lkml.org 
[lkml]   [1996]   [Dec]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: Proposal: restrict link(2)
Date

> If you think it's that easy and necessary, then do it yourself. You've
> got the kernel source, after all. I myself happen to think there's no
> good reason to change the behavior of link() in a way that makes it
> incompatible with other UNIX systems, or to make any such change part of
> a distribution Linux kernel.
>

Well, I've been reading this discussion and would like to state my (very
humble) opinion:
I think that the suggested patch should exist. I'm not sure about having it
in the mainstream kernel. I'm not kernel-knowledgeable enough to know if
there are already any options that deliberatedly break compatibility with other UN*X, but it doesn't sound like the right thing.

Of course, anyone should feel free to apply whatever patches he/she wants. But
as Linux becomes more and more popular, the .config options grow and grow,
maybe making the compiling process a little confusing. I think the kernel
documentation might include some weird patches, with BIG "do it at your risk"
warnings. I believe this is particularly important when addressing an issue as
sensitive as security. Compatibility/security tradeoff? I wouldn't use the
patch, but it wouldn't hurt to know that it exists and where to find it just in case.

But I wouldn't like going through lots of options for perhaps two users among
millions:

Parallel port Space Shuttle support (CONFIG_PARALLEL_SHUTTLE) [N/y/m/?]

Just my $0.00002, Diego.
--
-------------------------------------------------------------------------------
Diego M. Basch dmb@hsnce.gov.ar
PGP Key: http://hi.fi.uba.ar/~diego/pgpkey

\
 
 \ /
  Last update: 2005-03-22 13:38    [W:0.131 / U:1.484 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site