lkml.org 
[lkml]   [1996]   [Apr]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: Steve's crashing 1.3 machine, cured?
Jamie Lokier (jamie@rebellion.co.UK) wrote:

: As a special luxury, how about locating hardware probes in another
: segment (using compiler and linker directives) so they can be unloaded
: after use?

I tried it, and successfully. Unfortunately, it requires marking
all initialization procedures/variables with __attribute__(....)
and my patches are pretty difficult to support even without these marks.
If you saw them, you can imagine what I feel when Linus reformats
any part of source tree...
BTW, it appeared that initializing code is pretty small part of devices.

: d. does not require ksyms.c

: Does this mean that the run-time kernel image includes all of its
: symbols at run-time?

It does not!

I just generate aksyms.def from the currently installed modules
and from a default set (aksyms-std.def, it is empty now)
It is easy, because my modules contain all necessary information
about symbol relations and dependencies.
So that, initial module tree is consistent automatically.
If you made a standalone module that requires a symbol
currently not exported and does not want to recompile kernel,
you just load "bridge" module made from System.map or vmlinux.

Alexey Kuznetsov.


\
 
 \ /
  Last update: 2005-03-22 13:37    [W:0.012 / U:0.876 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site