fold_truthop patch causes sparcv9 testsuite regressions

[David S. Miller]

   From: Roger Sayle <roger@eyesopen.com>
   Date: Mon, 24 Jun 2002 07:42:31 -0600 (MDT)
   
   That's what surprises me.  I'd have thought there'd atleast be a test
   for the warning in gcc.dg.  Perhaps if this check, if it exists, was
   made a link test instead of just a compile test?
   
It should be doing both.  But anyways, a non-zero exit from the
compiler would have been caught unless the test is just doing a -c -o
compilation.