Date: Sun, 04 Mar 2001 09:34:33 -0800 (PST) From: John Baldwin <jhb@FreeBSD.org> To: Bruce Evans <bde@zeta.org.au> Cc: freebsd-current@FreeBSD.org, The Hermit Hacker <scrappy@hub.org> Subject: RE: Problem with sio in -current ... possible cause of hangs? Message-ID: <XFMail.010304093433.jhb@FreeBSD.org> In-Reply-To: <Pine.BSF.4.21.0103042008330.6902-100000@besplex.bde.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 04-Mar-01 Bruce Evans wrote: > On Sat, 3 Mar 2001, John Baldwin wrote: > >> On 03-Mar-01 The Hermit Hacker wrote: >> > If I plug my cable from /dev/ttyd0 -> /dev/ttyd1 on the same >> > machine, run getty on /dev/ttyd1 and use kermit to connect to /dev/cuaa0, >> > I get no response back, which is why I'm wondering about sio ... >> >> Try turning clocal off on the host and port you are running kermit on. Even > > Or just use /dev/cuaa*. Erm, well, that is what I use, but clocal defaults to off on those I think. >> then, I still have yet to get getty to work at all, it's always stuck in >> 'siodcd'. I've noticed via debugging output that the DCD change bit does >> raise >> for a read, but that teh DCD status bit stays at zero the entire time. The > > This seems to indicate a cabling problem. DCD and DCD change shouldn't > change for a read; they should only change when "carrier" changes. I meant 'read' as in a read of the status register containing the DCD. >> sio driver seems to ignore the change bit and only read the status bit, so >> it >> thinks DCD is never raised and hangs forever on open. Note that I can get a > > Very short transients in DCD would be missed by the driver. The DCD change > bit is for not missing transients, but in normal modem applications missing > transients is probably a feature. Anyway, it's not clear what state change > should occur in the driver if the DCD state hasn't changed when the driver > looks at it. > >> getty fine on a serial console, just not on a /dev/ttydX that's not also the >> serial console. :( I've had this problem since before PRE_SMPNG however. > > Certainly a hardware problem :-). Well, it seems to be now as serial console in the loader doesn't even work. The odd thing is that the same exact cables in the same exact setup worked on the SMPng tree a week before PRE_SMPNG because I used it for remote gdb and serial console. About a week before the commit I started getting the siodcd hangs and haven't been able to shake them ever since. :( > Bruce -- John Baldwin <jhb@FreeBSD.org> -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.baldwin.cx/~john/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.010304093433.jhb>