Re: JSON PMDA

[fche@xxxxxxxxxx (Frank Ch. Eigler)]

dsmith wrote:

> [...]
> I wonder if it wouldn't make sense to define something like the
> following instead:
>
>     int pmdaCacheOp2(pmInDom indom, int op, unsigned long val);
>
> with an associated op of PMDA_CACHE2_RESIZE. That would give more room
> for the addition of future operations.

FWIW, I'm not a fan of ioctl-like functions like this.  They provide
apparent API/ABI stability but not actual API/ABI stability.  (There
would be no way to verify op values or appropriateness of val for each
op, etc.)  Adding new functions as first class functions is easy.


- FChE