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

Benjamin Herrenschmidt benh at au1.ibm.com
Thu Feb 12 17:08:19 AEDT 2015

On Thu, 2015-02-12 at 11:05 +0530, Neelesh Gupta wrote:
> The node gets processed where linux makes an OPAL call, apparently it 
> will be
> immediate but depends on the linux when it consumes it. Moreover, I put the
> list thinking that there could be simultaneous ATTNs from multiple procs or
> interrupts of different 'prd_msg_type'..

Ah that's a good point, we might have attentions (and OCC resets) coming
simultaneously on multiple chips since the plan was to only mask the IPOLL
mask locally...

Jeremy, that changes things a bit for us... We can keep the code simpler
still I think, but we probably do need some kind of queue. The code is
simpler because userspace will only process one at a time and we
can bake that in the design on our side.

Unless we want to plan for a hypothetical day where HBRTD can process
multiple at a time...


More information about the Skiboot mailing list