Date: Wed, 26 Mar 2014 12:06:30 -0400 From: John Baldwin <jhb@freebsd.org> To: Koji Mori <mrkj.yy@gmail.com> Cc: freebsd-users-jp@freebsd.org Subject: [FreeBSD-users-jp 95175] Re: FreeBSD 10.0 =?iso-8859-15?q?=1B=24B=24G=1B?=(B NIC (em) =?iso-8859-15?q?=1B=24B=24?=, G'<=?iso-8859-15?q?1=245=24l=24J=1B?=( =?iso-8859-15?q?B=1B=24B=24=24=1B?=(B Message-ID: <201403261206.30698.jhb@freebsd.org> In-Reply-To: <53322870.7090600@gmail.com> References: <5331393E.5020007@gmail.com> <896B9B3C-AD2E-4337-B8AA-DFA0ECDF0126@bsdimp.com> <53322870.7090600@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday, March 25, 2014 9:08:00 pm Koji Mori wrote: > (2014/03/26 3:21), Warner Losh wrote: > > > Hi Mori-san. Is there a dmesg with bootverbose=1? It looks like none of the I/O ports are allocated. > > I forwarded your email to John Baldwin and he asked. > > Is is not with bootverbose=1. > > Then, I tried reboot, so this attached file is a whole dmesg with > bootverbose=1. > > Thanks, Warner. Ugh, I see the problem. pcib1 is marked as doing ISA decoding (so we jump through a whole lot of hoops to only allocate specific subranges), but pcib2 is behind pcib1 and is not marked as ISA decoding. Let me look to see what the standard says about this. The simplest fix might be to explicitly tag all bridges behind an ISA decoding PCI-PCI bridge as ISA decoding. -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201403261206.30698.jhb>