDR]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR_REF]] + // LAMBDA: [[SFVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 3 + // LAMBDA: store float* [[SFVAR_CONV]], float** [[SFVAR_PRIVATE_ADDR_REF]] + // LAMBDA: call{{.*}} void [[INNER_LAMBDA:@.+]](%{{.+}}* [[ARG]]) + // LAMBDA: call {{.*}}void @__kmpc_for_static_fini( + // LAMBDA: ret void + [&]() { + // LAMBDA: define {{.+}} void [[INNER_LAMBDA]](%{{.+}}* [[ARG_PTR:%.+]]) + // LAMBDA: store %{{.+}}* [[ARG_PTR]], %{{.+}}** [[ARG_PTR_REF:%.+]], + g = 2; + g1 = 2; + svar = 4; + sfvar = 8.0; + // LAMBDA: [[ARG_PTR:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_PTR_REF]] + // LAMBDA: [[G_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 + // LAMBDA: [[G_REF:%.+]] = load double*, double** [[G_PTR_REF]] + // LAMBDA: store double 2.0{{.+}}, double* [[G_REF]] + + // LAMBDA: [[TMP_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 1 + // LAMBDA: [[G1_REF:%.+]] = load double*, double** [[TMP_PTR_REF]] + // LAMBDA: store double 2.0{{.+}}, double* [[G1_REF]], + // LAMBDA: [[SVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 2 + // LAMBDA: [[SVAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PTR_REF]] + // LAMBDA: store i{{[0-9]+}} 4, i{{[0-9]+}}* [[SVAR_REF]] + // LAMBDA: [[SFVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 3 + // LAMBDA: [[SFVAR_REF:%.+]] = load float*, float** [[SFVAR_PTR_REF]] + // LAMBDA: store float 8.0{{.+}}, float* [[SFVAR_REF]] + }(); + } + }(); + return 0; + #else + S<float> test; + int t_var = 0; + int vec[] = {1, 2}; + S<float> s_arr[] = {1, 2}; + S<float> &var = test; + + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for firstprivate(t_var, vec, s_arr, s_arr, var, var, svar) + for (int i = 0; i < 2; ++i) { + vec[i] = t_var; + s_arr[i] = var; + } + return tmain<int>(); + #endif +} + +// CHECK-LABEL: define{{.*}} i{{[0-9]+}} @main() +// CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]], +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]]) +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOAD_FUN_0:@.+]]( +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR:@.+]]([[S_FLOAT_TY]]* [[TEST]]) + +// CHECK: define{{.+}} [[OFFLOAD_FUN_0]](i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], [2 x [[S_FLOAT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_FLOAT_TY]]* {{.+}} [[VAR_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]]) +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 5, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i{{[0-9]+}}, [2 x i{{[0-9]+}}]*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]*, i{{[0-9]+}})* [[OMP_OUTLINED_0:@.+]] to void +// CHECK: ret + +// CHECK: define internal void [[OMP_OUTLINED_0]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], [2 x [[S_FLOAT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_FLOAT_TY]]* {{.+}} [[VAR_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]]) + +// addr alloca's +// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*, +// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_FLOAT_TY]]*, +// CHECK: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[TMP:%.+]] = alloca [[S_FLOAT_TY]]*, + +// skip loop alloca's +// CHECK: [[OMP_IV:.omp.iv+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_LB:.omp.comb.lb+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_UB:.omp.comb.ub+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_ST:.omp.stride+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_IS_LAST:.omp.is_last+]] = alloca i{{[0-9]+}}, + +// private alloca's +// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]], +// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_FLOAT_TY]]*, +// CHECK: [[SVAR_PRIV:%.+]] = alloca i{{[0-9]+}}, + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]] + +// init addr alloca's with input values +// CHECK-DAG: store {{.+}} [[T_VAR_IN]], {{.+}}* [[T_VAR_ADDR]], +// CHECK-DAG: store {{.+}} [[VEC_IN]], {{.+}} [[VEC_ADDR]], +// CHECK-DAG: store {{.+}} [[S_ARR_IN]], {{.+}} [[S_ARR_ADDR]], +// CHECK-DAG: store {{.+}} [[VAR_IN]], {{.+}} [[VAR_ADDR]], +// CHECK-DAG: store {{.+}} [[SVAR_IN]], {{.+}} [[SVAR_ADDR]], + +// init private alloca's with addr alloca's +// t-var +// CHECK-64-DAG: [[T_VAR_CONV:%.+]] = bitcast {{.+}} [[T_VAR_ADDR]] to +// CHECK-64-DAG: [[T_VAR_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_CONV]], +// CHECK-32-DAG: [[T_VAR_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_ADDR]], +// CHECK-DAG: store {{.+}} [[T_VAR_ADDR_VAL]], {{.+}} [[T_VAR_PRIV]], + +// vec +// CHECK-DAG: [[VEC_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[VEC_ADDR]], +// CHECK-DAG: [[VEC_PRIV_BCAST:%.+]] = bitcast {{.+}} [[VEC_PRIV]] to +// CHECK-DAG: [[VEC_ADDR_BCAST:%.+]] = bitcast {{.+}} [[VEC_ADDR_VAL]] to +// CHECK-DAG: call void @llvm.memcpy{{.+}}({{.+}}* [[VEC_PRIV_BCAST]], {{.+}}* [[VEC_ADDR_BCAST]], + +// s_arr +// CHECK-DAG: [[S_ARR_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[S_ARR_ADDR]], +// CHECK-DAG: [[S_ARR_BGN:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_PRIV]], +// CHECK-DAG: [[S_ARR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[S_ARR_ADDR_VAL]] to +// CHECK-DAG: [[S_ARR_BGN_GEP:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_BGN]], +// CHECK-DAG: [[S_ARR_EMPTY:%.+]] = icmp {{.+}} [[S_ARR_BGN]], [[S_ARR_BGN_GEP]] +// CHECK-DAG: br {{.+}} [[S_ARR_EMPTY]], label %[[CPY_DONE:.+]], label %[[CPY_BODY:.+]] +// CHECK-DAG: [[CPY_BODY]]: +// CHECK-DAG: call void @llvm.memcpy{{.+}}( +// CHECK-DAG: [[CPY_DONE]]: + +// var +// CHECK-DAG: [[TMP_REF:%.+]] = load {{.+}}*, {{.+}}* [[TMP]], +// CHECK-DAG: [[VAR_PRIV_BCAST:%.+]] = bitcast {{.+}}* [[VAR_PRIV]] to +// CHECK-DAG: [[TMP_REF_BCAST:%.+]] = bitcast {{.+}}* [[TMP_REF]] to +// CHECK-DAG: call void @llvm.memcpy.{{.+}}({{.+}}* [[VAR_PRIV_BCAST]], {{.+}}* [[TMP_REF_BCAST]], +// CHECK-DAG: store {{.+}}* [[VAR_PRIV]], {{.+}}** [[TMP_PRIV]], + +// svar +// CHECK-64-DAG: [[SVAR_CONV:%.+]] = bitcast {{.+}}* [[SVAR_ADDR]] to +// CHECK-64-DAG: [[SVAR_CONV_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_CONV]], +// CHECK-32-DAG: [[SVAR_CONV_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_ADDR]], +// CHECK-DAG: store {{.+}} [[SVAR_CONV_VAL]], {{.+}}* [[SVAR_PRIV]], + +// CHECK: call void @__kmpc_for_static_init_4( +// pass private alloca's to fork +// CHECK-DAG: [[T_VAR_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_PRIV]], +// not dag to distinguish with S_VAR_CAST +// CHECK-64: [[T_VAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[T_VAR_CAST:%.+]] to +// CHECK-64-DAG: store {{.+}} [[T_VAR_PRIV_VAL]], {{.+}} [[T_VAR_CAST_CONV]], +// CHECK-32: store {{.+}} [[T_VAR_PRIV_VAL]], {{.+}} [[T_VAR_CAST:%.+]], +// CHECK-DAG: [[T_VAR_CAST_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_CAST]], +// CHECK-DAG: [[TMP_PRIV_VAL:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP_PRIV]], +// CHECK-DAG: [[SVAR_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_PRIV]], +// CHECK-64-DAG: [[SVAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[SVAR_CAST:%.+]] to +// CHECK-64-DAG: store {{.+}} [[SVAR_PRIV_VAL]], {{.+}}* [[SVAR_CAST_CONV]], +// CHECK-32-DAG: store {{.+}} [[SVAR_PRIV_VAL]], {{.+}}* [[SVAR_CAST:%.+]], +// CHECK-DAG: [[SVAR_CAST_VAL:%.+]] = load {{.+}}, {{.+}}* [[SVAR_CAST]], +// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_0:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, [2 x i{{[0-9]+}}]* [[VEC_PRIV]], i{{[0-9]+}} [[T_VAR_CAST_VAL]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], [[S_FLOAT_TY]]* [[TMP_PRIV_VAL]], i{{[0-9]+}} [[SVAR_CAST_VAL]]) +// CHECK: call void @__kmpc_for_static_fini( + +// call destructors: var.. +// CHECK-DAG: call {{.+}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]]) + +// ..and s_arr +// CHECK: {{.+}}: +// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_FLOAT_TY]]* +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]], +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]]) + +// CHECK: ret void + +// By OpenMP specifications, 'firstprivate' applies to both distribute and parallel for. +// However, the support for 'firstprivate' of 'parallel' is only used when 'parallel' +// is found alone. Therefore we only have one 'firstprivate' support for 'parallel for' +// in combination +// CHECK: define internal void [[OMP_PARFOR_OUTLINED_0]]({{.+}}, {{.+}}, {{.+}}, {{.+}}, [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x [[S_FLOAT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_FLOAT_TY]]* {{.+}} [[VAR_IN:%.+]], i{{[0-9]+}} [[SVAR_IN:%.+]]) + +// addr alloca's +// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*, +// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_FLOAT_TY]]*, +// CHECK: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}}, + +// skip loop alloca's +// CHECK: [[OMP_IV:.omp.iv+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_LB:.omp.lb+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_UB:.omp.ub+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_ST:.omp.stride+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_IS_LAST:.omp.is_last+]] = alloca i{{[0-9]+}}, + +// private alloca's +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]], +// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_FLOAT_TY]]*, + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]] + +// init addr alloca's with input values +// CHECK-DAG: store {{.+}} [[VEC_IN]], {{.+}} [[VEC_ADDR]], +// CHECK-DAG: store {{.+}} [[T_VAR_IN]], {{.+}}* [[T_VAR_ADDR]], +// CHECK-DAG: store {{.+}} [[S_ARR_IN]], {{.+}} [[S_ARR_ADDR]], +// CHECK-DAG: store {{.+}} [[VAR_IN]], {{.+}} [[VAR_ADDR]], +// CHECK-DAG: store {{.+}} [[SVAR_IN]], {{.+}} [[SVAR_ADDR]], + +// init private alloca's with addr alloca's +// vec +// CHECK-DAG: [[VEC_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[VEC_ADDR]], +// CHECK-DAG: [[VEC_PRIV_BCAST:%.+]] = bitcast {{.+}} [[VEC_PRIV]] to +// CHECK-DAG: [[VEC_ADDR_BCAST:%.+]] = bitcast {{.+}} [[VEC_ADDR_VAL]] to +// CHECK-DAG: call void @llvm.memcpy{{.+}}({{.+}}* [[VEC_PRIV_BCAST]], {{.+}}* [[VEC_ADDR_BCAST]], + +// s_arr +// CHECK-DAG: [[S_ARR_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[S_ARR_ADDR]], +// CHECK-DAG: [[S_ARR_BGN:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_PRIV]], +// CHECK-DAG: [[S_ARR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[S_ARR_ADDR_VAL]] to +// CHECK-DAG: [[S_ARR_BGN_GEP:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_BGN]], +// CHECK-DAG: [[S_ARR_EMPTY:%.+]] = icmp {{.+}} [[S_ARR_BGN]], [[S_ARR_BGN_GEP]] +// CHECK-DAG: br {{.+}} [[S_ARR_EMPTY]], label %[[CPY_DONE:.+]], label %[[CPY_BODY:.+]] +// CHECK-DAG: [[CPY_BODY]]: +// CHECK-DAG: call void @llvm.memcpy{{.+}}( +// CHECK-DAG: [[CPY_DONE]]: + +// var +// CHECK-DAG: [[VAR_ADDR_REF:%.+]] = load {{.+}}*, {{.+}}* [[VAR_ADDR]], +// CHECK-DAG: [[VAR_PRIV_BCAST:%.+]] = bitcast {{.+}}* [[VAR_PRIV]] to +// CHECK-DAG: [[VAR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[VAR_ADDR_REF]] to +// CHECK-DAG: call void @llvm.memcpy.{{.+}}({{.+}}* [[VAR_PRIV_BCAST]], {{.+}}* [[VAR_ADDR_BCAST]], +// CHECK-DAG: store {{.+}}* [[VAR_PRIV]], {{.+}}** [[TMP_PRIV]], + +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call void @__kmpc_for_static_fini( + +// call destructors: var.. +// CHECK-DAG: call {{.+}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]]) + +// ..and s_arr +// CHECK: {{.+}}: +// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_FLOAT_TY]]* +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]], +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]]) + +// CHECK: ret void + +// template tmain with S_INT_TY +// CHECK-LABEL: define{{.*}} i{{[0-9]+}} @{{.+}}tmain{{.+}}() +// CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]], +// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]]) +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOAD_FUN_0:@.+]]( +// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR:@.+]]([[S_INT_TY]]* [[TEST]]) + +// CHECK: define{{.+}} [[OFFLOAD_FUN_0]](i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], [2 x [[S_INT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_INT_TY]]* {{.+}} [[VAR_IN:%.+]]) +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i{{[0-9]+}}, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[OMP_OUTLINED_0:@.+]] to void +// CHECK: ret + +// CHECK: define internal void [[OMP_OUTLINED_0]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], [2 x [[S_INT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_INT_TY]]* {{.+}} [[VAR_IN:%.+]]) + +// addr alloca's +// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*, +// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_INT_TY]]]*, +// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_INT_TY]]*, +// CHECK: [[TMP:%.+]] = alloca [[S_INT_TY]]*, + +// skip loop alloca's +// CHECK: [[OMP_IV:.omp.iv+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_LB:.omp.comb.lb+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_UB:.omp.comb.ub+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_ST:.omp.stride+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_IS_LAST:.omp.is_last+]] = alloca i{{[0-9]+}}, + +// private alloca's +// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]], +// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_INT_TY]]*, + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]] + +// init addr alloca's with input values +// CHECK-DAG: store {{.