You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Reproduce (x86_64 linux):
% ./bin_linux/asan_test64 --gtest_filter=AddressSanitizerInterface.GetOwnershipStressTest:AddressSanitizer.OOB_*
2>&1 | grep OK
[ OK ] AddressSanitizerInterface.GetOwnershipStressTest (3854 ms)
[ OK ] AddressSanitizer.OOB_char (5224 ms)
[ OK ] AddressSanitizer.OOB_int (5208 ms)
./bin_linux/asan_test64 --gtest_filter=AddressSanitizer.OOB_* 2>&1 | grep OK
[ OK ] AddressSanitizer.OOB_char (760 ms)
[ OK ] AddressSanitizer.OOB_int (794 ms)
I suspect that multiple mallocs in AddressSanitizerInterface.GetOwnershipStressTest
interfere with the death tests in AddressSanitizer.OOB_char somehow.
Reported by konstantin.s.serebryany on 2012-06-05 13:23:41
The text was updated successfully, but these errors were encountered:
Simple experiment:
TEST(AddressSanitizer, MultipleDeathTest) {
for (int i = 0; i < 100; i++) {
EXPECT_DEATH(exit(1), "");
}
}
TEST(AddressSanitizer, MultipleDeathAndMallocTest) {
free_aaa(malloc_aaa(1 << 29));
for (int i = 0; i < 100; i++) {
EXPECT_DEATH(exit(1), "");
}
}
The first test runs < 2 seconds, the second test runs 10 seconds.
Looks like the cost of the death test depends on the amount of reserved memory.
Oh my...
Reported by konstantin.s.serebryany on 2012-06-05 13:36:27
Originally reported on Google Code with ID 77
Reported by
konstantin.s.serebryany
on 2012-06-05 13:23:41The text was updated successfully, but these errors were encountered: