Date: Wed, 08 Sep 1999 10:33:30 +0100 From: Alan Judge <Alan.Judge@indigo.ie> To: stable@freebsd.org Subject: Re: newly introduced repeatable SMP hang in STABLE/3.3-RC Message-ID: <19990908093405.BC82B1534C@hub.freebsd.org> In-Reply-To: Message from Jeremy McMillan dated Tuesday at 19:30.
next in thread | raw e-mail | index | archive | help
Jeremy> Isn't this *wedged* instead of *hung*? I guess it depends on your terminology. Wedged is probably more accurate though. Jeremy> and what *exactly* is wedged/hung? All user-land processes, though ICMP still works, you can still get into DDB, hitting a key on the console correctly disables the screen saver, and you can still switch between VTYs. And to answer Mike Smith's message as well, this makes it sound like it's not spltty. I don't know enough about the interrupt internals for FreeBSD, since most of my kernel hacking has been on other platforms. Is there any easy way to tell from a dump what the current spl was. From the code it looks like cpl is the right variable. It's set to all ones, which looks like splhigh or splclock. But that could be a result of being in DDB or a side effect of MPgetlock. Anyone know how to trace where MPgetlock is called from and what the other CPU is doing? -- Alan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990908093405.BC82B1534C>