+}} [[T_VAR_IN]], {{.+}}* [[T_VAR_ADDR]], +// CHECK-DAG: store {{.+}} [[VEC_IN]], {{.+}} [[VEC_ADDR]], +// CHECK-DAG: store {{.+}} [[S_ARR_IN]], {{.+}} [[S_ARR_ADDR]], +// CHECK-DAG: store {{.+}} [[VAR_IN]], {{.+}} [[VAR_ADDR]], + +// init private alloca's with addr alloca's +// t-var +// CHECK-64-DAG: [[T_VAR_CONV:%.+]] = bitcast {{.+}} [[T_VAR_ADDR]] to +// CHECK-64-DAG: [[T_VAR_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_CONV]], +// CHECK-32-DAG: [[T_VAR_ADDR_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_ADDR]], +// CHECK-DAG: store {{.+}} [[T_VAR_ADDR_VAL]], {{.+}} [[T_VAR_PRIV]], + +// vec +// CHECK-DAG: [[VEC_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[VEC_ADDR]], +// CHECK-DAG: [[VEC_PRIV_BCAST:%.+]] = bitcast {{.+}} [[VEC_PRIV]] to +// CHECK-DAG: [[VEC_ADDR_BCAST:%.+]] = bitcast {{.+}} [[VEC_ADDR_VAL]] to +// CHECK-DAG: call void @llvm.memcpy{{.+}}({{.+}}* [[VEC_PRIV_BCAST]], {{.+}}* [[VEC_ADDR_BCAST]], + +// s_arr +// CHECK-DAG: [[S_ARR_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[S_ARR_ADDR]], +// CHECK-DAG: [[S_ARR_BGN:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_PRIV]], +// CHECK-DAG: [[S_ARR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[S_ARR_ADDR_VAL]] to +// CHECK-DAG: [[S_ARR_BGN_GEP:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_BGN]], +// CHECK-DAG: [[S_ARR_EMPTY:%.+]] = icmp {{.+}} [[S_ARR_BGN]], [[S_ARR_BGN_GEP]] +// CHECK-DAG: br {{.+}} [[S_ARR_EMPTY]], label %[[CPY_DONE:.+]], label %[[CPY_BODY:.+]] +// CHECK-DAG: [[CPY_BODY]]: +// CHECK-DAG: call void @llvm.memcpy{{.+}}( +// CHECK-DAG: [[CPY_DONE]]: + +// var +// CHECK-DAG: [[TMP_REF:%.+]] = load {{.+}}*, {{.+}}* [[TMP]], +// CHECK-DAG: [[VAR_PRIV_BCAST:%.+]] = bitcast {{.+}}* [[VAR_PRIV]] to +// CHECK-DAG: [[TMP_REF_BCAST:%.+]] = bitcast {{.+}}* [[TMP_REF]] to +// CHECK-DAG: call void @llvm.memcpy.{{.+}}({{.+}}* [[VAR_PRIV_BCAST]], {{.+}}* [[TMP_REF_BCAST]], +// CHECK-DAG: store {{.+}}* [[VAR_PRIV]], {{.+}}** [[TMP_PRIV]], + +// CHECK: call void @__kmpc_for_static_init_4( +// pass private alloca's to fork +// CHECK-DAG: [[T_VAR_PRIV_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_PRIV]], +// not dag to distinguish with S_VAR_CAST +// CHECK-64: [[T_VAR_CAST_CONV:%.+]] = bitcast {{.+}}* [[T_VAR_CAST:%.+]] to +// CHECK-64-DAG: store {{.+}} [[T_VAR_PRIV_VAL]], {{.+}} [[T_VAR_CAST_CONV]], +// CHECK-32: store {{.+}} [[T_VAR_PRIV_VAL]], {{.+}} [[T_VAR_CAST:%.+]], +// CHECK-DAG: [[T_VAR_CAST_VAL:%.+]] = load {{.+}}, {{.+}}* [[T_VAR_CAST]], +// CHECK-DAG: [[TMP_PRIV_VAL:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP_PRIV]], +// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_0:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, [2 x i{{[0-9]+}}]* [[VEC_PRIV]], i{{[0-9]+}} [[T_VAR_CAST_VAL]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], [[S_INT_TY]]* [[TMP_PRIV_VAL]]) +// CHECK: call void @__kmpc_for_static_fini( + +// call destructors: var.. +// CHECK-DAG: call {{.+}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]]) + +// ..and s_arr +// CHECK: {{.+}}: +// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_INT_TY]]* +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]], +// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]]) + +// CHECK: ret void + +// By OpenMP specifications, 'firstprivate' applies to both distribute and parallel for. +// However, the support for 'firstprivate' of 'parallel' is only used when 'parallel' +// is found alone. Therefore we only have one 'firstprivate' support for 'parallel for' +// in combination +// CHECK: define internal void [[OMP_PARFOR_OUTLINED_0]]({{.+}}, {{.+}}, {{.+}}, {{.+}}, [2 x i{{[0-9]+}}]* {{.+}} [[VEC_IN:%.+]], i{{[0-9]+}} [[T_VAR_IN:%.+]], [2 x [[S_INT_TY]]]* {{.+}} [[S_ARR_IN:%.+]], [[S_INT_TY]]* {{.+}} [[VAR_IN:%.+]]) + +// addr alloca's +// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*, +// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_INT_TY]]]*, +// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_INT_TY]]*, + +// skip loop alloca's +// CHECK: [[OMP_IV:.omp.iv+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_LB:.omp.lb+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_UB:.omp.ub+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_ST:.omp.stride+]] = alloca i{{[0-9]+}}, +// CHECK: [[OMP_IS_LAST:.omp.is_last+]] = alloca i{{[0-9]+}}, + +// private alloca's +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]], +// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_INT_TY]]*, + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]] + +// init addr alloca's with input values +// CHECK-DAG: store {{.+}} [[VEC_IN]], {{.+}} [[VEC_ADDR]], +// CHECK-DAG: store {{.+}} [[T_VAR_IN]], {{.+}}* [[T_VAR_ADDR]], +// CHECK-DAG: store {{.+}} [[S_ARR_IN]], {{.+}} [[S_ARR_ADDR]], +// CHECK-DAG: store {{.+}} [[VAR_IN]], {{.+}} [[VAR_ADDR]], + +// init private alloca's with addr alloca's +// vec +// CHECK-DAG: [[VEC_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[VEC_ADDR]], +// CHECK-DAG: [[VEC_PRIV_BCAST:%.+]] = bitcast {{.+}} [[VEC_PRIV]] to +// CHECK-DAG: [[VEC_ADDR_BCAST:%.+]] = bitcast {{.+}} [[VEC_ADDR_VAL]] to +// CHECK-DAG: call void @llvm.memcpy{{.+}}({{.+}}* [[VEC_PRIV_BCAST]], {{.+}}* [[VEC_ADDR_BCAST]], + +// s_arr +// CHECK-DAG: [[S_ARR_ADDR_VAL:%.+]] = load {{.+}}*, {{.+}}** [[S_ARR_ADDR]], +// CHECK-DAG: [[S_ARR_BGN:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_PRIV]], +// CHECK-DAG: [[S_ARR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[S_ARR_ADDR_VAL]] to +// CHECK-DAG: [[S_ARR_BGN_GEP:%.+]] = getelementptr {{.+}}, {{.+}}* [[S_ARR_BGN]], +// CHECK-DAG: [[S_ARR_EMPTY:%.+]] = icmp {{.+}} [[S_ARR_BGN]], [[S_ARR_BGN_GEP]] +// CHECK-DAG: br {{.+}} [[S_ARR_EMPTY]], label %[[CPY_DONE:.+]], label %[[CPY_BODY:.+]] +// CHECK-DAG: [[CPY_BODY]]: +// CHECK-DAG: call void @llvm.memcpy{{.+}}( +// CHECK-DAG: [[CPY_DONE]]: + +// var +// CHECK-DAG: [[VAR_ADDR_REF:%.+]] = load {{.+}}*, {{.+}}* [[VAR_ADDR]], +// CHECK-DAG: [[VAR_PRIV_BCAST:%.+]] = bitcast {{.+}}* [[VAR_PRIV]] to +// CHECK-DAG: [[VAR_ADDR_BCAST:%.+]] = bitcast {{.+}}* [[VAR_ADDR_REF]] to +// CHECK-DAG: call void @llvm.memcpy.{{.+}}({{.+}}* [[VAR_PRIV_BCAST]], {{.+}}* [[VAR_ADDR_BCAST]], +// CHECK-DAG: store {{.+}}* [[VAR_PRIV]], {{.+}}** [[TMP_PRIV]], + +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call void @__kmpc_for_static_fini( + +// call destructors: var.. +// CHECK-DAG: call {{.+}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]]) + +// ..and s_arr +// CHECK: {{.+}}: +// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_INT_TY]]* +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]], +// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]]) + +// CHECK: ret void + +#endif diff --git a/test/OpenMP/distribute_parallel_for_if_codegen.cpp b/test/OpenMP/distribute_parallel_for_if_codegen.cpp new file mode 100644 index 0000000000000..e6cd7215f585c --- /dev/null +++ b/test/OpenMP/distribute_parallel_for_if_codegen.cpp @@ -0,0 +1,192 @@ +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -triple %itanium_abi_triple -emit-llvm %s -o - | FileCheck %s +// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -std=c++11 -triple %itanium_abi_triple -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -triple %itanium_abi_triple -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck --check-prefix=CHECK %s +// expected-no-diagnostics +#ifndef HEADER +#define HEADER + +void fn1(); +void fn2(); +void fn3(); +void fn4(); +void fn5(); +void fn6(); + +int Arg; + +// CHECK-LABEL: define {{.*}}void @{{.+}}gtid_test +void gtid_test() { +#pragma omp target +#pragma omp teams +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_0:@.+]]( +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_1:@.+]]( +#pragma omp distribute parallel for + for(int i = 0 ; i < 100; i++) {} + // CHECK: define internal void [[OFFLOADING_FUN_0]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_0:@.+]] to {{.+}}) + // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_0]]( + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, {{.+}}* [[OMP_OUTLINED_0:@.+]] to void + // CHECK: call void @__kmpc_for_static_fini( + + // CHECK: define{{.+}} void [[OMP_OUTLINED_0]]( + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call void @__kmpc_for_static_fini( + // CHECK: ret +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for if (parallel: false) + for(int i = 0 ; i < 100; i++) { + // CHECK: define internal void [[OFFLOADING_FUN_1]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_1:@.+]] to {{.+}}) + // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_1]]( + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call void @__kmpc_serialized_parallel( + // CHECK: call void [[OMP_OUTLINED_1:@.+]]( + // CHECK: call void @__kmpc_end_serialized_parallel( + // CHECK: call void @__kmpc_for_static_fini( + // CHECK: define{{.+}} void [[OMP_OUTLINED_1]]( + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call void @{{.+}}gtid_test + // CHECK: call void @__kmpc_for_static_fini( + // CHECK: ret + gtid_test(); + } +} + + +template <typename T> +int tmain(T Arg) { +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for if (true) + for(int i = 0 ; i < 100; i++) { + fn1(); + } +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for if (false) + for(int i = 0 ; i < 100; i++) { + fn2(); + } +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for if (parallel: Arg) + for(int i = 0 ; i < 100; i++) { + fn3(); + } + return 0; +} + +// CHECK-LABEL: define {{.*}}i{{[0-9]+}} @main() +int main() { +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_0:@.+]]( +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_1:@.+]]( +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_2:@.+]]( +// CHECK: = call {{.*}}i{{.+}} @{{.+}}tmain +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for if (true) + for(int i = 0 ; i < 100; i++) { + // CHECK: define internal void [[OFFLOADING_FUN_0]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_0:@.+]] to {{.+}}) + // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_0]]( + + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, {{.+}}* [[OMP_OUTLINED_2:@.+]] to void + // CHECK: call void @__kmpc_for_static_fini( + // CHECK: define{{.+}} void [[OMP_OUTLINED_2]]( + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call {{.*}}void @{{.+}}fn4 + // CHECK: call void @__kmpc_for_static_fini( + + fn4(); + } + +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for if (false) + for(int i = 0 ; i < 100; i++) { + // CHECK: define internal void [[OFFLOADING_FUN_1]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_1:@.+]] to {{.+}}) + // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_1]]( + + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call void @__kmpc_serialized_parallel( + // CHECK: call void [[OMP_OUTLINED_3:@.+]]( + // CHECK: call void @__kmpc_end_serialized_parallel( + // CHECK: call void @__kmpc_for_static_fini( + + // CHECK: define{{.+}} void [[OMP_OUTLINED_3]]( + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call {{.*}}void @{{.+}}fn5 + // CHECK: call void @__kmpc_for_static_fini( + fn5(); + } + +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for if (Arg) + for(int i = 0 ; i < 100; i++) { + // CHECK: define internal void [[OFFLOADING_FUN_2]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 1, {{.+}}* [[OMP_TEAMS_OUTLINED_2:@.+]] to {{.+}}) + // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_2]]( + + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, {{.+}}* [[OMP_OUTLINED_4:@.+]] to void + // CHECK: call void @__kmpc_serialized_parallel( + // CHECK: call void [[OMP_OUTLINED_4:@.+]]( + // CHECK: call void @__kmpc_end_serialized_parallel( + // CHECK: call void @__kmpc_for_static_fini( + + // CHECK: define{{.+}} void [[OMP_OUTLINED_4]]( + // CHECK: call void @__kmpc_for_static_init_4( + // CHECK: call {{.*}}void @{{.+}}fn6 + // CHECK: call void @__kmpc_for_static_fini( + fn6(); + } + + return tmain(Arg); +} + +// CHECK-LABEL: define {{.+}} @{{.+}}tmain + +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, void {{.+}}* [[T_OUTLINE_FUN_1:@.+]] to void +// CHECK: call void @__kmpc_for_static_fini( + +// CHECK: define internal {{.*}}void [[T_OUTLINE_FUN_1]] +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call {{.*}}void @{{.+}}fn1 +// CHECK: call void @__kmpc_for_static_fini( +// CHECK: ret void + +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call {{.*}}void @__kmpc_serialized_parallel( +// CHECK: call void [[T_OUTLINE_FUN_2:@.+]]( +// CHECK: call {{.*}}void @__kmpc_end_serialized_parallel( +// CHECK: call void @__kmpc_for_static_fini( + +// CHECK: define internal {{.*}}void [[T_OUTLINE_FUN_2]] +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call {{.*}}void @{{.+}}fn2 +// CHECK: call void @__kmpc_for_static_fini( +// CHECK: ret void + +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 2, void {{.+}}* [[T_OUTLINE_FUN_3:@.+]] to void +// CHECK: call {{.*}}void @__kmpc_serialized_parallel( +// call void [[T_OUTLINE_FUN_3:@.+]]( +// CHECK: call {{.*}}void @__kmpc_end_serialized_parallel( + +// CHECK: define internal {{.