lkml.org 
[lkml]   [1997]   [Nov]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: CONFIG_SMP patch updated for 2.1.61


On Wed, 5 Nov 1997, Werner Almesberger wrote:
>
> > And I believe this could (and should) be done without mkdep.c: gcc is
> > able to generate dependency information for us. (I already use
> > modified makefiles from mec@shout.net, which do not need mkdep.c -
> > dependency info are generated on the fly when compiling).
>
> Hmm, mkdep does its job pretty quickly and in this case, "normal" gcc
> couldn't easily do the same thing, so with my approach, mkdep becomes
> actually quite unavoidable.

Yes. "gcc -M" has various "features" that makes it completely useless for
any real dependancy generation.

- it's slow as molasses. I'm not talking "quite slow", I'm talking "so
slow that it's pretty much useless for any reasonably sized program".
It's essentially only useful for batch-processing.
- it doesn't know about configuration dependencies (ie the things that
mkdep does to warn you when the config file hasn't been properly
included)
- it actually parses C preprocessor stuff (notably #ifdef), and because
of that it often leaves out critical dependency information (and it
_requires_ that you know all of the configuration at mkdep time).
- it will do recursive dependencies (exactly because it parses the C
file), which is extremely inefficient (that's the whole point on having
"make" in the first place - to keep track of complex dependencies).

It becomes even more useless when trying to make a more intelligent
configuration setup.

Linus


\
 
 \ /
  Last update: 2005-03-22 13:40    [W:0.057 / U:16.888 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site