1 |
Am 09.03.2013 19:15, schrieb Florian Philipp: |
2 |
> Hi list! |
3 |
> |
4 |
> Whenever I do sequential IO for a long stretch of time (e.g. md5summing |
5 |
> 40 GB), I'm experiencing high load (ca. 6 on a 4 CPU system) and |
6 |
> temporary freezes of most applications. For example, switching between |
7 |
> tabs in konsole sometimes takes more than 2 seconds. |
8 |
> |
9 |
> When doing this on an ext4 filesystem, the load seems to result from |
10 |
> khugepaged and kswapd0 as well as some kworkers. |
11 |
> |
12 |
> I think I've had similar issues with NFS over wifi but I cannot test |
13 |
> this now. |
14 |
> |
15 |
> Today I copied 60GB from my hard disk to an USB disk formatted with |
16 |
> NTFS, issuing the copy command from KDE's dolphin. The freezes became so |
17 |
> long it was impossible to work and then X11 locked up and had to be killed. |
18 |
> |
19 |
> I tried using a preemptive kernel but that didn't seem to help. blkio |
20 |
> and cpu cgroups didn't help either. Ionice seems to be the only solution |
21 |
> but while I'm okay with that, my dad won't be. Can anyone tell me what |
22 |
> is causing this behavior? |
23 |
> |
24 |
> Throughput is good, by the way. That's why I don't suspect a driver issue. |
25 |
> |
26 |
> Thanks in advance! |
27 |
> Florian Philipp |
28 |
> |
29 |
congratulation. You hit 'the bug'. Been around for ages, but for magical |
30 |
reasons kernel dev are unable to see or unable to do something about it. |
31 |
If you are using a vanilla kernel, posting on lkml might be the right |
32 |
thing to do. |