*}}void [[T_OUTLINE_FUN_3]] +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call {{.*}}void @{{.+}}fn3 +// CHECK: call void @__kmpc_for_static_fini( +// CHECK: ret void +#endif diff --git a/test/OpenMP/distribute_parallel_for_lastprivate_codegen.cpp b/test/OpenMP/distribute_parallel_for_lastprivate_codegen.cpp new file mode 100644 index 0000000000000..2e8da79c03bd4 --- /dev/null +++ b/test/OpenMP/distribute_parallel_for_lastprivate_codegen.cpp @@ -0,0 +1,653 @@ +// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64 +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64 +// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32 +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32 + +// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64 +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64 +// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32 +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32 +// expected-no-diagnostics +#ifndef HEADER +#define HEADER + +template <class T> +struct S { + T f; + S(T a) : f(a) {} + S() : f() {} + operator T() { return T(); } + ~S() {} +}; + +// CHECK: [[S_FLOAT_TY:%.+]] = type { float } +// CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} } +template <typename T> +T tmain() { + S<T> test; + T t_var = T(); + T vec[] = {1, 2}; + S<T> s_arr[] = {1, 2}; + S<T> &var = test; + #pragma omp target + #pragma omp teams +#pragma omp distribute parallel for lastprivate(t_var, vec, s_arr, s_arr, var, var) + for (int i = 0; i < 2; ++i) { + vec[i] = t_var; + s_arr[i] = var; + } + return T(); +} + +int main() { + static int svar; + volatile double g; + volatile double &g1 = g; + + #ifdef LAMBDA + // LAMBDA-LABEL: @main + // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]]( + [&]() { + static float sfvar; + // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]]( + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN:@.+]]( + + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN]]( + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 4, {{.+}}* [[OMP_OUTLINED:@.+]] to {{.+}}) + #pragma omp target + #pragma omp teams +#pragma omp distribute parallel for lastprivate(g, g1, svar, sfvar) + for (int i = 0; i < 2; ++i) { + // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, double*{{.+}} [[G_IN:%.+]], double*{{.+}} [[G1_IN:%.+]], i{{[0-9]+}}*{{.+}} [[SVAR_IN:%.+]], float*{{.+}} [[SFVAR_IN:%.+]]) + // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double*, + // LAMBDA: [[G1_PRIVATE_ADDR:%.+]] = alloca double*, + // LAMBDA: [[SVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}}*, + // LAMBDA: [[SFVAR_PRIVATE_ADDR:%.+]] = alloca float*, + // LAMBDA: [[TMP_G1:%.+]] = alloca double*, + // loop variables + // LAMBDA: {{.+}} = alloca i{{[0-9]+}}, + // LAMBDA: {{.+}} = alloca i{{[0-9]+}}, + // LAMBDA: {{.+}} = alloca i{{[0-9]+}}, + // LAMBDA: {{.+}} = alloca i{{[0-9]+}}, + // LAMBDA: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[G_PRIVATE:%.+]] = alloca double, + // LAMBDA: [[G1_PRIVATE:%.+]] = alloca double, + // LAMBDA: [[TMP_G1_PRIVATE:%.+]] = alloca double*, + // LAMBDA: [[SVAR_PRIVATE:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SFVAR_PRIVATE:%.+]] = alloca float, + + // init addr alloca's + // LAMBDA: store double* [[G_IN]], double** [[G_PRIVATE_ADDR]], + // LAMBDA: store double* [[G1_IN]], double** [[G1_PRIVATE_ADDR]], + // LAMBDA: store i{{[0-9]+}}* [[SVAR_IN]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR]], + // LAMBDA: store float* [[SFVAR_IN]], float** [[SFVAR_PRIVATE_ADDR]], + + // init private variables + // LAMBDA: [[G_IN_REF:%.+]] = load double*, double** [[G_PRIVATE_ADDR]], + // LAMBDA: [[SVAR_IN_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR]], + // LAMBDA: [[SFVAR_IN_REF:%.+]] = load float*, float** [[SFVAR_PRIVATE_ADDR]], + // LAMBDA: [[G1_IN_REF:%.+]] = load double*, double** [[G1_PRIVATE_ADDR]], + // LAMBDA: store double* [[G1_IN_REF]], double** [[TMP_G1]], + // LAMBDA: [[TMP_G1_VAL:%.+]] = load double*, double** [[TMP_G1]], + // LAMBDA: store double* [[G1_PRIVATE]], double** [[TMP_G1_PRIVATE]], + + // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4( + // LAMBDA: [[G1_PAR:%.+]] = load{{.+}}, {{.+}} [[TMP_G1_PRIVATE]], + // LAMBDA-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[G_PRIVATE]], {{.+}} [[G1_PAR]], {{.+}} [[SVAR_PRIVATE]], {{.+}} [[SFVAR_PRIVATE]]) + // LAMBDA-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[G_PRIVATE]], {{.+}} [[G1_PAR]], {{.+}} [[SVAR_PRIVATE]], {{.+}} [[SFVAR_PRIVATE]]) + // LAMBDA: call {{.*}}void @__kmpc_for_static_fini( + + // lastprivate + // LAMBDA: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]], + // LAMBDA: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0 + // LAMBDA: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]] + + // LAMBDA: [[OMP_LASTPRIV_BLOCK]]: + // LAMBDA: [[G_PRIV_VAL:%.+]] = load double, double* [[G_PRIVATE]], + // LAMBDA: store{{.*}} double [[G_PRIV_VAL]], double* [[G_IN_REF]], + // LAMBDA: [[TMP_G1_PRIV_REF:%.+]] = load double*, double** [[TMP_G1_PRIVATE]], + // LAMBDA: [[TMP_G1_PRIV_VAL:%.+]] = load double, double* [[TMP_G1_PRIV_REF]], + // LAMBDA: store{{.*}} double [[TMP_G1_PRIV_VAL]], double* [[TMP_G1_VAL]], + + // LAMBDA: [[SVAR_PRIV_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[SVAR_PRIVATE]], + // LAMBDA: store i{{[0-9]+}} [[SVAR_PRIV_VAL]], i{{[0-9]+}}* [[SVAR_IN_REF]], + // LAMBDA: [[SFVAR_PRIV_VAL:%.+]] = load float, float* [[SFVAR_PRIVATE]], + // LAMBDA: store float [[SFVAR_PRIV_VAL]], float* [[SFVAR_IN_REF]], + // LAMBDA: br label %[[OMP_LASTPRIV_DONE]] + // LAMBDA: [[OMP_LASTPRIV_DONE]]: + // LAMBDA: ret + + g = 1; + g1 = 1; + svar = 3; + sfvar = 4.0; + // outlined function for 'parallel for' + // LAMBDA-64: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[G_IN:%.+]], {{.+}} [[G1_IN:%.+]], {{.+}} [[SVAR_IN:%.+]], {{.+}} [[SFVAR_IN:%.+]]) + // LAMBDA-32: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[G_IN:%.+]], {{.+}} [[G1_IN:%.+]], {{.+}} [[SVAR_IN:%.+]], {{.+}} [[SFVAR_IN:%.+]]) + + // addr alloca's + // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double*, + // LAMBDA: [[G1_PRIVATE_ADDR:%.+]] = alloca double*, + // LAMBDA: [[SVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}}*, + // LAMBDA: [[SFVAR_PRIVATE_ADDR:%.+]] = alloca float*, + + // loop variables + // LAMBDA: {{.+}} = alloca i{{[0-9]+}}, + // LAMBDA: {{.+}} = alloca i{{[0-9]+}}, + // LAMBDA: {{.+}} = alloca i{{[0-9]+}}, + // LAMBDA: {{.+}} = alloca i{{[0-9]+}}, + + // private alloca's + // LAMBDA: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[G_PRIVATE:%.+]] = alloca double, + // LAMBDA: [[G1_PRIVATE:%.+]] = alloca double, + // LAMBDA: [[TMP_G1_PRIVATE:%.+]] = alloca double*, + // LAMBDA: [[SVAR_PRIVATE:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SFVAR_PRIVATE:%.+]] = alloca float, + + // init addr alloca's + // LAMBDA: store double* [[G_IN]], double** [[G_PRIVATE_ADDR]], + // LAMBDA: store double* [[G1_IN]], double** [[G1_PRIVATE_ADDR]], + // LAMBDA: store i{{[0-9]+}}* [[SVAR_IN]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR]], + // LAMBDA: store float* [[SFVAR_IN]], float** [[SFVAR_PRIVATE_ADDR]], + + // init private variables + // LAMBDA: [[G_IN_REF:%.+]] = load double*, double** [[G_PRIVATE_ADDR]], + // LAMBDA: [[SVAR_IN_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR]], + // LAMBDA: [[SFVAR_IN_REF:%.+]] = load float*, float** [[SFVAR_PRIVATE_ADDR]], + + // LAMBDA: [[G1_IN_REF:%.+]] = load double*, double** [[G1_PRIVATE_ADDR]], + // LAMBDA: store double* [[G1_PRIVATE]], double** [[TMP_G1]], + + // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4( + + // loop body + // LAMBDA: store double 1.0{{.+}}, double* [[G_PRIVATE]], + // LAMBDA: [[TMP_G1_REF:%.+]] = load double*, double** [[TMP_G1_PRIVATE]], + // LAMBDA: store{{.+}} double 1.0{{.+}}, double* [[TMP_G1_REF]], + // LAMBDA: store i{{[0-9]+}} 3, i{{[0-9]+}}* [[SVAR_PRIVATE]], + // LAMBDA: store float 4.0{{.+}}, float* [[SFVAR_PRIVATE]], + // LAMBDA: [[G_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 + // LAMBDA: store double* [[G_PRIVATE]], double** [[G_PRIVATE_ADDR_REF]], + // LAMBDA: [[TMP_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1 + // LAMBDA: [[G1_PRIVATE_ADDR_FROM_TMP:%.+]] = load double*, double** [[TMP_G1_PRIVATE]], + // LAMBDA: store double* [[G1_PRIVATE_ADDR_FROM_TMP]], double** [[TMP_PRIVATE_ADDR_REF]], + // LAMBDA: [[SVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 2 + // LAMBDA: store i{{[0-9]+}}* [[SVAR_PRIVATE]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR_REF]] + // LAMBDA: [[SFVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 3 + // LAMBDA: store float* [[SFVAR_PRIVATE]], float** [[SFVAR_PRIVATE_ADDR_REF]] + // LAMBDA: call{{.*}} void [[INNER_LAMBDA:@.+]](%{{.+}}* [[ARG]]) + + // LAMBDA: call {{.*}}void @__kmpc_for_static_fini( + + // lastprivate + // LAMBDA: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]], + // LAMBDA: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0 + // LAMBDA: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]] + // LAMBDA: [[OMP_LASTPRIV_BLOCK]]: + // LAMBDA: [[G_PRIV_VAL:%.+]] = load double, double* [[G_PRIVATE]], + // LAMBDA: store{{.*}} double [[G_PRIV_VAL]], double* [[G_IN_REF]], + // LAMBDA: [[TMP_G1_PRIV_REF:%.+]] = load double*, double** [[TMP_G1_PRIVATE]], + // LAMBDA: [[TMP_G1_PRIV_VAL:%.+]] = load double, double* [[TMP_G1_PRIV_REF]], + // LAMBDA: store{{.*}} double [[TMP_G1_PRIV_VAL]], double* [[G1_IN_REF]], + // LAMBDA: [[SVAR_PRIV_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[SVAR_PRIVATE]], + // LAMBDA: store i{{[0-9]+}} [[SVAR_PRIV_VAL]], i{{[0-9]+}}* [[SVAR_IN_REF]], + // LAMBDA: [[SFVAR_PRIV_VAL:%.+]] = load float, float* [[SFVAR_PRIVATE]], + // LAMBDA: store float [[SFVAR_PRIV_VAL]], float* [[SFVAR_IN_REF]], + // LAMBDA: br label %[[OMP_LASTPRIV_DONE]] + // LAMBDA: [[OMP_LASTPRIV_DONE]]: + // LAMBDA: ret + + [&]() { + // LAMBDA: define {{.+}} void [[INNER_LAMBDA]](%{{.+}}* [[ARG_PTR:%.+]]) + // LAMBDA: store %{{.+}}* [[ARG_PTR]], %{{.+}}** [[ARG_PTR_REF:%.+]], + g = 2; + g1 = 2; + svar = 4; + sfvar = 8.0; + // LAMBDA: [[ARG_PTR:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_PTR_REF]] + // LAMBDA: [[G_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 + // LAMBDA: [[G_REF:%.+]] = load double*, double** [[G_PTR_REF]] + // LAMBDA: store double 2.0{{.+}}, double* [[G_REF]] + + // LAMBDA: [[TMP_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 1 + // LAMBDA: [[G1_REF:%.+]] = load double*, double** [[TMP_PTR_REF]] + // LAMBDA: store double 2.0{{.+}}, double* [[G1_REF]], + // LAMBDA: [[SVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 2 + // LAMBDA: [[SVAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PTR_REF]] + // LAMBDA: store i{{[0-9]+}} 4, i{{[0-9]+}}* [[SVAR_REF]] + // LAMBDA: [[SFVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 3 + // LAMBDA: [[SFVAR_REF:%.+]] = load float*, float** [[SFVAR_PTR_REF]] + // LAMBDA: store float 8.0{{.+}}, float* [[SFVAR_REF]] + }(); + } + }(); + return 0; + #else + S<float> test; + int t_var = 0; + int vec[] = {1, 2}; + S<float> s_arr[] = {1, 2}; + S<float> &var = test; + + #pragma omp target + #pragma omp teams +#pragma omp distribute parallel for lastprivate(t_var, vec, s_arr, s_arr, var, var, svar) + for (int i = 0; i < 2; ++i) { + vec[i] = t_var; + s_arr[i] = var; + } + int i; + + return tmain<int>(); + #endif +} + +// CHECK: define{{.*}} i{{[0-9]+}} @main() +// CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]], +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]]) +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOAD_FUN:@.+]](i{{[0-9]+}} {{.+}}, [2 x i{{[0-9]+}}]* {{.+}}, [2 x [[S_FLOAT_TY]]]* {{.+}}, [[S_FLOAT_TY]]* {{.+}}, i{{[0-9]+}} {{.+}}) +// CHECK: ret + +// CHECK: define{{.+}} [[OFFLOAD_FUN]](i{{[0-9]+}} {{.+}}, [2 x i{{[0-9]+}}]*{{.+}} {{.+}}, [2 x [[S_FLOAT_TY]]]*{{.+}} {{.+}}, [[S_FLOAT_TY]]*{{.+}} {{.+}}, i{{[0-9]+}} {{.+}}) +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams( +// CHECK: ret +// +// CHECK: define internal void [[OMP_OUTLINED:@.+]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i{{[0-9]+}}*{{.+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]*{{.+}} [[VEC_IN:%.+]], [2 x [[S_FLOAT_TY]]]*{{.+}} [[S_ARR_IN:%.+]], [[S_FLOAT_TY]]*{{.+}} [[VAR_IN:%.+]], i{{[0-9]+}}*{{.*}} [[S_VAR_IN:%.+]]) +// CHECK: {{.+}} = alloca i{{[0-9]+}}*, +// CHECK: {{.+}} = alloca i{{[0-9]+}}*, +// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}*, +// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*, +// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_FLOAT_TY]]*, +// CHECK: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}}*, +// CHECK: [[TMP:%.