| 1 |
I have try the BFQ patch outside of the kernel mainline, it works well. |
| 2 |
Maybe you would like to see: |
| 3 |
http://algo.ing.unimo.it/people/paolo/disk_sched/ |
| 4 |
|
| 5 |
Also there is a 15-minute demo of the performance of BFQ: |
| 6 |
http://youtu.be/J-e7LnJblm8 |
| 7 |
|
| 8 |
2013/3/10 Volker Armin Hemmann <volkerarmin@××××××××××.com> |
| 9 |
|
| 10 |
> Am 09.03.2013 19:15, schrieb Florian Philipp: |
| 11 |
> > Hi list! |
| 12 |
> > |
| 13 |
> > Whenever I do sequential IO for a long stretch of time (e.g. md5summing |
| 14 |
> > 40 GB), I'm experiencing high load (ca. 6 on a 4 CPU system) and |
| 15 |
> > temporary freezes of most applications. For example, switching between |
| 16 |
> > tabs in konsole sometimes takes more than 2 seconds. |
| 17 |
> > |
| 18 |
> > When doing this on an ext4 filesystem, the load seems to result from |
| 19 |
> > khugepaged and kswapd0 as well as some kworkers. |
| 20 |
> > |
| 21 |
> > I think I've had similar issues with NFS over wifi but I cannot test |
| 22 |
> > this now. |
| 23 |
> > |
| 24 |
> > Today I copied 60GB from my hard disk to an USB disk formatted with |
| 25 |
> > NTFS, issuing the copy command from KDE's dolphin. The freezes became so |
| 26 |
> > long it was impossible to work and then X11 locked up and had to be |
| 27 |
> killed. |
| 28 |
> > |
| 29 |
> > I tried using a preemptive kernel but that didn't seem to help. blkio |
| 30 |
> > and cpu cgroups didn't help either. Ionice seems to be the only solution |
| 31 |
> > but while I'm okay with that, my dad won't be. Can anyone tell me what |
| 32 |
> > is causing this behavior? |
| 33 |
> > |
| 34 |
> > Throughput is good, by the way. That's why I don't suspect a driver |
| 35 |
> issue. |
| 36 |
> > |
| 37 |
> > Thanks in advance! |
| 38 |
> > Florian Philipp |
| 39 |
> > |
| 40 |
> congratulation. You hit 'the bug'. Been around for ages, but for magical |
| 41 |
> reasons kernel dev are unable to see or unable to do something about it. |
| 42 |
> If you are using a vanilla kernel, posting on lkml might be the right |
| 43 |
> thing to do. |
| 44 |
> |
| 45 |
> |
| 46 |
|
| 47 |
|
| 48 |
-- |
| 49 |
|
| 50 |
Best regards! |
| 51 |
|
| 52 |
------------------------------------------------ |
| 53 |
|
| 54 |
Yu-yu Lin(林育宇) |
| 55 |
|
| 56 |
The Guangdong Key Laboratory of Information Security Technology (IST), |
| 57 |
School of Information Science and Technology, |
| 58 |
Sun Yat-sen (Zhongshan) University (中山大学), |
| 59 |
Guangzhou,P.R.China. |
| 60 |
|
| 61 |
Email: cosmoslx@×××××.com |
| 62 |
|
| 63 |
------------------------------------------------ <cosmoslx@×××××.com> |
Archives