RE: collapsing devices.

[Yinghai Lu]

Eric,

I wonder if ht_collapse_previous_enumeration need to the collapse from tail
to head.

For example if

8131(1)-8131-8111
Unit id 1-->3-->5

Then if clear the 8131(1) at first, can left chain still be accessed.....

YH

-----Original Message-----
From: [e-mail]
[mailto:[e-mail]] On Behalf Of Yinghai Lu
Sent: Sunday, December 12, 2004 5:45 PM
To: 'Stefan Reinauer'; [e-mail]
Subject: RE: collapsing devices.

Actually for cpu0 link1, in the auto.c stage, incoherent.c only need to
handle bus 0, because ht device only at bus 0 (8131 and 8111).

For cpu1 link2, in the auto.c stage, incoherent device only need to handle
to bus 4, because we put the device on bus 4 according to setting in 0xe4 of
PCI_dev.

The purpose of incoherent.c is set the HT-link and width and return
need_reset. Auto.c will according to need_reset from coherent.c and
incoherent.c to reset in auto.c stage.

So in the northbrige.c of hardwaremain stage will not need to reset and make
ht speed and width effective.

The HT devices in the same HT chain are always in the same bus.

YH

-----Original Message-----
From: [e-mail]
[mailto:[e-mail]] On Behalf Of Stefan Reinauer
Sent: Saturday, December 11, 2004 12:12 PM
To: [e-mail]
Subject: collapsing devices.

Hi,

I am desperate, still getting:

LinuxBIOS-1.1.7.0-fallback Sat Dec 11 20:30:55 CET 2004 starting...
setting up resource map....done.
04 nodes initialized.
udev=(0,0x18,0) 0xb4=00030000
Pretending to collapse device 74501022 on bus 00000000
udev=(0,0x19,0) 0xd4=000a0400
Pretending to collapse device 74501022 on bus 00000004

* Why would only bus 0 and 4 be collapsed, but not bus 1,2,3?

Stefan

_______________________________________________
Linuxbios mailing list
[e-mail]
http://www.clustermatic.org/mailman/listinfo/linuxbios

_______________________________________________
Linuxbios mailing list
[e-mail]
http://www.clustermatic.org/mailman/listinfo/linuxbios