*]] = alloca [[S_FLOAT_TY]]*, +// skip loop variables +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]], +// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_FLOAT_TY]]*, +// CHECK: [[S_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, + +// copy from parameters to local address variables +// CHECK: store i{{[0-9]+}}* [[T_VAR_IN]], i{{[0-9]+}}** [[T_VAR_ADDR]], +// CHECK: store [2 x i{{[0-9]+}}]* [[VEC_IN]], [2 x i{{[0-9]+}}]** [[VEC_ADDR]], +// CHECK: store [2 x [[S_FLOAT_TY]]]* [[S_ARR_IN]], [2 x [[S_FLOAT_TY]]]** [[S_ARR_ADDR]], +// CHECK: store [[S_FLOAT_TY]]* [[VAR_IN]], [[S_FLOAT_TY]]** [[VAR_ADDR]], +// CHECK: store i{{[0-9]+}}* [[S_VAR_IN]], i{{[0-9]+}}** [[SVAR_ADDR]], + +// load content of local address variables +// CHECK: [[T_VAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_ADDR]], +// CHECK: [[VEC_ADDR_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_ADDR]], +// CHECK: [[S_ARR_ADDR_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[S_ARR_ADDR]], +// CHECK: [[SVAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_ADDR]], +// CHECK: [[VAR_ADDR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_ADDR]], +// CHECK: store [[S_FLOAT_TY]]* [[VAR_ADDR_REF]], [[S_FLOAT_TY]]** [[TMP]], +// CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[OMP_IS_LAST]], + +// call constructor for s_arr +// CHECK: [[S_ARR_BGN:%.+]] = getelementptr{{.+}} [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], +// CHECK: [[S_ARR_END:%.+]] = getelementptr {{.+}} [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_BGN]], +// CHECK: br label %[[S_ARR_CST_LOOP:.+]] +// CHECK: [[S_ARR_CST_LOOP]]: +// CHECK: [[S_ARR_CTOR:%.+]] = phi {{.+}} +// CHECK: call void [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_CTOR]]) +// CHECK: [[S_ARR_NEXT:%.+]] = getelementptr {{.+}} [[S_ARR_CTOR]], +// CHECK: [[S_ARR_DONE:%.+]] = icmp {{.+}} [[S_ARR_NEXT]], [[S_ARR_END]] +// CHECK: br i1 [[S_ARR_DONE]], label %[[S_ARR_CST_END:.+]], label %[[S_ARR_CST_LOOP]] +// CHECK: [[S_ARR_CST_END]]: +// CHECK: [[TMP_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP]], +// CHECK: call void [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]]) +// CHECK: store [[S_FLOAT_TY]]* [[VAR_PRIV]], [[S_FLOAT_TY]]** [[TMP_PRIV]], + +// the distribute loop +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: [[TMP_PRIV_VAL:%.+]] = load {{.+}}, {{.+}} [[TMP_PRIV]], +// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[VEC_PRIV]], {{.+}} [[T_VAR_PRIV]], {{.+}} [[S_ARR_PRIV]], {{.+}} [[TMP_PRIV_VAL]], {{.+}} [[S_VAR_PRIV]]) +// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[VEC_PRIV]], {{.+}} [[T_VAR_PRIV]], {{.+}} [[S_ARR_PRIV]], {{.+}} [[TMP_PRIV_VAL]], {{.+}} [[S_VAR_PRIV]]) + +// CHECK: call void @__kmpc_for_static_fini( + +// lastprivates +// CHECK: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]], +// CHECK: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0 +// CHECK: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]] + +// CHECK: [[OMP_LASTPRIV_BLOCK]]: +// CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]], +// CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_ADDR_REF]], +// CHECK: [[BCAST_VEC_ADDR_REF:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_ADDR_REF]] to i8* +// CHECK: [[BCAST_VEC_PRIV:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[BCAST_VEC_ADDR_REF]], i8* [[BCAST_VEC_PRIV]], +// CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_ADDR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 +// CHECK: [[S_ARR_PRIV_BCAST:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]] to [[S_FLOAT_TY]]* +// CHECK: [[S_ARR_BEGIN_GEP:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_BEGIN]], i{{[0-9]+}} 2 +// CHECK: [[S_ARR_IS_EMPTY:%.+]] = icmp eq [[S_FLOAT_TY]]* [[S_ARR_BEGIN]], [[S_ARR_BEGIN_GEP]] +// CHECK: br i1 [[S_ARR_IS_EMPTY]], label %[[S_ARR_COPY_DONE:.+]], label %[[S_ARR_COPY_BLOCK:.+]] +// CHECK: [[S_ARR_COPY_BLOCK]]: +// CHECK: [[S_ARR_SRC_EL:%.+]] = phi [[S_FLOAT_TY]]*{{.+}} +// CHECK: [[S_ARR_DST_EL:%.+]] = phi [[S_FLOAT_TY]]*{{.+}} +// CHECK: [[S_ARR_DST_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_DST_EL]] to i8* +// CHECK: [[S_ARR_SRC_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_SRC_EL]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_DST_BCAST]], i8* [[S_ARR_SRC_BCAST]]{{.+}}) +// CHECK: [[S_ARR_DST_NEXT:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_DST_EL]], i{{[0-9]+}} 1 +// CHECK: [[S_ARR_SRC_NEXT:%.+]] = getelementptr{{.+}} +// CHECK: [[CPY_IS_FINISHED:%.+]] = icmp eq [[S_FLOAT_TY]]* [[S_ARR_DST_NEXT]], [[S_ARR_BEGIN_GEP]] +// CHECK: br i1 [[CPY_IS_FINISHED]], label %[[S_ARR_COPY_DONE]], label %[[S_ARR_COPY_BLOCK]] +// CHECK: [[S_ARR_COPY_DONE]]: +// CHECK: [[TMP_VAL1:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP_PRIV]], +// CHECK: [[VAR_ADDR_REF_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[TMP_REF]] to i8* +// CHECK: [[TMP_VAL1_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[TMP_VAL1]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VAR_ADDR_REF_BCAST]], i8* [[TMP_VAL1_BCAST]],{{.+}}) +// CHECK: [[SVAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[S_VAR_PRIV]], +// CHECK: store i{{[0-9]+}} [[SVAR_VAL]], i{{[0-9]+}}* [[SVAR_ADDR_REF]], +// CHECK: ret void + +// outlined function for 'parallel for' +// CHECK-64: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[VEC_IN:%.+]], {{.+}} [[T_VAR_IN:%.+]], {{.+}} [[S_ARR_IN:%.+]], {{.+}} [[VAR_IN:%.+]], {{.+}} [[SVAR_IN:%.+]]) +// CHECK-32: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[VEC_IN:%.+]], {{.+}} [[T_VAR_IN:%.+]], {{.+}} [[S_ARR_IN:%.+]], {{.+}} [[VAR_IN:%.+]], {{.+}} [[SVAR_IN:%.+]]) + +// CHECK: {{.+}} = alloca i{{[0-9]+}}*, +// CHECK: {{.+}} = alloca i{{[0-9]+}}*, +// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*, +// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}*, +// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_FLOAT_TY]]*, +// CHECK: [[SVAR_ADDR:%.+]] = alloca i{{[0-9]+}}*, +// skip loop variables +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]], +// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_FLOAT_TY]]*, +// CHECK: [[S_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, + +// copy from parameters to local address variables +// CHECK: store [2 x i{{[0-9]+}}]* [[VEC_IN]], [2 x i{{[0-9]+}}]** [[VEC_ADDR]], +// CHECK: store i{{[0-9]+}}* [[T_VAR_IN]], i{{[0-9]+}}** [[T_VAR_ADDR]], +// CHECK: store [2 x [[S_FLOAT_TY]]]* [[S_ARR_IN]], [2 x [[S_FLOAT_TY]]]** [[S_ARR_ADDR]], +// CHECK: store [[S_FLOAT_TY]]* [[VAR_IN]], [[S_FLOAT_TY]]** [[VAR_ADDR]], +// CHECK: store i{{[0-9]+}}* [[S_VAR_IN]], i{{[0-9]+}}** [[SVAR_ADDR]], + +// load content of local address variables +// CHECK: [[VEC_ADDR_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_ADDR]], +// CHECK: [[T_VAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_ADDR]], +// CHECK: [[S_ARR_ADDR_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[S_ARR_ADDR]], +// CHECK: [[SVAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_ADDR]], +// CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[OMP_IS_LAST]], + +// call constructor for s_arr +// CHECK: [[S_ARR_BGN:%.+]] = getelementptr{{.+}} [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], +// CHECK: [[S_ARR_END:%.+]] = getelementptr {{.+}} [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_BGN]], +// CHECK: br label %[[S_ARR_CST_LOOP:.+]] +// CHECK: [[S_ARR_CST_LOOP]]: +// CHECK: [[S_ARR_CTOR:%.+]] = phi {{.+}} +// CHECK: call void [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_CTOR]]) +// CHECK: [[S_ARR_NEXT:%.+]] = getelementptr {{.+}} [[S_ARR_CTOR]], +// CHECK: [[S_ARR_DONE:%.+]] = icmp {{.+}} [[S_ARR_NEXT]], [[S_ARR_END]] +// CHECK: br i1 [[S_ARR_DONE]], label %[[S_ARR_CST_END:.+]], label %[[S_ARR_CST_LOOP]] +// CHECK: [[S_ARR_CST_END]]: +// CHECK: [[VAR_ADDR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_ADDR]], +// CHECK: call void [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]]) +// CHECK: store [[S_FLOAT_TY]]* [[VAR_PRIV]], [[S_FLOAT_TY]]** [[TMP_PRIV]], + +// CHECK: call void @__kmpc_for_static_init_4( + +// loop body +// assignment: vec[i] = t_var; +// CHECK: [[T_VAR_PRIV_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]], +// CHECK: [[VEC_PTR:%.+]] = getelementptr inbounds [2 x i{{[0-9]+}}], [2 x i{{[0-9]+}}]* [[VEC_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} {{.+}} +// CHECK: store i{{[0-9]+}} [[T_VAR_PRIV_VAL]], i{{[0-9]+}}* [[VEC_PTR]], + +// assignment: s_arr[i] = var; +// CHECK-DAG: [[S_ARR_PTR:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], +// CHECK-DAG: [[TMP_VAL:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP_PRIV]], +// CHECK-DAG: [[S_ARR_PTR_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_PTR]] to i8* +// CHECK-DAG: [[TMP_VAL_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[TMP_VAL]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_PTR_BCAST]], i8* [[TMP_VAL_BCAST]], + +// CHECK: call void @__kmpc_for_static_fini( + +// lastprivates +// CHECK: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]], +// CHECK: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0 +// CHECK: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]] + +// CHECK: [[OMP_LASTPRIV_BLOCK]]: +// CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]], +// CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_ADDR_REF]], +// CHECK: [[BCAST_VEC_ADDR_REF:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_ADDR_REF]] to i8* +// CHECK: [[BCAST_VEC_PRIV:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[BCAST_VEC_ADDR_REF]], i8* [[BCAST_VEC_PRIV]], +// CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_ADDR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 +// CHECK: [[S_ARR_PRIV_BCAST:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]] to [[S_FLOAT_TY]]* +// CHECK: [[S_ARR_BEGIN_GEP:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_BEGIN]], i{{[0-9]+}} 2 +// CHECK: [[S_ARR_IS_EMPTY:%.+]] = icmp eq [[S_FLOAT_TY]]* [[S_ARR_BEGIN]], [[S_ARR_BEGIN_GEP]] +// CHECK: br i1 [[S_ARR_IS_EMPTY]], label %[[S_ARR_COPY_DONE:.+]], label %[[S_ARR_COPY_BLOCK:.+]] +// CHECK: [[S_ARR_COPY_BLOCK]]: +// CHECK: [[S_ARR_SRC_EL:%.+]] = phi [[S_FLOAT_TY]]*{{.+}} +// CHECK: [[S_ARR_DST_EL:%.+]] = phi [[S_FLOAT_TY]]*{{.+}} +// CHECK: [[S_ARR_DST_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_DST_EL]] to i8* +// CHECK: [[S_ARR_SRC_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[S_ARR_SRC_EL]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_DST_BCAST]], i8* [[S_ARR_SRC_BCAST]]{{.+}}) +// CHECK: [[S_ARR_DST_NEXT:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_DST_EL]], i{{[0-9]+}} 1 +// CHECK: [[S_ARR_SRC_NEXT:%.+]] = getelementptr{{.+}} +// CHECK: [[CPY_IS_FINISHED:%.+]] = icmp eq [[S_FLOAT_TY]]* [[S_ARR_DST_NEXT]], [[S_ARR_BEGIN_GEP]] +// CHECK: br i1 [[CPY_IS_FINISHED]], label %[[S_ARR_COPY_DONE]], label %[[S_ARR_COPY_BLOCK]] +// CHECK: [[S_ARR_COPY_DONE]]: +// CHECK: [[TMP_VAL1:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[TMP_PRIV]], +// CHECK: [[VAR_ADDR_REF_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[VAR_ADDR_REF]] to i8* +// CHECK: [[TMP_VAL1_BCAST:%.+]] = bitcast [[S_FLOAT_TY]]* [[TMP_VAL1]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VAR_ADDR_REF_BCAST]], i8* [[TMP_VAL1_BCAST]],{{.+}}) +// CHECK: [[SVAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[S_VAR_PRIV]], +// CHECK: store i{{[0-9]+}} [[SVAR_VAL]], i{{[0-9]+}}* [[SVAR_ADDR_REF]], +// CHECK: ret void + +// template tmain +// CHECK: define{{.*}} i{{[0-9]+}} [[TMAIN_INT:@.+]]() +// CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]], +// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]]) +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOAD_FUN_1:@.+]](i{{[0-9]+}} {{.+}}, [2 x i{{[0-9]+}}]* {{.+}}, [2 x [[S_INT_TY]]]* {{.+}}, [[S_INT_TY]]* {{.+}}) +// CHECK: ret + +// CHECK: define internal void [[OFFLOAD_FUN_1]]( +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, +// CHECK: ret + +// CHECK: define internal void [[OMP_OUTLINED_1:@.+]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i{{[0-9]+}}*{{.+}} [[T_VAR_IN:%.+]], [2 x i{{[0-9]+}}]*{{.+}} [[VEC_IN:%.+]], [2 x [[S_INT_TY]]]*{{.+}} [[S_ARR_IN:%.+]], [[S_INT_TY]]*{{.+}} [[VAR_IN:%.+]]) +// skip alloca of global_tid and bound_tid +// CHECK: {{.+}} = alloca i{{[0-9]+}}*, +// CHECK: {{.+}} = alloca i{{[0-9]+}}*, +// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}*, +// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*, +// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_INT_TY]]]*, +// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_INT_TY]]*, +// CHECK: [[TMP:%.+]] = alloca [[S_INT_TY]]*, +// skip loop variables +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]], +// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_INT_TY]]*, + +// skip init of bound and global tid +// CHECK: store i{{[0-9]+}}* {{.*}}, +// CHECK: store i{{[0-9]+}}* {{.*}}, +// copy from parameters to local address variables +// CHECK: store i{{[0-9]+}}* [[T_VAR_IN]], i{{[0-9]+}}** [[T_VAR_ADDR]], +// CHECK: store [2 x i{{[0-9]+}}]* [[VEC_IN]], [2 x i{{[0-9]+}}]** [[VEC_ADDR]], +// CHECK: store [2 x [[S_INT_TY]]]* [[S_ARR_IN]], [2 x [[S_INT_TY]]]** [[S_ARR_ADDR]], +// CHECK: store [[S_INT_TY]]* [[VAR_IN]], [[S_INT_TY]]** [[VAR_ADDR]], + +// load content of local address variables +// CHECK: [[T_VAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_ADDR]], +// CHECK: [[VEC_ADDR_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_ADDR]], +// CHECK: [[S_ARR_ADDR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_ADDR]], +// CHECK: [[VAR_ADDR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_ADDR]], +// CHECK-DAG: store [[S_INT_TY]]* [[VAR_ADDR_REF]], [[S_INT_TY]]** [[TMP]], +// CHECK-DAG: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[OMP_IS_LAST]], +// CHECK-DAG: [[TMP_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP]], + +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: [[TMP_PRIV_VAL:%.