Messages in this thread | | | Date | Tue, 4 Nov 1997 10:46:13 -0800 (PST) | From | Linus Torvalds <> | Subject | Re: CONFIG_SMP patch updated for 2.1.61 |
| |
On Tue, 4 Nov 1997, Erik Andersen wrote: > > Here it is: > > ftp://ftp.shout.net/pub/users/mec/patch/config-smp.2161 > > This updates my previous patch, config-smp-2154. There are no > substantial changes.
Note that the principal reason why CONFIG_SMP isn't an option is that it affects too much. It means that if you re-make the config file, _everything_ gets recompiled. It _used_ to be a config option, but I removed it because of this problem.
The only way I would like a CONFIG_SMP would be if the config process itself was made smarter, and did:
- separate config files (according to some simple rule). Instead of doing
#include <linux/config.h>
we'd do
#include <config/smp.h> #include <config/scsi.h> ...
(maybe 10 different config files - very few files would care about more than one or two)
- better CONFIG_XXX sanity checking that knew about the rules. Right now "make dep" can warn when somebody uses a CONFIG option without including the config.h file. If the above is done, then mkdep.c also needs to be smarter.
This could be done by selecting the config filename by looking at the name of the config option itself and making that a rule. For example:
CONFIG_SCSI_AHA1770 /* CONFIG_SCSI_ prefix -> <config/scsi.h> CONFIG_FS_MSDOS /* CONFIG_FS_ prefix -> <config/fs.h> ...
This way mkdep.c and the config scripts themselves would be able to determine which config file is implied by which config option.
In short, I think that we need a more clever config setup, and before we have that I don't want to see CONFIG_SMP as an option.
Linus
| |