This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: autogen -T ../../trunk/fixincludes/check.tpl ....


http://gcc.gnu.org/ml/gcc-patches/2006-03/msg00783.html

- Requiring autogen to test fixincludes doesn't seem like a big deal.
Death by a thousand cuts. It is a feature that one need not have other
packages on his system when doing a make check.

The one and only real purpose to doing the check is to verify that anybody making changes to the fixes has not broken other fixes. So, "make check" in the fixincludes directory is a validation that nothing is broken, not so much a validation that the compiler was built correctly (successfully). It is otherwise innocuous (but time consuming) to run it as a regular part of the compiler make check.

SO: if it is going to be run as part of a top level "make check", then
it is proper to put the checking script under source control.  If it is
only to be run after making changes to the fixes, then *perforce*
someone doing that must have autogen.  (Or more editing patience
than I do...)

I think the check.sh script should be removed from source control.
``make check'' in fixincludes should not be a default action for a
top level make check.

But it's no biggie either way either.

Cheers - Bruce


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]