[PATCH/RFC] PCI Error Recovery

Benjamin Herrenschmidt benh at kernel.crashing.org
Sat Mar 12 22:16:34 EST 2005

> e.g. on typical PCI Express chipsets there is no such concept.
> With error reporting there you just get told "one of your 
> previous accesses or DMAs failed", but the device is not
> completely gone.
> IMHO the concept of "slot freeze" is too PPC64 specific to
> be exposed like this. While there are other chipsets
> that do similar things a lot of widely used ones will not.
> Can you reformulate it in terms of "error states" ?
> Just report an error occurred and the device is unreliable.

I don't want to expose it that way neither, but the fact is we can't
just have "generic" states that apply to every architecture. I haven't
yet looked at Linas latest patch though, but at one point, we need to
define a few states that may or may not apply to a given architecture
and give enough info to drivers to deal with them as much as they can.
I'm afraid we can't completely avoid some of the complexity here.


More information about the Linuxppc64-dev mailing list