lkml.org 
[lkml]   [1999]   [Dec]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Per-Processor Data Paget

On Thu, 9 Dec 1999, Scott Lurndal wrote:

> > it's not even possible - think about threads on different processors
> > running at once. Which fixed address? Page tables are shared.

> It certainly can be done, and it has been done.
>
> Only the page table containing the virt addr 0xc0000000 need be
> different between multiple threads.
>
> Page tables don't _have_ to be shared, either, they could,
> instead be duplicated.
>
> In this proposal, however, only the page directory need be
> different, the page tables for all user virtual address pages
> can be shared between threads which shared address spaces.

clone()ing costs go up immediately, from the current 10 usecs cost to
create a process to 50-100 usecs, plus MM handling gets more complicated.
Plus in PAE mode this would be even more complicated (because two levels
need to be private to the thread). I dont think this brings any good.

what you propose is not possible without destroying true threads.

-- mingo


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:55    [W:0.089 / U:1.352 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site