1 |
Chris Gianelloni wrote: |
2 |
> On Fri, 2007-05-11 at 10:25 +1000, Paul de Vrieze wrote: |
3 |
> |
4 |
>> M. Edward (Ed) Borasky wrote: |
5 |
>> |
6 |
>>> I can't tell yet whether it's hardware or software, but I'm guessing the |
7 |
>>> kernel at this point rather than hardware -- I have some |
8 |
>>> /var/log/messages traces that don't look like hardware. Once I get a |
9 |
>>> stable OS, I'll load Gentoo from it and do the real debugging. The |
10 |
>>> 2.4.20 kernel in Gentoo stable has to be better that the 2.4.18s the |
11 |
>>> other distros seem to be carrying. |
12 |
>>> |
13 |
>> What I find to be a cause of a lot of stability issues. Probably even |
14 |
>> more so on multiprocessor/core systems is preemption. Voluntary |
15 |
>> preemption is OK, but other preemption still seems to be a bit flaky at |
16 |
>> places. |
17 |
>> |
18 |
> |
19 |
> Umm... What? |
20 |
> |
21 |
> I use preemption all the time on all of my machines, which are *all* |
22 |
> multi-processor or multi-core. The kernel preemption works just fine on |
23 |
> all of them. What tends to be the problem is shoddy APIC or ACPI |
24 |
> implementations on the cheaper (read, not server/workstation class) |
25 |
> motherboards. |
26 |
> |
27 |
> |
28 |
I'm still testing some things, but there are some bugs in the NVidia |
29 |
"sata_nv" part of the AMD64 Linux kernel that have showed up in the |
30 |
Debian bug archives. Apparently it worked in older kernels and got |
31 |
broken "recently". I haven't found any differences with or without |
32 |
either pre-emption, SMP vs. UP, ACPI or APIC. The only thing I've found |
33 |
so far is that 32-bit kernels seem to work and 64-bit ones don't. At the |
34 |
moment I've only got a gigabyte of RAM in the machine so I'm not losing |
35 |
a heck of a lot by running a 32-bit kernel. :) |
36 |
-- |
37 |
gentoo-releng@g.o mailing list |