comp.programming

kenobi

10/21/2015 7:45:00 AM

W dniu sroda, 21 pazdziernika 2015 01:48:50 UTC+2 uzytkownik Ramine napisal:
> Hello,
>
>
> As you have noticed i have wrote my new algorithm versoin 1.31
> of my scalable distributed reader-writer mutex, and just after that i
> have made it sequential consistent, but you have to understand ladies
> and gentlemen that i have implemented my new algorithm in Object Pascal
> and i have ported it to FreePascal and Delphi, and they both (Delphi and
> FreePascal compilers) respect the strong memory model of the x86
> architecture, so it was much easier for me to make it sequential
> consistent because you have only to take care of the following
> non-guarantee listed in the x86 spec:
>
> Loads may be reordered with older stores to different locations.
>
>
> So i have made my new algorithm version 1.31 of my scalable distributed
> reader-writer mutex sequential consistent with easy.
>
>
> That's not the case for C++, C++ have made it difficult for sequential
> consistency, because it respect the weak memory model that is much
> difficult to reason about and that is prone to mistakes by programmers,
> so becareful of C++ and languages that respect the weak memory model.
>
>
> And also my new algorithm of a scalable reader-writer mutex version 1.31
> take care of false-sharing and it is now sequential consistent and like
> in Seqlock or RCU , my new scalable distributed reader-writer mutex
> doesn't use any atomic operations and/or StoreLoad style memory barriers
> on the reader side, so it's very fast and scalable..but you have to use
> the define's option TLW_RWLockX or the define's option TRWLockX inside
> the defines1.inc file for that.
>
>
> You can download it from here:
>
>
> https://sites.google.com/site/aminer68/scalable-distributed-reader-wr...
>
>
>
> Thank you,
> Amine Moulay Ramdane.

sorry spam lama, it goes to spam bin without reading, yea ^^