Re: [resend][PATCH] Added PR_SET_PROCTITLE_AREA option for prctl()

[KOSAKI Motohiro <kosaki.motohiro-+CUm20s59erQFUHtdCDX3A@public.gmane.org>]

> On Tue, 13 Oct 2009 04:03:45 +0900 (JST)
> KOSAKI Motohiro <kosaki.motohiro-+CUm20s59erQFUHtdCDX3A@public.gmane.org> wrote:
> 
> > > Start simple.  What's wrong with mutex_lock() on the reader and writer
> > > sides?  rwsems might be OK too.
> > > 
> > > In both cases we should think about whether persistent readers can
> > > block the writer excessively though.
> > 
> > I thought your mention seems reasonable. then I mesured various locking
> > performance.
> > 
> > 		no-contention	read-read contetion	read-write contention
> > w/o patch	4627 ms		 7575 ms		 N/A
> > mutex		5717 ms		33872 ms (!)		14793 ms
> > rw-semaphoe	6846 ms		10734 ms		36156 ms (!)
> > seqlock		4754 ms		 7558 ms		 9373 ms
> > 
> > Umm, seqlock is significantly better than other.
> 
> Sure, but even the worst case there is 1,000,000 operations in 34
> seconds (yes?). 33 microseconds for a /proc read while under a specific
> local DoS attack is OK!
> 
> If so then all implementations are acceptable and we should choose the
> simplest, most-obviously-correct one.

Hm, ok!

I had guessed you don't accept this slowness. but my guess was wrong.
I have no objection to use rw-semaphoe if you accept it.