Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.programming.threads > #1734

Re: Ticket spinlock...

From aminer <aminer@toto.net>
Newsgroups comp.programming.threads
Subject Re: Ticket spinlock...
Date 2013-09-23 17:07 -0700
Organization albasani.net
Message-ID <l1qai8$7sg$1@news.albasani.net> (permalink)
References <l1q9mr$6dg$1@news.albasani.net>

Show all headers | View raw


I correct some english mistakes, pease read again...



Hello,

That's not the end of the story, i have benchmarked the ticket spinlock
that i am using and that you find here:

http://code.google.com/p/gpdelphiunits/source/browse/trunk/src/SpinLock.pas?r=37

and found that it has a poor performance cause in the case of a spinlock 
without tickets when it is unlocked() the first thread that comes first 
to the lock() will enter immediatly the locked section, but that's not 
the case with a Ticket spinlock cause if the first thread that enters 
the lock() first has not the ticket to enter, the other thread that have 
the ticket to enter and that is waiting for his turn will wait 
longer/more , so this is why the Ticket spinlock has a poor performance 
compared to a simple spinlock with a backoff.


So i will advise you to avoid the Ticket spinlock.




Thank you,
Amine Moulay Ramdane.


On 9/23/2013 4:52 PM, aminer wrote:
>
> Hello,
>
> That's not the end of the story, i have benchmarked the ticket spinlock
> that i am using and that you find here:
>
> http://code.google.com/p/gpdelphiunits/source/browse/trunk/src/SpinLock.pas?r=37
>
>
> and found that it has a poor performance cause in the case of a spinlock
> without tickets when it is unlocked() the first thread that comes first
> to the lock() will enter immediatly the locked section, but that's not
> the case with a Ticket spinlock cause if the first thread that enter the
> lock() have not the ticket the other threads that have the ticket and
> that are waiting for there turn will wait more , so this is why the
> Ticket spinlock have a poor performance compared to a simple spinlock
> with a backoff.
>
>
> So i will advice you to avoid the Ticket spinlock.
>
>
>
>
> Thank you,
> Amine Moulay Ramdane.

Back to comp.programming.threads | Previous | NextPrevious in thread | Find similar


Thread

Ticket spinlock... aminer <aminer@toto.net> - 2013-09-23 16:52 -0700
  Re: Ticket spinlock... aminer <aminer@toto.net> - 2013-09-23 17:07 -0700

csiph-web