Date: Sun, 26 Jun 2005 16:42:25 -0700 From: Eric Anholt <eta@lclark.edu> To: Martin Cracauer <cracauer@cons.org> Cc: freebsd-current@freebsd.org, Adam K Kirchhoff <adamk@voicenet.com>, Warner Losh <imp@bsdimp.com> Subject: Re: 6.0-current panic: loading radeon module Message-ID: <1119829345.3817.62.camel@leguin> In-Reply-To: <20050626191956.A59922@cons.org> References: <20050617173008.A11142@cons.org> <20050617194638.A13394@cons.org> <20050617.233055.41723867.imp@bsdimp.com> <200506241641.25433.jhb@FreeBSD.org> <1119650444.1173.105.camel@leguin> <20050624195906.A5702@cons.org> <1119816026.3817.20.camel@leguin> <20050626191956.A59922@cons.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 2005-06-26 at 19:19 -0400, Martin Cracauer wrote: > > OK, one more try, for everyone who's been having hangs/reboots on X > > startup since April. Attached is a patch that will hopefully kill the > > problem off. > > The candidate gets 100 points. It fixes the panic and hardware 3D > worked for a few tests. No other side effects observed. Thanks so > much! > > Out of curiosity, why does a 32 MB VRAM card try to map 128 MB? Do you only had 32MB? If so, oops on my part. Every other case of this I had seen was 128MB, so I was just assuming. Basically the issue (afaik) is that you're running out of KVA because framebuffers are large and KVA isn't all that huge. pmap_mapdev was not designed to handle failure to allocate memory, so it just paniced instead of returning NULL (which wouldn't have helped us anyway, due to other issues). -- Eric Anholt eta@lclark.edu http://people.freebsd.org/~anholt/ anholt@FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1119829345.3817.62.camel>