| 1 |
Am Montag, 7. Mai 2012, 15:56:20 schrieb walt: |
| 2 |
> On 05/07/2012 10:14 AM, Volker Armin Hemmann wrote: |
| 3 |
> > Am Montag, 7. Mai 2012, 18:44:05 schrieb Michael Hampicke: |
| 4 |
> >> Maybe changing the kernel io scheduler will help? |
| 5 |
> >> |
| 6 |
> >> # /usr/src/linux/Documentation/block/switching-sched.txt |
| 7 |
> >> # /usr/src/linux/Documentation/block/deadline-iosched.txt |
| 8 |
> >> # /usr/src/linux/Documentation/block/cfq-iosched.txt |
| 9 |
> > |
| 10 |
> > nice idea - but that didn't help in the past. Why should it help now? |
| 11 |
> > |
| 12 |
> > The question is - why does disk IO make the mouse jerky and delays |
| 13 |
> > keyboard |
| 14 |
> > input? That is just idiotic. |
| 15 |
> |
| 16 |
> Have you tested ck-sources for comparison? |
| 17 |
|
| 18 |
in the past, yes. Complete disappointment. |
| 19 |
|
| 20 |
But it is ok, if the io happens between two very fast drives (ssd - ssd or ssd |
| 21 |
- raid5). So the problem is: |
| 22 |
io with one slow drive being written to |
| 23 |
swapping |
| 24 |
|
| 25 |
those cases are just horrible. |
| 26 |
|
| 27 |
And vanilla-sources just work. Stable. Even with suspend to ram and ati |
| 28 |
drivers. |
| 29 |
|
| 30 |
-- |
| 31 |
#163933 |
Archives