[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-smp
Subject:    Re: idea - SGI's sproc() call
From:       hpa () transmeta ! com (H !  Peter Anvin)
Date:       1997-10-24 5:28:15
[Download RAW message or body]

Followup to:  <3447D58C.167E@nothingreal.com>
By author:    Emmanuel Mogenet <mgix@nothingreal.com>
In newsgroup: linux.dev.smp
> 
> True, featurewise clone is better than sproc.
> 
> But, there's a *major* downside to clone:
> 
> Clone is so barebone it still requires a piece of assembly code
> in userland mode to implement a clean sproc (in order to handle
> the stack pointer duplication atomically.).
> 
> Clone would definetly kick asses if it could handle that part of the
> job itself, cause right now any implementation of sproc on top
> of clone is:
> 
>         1. A pain in the neck to maintain
>         2. Not portable from one architecture to the other.
> 

That's what libraries are for.  I think glibc has a clone() wrapper
which handles the stack pointer issue.

	-hpa
-- 
    PGP: 2047/2A960705 BA 03 D3 2C 14 A8 A8 BD  1E DF FE 69 EE 35 BD 74
    See http://www.zytor.com/~hpa/ for web page and full PGP public key
        I am Bahá'í -- ask me about it or see http://www.bahai.org/
   "To love another person is to see the face of God." -- Les Misérables

[prev in list] [next in list] [prev in thread] [next in thread]