1 |
Norberto Bensa wrote: |
2 |
> Quoting Nikos Chantziaras <realnc@×××××.de>: |
3 |
> |
4 |
>> It got worse with kernel 2.6.28, btw, especially mouse cursor movement. |
5 |
>> It gets stuck and skips very noticeably. Fortunately, it's not a |
6 |
>> point yet where I would describe it as unusable, but if the trend |
7 |
>> continues, desktop will be totally unusable by 2.6.31/32 when doing |
8 |
>> something that produces load. |
9 |
> |
10 |
> Oh pleeeaase... |
11 |
> |
12 |
> You have a broken system. Don't blame Linux for that. |
13 |
|
14 |
Could be. But you don't know for sure :P |
15 |
|
16 |
|
17 |
> $ uname -a |
18 |
> Linux venkman 2.6.28-gentoo #1 SMP PREEMPT Thu Dec 25 16:48:22 ARST 2008 |
19 |
> x86_64 Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz GenuineIntel GNU/Linux |
20 |
> |
21 |
> I'm currently compiling gcc-4.3.2-r2, I have no cursor lagging, stuck, |
22 |
> or whatever. Everything is normal. Like if I were not compiling at all. |
23 |
|
24 |
I've just emerged glibc (with -j2). It's especially noticeable with it. |
25 |
Some source files, especially at the beginning of the build, are very |
26 |
short and result in a lot processes getting spawned and terminated very |
27 |
fast, along with files rapidly created on disk. Can you give glibc a test? |
28 |
|
29 |
I've checked my config and can't find anything wrong. I have tickless, |
30 |
preempt, 1000Hz, -O2, CFQ. The kernel is stripped off of everything not |
31 |
needed. |
32 |
|
33 |
Maybe it's due to fglrx? (ATI Catalyst driver.) |