[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 04 Apr 2008 16:59:27 +0200
From: Nadia Derbey <Nadia.Derbey@...l.net>
To: Mike Galbraith <efault@....de>
Cc: Manfred Spraul <manfred@...orfullife.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
paulmck@...ux.vnet.ibm.com,
Andrew Morton <akpm@...ux-foundation.org>,
Peter Zijlstra <peterz@...radead.org>,
Pavel Emelianov <xemul@...nvz.org>
Subject: Re: Scalability requirements for sysv ipc (+namespaces broken with
SEM_UNDO)
Mike Galbraith wrote:
> On Sun, 2008-03-30 at 16:12 +0200, Manfred Spraul wrote:
>
>>Mike Galbraith wrote:
>>
>>>On Sat, 2008-03-22 at 20:08 +0100, Manfred Spraul wrote:
>>>
>>>
>>>
>>>>just the normal performance of 2.6.25-rc3 is abyssimal, 55 to 60% slower
>>>>than 2.6.18.8:
>>>>
>>>
>>>After manually reverting 3e148c79938aa39035669c1cfa3ff60722134535,
>>>2.6.25.git scaled linearly
>>
>>We can't just revert that patch: with IDR, a global lock is mandatory :-(
>>We must either revert the whole idea of using IDR or live with the
>>reduced scalability.
>
>
> Yeah, I looked at the problem, but didn't know what the heck to do about
> it, so just grabbed my axe to verify/quantify.
>
>
>>Actually, there are further bugs: the undo structures are not
>>namespace-aware, thus semop with SEM_UNDO, unshare, create new array
>>with same id, but more semaphores, another semop with SEM_UNDO will
>>corrupt kernel memory :-(
>>I'll try to clean up the bugs first, then I'll look at the scalability
>>again.
>
>
> Great!
>
> -Mike
>
>
>
>
I could get better results with the following solution:
wrote an RCU-based idr api (layers allocation is managed similarly to
the radix-tree one)
Using it in the ipc code makes me get rid of the read lock taken in
ipc_lock() (the one introduced in 3e148c79938aa39035669c1cfa3ff60722134535).
You'll find the results in attachment (kernel is 2.6.25-rc3-mm1).
output.25_rc3_mm1.ref.8 --> pmsg output for the 2.6.25-rc3-mm1
plot.25_rc3_mm1.ref.8 --> previous file results for use by gnuplot
output.25_rc3_mm1.ridr.8 --> pmsg output for the 2.6.25-rc3-mm1
+ rcu-based idrs
plot.25_rc3_mm1.ridr.8 --> previous file results for use by gnuplot
I think I should be able to send a patch next week. It is presently an
uggly code: I copied idr.c and idr.h into ridr.c and ridr.h to go fast,
so didn't do any code factorization.
Regards
Nadia
View attachment "output.25_rc3_mm1.ref.8" of type "text/x-troff-man" (5846 bytes)
View attachment "plot.25_rc3_mm1.ref.8" of type "text/x-troff-man" (74 bytes)
View attachment "output.25_rc3_mm1.ridr.8" of type "text/x-troff-man" (5851 bytes)
View attachment "plot.25_rc3_mm1.ridr.8" of type "text/x-troff-man" (80 bytes)
Powered by blists - more mailing lists