[Devel] nptl perf bench and profiling with pidns patchsets

Pavel Emelianov xemul at openvz.org
Wed Jun 13 02:27:29 PDT 2007


Cedric Le Goater wrote:
> Pavel Emelianov wrote:
>> Cedric Le Goater wrote:
>>> Pavel and all,
>> [snip] 
>>
>>> findings are : 
>>>
>>> * definitely better results for suka's patchset. suka's patchset is 
>>>   also getting better results with unixbench on a 2.6.22-rc1-mm1 but 
>>>   the values are really dispersed. can you confirm ?
>>> * suka's patchset would benefit from some optimization in init_upid() 
>>>   and dup_struct_pid()  
>> We have found the reason why Suka's patches showed better performance.
>> Some time ago I sent a letter saying that proc_flush_task() actually
>> never worked with his patches - that's the main problem. After removing
>> this call from my patches the results turned to those similar to my.
>>
>> I'd also like to note that broken-out set of patches is not git bisect
>> safe at all. The very first patch of his own OOPSes the node. Some
>> subsequent patches contain misprints that break the compilation, etc.
>>
>> So I ask you again - let us prepare our patches again and compare the 
>> performance one more time.
> 
> OK. that's fine with me. 
> 
> I'm not exactly in a neutral zone but I have the blades ready for the 
> next drop of patches. I'll torture them if you don't mind.

I do not :) I am going to send my view of pid namespaces this evening
or tomorrow morning (I am in GMT+3 time zone :)). Are you going to fix
your patches for comparison?

> C.
> 

Thanks,
Pavel


More information about the Containers mailing list