Date: Sun, 28 Dec 1997 16:18:44 +1030 From: Greg Lehey <grog@lemis.com> To: Chris Timmons <skynyrd@opus.cts.cwu.edu> Cc: Bruce Evans <bde@zeta.org.au>, julian@whistle.com, freebsd-current@FreeBSD.ORG Subject: Re: Remote gdb (was: no boot: config -g and options DDB) Message-ID: <19971228161844.52116@lemis.com> In-Reply-To: <Pine.BSF.3.96.971227213721.8511A-100000@opus.cts.cwu.edu>; from Chris Timmons on Sat, Dec 27, 1997 at 09:40:36PM -0800 References: <19971228155633.57807@lemis.com> <Pine.BSF.3.96.971227213721.8511A-100000@opus.cts.cwu.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Dec 27, 1997 at 09:40:36PM -0800, Chris Timmons wrote: > On Sun, 28 Dec 1997, Greg Lehey wrote: > >> OK, I have it working now. On the face of it, it looks as if these >> fixes did it, but I couldn't be sure without going back and trying >> without. Thanks. >> >> In summary, it appears that you need to build a kernel with COMCONSOLE >> and at least one of the sio ports with flags 0x10. Contrary to my >> previous claim, it doesn't have to be sio0: the kernel chooses the >> first sio with flags 0x10. These fixes mean that it will work without >> the -h flag. > > COMCONSOLE exists only in our hearts now :) Oh. > Are you sure it chooses the first sio with flags 0x10, or does the first > boot prompt get squirted out all eligible sio units? It's in sioncprobe(), the function for which Bruce supplied the patches. It goes out and looks for the first "enabled" console, does its thing and breaks out of the loop. Its thing includes setting the variable siocniobase, which is used by the other serial console routines to specify the output port. This stuff's pretty basic. Don't expect it to be clever enough to maintain the pressure in multiple pipes :-) Greg
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19971228161844.52116>