| 1 |
Duncan posted <pan.2006.03.26.10.34.46.613270@×××.net>, excerpted below, |
| 2 |
on Sun, 26 Mar 2006 03:34:47 -0700: |
| 3 |
|
| 4 |
> With kernel 2.6.15, configuring IOMMU won't let me boot (without a mem= |
| 5 |
> kernel command line). Configuring it in but forcing it off (iommu=off on |
| 6 |
> the kernel command line) gives me some IOMMU needed errors and it still |
| 7 |
> won't boot (again, without a mem=). Configuring it out, the system boots |
| 8 |
> just fine. |
| 9 |
> |
| 10 |
> With kernel 2.6.16, configuring IOMMU in or out, on or off, won't boot |
| 11 |
> (without a mem=). |
| 12 |
|
| 13 |
For those that may be interested, I think I may have a defective hardware |
| 14 |
IOMMU. Setting iommu=soft on the kernel command line works just fine. |
| 15 |
That's a bit disappointing for AMD hardware, where the hardware IOMMU is |
| 16 |
supposed to be better, but I guess it's no worse than the Intel chips, |
| 17 |
which don't have an IOMMU so emulate it in software. |
| 18 |
|
| 19 |
It's still fairly likely that I have something not quite configured |
| 20 |
correctly, however, and that when I figure it out, it'll work the way |
| 21 |
it's supposed to. I have some research to do tho, to grok that. |
| 22 |
Meanwhile, with the iommu=soft switch, I can run 2.6.16 kernels with full |
| 23 |
memory, just as I could .15 kernels. (The .15 and apparently earlier |
| 24 |
kernels had bounce-buffers built-in for SCSI, while .16 took them out |
| 25 |
since there's the IOMMU in either software for Intel, or hardware for |
| 26 |
normal AMD, anyway.) |
| 27 |
|
| 28 |
BTW, merged KDE 3.5.2 last nite, upgraded from 3.5.1. That's the first |
| 29 |
time I've run a major emerge session with $PORTAGE_TMPDIR on a RAM backed |
| 30 |
tmpfs. VERY NICE! =8^) I /still/ haven't used the full 8 gigabit yet, |
| 31 |
tho. I got upto 6.5-ish when kmail was compiling, with four other compile |
| 32 |
jobs running in parallel (5 total), each of which was running |
| 33 |
MAKEOPTS="-j6 -l5". The -l5 limited load some, and KDE doesn't make as |
| 34 |
good a use of multiple compile jobs as it could (the reason I was running |
| 35 |
5 in parallel since the -j6 -l5 doesn't work so well with KDE), but I |
| 36 |
/was/ running 7-10 load average for several hours (on a dual Opteron 242). |
| 37 |
Hopefully by the time I'm compiling KDE4, I'll be doing it with dual |
| 38 |
cores, a bit faster than the 1.6 GHz of the Opteron 242s, as well. |
| 39 |
|
| 40 |
-- |
| 41 |
Duncan - List replies preferred. No HTML msgs. |
| 42 |
"Every nonfree program has a lord, a master -- |
| 43 |
and if you use the program, he is your master." Richard Stallman in |
| 44 |
http://www.linuxdevcenter.com/pub/a/linux/2004/12/22/rms_interview.html |
| 45 |
|
| 46 |
|
| 47 |
-- |
| 48 |
gentoo-amd64@g.o mailing list |
Archives