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/atomic-ops-libcall.c

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

121 lines
4.6 KiB
C
Raw Normal View History

Fix atomic libcall. This commit fixes a cast instruction assertion failure due to the incompatible type cast. This will only happen when the target requires atomic libcalls. llvm-svn: 204834
2014-03-27 01:35:01 +08:00
// RUN: %clang_cc1 < %s -triple armv5e-none-linux-gnueabi -emit-llvm -O1 | FileCheck %s
[CodeGen] remove/fix checks that will fail when r301923 is recommitted Don't test the optimizer as part of front-end verification. llvm-svn: 301928
2017-05-02 23:20:18 +08:00
// FIXME: This file should not be checking -O1 output.
// Ie, it is testing many IR optimizer passes as part of front-end verification.
Fix atomic libcall. This commit fixes a cast instruction assertion failure due to the incompatible type cast. This will only happen when the target requires atomic libcalls. llvm-svn: 204834
2014-03-27 01:35:01 +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
};
int *test_c11_atomic_fetch_add_int_ptr(_Atomic(int *) *p) {
// CHECK: test_c11_atomic_fetch_add_int_ptr
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_add_4(i8* {{%[0-9]+}}, i32 12, i32 5)
Fix atomic libcall. This commit fixes a cast instruction assertion failure due to the incompatible type cast. This will only happen when the target requires atomic libcalls. llvm-svn: 204834
2014-03-27 01:35:01 +08:00
return __c11_atomic_fetch_add(p, 3, memory_order_seq_cst);
}
int *test_c11_atomic_fetch_sub_int_ptr(_Atomic(int *) *p) {
// CHECK: test_c11_atomic_fetch_sub_int_ptr
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_sub_4(i8* {{%[0-9]+}}, i32 20, i32 5)
Fix atomic libcall. This commit fixes a cast instruction assertion failure due to the incompatible type cast. This will only happen when the target requires atomic libcalls. llvm-svn: 204834
2014-03-27 01:35:01 +08:00
return __c11_atomic_fetch_sub(p, 5, memory_order_seq_cst);
}
int test_c11_atomic_fetch_add_int(_Atomic(int) *p) {
// CHECK: test_c11_atomic_fetch_add_int
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_add_4(i8* {{%[0-9]+}}, i32 3, i32 5)
Fix atomic libcall. This commit fixes a cast instruction assertion failure due to the incompatible type cast. This will only happen when the target requires atomic libcalls. llvm-svn: 204834
2014-03-27 01:35:01 +08:00
return __c11_atomic_fetch_add(p, 3, memory_order_seq_cst);
}
int test_c11_atomic_fetch_sub_int(_Atomic(int) *p) {
// CHECK: test_c11_atomic_fetch_sub_int
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_sub_4(i8* {{%[0-9]+}}, i32 5, i32 5)
Fix atomic libcall. This commit fixes a cast instruction assertion failure due to the incompatible type cast. This will only happen when the target requires atomic libcalls. llvm-svn: 204834
2014-03-27 01:35:01 +08:00
return __c11_atomic_fetch_sub(p, 5, memory_order_seq_cst);
}
int *fp2a(int **p) {
// CHECK: @fp2a
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_sub_4(i8* {{%[0-9]+}}, i32 4, i32 0)
Fix atomic libcall. This commit fixes a cast instruction assertion failure due to the incompatible type cast. This will only happen when the target requires atomic libcalls. llvm-svn: 204834
2014-03-27 01:35:01 +08:00
// Note, the GNU builtins do not multiply by sizeof(T)!
return __atomic_fetch_sub(p, 4, memory_order_relaxed);
}
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
int test_atomic_fetch_add(int *p) {
// CHECK: test_atomic_fetch_add
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_add_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_fetch_add(p, 55, memory_order_seq_cst);
}
int test_atomic_fetch_sub(int *p) {
// CHECK: test_atomic_fetch_sub
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_sub_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_fetch_sub(p, 55, memory_order_seq_cst);
}
int test_atomic_fetch_and(int *p) {
// CHECK: test_atomic_fetch_and
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_and_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_fetch_and(p, 55, memory_order_seq_cst);
}
int test_atomic_fetch_or(int *p) {
// CHECK: test_atomic_fetch_or
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_or_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_fetch_or(p, 55, memory_order_seq_cst);
}
int test_atomic_fetch_xor(int *p) {
// CHECK: test_atomic_fetch_xor
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_xor_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_fetch_xor(p, 55, memory_order_seq_cst);
}
int test_atomic_fetch_nand(int *p) {
// CHECK: test_atomic_fetch_nand
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: {{%[^ ]*}} = call i32 @__atomic_fetch_nand_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_fetch_nand(p, 55, memory_order_seq_cst);
}
int test_atomic_add_fetch(int *p) {
// CHECK: test_atomic_add_fetch
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: [[CALL:%[^ ]*]] = call i32 @__atomic_fetch_add_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Correct atomic libcall support for __atomic_*_fetch builtins. In r244063, I had caused these builtins to call the same-named library functions, __atomic_*_fetch_SIZE. However, this was incorrect: while those functions are in fact supported by GCC's libatomic, they're not documented by the spec (and gcc doesn't ever call them). Instead, you're /supposed/ to call the __atomic_fetch_* builtins and then redo the operation inline to return the final value. Differential Revision: http://reviews.llvm.org/D14385 llvm-svn: 252920
2015-11-13 02:37:29 +08:00
// CHECK: {{%[^ ]*}} = add i32 [[CALL]], 55
