Bartlomiej Zolnierkiewicz wrote:
> On Tuesday 13 of January 2004 16:40, Jens Benecke wrote:
>> Bartlomiej Zolnierkiewicz wrote:
>> >> I have found a (perhaps THE) reason why my X is so jerky: the nforce2
>> >> chipset driver (amd74xx) doesn't load, because it "thinks" the BIOS
>> >> IDE ports are disabled - which is definitely not the case
>> >
>> > It doesn't load because IDE ports are already controlled by generic IDE
>> > code.
>> > Just use CONFIG_BLK_DEV_AMD74XX=y. I will fix this "BIOS" comment.
>>
>> I can't, because I (plan to) use this kernel on many different machines.
>> Not all of those (in fact: only one) uses the amd74xx module.
>
> So what? It won't be used on other machines, but it will eat a little
> kernel image space & memory.
Then I'd have to statically compile in *all* IDE modules.
The point is that I'm providing specially configured kernels for a large
group of users, most of which I don't even know (much less their hardware
configurations).
>> Is there a kernel parameter I can use to disable the generic IDE code on
>> boot?
>
> No, but I will later make patch to allow disabling/modularizing it.
That'd be great. :-)
Thanks!
PS: this worked in 2.4 (loading the IDE driver later as module, but booting
from IDE as well), why doesn't it work in 2.6 any more?
--
Jens Benecke (jens at spamfreemail.de)
http://www.hitchhikers.de - Europaweite kostenlose Mitfahrzentrale
http://www.spamfreemail.de - 100% saubere Postf?cher - garantiert!
http://www.rb-hosting.de - PHP ab 9? - SSH ab 19? - g?nstiger Traffic
On Tuesday 13 of January 2004 19:11, Jens Benecke wrote:
> PS: this worked in 2.4 (loading the IDE driver later as module, but booting
> from IDE as well), why doesn't it work in 2.6 any more?
Because 2.6.x is different (most host drivers probe for drives themselves)
and nobody fixed this issue :-).
--bart