Steve,
Should we commit Geoff's fix or is there a better way to fix it?
Ali
On Dec 29, 2007, at 1:39 PM, Steve Reinhardt wrote:
OK, great, glad you tracked that bug down. Your fix is a pretty good
one, but I think the right answer is that CPU3 in your example should
not issue a ReadEx... if
I'd say go ahead and commit it. I'm 98% sure that once we properly fail SCs
when they get invalidated it won't be necessary, but I can't promise I'll
get that in any time soon, and in the meantime it's a pretty good stopgap.
If you can put a comment in the flyspray task I mentioned earlier (
http
