[Skiboot] [PATCH v2 2/2] prd: PRD framework

[Benjamin Herrenschmidt]

On Thu, 2015-02-12 at 18:27 +1100, Stewart Smith wrote:

> This is a problem - a flood of these will hit ENOMEM and that won't be
> good for keeping the system running at all (many OPAL calls would fail
> without being able to allocate memory).

We can't flood them more than one per chip of each kind (OCC and ATTN)
I think, can we ? I mean the OCC one might be we could pace it ourselves
by ignoring subsequent interrupts from the OCC until the previous one is
complete.

Ben.