Date: Thu, 27 Jul 2006 17:35:54 -0400 From: John Baldwin <jhb@freebsd.org> To: freebsd-hackers@freebsd.org Cc: Dag-Erling =?iso-8859-1?q?Sm=F8rgrav?= <des@des.no>, Intron <mag@intron.ac>, Hans Petter Selasky <hselasky@c2i.net> Subject: Re: How to Use ddb(4)? Message-ID: <200607271735.55199.jhb@freebsd.org> In-Reply-To: <200607272104.13079.hselasky@c2i.net> References: <1153997829.12770@origin.intron.ac> <868xme28nt.fsf@xps.des.no> <200607272104.13079.hselasky@c2i.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday 27 July 2006 15:04, Hans Petter Selasky wrote: > On Thursday 27 July 2006 20:40, Dag-Erling Sm=F8rgrav wrote: > > maksim yevmenkin <maksim.yevmenkin@savvis.net> writes: > > > Dag-Erling Sm=F8rgrav <des@des.no> writes: > > > > No, kbdmux needs to be fixed so it works in DDB. > > > > > > actually, atkbd(4) needs to be fixed to support "polled" mode :) > > > > It used to work fine before kbdmux(4) came along... > > >=20 > Just a comment: Don't forget to lock Giant before calling anything in the= =20 > keyboard layer. That means DDB must lock Giant, else at least my new USB= =20 > keyboard driver will panic, saying Giant is not owned. The old USB keyboa= rd=20 > driver did not care about this. DDB should not be acquiring locks or depend on being able to do so. When y= ou=20 panic the machine is an unknown state for one, and you really want to leave= =20 the overall system state as undisturbed as possible while in the debugger t= o=20 aid in debugging. =2D-=20 John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200607271735.55199.jhb>