[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Anyone tried the ASUS A8V-E Deluxe?



Lennart Sorensen wrote:

On Sun, Mar 06, 2005 at 02:03:56AM +0200, Kyuu Eturautti wrote:
Thank you for the tip. I can now confirm that the A8V-E Deluxe onboard NIC will function with kernel versions up to 2.6.9 but not with 2.6.10 or .11. After I set up 2.6.9, I ran the install.sh script that came with the driver from syskonnect.com and generated a patch, which I applied on the kernel sources. The other installation options (expert and user) did not work. I tried all of the install options on 2.6.11, 2.6.10 and 2.6.9, and only 2.6.9 worked.

It seems I'm stuck with 2.6.9 for this particular motherboard. A newer linux driver is clearly needed from syskonnect / marvell. The nature of the errors (functions pci_save_state and pci_restore_state) might indicate functionality added in 2.6.10. I wouldn't know, I'm not a coder.

I hate saying it, but every time I buy an Asus motherboard, I face a lot of problems. It's happened now seven times in a row...twice this week alone. I guess this is what can be called human-hardware incompatibility.

Hmm, how odd.  I have never had a problem with Asus boards, and I have
quite a few (haven't bought anything else since 1992).

It seems odd that a driver working in 2.6.9 stops working in newer
versions, so someone changed something in a way that broke it, and
perhaps not just on Asus, although given how common they are, I guess
Asus being the best known for encountering this problem does make sense.
I wonder if there is anything different about how Asus connects this
chip, or if they are just using a newer/older revision than other boards
(assuming other boards have sk98lin onboard and not just cheap
realteks).

Len Sorensen
As I understand it, this Asus model connects to the NIC via PCI Express. That's obviously a (good) change and may have something to do with the incompatibilities.

However, the syskonnect.com driver fails in compilation already, so this is probably just something they need to modify there to support some functions that were changed fron 2.6.9 to 2.6.10. I'm assuming it won't be much trouble for them, but the better question is WHEN they'll be doing the fix. I've e-mailed them with details, let's hopt they have a chance to respond.

In the meantime, I think I'll condemn that particular motherboard to Windows work. I'll test this further on Linux when and if new driver versions are released. I'll inform the debian-amd64 list at that time, unless someone beats me to it.


/v\



Reply to: