https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=235640
--- Comment #3 from Mark Johnston <ma...@freebsd.org> --- This is caused by a race between one end of a pipe being closed and the registration of an EVFILT_WRITE knote for the other end of the pipe. Suppose f_attach (pipe_kqfilter()) succeeds, but before kn_list_lock() is called, the other end of the pipe is closed. pipeclose() removes the EVFILT_WRITE knote from that end's knlist, so kn_list_lock() returns NULL, and we call filt_pipewrite() without the pipe's mutex held. I think this simply means that the assertion in filt_pipewrite() is wrong. We can only assert that the pipe lock is held if wpipe->pipe_present == PIPE_ACTIVE. -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ freebsd-bugs@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"
