Alan Cox wrote: >I'm not up on the S/390 architecture code but the standard Linux >behaviour to generate page aging data in the absence of dirty flags is >to force pages read only and take the protection fault internally.
We did it that way in Linux 2.2, but since 2.4 we actually use the hardware dirty bits from the storage keys (i.e. per-physical-page instead of per-pte dirty bits). Nevertheless, we of course still use protection faults to implement copy-on-write, as Neale already pointed out. (For the S/390 folks, this means that we make essential use of the 'suppression-on-protection' facility; Linux definitely cannot run on machines lacking that facility, not even the 'vintage' patches will help you there ;-)) Bye, Ulrich -- Dr. Ulrich Weigand [EMAIL PROTECTED]
