Re: Unstable kernel question
2.2.14 should work with the athlon, what cpu arch are you telling the
config program you have? you probably need to tell it Pentium (unless
there is an athlon option). from what i saw 2.3.x didn't have all the
athlon patches from 2.2 yet but it was on alan cox's to-do list. if the
kernel doesn't detect the chipset that is not unusual it just can't
optimize for the chipset, standard drivers would work.
also check the compiler you are using (gcc -v) it is reccomended to use
2.7.2.3 when compiling a kernel, although there is some reports of success
with egcs 1.1 and very few success reports using egcs 2.95.
nate
On Tue, 8 Feb 2000, [iso-8859-1] José Luis Gómez Dans wrote:
j.l.go >Hi!
j.l.go > I don't know if you remember some query about compiling a new
j.l.go >kernel for Athlon based systems. Basically, my kernel just woudln't
j.l.go >work. In the end, I found out that this wasn't due to a wrong processor
j.l.go >or anything like that: it was due to the motherboard (WS-6167), and in
j.l.go >particular of its IDE controller (Viper 756, IIRC). Right, AFAIK, 2.2.14
j.l.go >does not have Viper 756 controller, so no wonder my kernels weren't
j.l.go >working! I downloaded 2.3.42 (the latest unstable), and had a go at it.
j.l.go >Now, it sort of boots up, and it detects the actual chipset. This is the
j.l.go >old output with 2.2.13 (as installed in the bootdisks)
j.l.go >
j.l.go >PCI_IDE: unknown IDE controller on PCI bus 00 device 39, VID=1022,
j.l.go >DID=7409
j.l.go >PCI_IDE: not 100% native mode: will probe irqs later
j.l.go >ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:pio
j.l.go >ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:DMA, hdd:pio
j.l.go >
j.l.go >and this is the log for the 2.3 kernel
j.l.go >AMD7409: IDE controller on PCI bus 00 dev 39
j.l.go >AMD7409: not 100% native mode: will probe irqs later
j.l.go >ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:pio
j.l.go >ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:DMA, hdd:pio
j.l.go >
j.l.go > However, the new kernel seems to splutter about. In particular,
j.l.go >I had loads of segfaults, so I rebooted into my old kernel, which
j.l.go >wouldn't boot up properly, as libm.so.6 didn't have a proper ELF header.
j.l.go >I solved this by installing libc6 again, as libm is quite widely used.
j.l.go >
j.l.go > My question is: I don't want to live on the bleeding edge kernel
j.l.go >stuff, but I'd like to compile a kernel with the Viper chip support (and
j.l.go >whatever else that needs). Is there any way to patch my 2.2.14 sources
j.l.go >so that this is possible?
j.l.go >
j.l.go > Thanks,
j.l.go > José
j.l.go >--
j.l.go >José L Gómez Dans PhD student
j.l.go > Radar & Communications Group
j.l.go > Department of Electronic Engineering
j.l.go > University of Sheffield UK
j.l.go >
j.l.go >
j.l.go >--
j.l.go >Unsubscribe? mail -s unsubscribe debian-user-request@lists.debian.org < /dev/null
j.l.go >
----------------------------------------[mailto:aphro@aphroland.org ]--
Vice President Network Operations http://www.firetrail.com/
Firetrail Internet Services Limited http://www.aphroland.org/
Everett, WA 425-348-7336 http://www.linuxpowered.net/
Powered By: http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMP http://yahoo.aphroland.org/
-----------------------------------------[mailto:aphro@netquest.net ]--
7:12am up 172 days, 19:26, 1 user, load average: 1.00, 1.02, 1.05
Reply to: