Re: failure of cross-compilation detection on BlueGene/L

[Steven G. Johnson]

Steven G. Johnson wrote:
> Steven G. Johnson wrote:
> > That's why I suggested something like powerpc64-ibm-bluegene, or 
> > possibly powerpc64-ibm-bluegene-{version} if a version can be detected 
> > and is useful.
>
> Whoops, that should be powerpc, not powerpc64; all of the BlueGene 
> machines apparently run 32-bit PowerPC 4xx processors.

Hmm, and the operating system should probably be "cnk" (for CNK = 
Compute Node Kernel, the term IBM uses to refer to the operating system 
on the BlueGene compute nodes), rather than "bluegene" (the name of the 
machine), to follow the usual naming conventions.

I'll send a patch to address@hidden as you suggested.

Steven