> Imagine everyone still including <linux/config.h>, but 'virtual' files
> linux/config/config_scsi_aha1770 etc. And then, when mkdep would see
> CONFIG_SCSI_AHA1770 in foo.c, it would make foo.c depend on
> CONFIG_SCSI_AHA1770. This could work.
That's how Werner Almesberger's patch works. Specifically, when foo.c
includes <linux/config.h> and uses the symbol CONFIG_SCSI_AHA1770,
then the dependency file does not say:
foo.c: include/linux/config.h /* not any more! */
Instead it says:
foo.c: include/config/scsi_aha1770.dep
'make config' updates the times on the individual files in
include/config/*.h. So the result is perfect granularity,
automatically, with no massive source-file changes.
> If someone comes with system that has better config_X dependencies
> (and read-only source tree etc.), but rewrites every Makefile in
> Linux, are you going to accept that patch? Is there at least chance?
These are two separate questions for me. In the short run,
Werner's patch is only 500 lines long and it does *not* rewrite all
the Makefiles or touch all the source files. I am going to focus
on this patch for now.
Michael Chastain
<mailto:mec@shout.net>
"love without fear"