runtime: TestFutexsleep fails when GOMAXPROCS>1 #7496
Labels
Comments
runtime/lock_futex.c says: // This implementation depends on OS-specific implementations of // // runtime·futexsleep(uint32 *addr, uint32 val, int64 ns) // Atomically, // if(*addr == val) sleep // Might be woken up spuriously; that's allowed. // Don't sleep longer than ns; ns < 0 means forever. // // runtime·futexwakeup(uint32 *addr, uint32 cnt) // If any procs are sleeping on addr, wake up at most cnt. |
|
This issue was closed by revision 4888781. Status changed to Fixed. |
|
This issue was updated by revision ae9b661. LGTM=iant R=golang-codereviews, gobot, iant, bradfitz CC=golang-codereviews https://golang.org/cl/72760043 |
This issue was closed.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
$ GOMAXPROCS=2 go test runtime -short -cpu=1,2,4,10 --- FAIL: TestFutexsleep-2 (0.00 seconds) futex_test.go:51: futexsleep test "before the year 2038" finished early after 74.681us --- FAIL: TestFutexsleep-4 (0.00 seconds) futex_test.go:61: futexsleep test "after the year 2038" finished early after 93.989us FAIL and http://build.golang.org/log/feb6cc2cb44943d70a431c8bba4ffdffc670a590 http://build.golang.org/log/c81fb002e52c602bfb4ba145a3e6fc9f87be1275 Looks there's some secret on MP stuff+userspace mutex.The text was updated successfully, but these errors were encountered: