4_2_0-0pre1v1 on hppa
I'm seeing problems where the sti framebuffer is not being found during
(==) Log file: "/var/log/XFree86.0.log", Time: Sun Jun 30 10:37:20 2002
(==) Using config file: "/etc/X11/XF86Config-4"
(EE) No OS PCI support available
(EE) FBDEV(0): Driver can't support depth 24
(EE) Screen(s) found, but none have a usable configuration.
What appears to be happening is that there's a scan of /proc/bus/pci to
find the video adapter. Which shouldn't work because my C110 doesn't
have a PCI bus (eisa, yes). I'm going to check the patches between 4.1.0
series, which did work fine, and 4.2.0. Maybe some got lost in the upgrade.
But assuming nothing was lost, and this is a change in the hardware
discovery algorithm... are we really pretending that all video cards
show up in /proc/bus/pci? Including isa, eisa, mca and vesa-bus
adapters? Because if we are, then I'm making fake PCI ids for HP cards
I'd rather follow the path taken by sparc/32, and read a different tree
(in their case, the openprom tree.) I'd have to modify the kernel to
expose "hardware found" information as a /proc file (but so what?)
____ .:. ____
Bryan W. Headley - firstname.lastname@example.org
To UNSUBSCRIBE, email to email@example.com
with a subject of "unsubscribe". Trouble? Contact firstname.lastname@example.org