Running with or without kern.smp.disabled, with a very up-to-date kernel, I can easily lock up the system by running gaithrstress -t 50 a few times at once. There is nothing special that has changed in the last few weeks, though I have extensive changes to kqueue and VM which I consider entirely stable at this point. Anything causing a problem like this seems like it's due to changes within the last few weeks. The only clue I have comes from a previous panic() (which promptly just locked up going into "KDB: enter: panic"), "kse/thread mismatch" as generated by choosethread(). The connect() system call was trying to do an mi_switch() inside a mtx_lock() at the time. -- Brian Fundakowski Feldman \'[ FreeBSD ]''''''''''\ <> green_at_FreeBSD.org \ The Power to Serve! \ Opinions expressed are my own. \,,,,,,,,,,,,,,,,,,,,,,\Received on Mon Jul 19 2004 - 03:25:36 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:02 UTC