Hi! On Thu, Jul 15, 1999 at 08:03:33AM +1000, Dancer wrote: > This isn't technically a security issue. It's an issue of resource > management. Most unices give the administrator the ability to deal with > this (setrlimit/getrlimit), but it's up to the administrator to do it. > Limitations that will protect from the above may hamper legitimate > applications at some deployments. Well, on MiNT, due to the lack of a virtual address space, the problem is even worse as all memory regions attached to a process at the time of the call to Pfork() have to be copied on each context switch that activates one of the Pfork()ed processes when any of the others was last active. This gets even more worse with active memory protection, though I discussed a way to bypass this (the slowdown with memory protection, not the performance problem of Pfork() itself) with Frank on IRC this week. It should make it into one of the next betas. Ciao Thomas -- Thomas Binder (Gryf @ IRCNet) gryf@hrzpub.tu-darmstadt.de PGP-key available on request! binder@rbg.informatik.tu-darmstadt.de Vote against SPAM: http://www.politik-digital.de/spam/
Attachment:
pgpASmerpUE9s.pgp
Description: PGP signature