[PATCH 3/6] 8xx: invalidate non present TLBs

Benjamin Herrenschmidt benh at kernel.crashing.org
Fri Oct 9 07:18:14 EST 2009


On Thu, 2009-10-08 at 13:11 -0700, Dan Malek wrote:
> 
> There are many comments written about 8xx as various
> behavior was discovered.  Worse, some of these details
> would be different among the different processor versions.
> You need to be careful and test as many different part
> versions as possible to ensure you have everything
> covered.....  then someone will find a part that doesn't
> quite work, "fix" it, and break others :-)
> 
> In this particular case, the PEM does state dcbst is treated
> as a load, but from experience we know 8xx doesn't work
> that way.  Of course, since dcbst is a store operation,
> you could argue that 8xx got it correct :-) 

Hehe. Well, it's architecturally incorrect, as dcbst is not really a
store operation in the sense that it doesn't modify the target cache
line, and as such doesn't (mustn't) be covered by write access
protection, shouldn't set DIRTY, etc...

So I would argue that 8xx got it wrong either way :-)

Cheers,
Ben.




More information about the Linuxppc-dev mailing list