On Tue, 2016-10-25 at 09:08 +0200, Lukas Slebodnik wrote:
On (25/10/16 14:23), William Brown wrote:
>Hi,
>
>I want to get feedback from the team about a small change I would like
>to make to our configure.ac. I want to add the following to
>configure.ac.
>
>m4_ifdef([AM_SILENT_RULES], [
> AM_SILENT_RULES([yes])
>])
>
We(sssd) do not use it by default for few reasons.
It need to be expilicitely disabled if you want to see failures
in CI/mock build. Otherwise, it's imposible to check whether
compilation/linking had correct flags on different distributions
fedora/rhel/debian/opensuse and troubleshooting is much harder.
Hmmm that's a good point. I didn't think of this.
This is a reason why we prefer to explicitely enable it at configure
time or at build time
./configure --enable-silent-rules
or at build time
make V=0 (make V=1)
I didn't realise you could do that.
Both versions are independent.
But I agree that thos posibility is very usefull for development.
IMHO, it might be better not to use "yes" as a default
m4_ifdef([AM_SILENT_RULES], [AM_SILENT_RULES])
BTW. I can already see silent-rules in configure help.
[root@7380c2d8ecf1 ds]# ./configure --help | grep silent
-q, --quiet, --silent do not print `checking ...' messages
--enable-silent-rules less verbose build output (undo: "make V=1")
--disable-silent-rules verbose build output (undo: "make V=0")
Thanks, I might just leave this one alone then.
--
Sincerely,
William Brown
Software Engineer
Red Hat, Brisbane