This is the mail archive of the
mailing list for the elfutils project.
Re: [PATCH] Improve AM_SILENT_RULES coverage
- From: Josh Stone <jistone at redhat dot com>
- To: elfutils-devel at lists dot fedorahosted dot org
- Date: Tue, 06 Oct 2015 10:23:06 -0700
- Subject: Re: [PATCH] Improve AM_SILENT_RULES coverage
On 10/06/2015 06:00 AM, Mark Wielaard wrote:
> On Mon, 2015-10-05 at 11:15 -0700, Josh Stone wrote:
>> Note, elfutils does not explicitly enable AM_SILENT_RULES. It's only
>> available starting from automake 1.11
> Note that we already require 1.11 or higher for parallel-tests.
> I wouldn't mind if we also just added AM_SILENT_RULES([yes]) by default.
I do prefer silent builds too. You'll probably want to force V=1 for
automated builds like rpm though. I leave the default choice to you.
>> , but starting from automake 1.13
>> silent rules are always generated, defaulting to verbose. $(AM_V_foo)
>> additions should be no-ops on systems that don't support silent rules.
> I like this patch. Just one question, what defines the $(AM_V_foo)
> (specifically AM_V_CC and AM_V_CCLD) variables? I only see AM_V_at and
> AM_V_GEN described in the automake manual. Are they officially OK to
> use? OK to push if they are. Otherwise I think we should define them (or
> some renamed version ourselves based on the official AM_V_GEN macro).
It's easy to answer "what defines" -- they're set in /usr/bin/automake
handle_languages() based on $lang->ccer and $lang->lder, which are names
declared in earlier register_language() calls.
I have no idea if they're "officially" OK though. It's strange that
these aren't documented at all. In practice it should be fine, I think,
but if you want to be really careful we can define our own renamed variants.