[lkml]   [2004]   [Jun]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH] Symlinks for building external modules
Sam Ravnborg wrote:
> On Fri, Jun 04, 2004 at 08:15:02PM +0300, Jari Ruusu wrote:
> > How long have you recommended building external modules like this:
> >
> > make -C /lib/modules/`uname -r`/build modules SUBDIRS=`pwd`
> > or
> > make -C /lib/modules/`uname -r`/build modules M=`pwd`
> >
> > Now they all have to be changed to:
> >
> > make -C /lib/modules/`uname -r`/source modules SUBDIRS=`pwd`
> > or
> > make -C /lib/modules/`uname -r`/source modules M=`pwd`
> That would not work either.
> You need to tell kbuild both where to find the source and the output files.
> So you have to specify both -C ... and O=...
> KERNEL=/lib/modules/`uname -r`
> make -C $KERNEL/source O=$KERNEL/build M=`pwd`
> There is no way to do the proposed chang and be backward compatible
> when the kernel is build using seperate output directories.

What I and Måns Rullgård are saying is this: A symlink to object directory
is missing. The fix is to add the damn symlink. However, breaking symlink to
source directory at the same time is unacceptable.

In other words:
build -> Symlink to kernel sources and build machinery, Makefiles and
scripts. No change here.
object -> Symlink to separate object tree. This is new.

Existing functionality when using separate source and object directories is
that only the 'build' symlink exists, and that points to kernel build
machinery. And that 'build' symlink must stay that way.

> Plese note that the patch Andreas made did not break existing setups
> if a seperate output directory was not used. The only effect
> would be an additional symlink to the same dir. (build and source would
> be links to the same dir).

All code that uses the 'build' symlink needs to be changed if Andreas' patch
is merged to mainline. That is quite breakage.

Sam, you are wrong here. I hope you have enough balls to admit that instead
of forcing everyone to deal with the proposed breakage.

Jari Ruusu 1024R/3A220F51 5B 4B F9 BB D3 3F 52 E9 DB 1D EB E3 24 0E A9 DD
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 14:03    [W:0.047 / U:10.240 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site