RFA: Fix for PR 24376

[Janis Johnson]

On Mon, Jan 30, 2006 at 03:51:29PM +0000, Nick Clifton wrote:
> Hi Gabriel,
> 
> >Is this a regression?  
> 
> I am not quite sure if it counts as a regression.  The PR includes a 
> test case which demonstrates the problem, but I am not sure that there 
> any gcc testcases which also suffer from the defect.  (But to be honest 
> I have not investigated all of the unexpected failures in the gcc 
> testsuite for a v850 toolchain to find out if this particular PR is the 
> cause).

It's a regression if the testcase passes with an earlier release of GCC.
Whether or not the bug is already covered by the testsuite is irrelevant.

Janis