maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/CodeGen/atomics-inlining.c

94 lines
5.3 KiB
C
Raw Normal View History

Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// RUN: %clang_cc1 -triple arm-linux-gnu -emit-llvm %s -o - | FileCheck %s -check-prefix=ARM
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
// RUN: %clang_cc1 -triple powerpc-linux-gnu -emit-llvm %s -o - | FileCheck %s -check-prefix=PPC32
// RUN: %clang_cc1 -triple powerpc64-linux-gnu -emit-llvm %s -o - | FileCheck %s -check-prefix=PPC64
// RUN: %clang_cc1 -triple mipsel-linux-gnu -emit-llvm %s -o - | FileCheck %s -check-prefix=MIPS32
// RUN: %clang_cc1 -triple mips64el-linux-gnu -emit-llvm %s -o - | FileCheck %s -check-prefix=MIPS64
unsigned char c1, c2;
unsigned short s1, s2;
unsigned int i1, i2;
unsigned long long ll1, ll2;
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
unsigned char a1[100], a2[100];
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
enum memory_order {
memory_order_relaxed,
memory_order_consume,
memory_order_acquire,
memory_order_release,
memory_order_acq_rel,
memory_order_seq_cst
};
void test1(void) {
(void)__atomic_load(&c1, &c2, memory_order_seq_cst);
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
(void)__atomic_store(&c1, &c2, memory_order_seq_cst);
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
(void)__atomic_load(&s1, &s2, memory_order_seq_cst);
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
(void)__atomic_store(&s1, &s2, memory_order_seq_cst);
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
(void)__atomic_load(&i1, &i2, memory_order_seq_cst);
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
(void)__atomic_store(&i1, &i2, memory_order_seq_cst);
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
(void)__atomic_load(&ll1, &ll2, memory_order_seq_cst);
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
(void)__atomic_store(&ll1, &ll2, memory_order_seq_cst);
(void)__atomic_load(&a1, &a2, memory_order_seq_cst);
(void)__atomic_store(&a1, &a2, memory_order_seq_cst);
CHECK-LABEL-ify some code gen tests to improve diagnostic experience when tests fail. llvm-svn: 188447
2013-08-15 14:47:53 +08:00
// ARM-LABEL: define arm_aapcscc void @test1
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// ARM: = call arm_aapcscc zeroext i8 @__atomic_load_1(i8* @c1
// ARM: call arm_aapcscc void @__atomic_store_1(i8* @c1, i8 zeroext
// ARM: = call arm_aapcscc zeroext i16 @__atomic_load_2(i8* bitcast (i16* @s1 to i8*)
// ARM: call arm_aapcscc void @__atomic_store_2(i8* bitcast (i16* @s1 to i8*), i16 zeroext
// ARM: = call arm_aapcscc i32 @__atomic_load_4(i8* bitcast (i32* @i1 to i8*)
// ARM: call arm_aapcscc void @__atomic_store_4(i8* bitcast (i32* @i1 to i8*), i32
// ARM: = call arm_aapcscc i64 @__atomic_load_8(i8* bitcast (i64* @ll1 to i8*)
// ARM: call arm_aapcscc void @__atomic_store_8(i8* bitcast (i64* @ll1 to i8*), i64
// ARM: call arm_aapcscc void @__atomic_load(i32 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
// ARM: call arm_aapcscc void @__atomic_store(i32 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
CHECK-LABEL-ify some code gen tests to improve diagnostic experience when tests fail. llvm-svn: 188447
2013-08-15 14:47:53 +08:00
// PPC32-LABEL: define void @test1
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// PPC32: = load atomic i8* @c1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// PPC32: store atomic i8 {{.*}}, i8* @c1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// PPC32: = load atomic i16* @s1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// PPC32: store atomic i16 {{.*}}, i16* @s1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// PPC32: = load atomic i32* @i1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// PPC32: store atomic i32 {{.*}}, i32* @i1 seq_cst
// PPC32: = call i64 @__atomic_load_8(i8* bitcast (i64* @ll1 to i8*)
// PPC32: call void @__atomic_store_8(i8* bitcast (i64* @ll1 to i8*), i64
// PPC32: call void @__atomic_load(i32 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
// PPC32: call void @__atomic_store(i32 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
CHECK-LABEL-ify some code gen tests to improve diagnostic experience when tests fail. llvm-svn: 188447
2013-08-15 14:47:53 +08:00
// PPC64-LABEL: define void @test1