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_add_fetch(p, 55, memory_order_seq_cst);
}
int test_atomic_sub_fetch(int *p) {
// CHECK: test_atomic_sub_fetch
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: [[CALL:%[^ ]*]] = call i32 @__atomic_fetch_sub_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Correct atomic libcall support for __atomic_*_fetch builtins. In r244063, I had caused these builtins to call the same-named library functions, __atomic_*_fetch_SIZE. However, this was incorrect: while those functions are in fact supported by GCC's libatomic, they're not documented by the spec (and gcc doesn't ever call them). Instead, you're /supposed/ to call the __atomic_fetch_* builtins and then redo the operation inline to return the final value. Differential Revision: http://reviews.llvm.org/D14385 llvm-svn: 252920
2015-11-13 02:37:29 +08:00
// CHECK: {{%[^ ]*}} = add i32 [[CALL]], -55
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_sub_fetch(p, 55, memory_order_seq_cst);
}
int test_atomic_and_fetch(int *p) {
// CHECK: test_atomic_and_fetch
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: [[CALL:%[^ ]*]] = call i32 @__atomic_fetch_and_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Correct atomic libcall support for __atomic_*_fetch builtins. In r244063, I had caused these builtins to call the same-named library functions, __atomic_*_fetch_SIZE. However, this was incorrect: while those functions are in fact supported by GCC's libatomic, they're not documented by the spec (and gcc doesn't ever call them). Instead, you're /supposed/ to call the __atomic_fetch_* builtins and then redo the operation inline to return the final value. Differential Revision: http://reviews.llvm.org/D14385 llvm-svn: 252920
2015-11-13 02:37:29 +08:00
// CHECK: {{%[^ ]*}} = and i32 [[CALL]], 55
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_and_fetch(p, 55, memory_order_seq_cst);
}
int test_atomic_or_fetch(int *p) {
// CHECK: test_atomic_or_fetch
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: [[CALL:%[^ ]*]] = call i32 @__atomic_fetch_or_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Correct atomic libcall support for __atomic_*_fetch builtins. In r244063, I had caused these builtins to call the same-named library functions, __atomic_*_fetch_SIZE. However, this was incorrect: while those functions are in fact supported by GCC's libatomic, they're not documented by the spec (and gcc doesn't ever call them). Instead, you're /supposed/ to call the __atomic_fetch_* builtins and then redo the operation inline to return the final value. Differential Revision: http://reviews.llvm.org/D14385 llvm-svn: 252920
2015-11-13 02:37:29 +08:00
// CHECK: {{%[^ ]*}} = or i32 [[CALL]], 55
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_or_fetch(p, 55, memory_order_seq_cst);
}
int test_atomic_xor_fetch(int *p) {
// CHECK: test_atomic_xor_fetch
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: [[CALL:%[^ ]*]] = call i32 @__atomic_fetch_xor_4(i8* {{%[0-9]+}}, i32 55, i32 5)
Correct atomic libcall support for __atomic_*_fetch builtins. In r244063, I had caused these builtins to call the same-named library functions, __atomic_*_fetch_SIZE. However, this was incorrect: while those functions are in fact supported by GCC's libatomic, they're not documented by the spec (and gcc doesn't ever call them). Instead, you're /supposed/ to call the __atomic_fetch_* builtins and then redo the operation inline to return the final value. Differential Revision: http://reviews.llvm.org/D14385 llvm-svn: 252920
2015-11-13 02:37:29 +08:00
// CHECK: {{%[^ ]*}} = xor i32 [[CALL]], 55
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_xor_fetch(p, 55, memory_order_seq_cst);
}
int test_atomic_nand_fetch(int *p) {
// CHECK: test_atomic_nand_fetch
Revert "Revert "As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there."" This reapplies: 8ff85ed905a7306977d07a5cd67ab4d5a56fafb4 Original commit message: As a follow-up to my initial mail to llvm-dev here's a first pass at the O1 described there. This change doesn't include any change to move from selection dag to fast isel and that will come with other numbers that should help inform that decision. There also haven't been any real debuggability studies with this pipeline yet, this is just the initial start done so that people could see it and we could start tweaking after. Test updates: Outside of the newpm tests most of the updates are coming from either optimization passes not run anymore (and without a compelling argument at the moment) that were largely used for canonicalization in clang. Original post: http://lists.llvm.org/pipermail/llvm-dev/2019-April/131494.html Tags: #llvm Differential Revision: https://reviews.llvm.org/D65410 This reverts commit c9ddb02659e3ece7a0d9d6b4dac7ceea4ae46e6d.
2019-11-27 12:28:52 +08:00
// CHECK: [[CALL:%[^ ]*]] = call i32 @__atomic_fetch_nand_4(i8* {{%[0-9]+}}, i32 55, i32 5)
[CodeGen] remove/fix checks that will fail when r301923 is recommitted Don't test the optimizer as part of front-end verification. llvm-svn: 301928
2017-05-02 23:20:18 +08:00
// FIXME: We should not be checking optimized IR. It changes independently of clang.
// FIXME-CHECK: [[AND:%[^ ]*]] = and i32 [[CALL]], 55
// FIXME-CHECK: {{%[^ ]*}} = xor i32 [[AND]], -1
Add missing atomic libcall support. Support for emitting libcalls for __atomic_fetch_nand and __atomic_{add,sub,and,or,xor,nand}_fetch was missing; add it, and some test cases. Differential Revision: http://reviews.llvm.org/D10847 llvm-svn: 244063
2015-08-06 00:57:36 +08:00
return __atomic_nand_fetch(p, 55, memory_order_seq_cst);
}