Date: Sun, 8 Feb 1998 16:25:09 +1030 From: Greg Lehey <grog@lemis.com> To: dyson@FreeBSD.ORG Cc: hackers@FreeBSD.ORG Subject: Re: ddb/kgdb: How do I find out what a process is doing? Message-ID: <19980208162509.45093@freebie.lemis.com> In-Reply-To: <199802080522.AAA12270@dyson.iquest.net>; from John S. Dyson on Sun, Feb 08, 1998 at 12:22:56AM -0500 References: <19980208152706.59008@freebie.lemis.com> <199802080522.AAA12270@dyson.iquest.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 8 February 1998 at 0:22:56 -0500, John S. Dyson wrote: > Greg Lehey said: >> I'm doing some kernel development work, and I'm using remote kgdb to >> debug it with. I currently have a process which is sleeping >> (eternally) in biord. How do I do a backtrace on it? > > If the debugger doesn't support using alternate kernel stacks, That's a question. Does it or doesn't it? I haven't been able to find any evidence. > I usually put a timeout in the tsleep, and call the debugger from > the tsleep error return. Seldom should a biord take longer than a > few seconds, so perhaps a timeout value of 5*hz should be safe. > This, of course, works well only for reproduceable problems. Sure, and only for biord. In this particular case, I know that it's broken, I was just wondering whether I should take a look at the nature of the breakage before rebooting the machine. Recoding the call to tsleep is overkill. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19980208162509.45093>