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// PPC64: = load atomic i8* @c1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// PPC64: store atomic i8 {{.*}}, i8* @c1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// PPC64: = load atomic i16* @s1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// PPC64: store atomic i16 {{.*}}, i16* @s1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// PPC64: = load atomic i32* @i1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// PPC64: store atomic i32 {{.*}}, i32* @i1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// PPC64: = load atomic i64* @ll1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// PPC64: store atomic i64 {{.*}}, i64* @ll1 seq_cst
// PPC64: call void @__atomic_load(i64 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
// PPC64: call void @__atomic_store(i64 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
CHECK-LABEL-ify some code gen tests to improve diagnostic experience when tests fail. llvm-svn: 188447
2013-08-15 14:47:53 +08:00
// MIPS32-LABEL: define void @test1
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// MIPS32: = load atomic i8* @c1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// MIPS32: store atomic i8 {{.*}}, i8* @c1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// MIPS32: = load atomic i16* @s1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// MIPS32: store atomic i16 {{.*}}, i16* @s1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// MIPS32: = load atomic i32* @i1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// MIPS32: store atomic i32 {{.*}}, i32* @i1 seq_cst
// MIPS32: call i64 @__atomic_load_8(i8* bitcast (i64* @ll1 to i8*)
// MIPS32: call void @__atomic_store_8(i8* bitcast (i64* @ll1 to i8*), i64
[mips] Promote all integral/enumeration types to the GPR width Summary: Ensure all integral/enumeration types are appropriately annotated with signext/zeroext. In particular, i32 now has these attributes when using the N32/N64 ABI. This paves the way for accurately representing the way the N32/N64 ABI's promotes integer arguments to i64. Reviewers: atanasyan Reviewed By: atanasyan Subscribers: cfe-commits, theraven Differential Revision: http://reviews.llvm.org/D5961 llvm-svn: 220563
2014-10-24 22:42:42 +08:00
// MIPS32: call void @__atomic_load(i32 zeroext 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
// MIPS32: call void @__atomic_store(i32 zeroext 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
CHECK-LABEL-ify some code gen tests to improve diagnostic experience when tests fail. llvm-svn: 188447
2013-08-15 14:47:53 +08:00
// MIPS64-LABEL: define void @test1
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// MIPS64: = load atomic i8* @c1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// MIPS64: store atomic i8 {{.*}}, i8* @c1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// MIPS64: = load atomic i16* @s1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// MIPS64: store atomic i16 {{.*}}, i16* @s1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// MIPS64: = load atomic i32* @i1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// MIPS64: store atomic i32 {{.*}}, i32* @i1 seq_cst
Slightly extend matching for atomic loads. Also add the "=" to the matched pattern, to see whether we actually save the loaded value. llvm-svn: 183029
2013-06-01 03:18:14 +08:00
// MIPS64: = load atomic i64* @ll1 seq_cst
Add support for optimized (non-generic) atomic libcalls. For integer types of sizes 1, 2, 4 and 8, libcompiler-rt (and libgcc) provide atomic functions that pass parameters by value and return results directly. libgcc and libcompiler-rt only provide optimized libcalls for __atomic_fetch_*, as generic libcalls on non-integer types would make little sense. This means that we can finally make __atomic_fetch_* work on architectures for which we don't provide these operations as builtins (e.g. ARM). This should fix the dreaded "cannot compile this atomic library call yet" error that would pop up once every while. llvm-svn: 183033
2013-06-01 03:27:59 +08:00
// MIPS64: store atomic i64 {{.*}}, i64* @ll1 seq_cst
[mips] Promote all integral/enumeration types to the GPR width Summary: Ensure all integral/enumeration types are appropriately annotated with signext/zeroext. In particular, i32 now has these attributes when using the N32/N64 ABI. This paves the way for accurately representing the way the N32/N64 ABI's promotes integer arguments to i64. Reviewers: atanasyan Reviewed By: atanasyan Subscribers: cfe-commits, theraven Differential Revision: http://reviews.llvm.org/D5961 llvm-svn: 220563
2014-10-24 22:42:42 +08:00
// MIPS64: call void @__atomic_load(i64 zeroext 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0)
// MIPS64: call void @__atomic_store(i64 zeroext 100, i8* getelementptr inbounds ([100 x i8]* @a1, i32 0, i32 0), i8* getelementptr inbounds ([100 x i8]* @a2, i32 0, i32 0)
[mips] Enable inlining of atomic ops on mips32 and mips64. llvm-svn: 172855
2013-01-19 05:58:11 +08:00
}