+]] = load {{.+}}, {{.+}} [[TMP_PRIV]], +// CHECK-64: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[VEC_PRIV]], {{.+}} [[T_VAR_PRIV]], {{.+}} [[S_ARR_PRIV]], {{.+}} [[TMP_PRIV_VAL]]) +// CHECK-32: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to void ({{.+}})*), {{.+}}, {{.+}}, {{.+}} [[VEC_PRIV]], {{.+}} [[T_VAR_PRIV]], {{.+}} [[S_ARR_PRIV]], {{.+}} [[TMP_PRIV_VAL]]) + +// CHECK: call void @__kmpc_for_static_fini( + +// lastprivates +// CHECK: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]], +// CHECK: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0 +// CHECK: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]] + +// CHECK: [[OMP_LASTPRIV_BLOCK]]: +// CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]], +// CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_ADDR_REF]], +// CHECK: [[BCAST_VEC_ADDR_REF:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_ADDR_REF]] to i8* +// CHECK: [[BCAST_VEC_PRIV:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[BCAST_VEC_ADDR_REF]], i8* [[BCAST_VEC_PRIV]], +// CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_ADDR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 +// CHECK: [[S_ARR_PRIV_BCAST:%.+]] = bitcast [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]] to [[S_INT_TY]]* +// CHECK: [[S_ARR_BEGIN_GEP:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_BEGIN]], i{{[0-9]+}} 2 +// CHECK: [[S_ARR_IS_EMPTY:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_BEGIN]], [[S_ARR_BEGIN_GEP]] +// CHECK: br i1 [[S_ARR_IS_EMPTY]], label %[[S_ARR_COPY_DONE:.+]], label %[[S_ARR_COPY_BLOCK:.+]] +// CHECK: [[S_ARR_COPY_BLOCK]]: +// CHECK: [[S_ARR_SRC_EL:%.+]] = phi [[S_INT_TY]]*{{.+}} +// CHECK: [[S_ARR_DST_EL:%.+]] = phi [[S_INT_TY]]*{{.+}} +// CHECK: [[S_ARR_DST_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_DST_EL]] to i8* +// CHECK: [[S_ARR_SRC_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_SRC_EL]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_DST_BCAST]], i8* [[S_ARR_SRC_BCAST]]{{.+}}) +// CHECK: [[S_ARR_DST_NEXT:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_DST_EL]], i{{[0-9]+}} 1 +// CHECK: [[S_ARR_SRC_NEXT:%.+]] = getelementptr{{.+}} +// CHECK: [[CPY_IS_FINISHED:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_DST_NEXT]], [[S_ARR_BEGIN_GEP]] +// CHECK: br i1 [[CPY_IS_FINISHED]], label %[[S_ARR_COPY_DONE]], label %[[S_ARR_COPY_BLOCK]] +// CHECK: [[S_ARR_COPY_DONE]]: +// CHECK: [[TMP_VAL:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP_PRIV]], +// CHECK: [[VAR_ADDR_REF_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[TMP_REF]] to i8* +// CHECK: [[TMP_VAL_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[TMP_VAL]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VAR_ADDR_REF_BCAST]], i8* [[TMP_VAL_BCAST]],{{.+}}) +// CHECK: ret void + +// outlined function for 'parallel for' +// CHECK-64: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[VEC_IN:%.+]], {{.+}} [[T_VAR_IN:%.+]], {{.+}} [[S_ARR_IN:%.+]], {{.+}} [[VAR_IN:%.+]]) +// CHECK-32: define{{.+}} void [[OMP_PARFOR_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, {{.+}}, {{.+}}, {{.+}} [[VEC_IN:%.+]], {{.+}} [[T_VAR_IN:%.+]], {{.+}} [[S_ARR_IN:%.+]], {{.+}} [[VAR_IN:%.+]]) + +// CHECK: {{.+}} = alloca i{{[0-9]+}}*, +// CHECK: {{.+}} = alloca i{{[0-9]+}}*, +// CHECK: [[VEC_ADDR:%.+]] = alloca [2 x i{{[0-9]+}}]*, +// CHECK: [[T_VAR_ADDR:%.+]] = alloca i{{[0-9]+}}*, +// CHECK: [[S_ARR_ADDR:%.+]] = alloca [2 x [[S_INT_TY]]]*, +// CHECK: [[VAR_ADDR:%.+]] = alloca [[S_INT_TY]]*, +// skip loop variables +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: {{.+}} = alloca i{{[0-9]+}}, +// CHECK: [[OMP_IS_LAST:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]], +// CHECK: [[TMP_PRIV:%.+]] = alloca [[S_INT_TY]]*, + +// copy from parameters to local address variables +// CHECK: store [2 x i{{[0-9]+}}]* [[VEC_IN]], [2 x i{{[0-9]+}}]** [[VEC_ADDR]], +// CHECK: store i{{[0-9]+}}* [[T_VAR_IN]], i{{[0-9]+}}** [[T_VAR_ADDR]], +// CHECK: store [2 x [[S_INT_TY]]]* [[S_ARR_IN]], [2 x [[S_INT_TY]]]** [[S_ARR_ADDR]], +// CHECK: store [[S_INT_TY]]* [[VAR_IN]], [[S_INT_TY]]** [[VAR_ADDR]], + +// load content of local address variables +// CHECK: [[VEC_ADDR_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_ADDR]], +// CHECK: [[T_VAR_ADDR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_ADDR]], +// CHECK: [[S_ARR_ADDR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_ADDR]], +// CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[OMP_IS_LAST]], + +// call constructor for s_arr +// CHECK: [[S_ARR_BGN:%.+]] = getelementptr{{.+}} [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], +// CHECK: [[S_ARR_END:%.+]] = getelementptr {{.+}} [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_BGN]], +// CHECK: br label %[[S_ARR_CST_LOOP:.+]] +// CHECK: [[S_ARR_CST_LOOP]]: +// CHECK: [[S_ARR_CTOR:%.+]] = phi {{.+}} +// CHECK: call void [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_CTOR]]) +// CHECK: [[S_ARR_NEXT:%.+]] = getelementptr {{.+}} [[S_ARR_CTOR]], +// CHECK: [[S_ARR_DONE:%.+]] = icmp {{.+}} [[S_ARR_NEXT]], [[S_ARR_END]] +// CHECK: br i1 [[S_ARR_DONE]], label %[[S_ARR_CST_END:.+]], label %[[S_ARR_CST_LOOP]] +// CHECK: [[S_ARR_CST_END]]: +// CHECK: [[VAR_ADDR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_ADDR]], +// CHECK: call void [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]]) +// CHECK: store [[S_INT_TY]]* [[VAR_PRIV]], [[S_INT_TY]]** [[TMP_PRIV]], + +// CHECK: call void @__kmpc_for_static_init_4( + +// assignment: vec[i] = t_var; +// CHECK: [[IV_VAL:%.+]] = +// CHECK: [[T_VAR_PRIV_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]], +// CHECK: [[VEC_PTR:%.+]] = getelementptr inbounds [2 x i{{[0-9]+}}], [2 x i{{[0-9]+}}]* [[VEC_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} {{.+}} +// CHECK: store i{{[0-9]+}} [[T_VAR_PRIV_VAL]], i{{[0-9]+}}* [[VEC_PTR]], + +// assignment: s_arr[i] = var; +// CHECK-DAG: [[S_ARR_PTR:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], +// CHECK-DAG: [[TMP_VAL:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP_PRIV]], +// CHECK-DAG: [[S_ARR_PTR_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_PTR]] to i8* +// CHECK-DAG: [[TMP_VAL_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[TMP_VAL]] to i8* +// CHECK-DAG: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_PTR_BCAST]], i8* [[TMP_VAL_BCAST]], + +// CHECK: call void @__kmpc_for_static_fini( + +// lastprivates +// CHECK: [[OMP_IS_LAST_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[OMP_IS_LAST]], +// CHECK: [[IS_LAST_IT:%.+]] = icmp ne i{{[0-9]+}} [[OMP_IS_LAST_VAL]], 0 +// CHECK: br i1 [[IS_LAST_IT]], label %[[OMP_LASTPRIV_BLOCK:.+]], label %[[OMP_LASTPRIV_DONE:.+]] + +// CHECK: [[OMP_LASTPRIV_BLOCK]]: +// CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_PRIV]], +// CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_ADDR_REF]], +// CHECK: [[BCAST_VEC_ADDR_REF:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_ADDR_REF]] to i8* +// CHECK: [[BCAST_VEC_PRIV:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[BCAST_VEC_ADDR_REF]], i8* [[BCAST_VEC_PRIV]], +// CHECK: [[S_ARR_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_ADDR_REF]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 +// CHECK: [[S_ARR_PRIV_BCAST:%.+]] = bitcast [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]] to [[S_INT_TY]]* +// CHECK: [[S_ARR_BEGIN_GEP:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_BEGIN]], i{{[0-9]+}} 2 +// CHECK: [[S_ARR_IS_EMPTY:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_BEGIN]], [[S_ARR_BEGIN_GEP]] +// CHECK: br i1 [[S_ARR_IS_EMPTY]], label %[[S_ARR_COPY_DONE:.+]], label %[[S_ARR_COPY_BLOCK:.+]] +// CHECK: [[S_ARR_COPY_BLOCK]]: +// CHECK: [[S_ARR_SRC_EL:%.+]] = phi [[S_INT_TY]]*{{.+}} +// CHECK: [[S_ARR_DST_EL:%.+]] = phi [[S_INT_TY]]*{{.+}} +// CHECK: [[S_ARR_DST_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_DST_EL]] to i8* +// CHECK: [[S_ARR_SRC_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[S_ARR_SRC_EL]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[S_ARR_DST_BCAST]], i8* [[S_ARR_SRC_BCAST]]{{.+}}) +// CHECK: [[S_ARR_DST_NEXT:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_DST_EL]], i{{[0-9]+}} 1 +// CHECK: [[S_ARR_SRC_NEXT:%.+]] = getelementptr{{.+}} +// CHECK: [[CPY_IS_FINISHED:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_DST_NEXT]], [[S_ARR_BEGIN_GEP]] +// CHECK: br i1 [[CPY_IS_FINISHED]], label %[[S_ARR_COPY_DONE]], label %[[S_ARR_COPY_BLOCK]] +// CHECK: [[S_ARR_COPY_DONE]]: +// CHECK: [[TMP_VAL1:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[TMP_PRIV]], +// CHECK: [[VAR_ADDR_REF_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[VAR_ADDR_REF]] to i8* +// CHECK: [[TMP_VAL1_BCAST:%.+]] = bitcast [[S_INT_TY]]* [[TMP_VAL1]] to i8* +// CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VAR_ADDR_REF_BCAST]], i8* [[TMP_VAL1_BCAST]],{{.+}}) +// CHECK: ret void + +#endif diff --git a/test/OpenMP/distribute_parallel_for_num_threads_codegen.cpp b/test/OpenMP/distribute_parallel_for_num_threads_codegen.cpp new file mode 100644 index 0000000000000..89e9905044399 --- /dev/null +++ b/test/OpenMP/distribute_parallel_for_num_threads_codegen.cpp @@ -0,0 +1,121 @@ +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -triple %itanium_abi_triple -emit-llvm %s -fexceptions -fcxx-exceptions -o - | FileCheck %s +// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -std=c++11 -triple %itanium_abi_triple -fexceptions -fcxx-exceptions -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -fopenmp-targets=powerpc64le-ibm-linux-gnu -x c++ -triple %itanium_abi_triple -fexceptions -fcxx-exceptions -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s +// expected-no-diagnostics +#ifndef HEADER +#define HEADER + +typedef __INTPTR_TYPE__ intptr_t; + +// CHECK-DAG: [[IDENT_T_TY:%.+]] = type { i32, i32, i32, i32, i8* } +// CHECK-DAG: [[S_TY:%.+]] = type { [[INTPTR_T_TY:i[0-9]+]], [[INTPTR_T_TY]], [[INTPTR_T_TY]] } +// CHECK-DAG: [[STR:@.+]] = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00" +// CHECK-DAG: [[DEF_LOC_2:@.+]] = private unnamed_addr constant [[IDENT_T_TY]] { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* [[STR]], i32 0, i32 0) } + +void foo(); + +struct S { + intptr_t a, b, c; + S(intptr_t a) : a(a) {} + operator char() { return a; } + ~S() {} +}; + +template <typename T, int C> +int tmain() { +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for num_threads(C) + for (int i = 0; i < 100; i++) + foo(); +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for num_threads(T(23)) + for (int i = 0; i < 100; i++) + foo(); + return 0; +} + +int main() { + S s(0); + char a = s; +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_0:@.+]]( +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOADING_FUN_1:@.+]]( +// CHECK: invoke{{.+}} [[TMAIN_5:@.+]]() +// CHECK: invoke{{.+}} [[TMAIN_1:@.+]]() +#pragma omp target +#pragma omp teams + // CHECK: define internal void [[OFFLOADING_FUN_0]]( + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_TEAMS_OUTLINED_0:@.+]] to {{.+}}) +#pragma omp distribute parallel for num_threads(2) + for (int i = 0; i < 100; i++) { + // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_0]]( + // CHECK: call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 2) + // CHECK: call {{.*}}void {{.*}} @__kmpc_fork_call( + foo(); + } +#pragma omp target +#pragma omp teams + // CHECK: define internal void [[OFFLOADING_FUN_1]]( + + // CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 1, {{.+}}* [[OMP_TEAMS_OUTLINED_1:@.+]] to {{.+}}) +#pragma omp distribute parallel for num_threads(a) + for (int i = 0; i < 100; i++) { + // CHECK: define{{.+}} void [[OMP_TEAMS_OUTLINED_1]]( + // CHECK-DAG: [[A_ADDR:%.+]] = alloca i8*, + // CHECK-DAG: [[A_REF:%.+]] = load i8*, i8** [[A_ADDR]], + // CHECK-DAG: [[A_VAL:%.+]] = load i8, i8* [[A_REF]], + // CHECK-DAG: [[A_EXT:%.+]] = sext i8 [[A_VAL]] to {{.+}} + // CHECK: call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 [[A_EXT]]) + // CHECK: call {{.*}}void {{.*}} @__kmpc_fork_call( + foo(); + } + return a + tmain<char, 5>() + tmain<S, 1>(); +} + +// tmain 5 +// CHECK-DAG: define {{.*}}i{{[0-9]+}} [[TMAIN_5]]() +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[T_OFFLOADING_FUN_0:@.+]]( +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[T_OFFLOADING_FUN_1:@.+]]( + +// tmain 1 +// CHECK-DAG: define {{.*}}i{{[0-9]+}} [[TMAIN_1]]() +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[T_OFFLOADING_FUN_2:@.+]]( +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[T_OFFLOADING_FUN_3:@.+]]( + +// CHECK: define internal void [[T_OFFLOADING_FUN_0]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[T_OMP_TEAMS_OUTLINED_0:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[T_OMP_TEAMS_OUTLINED_0]]( +// CHECK: call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 5) +// CHECK: call {{.*}}void {{.*}} @__kmpc_fork_call( + +// CHECK: define internal void [[T_OFFLOADING_FUN_1]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[T_OMP_TEAMS_OUTLINED_1:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[T_OMP_TEAMS_OUTLINED_1]]( +// CHECK: call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 23) +// CHECK: call {{.