Date: Sat, 20 Dec 1997 01:00:01 -0800 (PST) From: Stefan Esser <se@FreeBSD.ORG> To: freebsd-bugs Subject: Re: i386/5300: /sys/i386/isa/pcibus.c does'nt discovered Cyclades Cyclom-16YeP Message-ID: <199712200900.BAA00302@hub.freebsd.org>
index | next in thread | raw e-mail
The following reply was made to PR i386/5300; it has been noted by GNATS.
From: Stefan Esser <se@FreeBSD.ORG>
To: nnd@itfs.nsk.su
Cc: FreeBSD-gnats-submit@freebsd.org, Stefan Esser <se@freebsd.org>
Subject: Re: i386/5300: /sys/i386/isa/pcibus.c does'nt discovered Cyclades Cyclom-16YeP
Date: Sat, 20 Dec 1997 09:52:29 +0100
On 1997-12-15 12:59 +0600, nnd@itfs.nsk.su wrote:
> Here is the (extract from) boot:-v dmesg for one of my PCs
> before and after the next "patch" was applied to /sys/i386/isa/pcibus.c
> class = inl(pci_cfgenable(0, device, 0, 8, 4)) >> 8;
> if (bootverbose)
> printf("[class=%06x] ", class);
> - if (class == 0 || (class & 0xf8f0ff) != 0)
> + if (class == 0 || (class & 0xf870ff) != 0)
Hmmm, I had to discard the class == 0 case, since that value
is often returned even if there is no PCI device ...
Since 0 is an obsolete class code ("old"), I decided to test
for at least one other PCI device with a non-zero class, and
in general there is at least a VGA card (and possibly disk
controllers and network cards) on PCI bus 0, which satisfy
that condition.
Your case is special, and I agree, that the test must be made
less strict.
OK. The bit-mask you suggest seems reasonable, and I'll commit
your patch to the -current tree!
Thanks for sending the patch, and sorry for the inconvenience!
Regards, STefan
help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199712200900.BAA00302>