Monday, March 20, 2017

Digest for comp.programming.threads@googlegroups.com - 2 updates in 1 topic

Ramine <toto@toto.net>: Mar 19 02:56PM -0400

Hello......
 
You don't understand, i was aware that memory ordering is not the only
issue and that you have to use fences and memory barriers but
what i am explaining to you is that my implementation of my algorithms
like my MLock and AMLock etc. already uses the correct fences and memory
barriers to ensure sequential consistency in the hardware side, but what
i need is to switch the optimization off to avoid memory ordering of the
Delphi and FreePascal compilers.
 
That's easy to do, i have thought that Delphi and FreePascal doesn't
reorder loads and stores, but this was a mistake, but this mistake is
easy to correct, but mistakes in software programming can happen, but i
am confident that as soon as i correct this mistake, you will be happy
with my C++ Libraries.
 
 
So stay tuned !
 
 
Thank you,
Amine Moulay Ramdne.
Ramine <toto@toto.net>: Mar 19 03:03PM -0400

On 3/19/2017 2:56 PM, Ramine wrote:
> barriers to ensure sequential consistency in the hardware side, but what
> i need is to switch the optimization off to avoid memory ordering of the
> Delphi and FreePascal compilers.
 
I will switch the optimization off just locally on my MLock and AMLock
and some others, that's easy to do to avoid memory reordering of Delphi
and FreePascal..
 
 
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to comp.programming.threads+unsubscribe@googlegroups.com.

No comments: