One tiny question first: ************************************************************** * Any pointers to a working ftp mailserver?? I can't ftp!! * ************************************************************** About swapping/paging: I think demand paged virtual memory is a requirement for a real OS. No more crammed "just a bit better than DOS" OS'es, please! After all, disk memory is still cheaper than RAM. Peter MacDonald asks if it's possible to demand pages from the load-module: yes in deed. This is done in lots of commercial systems; one reason being the startup time for large applica-- tions; e.g. emacs :-). You only need a few pages of 4kb to start the monster. The system will seem much faster. For this to work properly, a few things are needed. For in-- stance, the code segment of the load module must start at a block boundary in the file system, eg. offset 2kb from the start. The file system must also be fast - pointers to physical disk locations should be kept in RAM - in this way, it will not be (much) slower than reading from a swap-disk. Of course the load module must be locked while it's being executed. I still think we need to be able to swap out data - think about running X-applications, with tons of data almost never being used. Kind regards Kurt \/\/achmann, M.Sc. EE/SE
| Amit K. Arora | [RFC] Heads up on sys_fallocate() |
| H. Peter Anvin | Re: [RFC 00/15] x86_64: Optimize percpu accesses |
| Nicolas Pitre | Re: [RFC patch 08/18] cnt32_to_63 should use smp_rmb() |
| Bart Van Assche | Integration of SCST in the mainstream Linux kernel |
git: | |
| Jarek Poplawski | [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| David Miller | [GIT]: Networking |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| Natalie Protasevich | [BUG] New Kernel Bugs |
