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

Groups > comp.programming.threads > #1562 > unrolled thread

Re: RWLock explanation

Back to article view | Back to comp.programming.threads

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: RWLock explanation "Chris M. Thomasson" <no@spam.invalid> - 2013-06-19 17:45 -0700
    Re: RWLock explanation "Chris M. Thomasson" <no@spam.invalid> - 2013-06-19 17:54 -0700

#1562 — Re: RWLock explanation

"aminer"  wrote in message news:kpt7eo$5o1$1@dont-email.me...

> Hello,
> I have wrote a webpage that explains my RWLock algorithm, here it is:
> http://pages.videotron.com/aminer/rwlock.html

Have you tried to run this though a race detection algorithm?

FWIW, here is a solid fast-path starvation-free bounded time read-write 
mutex
algorithm I invented and posted here a while back:

http://software.intel.com/en-us/forums/topic/296471
(read all please)

AFAICT, it outperformed all native OS rwmutex implementations I have tested
against. It probably still does. At least it beats windows native rw mutex 
algorithm.


BTW, I have run this algorithm through Relacy Race Detector, and others,
many, many times indeed!

;^)



Also, here is the most elegant and compact rwmutex spinlock algorithm I have
ever seen:

https://groups.google.com/forum/?fromgroups#!msg/comp.programming.threads/_FtyJJAErGU/GvksDOZ2Pp8J


Bakery algorithms are simply the best!   :^) 

[toc] | [next] | [standalone]

#1564

"Chris M. Thomasson"  wrote in message 
news:kptjc4$glf$1@speranza.aioe.org...

> "aminer"  wrote in message news:kpt7eo$5o1$1@dont-email.me...

> > Hello,
> > I have wrote a webpage that explains my RWLock algorithm, here it is:
> > http://pages.videotron.com/aminer/rwlock.html

> Have you tried to run this though a race detection algorithm?

FWIW, here is a solid fast-path starvation-free bounded time read-write
> mutex
> algorithm I invented and posted here a while back:

> http://software.intel.com/en-us/forums/topic/296471
> (read all please)

> AFAICT, it outperformed all native OS rwmutex implementations I have 
> tested
> against. It probably still does. At least it beats windows native rw mutex 
> algorithm.

> BTW, I have run this algorithm through Relacy Race Detector, and others,
> many, many times indeed!

I think my algorithm is being used by UltimatePP (U++) for several years:

http://www.ultimatepp.org/forum/index.php?t=msg&goto=14993

http://ultimatepp.org/redmine/projects/upp/repository/entry/tags/n-2011-14/uppsrc/Core/Mt.cpp
(search for my last name to fine the code)


I hope it has generated a good experience for that project!

Its nice to know that this can be useful.   :^)

[...] 

[toc] | [prev] | [standalone]

Back to top | Article view | comp.programming.threads

csiph-web