[lkml]   [1997]   [Apr]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: thousands of threads & kernel continuations
> The Utah Mach project went further and observed that clients can lend
> their threads to the servers which process their RPC calls. Not only do
> you not have to keep all those kernel stacks around, but servers don't
> even need to create service threads in the first place, because their
> clients will lend them ones when the time comes. Furthermore, the latency

It was about this point why I finally decided the utah mach people had
gone totally off their trolley (Apart from the fact many of the papers
are great where code isnt). Making a call where you block and lend
the person your stack is known in normal programming as

a function call


 \ /
  Last update: 2005-03-22 13:39    [W:0.036 / U:4.560 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site