1 |
On Saturday 25 March 2006 17:37, Duncan <1i5t5.duncan@×××.net> wrote |
2 |
about '[gentoo-amd64] Re: 2.6.16 and ndiswrapper': |
3 |
> I only found another late in the cycle, when I upgraded to 8 gigs of |
4 |
> memory from 1 gig. .15 and stable releases thereof work. .16 doesn't. |
5 |
> I haven't had time to trace that one further, however, and it's possible |
6 |
> I just don't quite have the kernel configured correctly and .15 just |
7 |
> happens to work anyway. This issue has to do with SATA, actually, I |
8 |
> think SCSI. With 8 gig of memory I'd normally configure IOMMU on but |
9 |
> that doesn't work with .15 or .16. With it off, .15 works, but .16 |
10 |
> fails when it tries to load the (libata based therefore SCSI based) SATA |
11 |
> RAID, saying the SCSI device nodes don't exist! My root is on RAID so |
12 |
> this is early kernel, where it first tries to load |
13 |
> SCSI-then-RAID-then-rootfs-read-only, so it's well before anything |
14 |
> userspace side is running, so it /has/ to be a kernel issue. There's a |
15 |
> changelog entry saying they eliminated bounce-buffers for SCSI that I |
16 |
> think is the problem, since bounce-buffers are >4 gig memory related, |
17 |
> but as I said, I haven't traced it yet, so I can't say for sure. |
18 |
|
19 |
I had something like this happening on my 4GB system. With IOMMU off, I'd |
20 |
only have access to like 2G of memory. With IOMMU on, I had various |
21 |
results, ranging from kernel panic before initrd is loaded to missing |
22 |
memory (small, like maybe the size of my PCI IO window?) to a completely |
23 |
working system with devices properly mapped to memory addresses beyond my |
24 |
physical RAM. These various behaviors were controlled, oddly enough, by |
25 |
my BIOS settings. |
26 |
|
27 |
I'm running a Tyan Dual-Opteron Dual-PCIe board... If you've got a similar |
28 |
board, I can share my BIOS and kernel settings and maybe you can resolve |
29 |
your IOMMU issues. |
30 |
|
31 |
-- |
32 |
"If there's one thing we've established over the years, |
33 |
it's that the vast majority of our users don't have the slightest |
34 |
clue what's best for them in terms of package stability." |
35 |
-- Gentoo Developer Ciaran McCreesh |