g77 tests for formatted I/O

[Joseph S. Myers]

On Sat, 26 Jan 2002, Toon Moene wrote:

> Well, I'm certainly fine with them in that respect.  However, I thought
> that we wanted to keep the testsuite a regression suite; in which case
> only entries should be allowed that actually test for a bug in the
> compiler.

No, codingconventions.html clearly states:

   Every language or library feature, whether standard or a GNU
   extension, and every warning GCC can give, should have testcases
   thoroughly covering both its specification and its implementation.
   Every bug fixed should have a testcase to detect if the bug recurs.

-- 
Joseph S. Myers
jsm28@cam.ac.uk