Re: rwlock implementation

["dada1" <dada1 cosmosbay com>]

>
> > Last month, I did a remark on this mailing list about the rwlock
> > implementation that unconditionnaly used a futex syscall at the unlock()
> > stage.
> > So far I received no comment.
> > Must I conclude nobody uses rwlocks in real applications ?
>
> I don't know about that but I can say that I haven't looked at it yet
> since I'm still chasing bugs and trying to extend the test suite.  Since
> I'm alone doing this it takes a while.
>
> None of the items posted here are lost 8this includes Jerry's debugging
> proposal, soory about that).  There is just a natural order: correctness
> first, then improvements.

Ok Ulrich.
I had to code an efficient rwlock implementation (at least for my needs)
Both in C and ia32 ASM.
Yould you like to get/comment my code ?