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

Groups > comp.programming.threads > #2788

Sequantial consistency

Show all headers | View raw

Hello...


As you have seen me doing in this forum, i have invented in front of you 
my Scalable distributed sequential lock and i have also did a
proof of correctness and a proof sequential consistency correctness,
but my proof apply only to FreePascal and Lazarus and Delphi compilers
that follow and respect the strong memory model of the x86 architecture 
and that don't force on you a weak memory model as is doing C++, but if 
you are using GCC C++ or Intel C++ or Microsoft VC C++ , you have to be 
carefull cause at the -O1 and -O2 and -O3  levels of optimization those 
C++ compilers do reorder instructions, so if you need to port my 
algorithm to C++ you have to insert memory ordering constraints 
instructions... but i have to be frank with you, cause i have took a 
look at the C++11 and what it is doing is forcing on us the weak memory 
model even if we are running on x86, and that's too sad ! cause by doing 
so the C++ compilers are complicating what have tried to easy and 
simplify the x86 architecture that is using a strong memory model.

But anyway hope you have undertood this post correctly...


You can download my scalable distributed sequential lock from:

https://sites.google.com/site/aminer68/scalable-distributed-sequential-lock


Thank you for your time.


Amine Moulay Ramdane.

Back to comp.programming.threads | Previous | Next | Find similar | Unroll thread

Thread

Sequantial consistency Ramine <ramine@1.1> - 2015-01-02 16:37 -0800

csiph-web