I'd like to suggest deprecating netsyms.c and ipv6_syms.c
in linux-2.5. Instead, as people write new code, they should put
the EXPORT_SYMBOL declarations where the symbol is actually defined.
Please correct me if I'm wrong, but my understanding is that
the 2.5 build process gets no build speed advantage from reducing the
number of files that export symbols.
In comparison, I see several advantages to making the
EXPORT_SYMBOL declarations in the files where the symbols are defined,
even if the advantages are modest:
1. Patches will not conflict as often.
2. It will elimaintes a need to #include <config.h>, making
builds after configuration changes a fraction of a second faster.
2a. Side note: Someday, I'd like to see <config.h> only
include boolean variables.
3. It should reduce bugs due to complex nested "#ifdef CONFIG_xxx"
sections or missing "CONFIG_xxx_MODULE" tests, which are often
only detectable in rare configuration combinations.
4. Eliminating if "#ifdef...#endif" pairs will reduce the kernel
line count by up to 88 lines.
5. Simplifies modularization efforts
5a. Simplified making facilities modular at all.
5b. Simplified chopping big modules into small ones.
Comments are welcome.
Adam J. Richter __ ______________ 575 Oroville Road
adam@yggdrasil.com \ / Milpitas, California 95035
+1 408 309-6081 | g g d r a s i l United States of America
"Free Software For The Rest Of Us."
-
: send the line "unsubscribe linux-net" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
