Re: [PATCH,testsuite]Disable warning for _Pragma3.c to avoid timestamp failures

[Janis Johnson <janis187 at us dot ibm dot com>]

On Wed, 2008-06-18 at 09:24 -0400, hutchinsonandy@aim.com wrote:
> Exactly my point. The test depends on update method and host tools - 
> and clearly it should not.
> Also it would break if the tested file were every revised.
> 
> If the purpose of the test were to test dependencies, there should be a 
> "generator" to take care of relative file ages.
> But is it a test is looking for absence of segmentation failure, so 
> that complexity can be avoided with a simple -w.
> 
> I briefly considered changing test to check it's own file timestamp - 
> but thought that would be slightly bizarre and might prompt some other 
> helpful warning.

Let's leave the test as it is.  http://gcc.gnu.org/svn.html recommends
the use of contrib/gcc_update, which solves many dependency issues.

Janis