maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/compiler-rt/test/tsan/thread_detach.c

21 lines
399 B
C
Raw Normal View History

tsan: allocate vector clocks using slab allocator Vector clocks is the most actively allocated object in tsan runtime. Current internal allocator is not scalable enough to handle allocation of clocks in scalable way (too small caches). This changes transforms clocks to 2-level array with 512-byte blocks. Since all blocks are of the same size, it's possible to cache them more efficiently in per-thread caches. llvm-svn: 214912
2014-08-06 02:45:02 +08:00
// RUN: %clang_tsan -O1 %s -o %t && %run %t 2>&1 | FileCheck %s
tsan: remove sleeps from tests Even sleep(1) lead to episodical flakes on some machines. Use an invisible by tsan barrier to enforce required execution order instead. This makes the tests deterministic and faster. llvm-svn: 226659
2015-01-21 21:50:02 +08:00
#include "test.h"
tsan: allocate vector clocks using slab allocator Vector clocks is the most actively allocated object in tsan runtime. Current internal allocator is not scalable enough to handle allocation of clocks in scalable way (too small caches). This changes transforms clocks to 2-level array with 512-byte blocks. Since all blocks are of the same size, it's possible to cache them more efficiently in per-thread caches. llvm-svn: 214912
2014-08-06 02:45:02 +08:00
void *Thread(void *x) {
tsan: remove sleeps from tests Even sleep(1) lead to episodical flakes on some machines. Use an invisible by tsan barrier to enforce required execution order instead. This makes the tests deterministic and faster. llvm-svn: 226659
2015-01-21 21:50:02 +08:00
barrier_wait(&barrier);
tsan: allocate vector clocks using slab allocator Vector clocks is the most actively allocated object in tsan runtime. Current internal allocator is not scalable enough to handle allocation of clocks in scalable way (too small caches). This changes transforms clocks to 2-level array with 512-byte blocks. Since all blocks are of the same size, it's possible to cache them more efficiently in per-thread caches. llvm-svn: 214912
2014-08-06 02:45:02 +08:00
return 0;
}
int main() {
tsan: remove sleeps from tests Even sleep(1) lead to episodical flakes on some machines. Use an invisible by tsan barrier to enforce required execution order instead. This makes the tests deterministic and faster. llvm-svn: 226659
2015-01-21 21:50:02 +08:00
barrier_init(&barrier, 2);
tsan: allocate vector clocks using slab allocator Vector clocks is the most actively allocated object in tsan runtime. Current internal allocator is not scalable enough to handle allocation of clocks in scalable way (too small caches). This changes transforms clocks to 2-level array with 512-byte blocks. Since all blocks are of the same size, it's possible to cache them more efficiently in per-thread caches. llvm-svn: 214912
2014-08-06 02:45:02 +08:00
pthread_t t;
pthread_create(&t, 0, Thread, 0);
tsan: remove sleeps from tests Even sleep(1) lead to episodical flakes on some machines. Use an invisible by tsan barrier to enforce required execution order instead. This makes the tests deterministic and faster. llvm-svn: 226659
2015-01-21 21:50:02 +08:00
barrier_wait(&barrier);
tsan: allocate vector clocks using slab allocator Vector clocks is the most actively allocated object in tsan runtime. Current internal allocator is not scalable enough to handle allocation of clocks in scalable way (too small caches). This changes transforms clocks to 2-level array with 512-byte blocks. Since all blocks are of the same size, it's possible to cache them more efficiently in per-thread caches. llvm-svn: 214912
2014-08-06 02:45:02 +08:00
pthread_detach(t);
[TSAN] Make tsan tests more portable (take 3) Some tests didn't merge stderr with stdout. Patch by Maxim Kuvyrkov. llvm-svn: 266426
2016-04-15 20:34:00 +08:00
fprintf(stderr, "PASS\n");
tsan: allocate vector clocks using slab allocator Vector clocks is the most actively allocated object in tsan runtime. Current internal allocator is not scalable enough to handle allocation of clocks in scalable way (too small caches). This changes transforms clocks to 2-level array with 512-byte blocks. Since all blocks are of the same size, it's possible to cache them more efficiently in per-thread caches. llvm-svn: 214912
2014-08-06 02:45:02 +08:00
return 0;
}
// CHECK-NOT: WARNING: ThreadSanitizer: thread leak
// CHECK: PASS