From: JS <lis...@os...> - 2001-02-19 15:01:40
|
At Sunday, 18 February 2001, you wrote: >lis...@os... said: >> The time will come when you descover that stopping the whole VM for a >> page fault in a single process isn't a good idea (is that what >> happens?) > >Or were you talking about a fault in the host kernel? > >If so, you're right. That's something I'm planning on bring up for 2.5. I >want UML to know when the host faulted, and go on to do something else. Anywhere. Is it possible to fix the kernel's code in real memory so it's not subject to paging? This would alleviate the problem, especially if the code is sharable (and shared). Cheers J =================================================================== EASY and FREE access to your email anywhere: http://Mailreader.com/ =================================================================== |