Don't corrupt the thread list if the main thread exits.

...and don't pass a non-heap pointer to free(3), either.

This patch replaces the "node** prev" with the clearer "node* prev"
style and fixes the null pointer dereference in the old code. That's
not sufficient to fix the reporter's bug, though. The pthread_internal_t*
for the main thread isn't heap-allocated --- __libc_init_tls causes a
pointer to a statically-allocated pthread_internal_t to be added to
the thread list.

Bug: http://code.google.com/p/android/issues/detail?id=37410
Change-Id: I112b7f22782fc789d58f9c783f7b323bda8fb8b7
This commit is contained in:
Elliott Hughes
2012-11-01 16:33:29 -07:00
parent 064f862d55
commit 4f251bee5d
4 changed files with 109 additions and 98 deletions

View File

@@ -109,3 +109,20 @@ TEST(pthread, pthread_join_self) {
void* result;
ASSERT_EQ(EDEADLK, pthread_join(pthread_self(), &result));
}
#if __BIONIC__ // For some reason, gtest on bionic can cope with this but gtest on glibc can't.
static void TestBug37410() {
pthread_t t1;
ASSERT_EQ(0, pthread_create(&t1, NULL, JoinFn, reinterpret_cast<void*>(pthread_self())));
pthread_exit(NULL);
}
// We have to say "DeathTest" here so gtest knows to run this test (which exits)
// in its own process.
TEST(pthread_DeathTest, pthread_bug_37410) {
// http://code.google.com/p/android/issues/detail?id=37410
::testing::FLAGS_gtest_death_test_style = "threadsafe";
EXPECT_EXIT(TestBug37410(), ::testing::ExitedWithCode(0), "");
}
#endif