Date: Mon, 13 Feb 2006 17:44:14 +0200 From: Giorgos Keramidas <keramida@ceid.upatras.gr> To: Robert Watson <rwatson@freebsd.org> Cc: freebsd-amd64@freebsd.org, freebsd-current@freebsd.org, Olivier Houchard <cognet@freebsd.org>, Steve Kargl <sgk@troutmask.apl.washington.edu> Subject: Re: HEADSUP: New pts code triggers panics on amd64 systems. Message-ID: <20060213154414.GA909@flame.pc> In-Reply-To: <20060210230950.GA938@flame.pc> References: <20060201235556.GA708@troutmask.apl.washington.edu> <20060207104411.GA1067@flame.pc> <20060207132335.W37594@fledge.watson.org> <20060210230950.GA938@flame.pc>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2006-02-11 01:09, Giorgos Keramidas <keramida@ceid.upatras.gr> wrote: > I'm not sure if what I see is a pts side-effect. I rebuilt a snapshot > of HEAD again today, and discovered something else that may help us > track this down. > > A few more observations, in case they prove helpful to someone more > acquainted with the way syscons output is supposed to work: > > - If I keep hitting Scroll-Lock again and again, then syscons output > *does* eventually appear. > > - If I type stuff without seeing it and then press RETURN twice, the > previous from last line *does* appear in my terminal. > > - When pressing CTRL-ALT-ESC, the debugger starts normally, but after > typing many times commands that have large output, i.e.: > > show witness > > the console locks up entirely. > > This looks like a locking problem, instead of a pts/syscons one :-/ More info available. I finally managed to run top(1) in multiuser mode with a HEAD kernel that was updated this morning, but without the 'pts' bits in kernel or userland. The console is still unusable and everything feels very sluggish. Looking at top output, when it eventually updated itself I saw 75% interrupt time and 25% idle time? Any ideas why this would happen with recent kernels? - Giorgos
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060213154414.GA909>