LW_RWLockX algorithm...

[aminer]

Hello,


I have updated scalable RWLock to version 2.14...

I have just implemented a new LW_RWLockX algorithm (a scalable RWLock),
it is a lightweight version that uses spin-wait but it is
starvation-free and it supports the following requirements:

1- It scales on multicores

2- It is starvation-free

3- it is portable.


You can download all the variants of my scalable RWLock 2.14 from:

http://pages.videotron.com/aminer/


And please take a look at my new algorithm inside LW_RWLockX.pas
algorithm inside the zip file.




Thank you,
Amine Moulay Ramdane.