On Thu, Dec 18, 2008 at 10:28 AM, John David Anglin <dave@xxxxxxxxxxxxxxxxxx> wrote: > From the documentation, it appears to me that a protection id trap should > be treated differently from a data TLB miss/data page fault. The latter can > potentially be fixed with TLB insertion. I don't think a protection ID trap > can be fixed... Yes, I think that is also true. What if, after a context switch we have stale TLB entry with an access ID from another process? When the kernel tries to write to the user address the current protection ids don't match the page's access id and we generate a protection id trap. Cheers, Carlos. -- To unsubscribe from this list: send the line "unsubscribe linux-parisc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
