Messages in this thread | | | Date | Fri, 19 Nov 2010 08:54:40 +0530 | From | Bharata B Rao <> | Subject | Re: [PATCH v3 0/7] CFS Bandwidth Control |
| |
On Wed, Nov 17, 2010 at 04:32:29PM +0800, Lai Jiangshan wrote: > On 10/12/2010 03:49 PM, Bharata B Rao wrote: > > Hi, > > > > Its been a while since we posted CPS hard limits (aka CFS bandwidth control > > now) patches, hence a quick recap first: > > > > - I have been working on CFS hard limits since last year and have posted > > a few versions of the same (last post: http://lkml.org/lkml/2010/1/5/44) > > - Paul Turner and Nikhil Rao meanwhile started working on CFS bandwidth > > control and have posted a couple of versions. > > (last post v2: http://lwn.net/Articles/385055/) > > > > Hi, All > > I tried it, it works very good! It adds a very useful feature. > It is very helpful for virtualization or cloud-computing I think. > I hope the next version patchset of it will be sent and get merged soon. > > When I tested it, my box became dead several times, infrequently, and > I don't know how to reproduce it, but next steps may help to reproduce it: > > # mount -t cgroup -o cpu xxx /mnt > # cd /mnt > # mkdir test > # cd test > # echo 5000000 > cpu.cfs_period_us > # echo 1000000 > cpu.cfs_quota_us > # cat /dev/zero > /dev/null & > # echo $! > tasks > # #make the system busy, browse websites and open many tabs in your browser etc > # #but do NOT do things important. > # #after some time, the system may dead. (can not respond even you type AltSysrqC) >
Hi Lai, Thanks for testing the patches. I haven't seen this behaviour still. But will try out the steps you mention and see if I can reproduce it here. Meanwhile if you find more information about the problem, do let us know.
Regards, Bharata.
| |