This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] Remove symlinks created in argv0-symlink.exp and general cleanup
- From: Doug Evans <dje at google dot com>
- To: Joel Brobecker <brobecker at adacore dot com>
- Cc: Sergio Durigan Junior <sergiodj at redhat dot com>, Pedro Alves <palves at redhat dot com>, Simon Marchi <simon dot marchi at ericsson dot com>, gdb-patches <gdb-patches at sourceware dot org>, Yao Qi <qiyaoltc at gmail dot com>
- Date: Wed, 12 Aug 2015 10:37:42 -0700
- Subject: Re: [PATCH] Remove symlinks created in argv0-symlink.exp and general cleanup
- Authentication-results: sourceware.org; auth=none
- References: <1438287227-11303-1-git-send-email-simon dot marchi at ericsson dot com> <55BF7962 dot 3060106 at redhat dot com> <55BF85BC dot 8040102 at ericsson dot com> <55BF9811 dot 50202 at redhat dot com> <20150804172110 dot GI4777 at adacore dot com> <CADPb22TsqOiq6AM9h_0A-a7rp=dX6ds-E8K21goX+AK5UMjadA at mail dot gmail dot com> <55CB123F dot 70603 at redhat dot com> <87r3n85ukx dot fsf at redhat dot com> <20150812172225 dot GE22245 at adacore dot com>
On Wed, Aug 12, 2015 at 10:22 AM, Joel Brobecker <brobecker@adacore.com> wrote:
>> FWIW, this would be awesome. I am not sure about forcing the dependency
>> on GNU Make for the testing (this would be OK by me, but maybe someone
>> has a case against), but defaulting to FORCE_PARALLEL=1 would be a step
>> forward, indeed. While hacking our testsuite to implement a way to
>> verify the racy tests we have, I could see how the fact that each test
>> has its own directory makes a difference.
>
> The only argument against GNU/Make that I remember is that on the *BSDs,
> GNU/Make is not the default. But the benefits we are getting from using
> GNU/Make continues to increase, so I think we're reaching a point where
> it seems to make better sense to me to require it. Doesn't GCC already
> do anyways?
I suspect, though I haven't tried, if we remove the check-parallel/check-single
test, it would be straightforward to remove the GNU make requirement
and just move the details to a wrapper script.
Always invoking runtest by a wrapper script has other benefits.
I'm not advocating for this, I only ever use GNU make.
But if the topic comes up, it's not clear to me GNU make
is an absolute requirement.