[LLVMdev] Atomic Operation and Synchronization Proposal v2

[David A. Greene]

On Thursday 12 July 2007 16:51, David Greene wrote:

> > You're still ordering load-store pairings, there is juts the added
> > dimensionality of types. This seems like an easy extension to the
> > existing proposal to combine the load and store pairings with a type
> > dimension to achieve finer-grained control. Does this make sense as an
> > incremental step from your end with much more experience comparing
> > your hardware to LLVM's IR?
>
> This would work for X1-style lsyncs

I take that back.  Maybe.

If by "type" you literally mean the data type of the value (int, float, etc.)
and the extent of the data (vector or scalar), then it won't handle the
X1 case where integer scalar instructions feed floating point vector 
instructions and similar combinations.

If by "type" you only mean the extent of the data, then it would work fine.

                                         -Dave