*}}void {{.*}} @__kmpc_fork_call( + +// CHECK: define internal void [[T_OFFLOADING_FUN_2]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[T_OMP_TEAMS_OUTLINED_2:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[T_OMP_TEAMS_OUTLINED_2]]( +// CHECK: call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 1) +// CHECK: call {{.*}}void {{.*}} @__kmpc_fork_call( + +// CHECK: define internal void [[T_OFFLOADING_FUN_3]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[T_OMP_TEAMS_OUTLINED_3:@.+]] to {{.+}}) + +// CHECK: define{{.+}} void [[T_OMP_TEAMS_OUTLINED_3]]( +// CHECK-DAG: [[CALL_RES:%.+]] = invoke{{.*}} i8 [[S_TY_CHAR_OP:@.+]]([[S_TY]]* {{.+}}) +// CHECK-DAG: [[CALL_RES_SEXT:%.+]] = sext i8 [[CALL_RES]] to {{.+}} +// CHECK: call {{.*}}void @__kmpc_push_num_threads([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 {{.+}}, i32 [[CALL_RES_SEXT]]) +// CHECK: call {{.*}}void {{.*}} @__kmpc_fork_call( +#endif diff --git a/test/OpenMP/distribute_parallel_for_private_codegen.cpp b/test/OpenMP/distribute_parallel_for_private_codegen.cpp new file mode 100644 index 0000000000000..d8d8a9a1edec9 --- /dev/null +++ b/test/OpenMP/distribute_parallel_for_private_codegen.cpp @@ -0,0 +1,297 @@ +// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64 +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-64 +// RUN: %clang_cc1 -DLAMBDA -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32 +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -DLAMBDA -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix LAMBDA --check-prefix LAMBDA-32 + +// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64 +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64 +// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32 +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32 +// expected-no-diagnostics +#ifndef HEADER +#define HEADER + +template <class T> +struct S { + T f; + S(T a) : f(a) {} + S() : f() {} + operator T() { return T(); } + ~S() {} +}; + +// CHECK: [[S_FLOAT_TY:%.+]] = type { float } +// CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} } +template <typename T> +T tmain() { + S<T> test; + T t_var = T(); + T vec[] = {1, 2}; + S<T> s_arr[] = {1, 2}; + S<T> &var = test; + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for private(t_var, vec, s_arr, s_arr, var, var) + for (int i = 0; i < 2; ++i) { + vec[i] = t_var; + s_arr[i] = var; + } + return T(); +} + +int main() { + static int svar; + volatile double g; + volatile double &g1 = g; + + #ifdef LAMBDA + // LAMBDA-LABEL: @main + // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]]( + [&]() { + static float sfvar; + // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]]( + // LAMBDA: call i{{[0-9]+}} @__tgt_target_teams( + // LAMBDA: call void [[OFFLOADING_FUN:@.+]]( + + // LAMBDA: define{{.+}} void [[OFFLOADING_FUN]]() + // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, i32 0, {{.+}}* [[OMP_OUTLINED:@.+]] to {{.+}}) + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for private(g, g1, svar, sfvar) + for (int i = 0; i < 2; ++i) { + // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_OUTLINED]](i32* noalias %{{.+}}, i32* noalias %{{.+}}) + // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double, + // LAMBDA: [[G1_PRIVATE_ADDR:%.+]] = alloca double, + // LAMBDA: [[TMP_PRIVATE_ADDR:%.+]] = alloca double*, + // LAMBDA: [[SVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SFVAR_PRIVATE_ADDR:%.+]] = alloca float, + // LAMBDA: store double* [[G1_PRIVATE_ADDR]], double** [[TMP_PRIVATE_ADDR]], + // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4( + // LAMBDA: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED:@.+]] to {{.+}}, + // LAMBDA: call {{.*}}void @__kmpc_for_static_fini( + // LAMBDA: ret void + + // LAMBDA: define{{.+}} void [[OMP_PARFOR_OUTLINED]]( + // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double, + // LAMBDA: [[G1_PRIVATE_ADDR:%.+]] = alloca double, + // LAMBDA: [[TMP_PRIVATE_ADDR:%.+]] = alloca double*, + // LAMBDA: [[SVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}}, + // LAMBDA: [[SFVAR_PRIVATE_ADDR:%.+]] = alloca float, + + g = 1; + g1 = 1; + svar = 3; + sfvar = 4.0; + // LAMBDA: store double* [[G1_PRIVATE_ADDR]], double** [[TMP_PRIVATE_ADDR]], + // LAMBDA: store double 1.0{{.+}}, double* [[G_PRIVATE_ADDR]], + // LAMBDA: store i{{[0-9]+}} 3, i{{[0-9]+}}* [[SVAR_PRIVATE_ADDR]], + // LAMBDA: store float 4.0{{.+}}, float* [[SFVAR_PRIVATE_ADDR]], + // LAMBDA: [[G_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 + // LAMBDA: store double* [[G_PRIVATE_ADDR]], double** [[G_PRIVATE_ADDR_REF]], + // LAMBDA: [[TMP_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1 + // LAMBDA: [[G1_PRIVATE_ADDR_FROM_TMP:%.+]] = load double*, double** [[TMP_PRIVATE_ADDR]], + // LAMBDA: store double* [[G1_PRIVATE_ADDR_FROM_TMP]], double** [[TMP_PRIVATE_ADDR_REF]], + // LAMBDA: [[SVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 2 + // LAMBDA: store i{{[0-9]+}}* [[SVAR_PRIVATE_ADDR]], i{{[0-9]+}}** [[SVAR_PRIVATE_ADDR_REF]] + // LAMBDA: [[SFVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 3 + // LAMBDA: store float* [[SFVAR_PRIVATE_ADDR]], float** [[SFVAR_PRIVATE_ADDR_REF]] + // LAMBDA: call{{.*}} void [[INNER_LAMBDA:@.+]](%{{.+}}* [[ARG]]) + // LAMBDA: call {{.*}}void @__kmpc_for_static_fini( + // LAMBDA: ret void + [&]() { + // LAMBDA: define {{.+}} void [[INNER_LAMBDA]](%{{.+}}* [[ARG_PTR:%.+]]) + // LAMBDA: store %{{.+}}* [[ARG_PTR]], %{{.+}}** [[ARG_PTR_REF:%.+]], + g = 2; + g1 = 2; + svar = 4; + sfvar = 8.0; + // LAMBDA: [[ARG_PTR:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_PTR_REF]] + // LAMBDA: [[G_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0 + // LAMBDA: [[G_REF:%.+]] = load double*, double** [[G_PTR_REF]] + // LAMBDA: store double 2.0{{.+}}, double* [[G_REF]] + + // LAMBDA: [[TMP_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 1 + // LAMBDA: [[G1_REF:%.+]] = load double*, double** [[TMP_PTR_REF]] + // LAMBDA: store double 2.0{{.+}}, double* [[G1_REF]], + // LAMBDA: [[SVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 2 + // LAMBDA: [[SVAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SVAR_PTR_REF]] + // LAMBDA: store i{{[0-9]+}} 4, i{{[0-9]+}}* [[SVAR_REF]] + // LAMBDA: [[SFVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 3 + // LAMBDA: [[SFVAR_REF:%.+]] = load float*, float** [[SFVAR_PTR_REF]] + // LAMBDA: store float 8.0{{.+}}, float* [[SFVAR_REF]] + }(); + } + }(); + return 0; + #else + S<float> test; + int t_var = 0; + int vec[] = {1, 2}; + S<float> s_arr[] = {1, 2}; + S<float> &var = test; + + #pragma omp target + #pragma omp teams + #pragma omp distribute parallel for private(t_var, vec, s_arr, s_arr, var, var, svar) + for (int i = 0; i < 2; ++i) { + vec[i] = t_var; + s_arr[i] = var; + } + return tmain<int>(); + #endif +} + +// CHECK: define{{.*}} i{{[0-9]+}} @main() +// CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]], +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]]) +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOAD_FUN_0:@.+]]( + +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR:@.+]]([[S_FLOAT_TY]]* [[TEST]]) +// CHECK: ret + +// CHECK: define{{.+}} [[OFFLOAD_FUN_0]]() +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[OMP_OUTLINED_0:@.+]] to void +// CHECK: ret +// +// CHECK: define internal void [[OMP_OUTLINED_0]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}) +// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]], +// CHECK-NOT: alloca [2 x [[S_FLOAT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]], +// CHECK-NOT: alloca [[S_FLOAT_TY]], +// CHECK: [[S_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]] +// CHECK-NOT: [[T_VAR_PRIV]] +// CHECK-NOT: [[VEC_PRIV]] +// this is the ctor loop +// CHECK: {{.+}}: +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_FLOAT_TY]]* +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]]) +// CHECK-NOT: [[T_VAR_PRIV]] +// CHECK-NOT: [[VEC_PRIV]] +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]]) +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_0:@.+]] to {{.+}}, +// CHECK: call void @__kmpc_for_static_fini( + +// call destructors: var.. +// CHECK-DAG: call {{.+}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]]) + +// ..and s_arr +// CHECK: {{.+}}: +// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_FLOAT_TY]]* +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]], +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]]) + +// CHECK: ret void + +// By OpenMP specifications, private applies to both distribute and parallel for. +// However, the support for 'private' of 'parallel' is only used when 'parallel' +// is found alone. Therefore we only have one 'private' support for 'parallel for' +// in combination +// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_0]]( +// CHECK: [[T_VAR_PRIV:%t_var+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%vec+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%s_arr+]] = alloca [2 x [[S_FLOAT_TY]]], +// CHECK-NOT: alloca [2 x [[S_FLOAT_TY]]], +// CHECK: [[VAR_PRIV:%var+]] = alloca [[S_FLOAT_TY]], +// CHECK-NOT: alloca [[S_FLOAT_TY]], +// CHECK: [[S_VAR_PRIV:%svar+]] = alloca i{{[0-9]+}}, +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]] +// CHECK-NOT: [[T_VAR_PRIV]] +// CHECK-NOT: [[VEC_PRIV]] +// this is the ctor loop +// CHECK: {{.+}}: +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_FLOAT_TY]]* +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]]) +// CHECK-NOT: [[T_VAR_PRIV]] +// CHECK-NOT: [[VEC_PRIV]] +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]]) +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call void @__kmpc_for_static_fini( + +// call destructors: var.. +// CHECK-DAG: call {{.+}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]]) + +// ..and s_arr +// CHECK: {{.+}}: +// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_FLOAT_TY]]* +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]], +// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_DESTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]]) + +// CHECK: ret void + +// template tmain with S_INT_TY +// CHECK: define{{.*}} i{{[0-9]+}} [[TMAIN_INT:@.+]]() +// CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]], +// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]]) +// CHECK: call i{{[0-9]+}} @__tgt_target_teams( +// CHECK: call void [[OFFLOAD_FUN_1:@.+]]( +// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR:@.+]]([[S_INT_TY]]* [[TEST]]) +// CHECK: ret + +// CHECK: ret + +// CHECK: define internal void [[OFFLOAD_FUN_1]]() +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_teams(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[OMP_OUTLINED_1:@.+]] to void +// CHECK: ret +// +// CHECK: define internal void [[OMP_OUTLINED_1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}) +// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]], +// CHECK-NOT: alloca [2 x [[S_INT_TY]]], +// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]], +// CHECK-NOT: alloca [[S_INT_TY]], +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]] +// CHECK-NOT: [[T_VAR_PRIV]] +// CHECK-NOT: [[VEC_PRIV]] +// CHECK: {{.+}}: +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_INT_TY]]* +// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]]) +// CHECK-NOT: [[T_VAR_PRIV]] +// CHECK-NOT: [[VEC_PRIV]] +// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]]) +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call{{.+}} @__kmpc_fork_call({{.+}}, {{.+}}, {{.+}}[[OMP_PARFOR_OUTLINED_1:@.+]] to {{.+}}, +// CHECK: call void @__kmpc_for_static_fini( +// CHECK: ret void + +// CHECK: define{{.+}} void [[OMP_PARFOR_OUTLINED_1]]( +// CHECK: [[T_VAR_PRIV:%t_var+]] = alloca i{{[0-9]+}}, +// CHECK: [[VEC_PRIV:%vec+]] = alloca [2 x i{{[0-9]+}}], +// CHECK: [[S_ARR_PRIV:%s_arr+]] = alloca [2 x [[S_INT_TY]]], +// CHECK-NOT: alloca [2 x [[S_INT_TY]]], +// CHECK: [[VAR_PRIV:%var+]] = alloca [[S_INT_TY]], +// CHECK-NOT: alloca [[S_INT_TY]], +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]] +// CHECK-NOT: [[T_VAR_PRIV]] +// CHECK-NOT: [[VEC_PRIV]] +// this is the ctor loop +// CHECK: {{.+}}: +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_INT_TY]]* +// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]]) +// CHECK-NOT: [[T_VAR_PRIV]] +// CHECK-NOT: [[VEC_PRIV]] +// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]]) +// CHECK: call void @__kmpc_for_static_init_4( +// CHECK: call void @__kmpc_for_static_fini( + +// call destructors: var.. +// CHECK-DAG: call {{.+}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]]) + +// ..and s_arr +// CHECK: {{.+}}: +// CHECK: [[S_ARR_EL_PAST:%.+]] = phi [[S_INT_TY]]* +// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = getelementptr {{.+}}, {{.+}} [[S_ARR_EL_PAST]], +// CHECK: call {{.