Re: pre-2.1.126-2 sd.c does not build as module

ketil@ii.uib.no
20 Oct 1998 10:51:22 +0200


Michael Elizabeth Chastain <mec@shout.net> writes:

> I have an idea for compilation testing:

> for i in 0 1 2 3 4 5 6 7 8 9 ; do
> for j in 0 1 2 3 4 5 6 7 8 9 ; do
> make randomconfig
> make vmlinux modules || (echo 'FAILED' ; exit 1)

Perhaps something like

make randomconfig $i $j
make... || echo $i $j FAILED
make clean

> done
> done
>
> The missing piece is a 'make randomconfig' option that answers
> all the questions randomly without human intervention.

<thought source="shower">
Could it be possible using an (SG|X)ML document for the configuration,
so that we could ensure only compatible options could legally be
selected with the DTD?

It is a pretty straightforward thing to parse such a file with SAX and
produce the right CONFIG_* options, and as an added bonus you get to use
emacs for editing your configuration. The business logic embedded in
config.in (
if [ "$CONFIG_..." = "y" ]; then
bool 'blah blah' CONFIG_....
) could be embedded in a DTD instead, and it would be pretty
straightforward to verify that a config is a legal combination of
options for that particular kernel version.

I guess it was a silly idea, really. Tell me if you want a suggestion
for a DTD.
</thought>

~kzm

-- 
If I haven't seen further, it is by standing in the footprints of giants

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/