Re: [Gnu-arch-users] GCC v. Arch address@hidden: Regressions on mainline]

[Andrew Suffield]

On Wed, Jun 23, 2004 at 12:10:28PM -0400, Colin Walters wrote:
> On Wed, 2004-06-23 at 16:57 +0100, Andrew Suffield wrote:
> 
> > This is almost, but not quite, precisely what you don't want. It only
> > works if you have an essentially static test suite in advance. Large
> > projects (such as gcc) have a test suite that changes often, and
> > variable requirements on which tests must pass.
> 
> I imagine that gcc has a fairly large portion of its test suite which is
> essentially static.  

Not useful, and...

> You could also imagine a separate category for the test suite, which
> developers could also commit to.

...yes, that emphasises the problem.

The *result* of 'make check' is not static at all. Consider the case
where a new test has just been added, which (a) fails, and (b) is a
regression from the previous release.

Your model now prohibits any further commits. This is totally not what
you wanted.

(Example, not specification of the problem. Most modifications to the
test suite create problems along these lines)

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'                          |
   `-             -><-          |

signature.asc
Description: Digital signature