Comment 208 for bug 1512120

(In reply to Andreas Böhler from comment #134)
> I forgot to mention that I am on an x64 Arch Linux machine (current as of
> today). an strace (with patch #6779 applied) gives:
>
> write(4, "\1\0\0\0\0\0\0\0", 8) = 8
> recvmsg(5, {msg_namelen=0}, MSG_CMSG_CLOEXEC) = -1 EAGAIN (Resource
> temporarily unavailable)
> inotify_rm_watch(14, 5) = 0
> futex(0x7fa94ca47a60, FUTEX_WAKE_PRIVATE, 1) = 1
> inotify_rm_watch(14, 6) = 0
> futex(0x7fa94ca47a60, FUTEX_WAKE_PRIVATE, 1) = 1
> lstat("/path/to/directory", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
> inotify_add_watch(14, "/path/to/directory",
> IN_MODIFY|IN_ATTRIB|IN_CLOSE_WRITE|IN_MOVED_FROM|IN_MOVED_TO|IN_CREATE|IN_DEL
> ETE|IN_DELETE_SELF|IN_MOVE_SELF|IN_UNMOUNT|IN_ONLYDIR) = 18
> futex(0x69f4f8, FUTEX_WAIT_PRIVATE, 2, NULL
>
> And there it stops and becomes unresponsive.

Yup, I can confirm that. I've pushed the attachment #6779 to Debian and got similar reports, and can confirm it myself. It happens when accessing a removable device, for example.