http://www.teq.org/jd/pciwalk.out.clean.bz2 11M compressed. I'd suspect the bridge chips are being found, since each PCI-X slot is basically one bus and all of the addon cards in there come up. regards, John On Fri, 23 May 2003, mark tinguely wrote: > If FreeBSD 4.8 does not know about the ServerWorks PCI bridge chips, then > devices on the higher buses are not found. I am surprised that the ACPI > in FreeBSD 5.x did not find them though. > > If the PCI bridge chips are not being discovered, I know in FreeBSD 4.8 > we can add the entry. I am not a real expert in the ACPI code though in > FreeBSD 5.x. > > If you would do an extensive PCI bus search (with pciwalk script below), > it will tell us what the the PCI bridge chips, and which bus holds the > missing GigE card. > > ---pciwalk script--- > from() { > usage="usage: from <start> to <finish> [by <increm>]" > let i=${1?$usage} > let j=${3?$usage} > let k=${5-1} > > if (( i <=j )) ; then > while (( i<=j )) ; do > echo $i > let i=i+k > done > else > while (( i>=j )) ; do > echo $i > let i=i+k > done > fi > } > > > for pci in $(from 0 to 255); do > for slot in $(from 0 to 31); do > for func in $(from 0 to 15); do > zero=$(pciconf -r pci$pci:$slot:$func 0 | sed 's/ //g') > > [[ $zero = 0xffffffff ]] && { continue; } > [[ $zero = 0x00000000 ]] && { continue; } > > echo "$pci:$slot:$func 0 " :$zero: > > for addr in $(from 4 to 255 by 4); do > echo -n "$pci:$slot:$func $addr " > pciconf -r pci$pci:$slot:$func $addr > done | grep -v -e 0xffffffff -e 0x00000000 > done > done > done > --end pciwalk script-- > > --Mark Tinguely >Received on Sun May 25 2003 - 17:37:44 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:09 UTC