Re: hook question

[David Jones <drj@...>]

On 10 Oct 2006, at 13:07, Mike Pall wrote:

> Hi,
>
> David Jones wrote:
> > This assumes that updates to each of hookmask and hook will take
> > place atomically.  There's no guarantee of that.  On some
> > architectures an update of hook may consist of two writes (to two
> > halves of a 64-bit quantity say) and a re-ordered write to hookmask
> > may appear in between them.
>
> Please. If updates to a pointer sized quantity (32 bit on 32 bit
> systems, 64 bit on 64 bit systems) are non-atomic, fire the CPU
> designer team. AFAIK none of the 64 bit multi-core machines on
> the market do this.

Oh I see.  When you said thread-safe you meant thread-safe on Intel  
Linux.  Right.  All The World's A Vax.  Well, I totally agree with  
you in that case.

C on AS/400 is typically implemented with 16-byte pointer values.  I  
doubt that they're atomic.

The charming world of C in the DOS era had plenty of examples where  
the pointer size was 32-bits but the word size was 16-bits.

It's also not unusual for function pointers to be twice as wide as  
your average data pointer.

drj