On Nov 9, 2006, at 4:35 PM, Amos Waterland wrote:

Flush the ERAT very early on secondary processors.

Because Jimi has expressed skepticism about the need to do this, we
provide the following empirical and statistical arguments.

I have personally experienced this issue myself, since using the patch I have _not_. I'm excepting the patch as a "FW workaround" and simply as yet another thing we need to reset on the chip. Having said that, and the evidence of experienced stability, I would like to explore the possibility that this is indeed some ERAT pollution issue.

I don't know enough about Low Level FW (LLFW) not the POR cycle on the secondary processors, but I'd like to know more :) any insight out there?


Xen-ppc-devel mailing list

Reply via email to