lkml.org 
[lkml]   [1998]   [Jul]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: STILL NEEDED PATCH-2.1.107
dalecki@cs.net.pl wrote:
>> >of CONFIG_UNIX the kernel will don't compile correctly. By the way: Why
>> >isn't then SysV IPC supposed to be dependant upon CONFIG_NET???!!!
>>
>> Because SysV IPC doesn't use sockets. It has its own set of syscalls.
>
>sockets have special syscalls too. And please don't tell me.... What You
>said is effectively : IPCs haves the logically correct placement in the
>setup becouse they have. The compleatly analogous socket mechanism doesn't
>becouse it doesn't...

CONFIG_SYSVIPC contrls the SysV IPC, i.e. shmget, msgget, and semget.

CONFIG_NET controls the socket api, i.e. the syscalls called socket,
bind, send and so on.

There is nothing in common between the SysV IPC and the socket API,
therefore they have different (unrelated) config options.

CONFIG_UNIX and CONFIG_INET both require the socket API and thus
require CONFIG_NET to be set.

Apparently from what you're saying there is still some code which
won't work properly if CONFIG_NET=y and CONFIG_UNIX!=n and
CONFIG_INET=n, and that is a bug. You're quite welcome to fix that if
you want to. Just don't expect Alan or anybody else to fix it soon,
since there are much more ugent problems elsewhere. And fixing it by
allowing CONFIG_NET=n and CONFIG_UNIX=y is not the correct solution.

/Christer (stating the obvious?)

--
If it's tourist season, why can't we shoot them?

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 13:43    [W:0.654 / U:0.140 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site