*}} [[S_INT_TY_DEF_DESTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]]) + +// CHECK: ret void + +#endif diff --git a/test/OpenMP/distribute_parallel_for_proc_bind_codegen.cpp b/test/OpenMP/distribute_parallel_for_proc_bind_codegen.cpp new file mode 100644 index 0000000000000..958b4e782f109 --- /dev/null +++ b/test/OpenMP/distribute_parallel_for_proc_bind_codegen.cpp @@ -0,0 +1,93 @@ +// add -fopenmp-targets + +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s +// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s +// expected-no-diagnostics +#ifndef HEADER +#define HEADER + +typedef __INTPTR_TYPE__ intptr_t; + +// CHECK-DAG: [[IDENT_T_TY:%.+]] = type { i32, i32, i32, i32, i8* } +// CHECK-DAG: [[STR:@.+]] = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00" +// CHECK-DAG: [[DEF_LOC_2:@.+]] = private unnamed_addr constant [[IDENT_T_TY]] { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* [[STR]], i32 0, i32 0) } + +void foo(); + +struct S { + intptr_t a, b, c; + S(intptr_t a) : a(a) {} + operator char() { return a; } + ~S() {} +}; + +template <typename T> +T tmain() { +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for proc_bind(master) + for(int i = 0; i < 1000; i++) {} + return T(); +} + +int main() { + // CHECK-LABEL: @main +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for proc_bind(spread) + for(int i = 0; i < 1000; i++) {} +#pragma omp target +#pragma omp teams +#pragma omp distribute parallel for proc_bind(close) + for(int i = 0; i < 1000; i++) {} + return tmain<int>(); +} + +// CHECK: call {{.*}}@__tgt_target_teams({{.+}}) +// CHECK: call void [[OFFL1:@.+]]() +// CHECK: call {{.*}}@__tgt_target_teams({{.+}}) +// CHECK: call void [[OFFL2:@.+]]() +// CHECK: [[CALL_RET:%.+]] = call{{.+}} i32 [[TMAIN:@.+]]() +// CHECK: ret i32 [[CALL_RET]] + +// CHECK: define{{.+}} void [[OFFL1]]( +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, {{.+}}, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}}) + +// CHECK: define{{.+}} [[OMP_OUTLINED_1]](i32* {{.+}} [[GTID_IN:%.+]], +// CHECK: [[GTID_ADDR:%.+]] = alloca i32*, +// CHECK: store i32* [[GTID_IN]], i32** [[GTID_ADDR]], +// CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_ADDR]], +// CHECK: [[GTID_VAL:%.+]] = load i32, i32* [[GTID_REF]], +// CHECK: call {{.*}}void @__kmpc_push_proc_bind([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 [[GTID_VAL]], i32 4) +// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call( +// CHECK: ret void + +// CHECK: define{{.+}} [[OFFL2]]() +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, {{.+}}, {{.+}}* [[OMP_OUTLINED_1:@.+]] to {{.+}}) + +// CHECK: define{{.+}} [[OMP_OUTLINED_1]](i32* {{.+}} [[GTID_IN:%.+]], +// CHECK: [[GTID_ADDR:%.+]] = alloca i32*, +// CHECK: store i32* [[GTID_IN]], i32** [[GTID_ADDR]], +// CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_ADDR]], +// CHECK: [[GTID_VAL:%.+]] = load i32, i32* [[GTID_REF]], +// CHECK: call {{.*}}void @__kmpc_push_proc_bind([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 [[GTID_VAL]], i32 3) +// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call( +// CHECK: ret void + +// CHECK: define{{.+}} [[TMAIN]]() +// CHECK: call {{.*}}@__tgt_target_teams({{.+}}) +// CHECK: call void [[OFFL3:@.+]]() + +// CHECK: define{{.+}} [[OFFL3]]() +// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_teams({{.+}}, {{.+}}, {{.+}}* [[OMP_OUTLINED_3:@.+]] to {{.+}}) + +// CHECK: define{{.+}} [[OMP_OUTLINED_3]](i32* {{.+}} [[GTID_IN:%.+]], +// CHECK: [[GTID_ADDR:%.+]] = alloca i32*, +// CHECK: store i32* [[GTID_IN]], i32** [[GTID_ADDR]], +// CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_ADDR]], +// CHECK: [[GTID_VAL:%.+]] = load i32, i32* [[GTID_REF]], +// CHECK: call {{.*}}void @__kmpc_push_proc_bind([[IDENT_T_TY]]* [[DEF_LOC_2]], i32 [[GTID_VAL]], i32 2) +// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call( +// CHECK: ret void +#endif diff --git a/test/Parser/cxx-modules-import.cpp b/test/Parser/cxx-modules-import.cpp deleted file mode 100644 index 0600eddecb834..0000000000000 --- a/test/Parser/cxx-modules-import.cpp +++ /dev/null @@ -1,58 +0,0 @@ -// RUN: rm -rf %t -// RUN: mkdir -p %t -// RUN: echo 'module x; int a, b;' > %t/x.cppm -// RUN: echo 'module x.y; int c;' > %t/x.y.cppm -// -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %t/x.cppm -o %t/x.pcm -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface -fmodule-file=%t/x.pcm %t/x.y.cppm -o %t/x.y.pcm -// -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ -// RUN: -DTEST=1 -DMODULE_KIND=implementation -DMODULE_NAME=z -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ -// RUN: -DTEST=2 -DMODULE_KIND=implementation -DMODULE_NAME=x -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ -// RUN: -DTEST=3 -DMODULE_KIND= -DMODULE_NAME=z -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ -// RUN: -DTEST=4 -DMODULE_KIND=partition -DMODULE_NAME=z -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ -// RUN: -DTEST=5 -DMODULE_KIND=elderberry -DMODULE_NAME=z -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ -// RUN: -DTEST=1 -DMODULE_KIND=implementation -DMODULE_NAME='z [[]]' -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ -// RUN: -DTEST=6 -DMODULE_KIND=implementation -DMODULE_NAME='z [[fancy]]' -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -I%t -fmodule-file=%t/x.y.pcm -verify %s \ -// RUN: -DTEST=7 -DMODULE_KIND=implementation -DMODULE_NAME='z [[maybe_unused]]' - -module MODULE_KIND MODULE_NAME; -#if TEST == 3 -// expected-error@-2 {{'module' declaration found while not building module interface}} -#elif TEST == 4 -// expected-error@-4 {{'module partition' declaration found while not building module interface}} -#elif TEST == 5 -// expected-error@-6 {{unexpected module kind 'elderberry'}} -#elif TEST == 6 -// expected-warning@-8 {{unknown attribute 'fancy' ignored}} -#elif TEST == 7 -// expected-error-re@-10 {{'maybe_unused' attribute cannot be applied to a module{{$}}}} -#endif - -int use_1 = a; -#if TEST != 2 -// expected-error@-2 {{declaration of 'a' must be imported from module 'x' before it is required}} -// expected-note@x.cppm:1 {{here}} -#endif - -import x; - -int use_2 = b; // ok - -import x [[]]; -import x [[foo]]; // expected-warning {{unknown attribute 'foo' ignored}} -import x [[noreturn]]; // expected-error {{'noreturn' attribute cannot be applied to a module import}} -import x [[blarg::noreturn]]; // expected-warning {{unknown attribute 'noreturn' ignored}} - -import x.y; -import x.; // expected-error {{expected a module name after module import}} -import .x; // expected-error {{expected a module name after module import}} - -import blarg; // expected-error {{module 'blarg' not found}} diff --git a/test/Parser/cxx-modules-interface.cppm b/test/Parser/cxx-modules-interface.cppm index f7835bd167b20..0e7c746833534 100644 --- a/test/Parser/cxx-modules-interface.cppm +++ b/test/Parser/cxx-modules-interface.cppm @@ -1,13 +1,12 @@ // RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -Dmodule=int -DERRORS +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -DERRORS -module foo; +export module foo; #ifndef ERRORS // expected-no-diagnostics #else -// expected-error@-4 {{expected module declaration at start of module interface}} +// FIXME: diagnose missing module-declaration when building module interface -// FIXME: support 'export module X;' and 'export { int n; module X; }' // FIXME: proclaimed-ownership-declarations? export { diff --git a/test/Preprocessor/init.c b/test/Preprocessor/init.c index 7548fbaae2eb7..5fb2011cd0250 100644 --- a/test/Preprocessor/init.c +++ b/test/Preprocessor/init.c @@ -8846,6 +8846,16 @@ // WEBASSEMBLY32-NEXT:#define __CHAR32_TYPE__ unsigned int // WEBASSEMBLY32-NEXT:#define __CHAR_BIT__ 8 // WEBASSEMBLY32-NOT:#define __CHAR_UNSIGNED__ +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_BOOL_LOCK_FREE 2 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_CHAR16_T_LOCK_FREE 2 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_CHAR32_T_LOCK_FREE 2 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_CHAR_LOCK_FREE 2 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_INT_LOCK_FREE 2 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_LLONG_LOCK_FREE 1 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_LONG_LOCK_FREE 2 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_POINTER_LOCK_FREE 2 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_SHORT_LOCK_FREE 2 +// WEBASSEMBLY32-NEXT:#define __CLANG_ATOMIC_WCHAR_T_LOCK_FREE 2 // WEBASSEMBLY32-NEXT:#define __CONSTANT_CFSTRINGS__ 1 // WEBASSEMBLY32-NEXT:#define __DBL_DECIMAL_DIG__ 17 // WEBASSEMBLY32-NEXT:#define __DBL_DENORM_MIN__ 4.9406564584124654e-324 @@ -9162,6 +9172,16 @@ // WEBASSEMBLY64-NEXT:#define __CHAR32_TYPE__ unsigned int // WEBASSEMBLY64-NEXT:#define __CHAR_BIT__ 8 // WEBASSEMBLY64-NOT:#define __CHAR_UNSIGNED__ +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_BOOL_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_CHAR16_T_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_CHAR32_T_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_CHAR_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_INT_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_LLONG_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_LONG_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_POINTER_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_SHORT_LOCK_FREE 2 +// WEBASSEMBLY64-NEXT:#define __CLANG_ATOMIC_WCHAR_T_LOCK_FREE 2 // WEBASSEMBLY64-NEXT:#define __CONSTANT_CFSTRINGS__ 1 // WEBASSEMBLY64-NEXT:#define __DBL_DECIMAL_DIG__ 17 // WEBASSEMBLY64-NEXT:#define __DBL_DENORM_MIN__ 4.9406564584124654e-324 @@ -9637,3 +9657,36 @@ // AVR:#define __WCHAR_MAX__ 32767 // AVR:#define __WCHAR_TYPE__ int // AVR:#define __WINT_TYPE__ int + + +// RUN: %clang_cc1 -E -dM -ffreestanding \ +// RUN: -triple i686-windows-msvc -fms-compatibility < /dev/null \ +// RUN: | FileCheck -match-full-lines -check-prefix MSVC-X32 %s + +// RUN: %clang_cc1 -E -dM -ffreestanding \ +// RUN: -triple x86_64-windows-msvc -fms-compatibility < /dev/null \ +// RUN: | FileCheck -match-full-lines -check-prefix MSVC-X64 %s + +// MSVC-X32:#define __CLANG_ATOMIC_BOOL_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_CHAR16_T_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_CHAR32_T_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_CHAR_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_INT_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_LLONG_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_LONG_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_POINTER_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_SHORT_LOCK_FREE 2 +// MSVC-X32-NEXT:#define __CLANG_ATOMIC_WCHAR_T_LOCK_FREE 2 +// MSVC-X32-NOT:#define __GCC_ATOMIC{{.*}} + +// MSVC-X64:#define __CLANG_ATOMIC_BOOL_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_CHAR16_T_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_CHAR32_T_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_CHAR_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_INT_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_LLONG_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_LONG_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_POINTER_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_SHORT_LOCK_FREE 2 +// MSVC-X64-NEXT:#define __CLANG_ATOMIC_WCHAR_T_LOCK_FREE 2 +// MSVC-X86-NOT:#define __GCC_ATOMIC{{.*}} diff --git a/test/Sema/atomic-ops.c b/test/Sema/atomic-ops.c index 8ebf3eaed4af2..eee1cda8635ea 100644 --- a/test/Sema/atomic-ops.c +++ b/test/Sema/atomic-ops.c @@ -7,19 +7,29 @@ struct S { char c[3]; }; _Static_assert(__GCC_ATOMIC_BOOL_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_BOOL_LOCK_FREE == __CLANG_ATOMIC_BOOL_LOCK_FREE, ""); _Static_assert(__GCC_ATOMIC_CHAR_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_CHAR_LOCK_FREE == __CLANG_ATOMIC_CHAR_LOCK_FREE, ""); _Static_assert(__GCC_ATOMIC_CHAR16_T_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_CHAR16_T_LOCK_FREE == __CLANG_ATOMIC_CHAR16_T_LOCK_FREE, ""); _Static_assert(__GCC_ATOMIC_CHAR32_T_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_CHAR32_T_LOCK_FREE == __CLANG_ATOMIC_CHAR32_T_LOCK_FREE, ""); _Static_assert(__GCC_ATOMIC_WCHAR_T_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_WCHAR_T_LOCK_FREE == __CLANG_ATOMIC_WCHAR_T_LOCK_FREE, ""); _Static_assert(__GCC_ATOMIC_SHORT_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_SHORT_LOCK_FREE == __CLANG_ATOMIC_SHORT_LOCK_FREE, ""); _Static_assert(__GCC_ATOMIC_INT_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_INT_LOCK_FREE == __CLANG_ATOMIC_INT_LOCK_FREE, ""); _Static_assert(__GCC_ATOMIC_LONG_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_LONG_LOCK_FREE == __CLANG_ATOMIC_LONG_LOCK_FREE, ""); #ifdef __i386__ _Static_assert(__GCC_ATOMIC_LLONG_LOCK_FREE == 1, ""); #else _Static_assert(__GCC_ATOMIC_LLONG_LOCK_FREE == 2, ""); #endif +_Static_assert(__GCC_ATOMIC_LLONG_LOCK_FREE == __CLANG_ATOMIC_LLONG_LOCK_FREE, ""); _Static_assert(__GCC_ATOMIC_POINTER_LOCK_FREE == 2, ""); +_Static_assert(__GCC_ATOMIC_POINTER_LOCK_FREE == __CLANG_ATOMIC_POINTER_LOCK_FREE, ""); _Static_assert(__c11_atomic_is_lock_free(1), ""); _Static_assert(__c11_atomic_is_lock_free(2), ""); diff --git a/test/Sema/warn-documentation.cpp b/test/Sema/warn-documentation.cpp index 34d8f5fd2da28..0c92b2aa029fe 100644 --- a/test/Sema/warn-documentation.cpp +++ b/test/Sema/warn-documentation.cpp @@ -1210,3 +1210,75 @@ template <class T> T test_function (T arg); /*! @function