On Wed, Feb 27, 2008 at 2:20 PM, Heiko Carstens <[EMAIL PROTECTED]> wrote:
> > I have no idea what's possible with an undocumented instruction ;-) > > If nothing else, we can get a CP PTF to only enable it on the 2nd ESSA > > instruction... > > Ah no... we can fix that. If you like to patch your kernel the patch below > should fix the incosistency between what z/VM and Linux think about cmma > being active. I just gave it a quick run and it seems to work. > Not guaranteed to work however :) > > For the rest I leave it up to Martin to comment, since I its his code. I am sure you got the wink on my suggestion to make CP count the 2nd time you try the ESSA... There's different ways to correct bad defaults: "Question the Defaults and Resist Change" :-) The customers I talk to do not wish to compile their own kernel - many will even need some time before they can pick up a new kernel provided by the distributor. My recommendation still stands for those customers. Oh, and I doubt your patch does the trick. The ESSA will program check when not there. I don't think fault.c will like that. PS I did read that Brian Wade's experiments were done with cmma=on and use the CP settings to enable / disable the feature. When my assumptions about the implementation are correct, then that was indeed the right way to do it. But it does not reveal what will happen to customers who end up with a mix of penguins. Rob ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
