head branch broken (SH4 at least)

[Will Newton]

On Fri, Mar 4, 2011 at 12:37 PM, Thomas Petazzoni
<thomas.petazzoni at free-electrons.com> wrote:
> Hello,
>
> On Fri, 4 Mar 2011 13:14:24 +0100
> Carmelo AMOROSO <carmelo.amoroso at st.com> wrote:
>
>> I've pull it this morning getting a long list of Peter's patch and
>> (using the same .config as before) I'm having a lot of build problems.
>
> I have also seen all these commits, but I don't remember seeing the
> corresponding patches being sent on the list. Shouldn't all patches be
> sent to the list for community approval before being applied ?

It would be nice to be able to get some more review on changes like this one:

http://git.uclibc.org/uClibc/commit/?id=152e66242f2af7cd26e29d59a31f8bb8479595e5

Which has only the comment "update some headers" most of which does
seem to be minor updates from glibc, but also adds a number of #if 0
to the code and a #warning to a .c file.

The next commit to locale.h is this one:

http://git.uclibc.org/uClibc/commit/?id=2a0c568ddc7b6b142dc049d8cc9cf32eb38e2321

"fix locale build", which should really be squashed into the previous
commit before pushing, this is what rebase is for, it makes history
easier to read and avoids breaking bisection.

Also subject lines of patches should be kernel style compliant so in
gitweb instead of:

4 days	style update
4 days	syntax fixes
4 days	avoid warnings on _STACK_GROWS_* usage
4 days	syntax fix

We can actually tell which files (or area of functionality) were modified.