test_function<int> */ template <> int test_function<int> (int arg); + +namespace AllowParamAndReturnsOnFunctionPointerVars { + +/** + * functionPointerVariable + * + * @param i is integer. + * @returns integer. + */ +int (*functionPointerVariable)(int i); + +struct HasFields { + /** + * functionPointerField + * + * @param i is integer. + * @returns integer. + */ + int (*functionPointerField)(int i); +}; + +// expected-warning@+5 {{'\returns' command used in a comment that is attached to a function returning void}} +/** + * functionPointerVariable + * + * \param p not here. + * \returns integer. + */ +void (*functionPointerVariableThatLeadsNowhere)(); + +// Still warn about param/returns commands for variables that don't specify +// the type directly: + +/** + * FunctionPointerTypedef + * + * \param i is integer. + * \returns integer. + */ +typedef int (*FunctionPointerTypedef)(int i); + +/** + * FunctionPointerTypealias + * + * \param i is integer. + * \returns integer. + */ +using FunctionPointerTypealias = int (*)(int i); + +// expected-warning@+5 {{'@param' command used in a comment that is not attached to a function declaration}} +// expected-warning@+5 {{'@returns' command used in a comment that is not attached to a function or method declaration}} +/** + * functionPointerVariable + * + * @param i is integer. + * @returns integer. + */ +FunctionPointerTypedef functionPointerTypedefVariable; + +struct HasMoreFields { + // expected-warning@+5 {{'\param' command used in a comment that is not attached to a function declaration}} + // expected-warning@+5 {{'\returns' command used in a comment that is not attached to a function or method declaration}} + /** + * functionPointerTypealiasField + * + * \param i is integer. + * \returns integer. + */ + FunctionPointerTypealias functionPointerTypealiasField; +}; + +} diff --git a/test/Sema/warn-documentation.m b/test/Sema/warn-documentation.m index 5e95e2a1e8a2f..3c1a369c4b14b 100644 --- a/test/Sema/warn-documentation.m +++ b/test/Sema/warn-documentation.m @@ -229,3 +229,73 @@ int FooBar(); - (void) VarArgMeth : (id)arg, ... {} @end +/** + * blockPointerVariable + * + * @param i is integer. + * @returns integer. + */ +int (^blockPointerVariable)(int i); + +struct HasFields { + /** + * blockPointerField + * + * \param i is integer. + * \returns integer. + */ + int (^blockPointerFields)(int i); +}; + +// expected-warning@+5 {{'\returns' command used in a comment that is attached to a function returning void}} +/** + * functionPointerVariable + * + * \param p not here. + * \returns integer. + */ +void (^_Nullable blockPointerVariableThatLeadsNowhere)(); + +@interface CheckFunctionBlockPointerVars { + /** + * functionPointerIVar + * + * @param i is integer. + * @returns integer. + */ + int (*functionPointerIVar)(int i); + + /** + * blockPointerIVar + * + * \param i is integer. + * \returns integer. + */ + int (^blockPointerIVar)(int i); +} + +/** + * functionPointerProperty + * + * @param i is integer. + * @returns integer. + */ +@property int (*functionPointerProperty)(int i); + +/** + * blockPointerProperty + * + * \param i is integer. + * \returns integer. + */ +@property int (^blockPointerProperty)(int i); + +/** + * blockReturnsNothing + * + * \returns Nothing, but can allow this as this pattern is used to document the + * value that the property getter returns. + */ +@property void (^blockReturnsNothing)(); + +@end diff --git a/test/SemaCXX/MicrosoftCompatibility-cxx98.cpp b/test/SemaCXX/MicrosoftCompatibility-cxx98.cpp deleted file mode 100644 index 3b80d0937e231..0000000000000 --- a/test/SemaCXX/MicrosoftCompatibility-cxx98.cpp +++ /dev/null @@ -1,23 +0,0 @@ -// RUN: %clang_cc1 %s -triple i686-pc-win32 -fsyntax-only -std=c++98 -Wmicrosoft -verify -fms-compatibility -fexceptions -fcxx-exceptions - - -//MSVC allows forward enum declaration -enum ENUM; // expected-warning {{forward references to 'enum' types are a Microsoft extension}} -ENUM *var = 0; -ENUM var2 = (ENUM)3; -enum ENUM1* var3 = 0;// expected-warning {{forward references to 'enum' types are a Microsoft extension}} - -typedef void (*FnPtrTy)(); -void (*PR23733_1)() = static_cast<FnPtrTy>((void *)0); // expected-warning {{static_cast between pointer-to-function and pointer-to-object is a Microsoft extension}} -void (*PR23733_2)() = FnPtrTy((void *)0); -void (*PR23733_3)() = (FnPtrTy)((void *)0); -void (*PR23733_4)() = reinterpret_cast<FnPtrTy>((void *)0); - -long function_prototype(int a); -long (*function_ptr)(int a); - -void function_to_voidptr_conv() { - void *a1 = function_prototype; // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}} - void *a2 = &function_prototype; // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}} - void *a3 = function_ptr; // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}} -} diff --git a/test/SemaCXX/MicrosoftCompatibility.cpp b/test/SemaCXX/MicrosoftCompatibility.cpp index 26cd7825ee87a..d095f6edc6179 100644 --- a/test/SemaCXX/MicrosoftCompatibility.cpp +++ b/test/SemaCXX/MicrosoftCompatibility.cpp @@ -224,6 +224,15 @@ template void function_missing_typename<D>(const D::Type param); } +//MSVC allows forward enum declaration +enum ENUM; // expected-warning {{forward references to 'enum' types are a Microsoft extension}} +ENUM *var = 0; +ENUM var2 = (ENUM)3; +enum ENUM1* var3 = 0;// expected-warning {{forward references to 'enum' types are a Microsoft extension}} + +enum ENUM1 { kA }; +enum ENUM1; // This way round is fine. + enum ENUM2 { ENUM2_a = (enum ENUM2) 4, ENUM2_b = 0x9FFFFFFF, // expected-warning {{enumerator value is not representable in the underlying type 'int'}} @@ -269,3 +278,18 @@ void g() { f(0xffffffffffffffffi64); } } + +typedef void (*FnPtrTy)(); +void (*PR23733_1)() = static_cast<FnPtrTy>((void *)0); // expected-warning {{static_cast between pointer-to-function and pointer-to-object is a Microsoft extension}} +void (*PR23733_2)() = FnPtrTy((void *)0); +void (*PR23733_3)() = (FnPtrTy)((void *)0); +void (*PR23733_4)() = reinterpret_cast<FnPtrTy>((void *)0); + +long function_prototype(int a); +long (*function_ptr)(int a); + +void function_to_voidptr_conv() { + void *a1 = function_prototype; // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}} + void *a2 = &function_prototype; // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}} + void *a3 = function_ptr; // expected-warning {{implicit conversion between pointer-to-function and pointer-to-object is a Microsoft extension}} +} diff --git a/test/SemaCXX/MicrosoftExtensions.cpp b/test/SemaCXX/MicrosoftExtensions.cpp index 96088a084558a..38949e11376d5 100644 --- a/test/SemaCXX/MicrosoftExtensions.cpp +++ b/test/SemaCXX/MicrosoftExtensions.cpp @@ -2,6 +2,7 @@ // RUN: %clang_cc1 -std=c++98 %s -triple i686-pc-win32 -fsyntax-only -Wmicrosoft -Wc++11-extensions -Wno-long-long -verify -fms-extensions -fexceptions -fcxx-exceptions -DTEST1 // RUN: %clang_cc1 -std=c++11 %s -triple i686-pc-win32 -fsyntax-only -Wmicrosoft -Wc++11-extensions -Wno-long-long -verify -fms-extensions -fexceptions -fcxx-exceptions -DTEST1 // RUN: %clang_cc1 %s -triple i686-pc-win32 -fsyntax-only -Wmicrosoft -Wc++11-extensions -Wno-long-long -verify -fexceptions -fcxx-exceptions -DTEST2 +// RUN: %clang_cc1 %s -triple i686-pc-win32 -fsyntax-only -std=c++11 -fms-compatibility -verify -DTEST3 #if TEST1 @@ -508,6 +509,13 @@ int S::fn() { return 0; } // expected-warning {{is missing exception specificati // Check that __unaligned is not recognized if MS extensions are not enabled typedef char __unaligned *aligned_type; // expected-error {{expected ';' after top level declarator}} +#elif TEST3 + +namespace PR32750 { +template<typename T> struct A {}; +template<typename T> struct B : A<A<T>> { A<T>::C::D d; }; // expected-error {{missing 'typename' prior to dependent type name 'A<T>::C::D'}} +} + #else #error Unknown test mode diff --git a/test/SemaCXX/modules-ts.cppm b/test/SemaCXX/modules-ts.cppm index 71c09d3bfe6cb..16695f6463a82 100644 --- a/test/SemaCXX/modules-ts.cppm +++ b/test/SemaCXX/modules-ts.cppm @@ -1,22 +1,21 @@ // RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -DTEST=0 -// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -Dmodule=int -DTEST=1 -// RUN: not %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -fmodule-file=%t.pcm -o %t.pcm -DTEST=2 2>&1 | FileCheck %s --check-prefix=CHECK-2 +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -o %t.pcm -verify -DTEST=1 +// RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -fmodule-file=%t.pcm -o %t.pcm -verify -DTEST=2 // RUN: %clang_cc1 -std=c++1z -fmodules-ts -emit-module-interface %s -fmodule-file=%t.pcm -o %t.pcm -verify -Dfoo=bar -DTEST=3 #if TEST == 0 // expected-no-diagnostics #endif -module foo; -#if TEST == 1 -// expected-error@-2 {{expected module declaration at start of module interface}} -#elif TEST == 2 -// CHECK-2: error: redefinition of module 'foo' +export module foo; +#if TEST == 2 +// expected-error@-2 {{redefinition of module 'foo'}} +// expected-note@modules-ts.cppm:* {{loaded from}} #endif static int m; // ok, internal linkage, so no redefinition error int n; -#if TEST == 3 +#if TEST >= 2 // expected-error@-2 {{redefinition of '}} // expected-note@-3 {{previous}} #endif diff --git a/test/SemaObjC/unguarded-availability.m b/test/SemaObjC/unguarded-availability.m index 4369a0e329d29..ae921f4a27b3e 100644 --- a/test/SemaObjC/unguarded-availability.m +++ b/test/SemaObjC/unguarded-availability.m @@ -1,5 +1,5 @@ // RUN: %clang_cc1 -triple x86_64-apple-macosx-10.9 -Wunguarded-availability -fblocks -fsyntax-only -verify %s -// RUN: %clang_cc1 -xobjective-c++ -DOBJCPP -triple x86_64-apple-macosx-10.9 -Wunguarded-availability -fblocks -fsyntax-only -verify %s +// RUN: %clang_cc1 -xobjective-c++ -std=c++11 -DOBJCPP -triple x86_64-apple-macosx-10.9 -Wunguarded-availability -fblocks -fsyntax-only -verify %s #define AVAILABLE_10_0 __attribute__((availability(macos, introduced = 10.0))) #define AVAILABLE_10_11 __attribute__((availability(macos, introduced = 10.11))) @@ -8,9 +8,9 @@ int func_10_11() AVAILABLE_10_11; // expected-note 4 {{'func_10_11' has been explicitly marked partial here}} #ifdef OBJCPP -// expected-note@+2 {{marked partial here}} +// expected-note@+2 2 {{marked partial here}} #endif -int func_10_12() AVAILABLE_10_12; // expected-note 5 {{'func_10_12' has been explicitly marked partial here}} +int func_10_12() AVAILABLE_10_12; // expected-note 6 {{'func_10_12' has been explicitly marked partial here}} int func_10_0() AVAILABLE_10_0; @@ -129,6 +129,12 @@ void test_params(int_10_12 x); // expected-warning {{'int_10_12' is partial: int void test_params2(int_10_12 x) AVAILABLE_10_12; // no warn +void (^topLevelBlockDecl)() = ^ { + func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{@available}} + if (@available(macos 10.12, *)) + func_10_12(); +}; + #ifdef OBJCPP int f(char) AVAILABLE_10_12; @@ -176,4 +182,10 @@ int instantiate_availability() { with_availability_attr<int_10_12>(); // expected-warning{{'with_availability_attr<int>' is only available on macOS 10.11 or newer}} expected-warning{{'int_10_12' is only available on macOS 10.12 or newer}} expected-note 2 {{enclose}} } +auto topLevelLambda = [] () { + func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{@available}} + if (@available(macos 10.12, *)) + func_10_12(); +}; + #endif diff --git a/test/SemaObjCXX/pr32725.mm b/test/SemaObjCXX/pr32725.mm new file mode 100644 index 0000000000000..8d7d1163250f3 --- /dev/null +++ b/test/SemaObjCXX/pr32725.mm @@ -0,0 +1,9 @@ +// RUN: %clang_cc1 -triple x86_64-apple-macosx -fsyntax-only -verify -x objective-c++ %s -o /dev/null +// REQUIRES: asserts + +struct objc_class { + unsigned long long bits; +}; +typedef struct objc_class *Class; +static void f(Class c) { (void)(c->bits & RW_HAS_OVERFLOW_REFCOUNT); } +// expected-error@-1{{use of undeclared identifier 'RW_HAS_OVERFLOW_REFCOUNT}} diff --git a/test/SemaOpenCL/cl20-device-side-enqueue.cl b/test/SemaOpenCL/cl20-device-side-enqueue.cl index 3e87cfcddf661..bafbe447ee286 100644 --- a/test/SemaOpenCL/cl20-device-side-enqueue.cl +++ b/test/SemaOpenCL/cl20-device-side-enqueue.cl @@ -1,12 +1,14 @@ -// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir-unknown-unknown" -verify -pedantic -fsyntax-only -DB32 -// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir64-unknown-unknown" -verify -pedantic -fsyntax-only -Wconversion -DWCONV +// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir-unknown-unknown" -verify -pedantic -fsyntax-only -DB32 -DQUALS= +// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir-unknown-unknown" -verify -pedantic -fsyntax-only -DB32 -DQUALS="const volatile" +// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir64-unknown-unknown" -verify -pedantic -fsyntax-only -Wconversion -DWCONV -DQUALS= +// RUN: %clang_cc1 %s -cl-std=CL2.0 -triple "spir64-unknown-unknown" -verify -pedantic -fsyntax-only -Wconversion -DWCONV -DQUALS="const volatile" typedef struct {int a;} ndrange_t; // Diagnostic tests for different overloads of enqueue_kernel from Table 6.13.17.1 of OpenCL 2.0 Spec. kernel void enqueue_kernel_tests() { queue_t default_queue; unsigned flags = 0; - ndrange_t ndrange; + QUALS ndrange_t ndrange; clk_event_t evt; clk_event_t event_wait_list; clk_event_t event_wait_list2[] = {evt, evt}; |