This is the mail archive of the libc-hacker@sourceware.cygnus.com mailing list for the glibc project.

Note that libc-hacker is a closed list. You may look at the archives of this list, but subscription and posting are not open.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Patch: HAS_COMPARE_AND_SWAP_WITH_RELEASE_SEMANTICS


On Wed, Jun 21, 2000 at 04:04:29PM -0700, Richard Henderson wrote:
> On Mon, Jun 19, 2000 at 10:54:19AM -0700, H . J . Lu wrote:
> > That is from the discussion of linuxthreads under ia64:
> > 
> > ---
> > The fundamental problem is that bith __pthread_lock and __pthread_unlock use
> > compare_and_swap as the underlying synchronization primitive.  The problem
> > is that, in IA64 terms, __pthread_lock wants acquire semantics (i.e. it
> > completes befor subsequent writes) where __pthread_unlock wants release
> > semantics (it completes after previous writes).  Currently there is only one
> > compare_and_swap, and it has release semantics.
> > ---
> 
> Actually, this is isn't right -- unless we give folks access to the
> full complement of lock and unlock routines, we want full-stop semantics.
> 

What folks? Those macros are internal to glibc. I don't see what is
wrong with them.


-- 
H.J. Lu (hjl@gnu.org)

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]