Sam Ravnborg wrote:
On Wed, Apr 13, 2005 at 01:49:10PM +0200, Toralf Lund wrote:
Yes. As I've (also) already said elsewhere, I knew that, really. The current build setup fails to do this partly for historical reasons, partly because the driver also supports different OSes. (And is still expected to build correctly with Linux 2.4, not just 2.6.)
Hmmm. Seems like my original reply to this message got lost...
What? Write a simple makefile that a normal human being may actually understand? Not autogenerate something utterly unreadable from something that's autogenerated from something that's ... ???Following trick works with both 2.4 and 2.6:
makefile: all: $(MAKE) -C Kernel_src_path SUBDIRS=$(PWD) modules
Makefile:
obj-m := mymodule.o
It obtains CFLAGS as expected etc.
People seems to do their best to avoid such a simple setup :-(
It's not *quite* that simple with the module I'm talking about, though, as the source code is split into several files. Which is a Good Thing, IMO.
Also, I was unsure if this would work with Linux 2.4 (but I was going to test it)...
Sam
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
