[PATCH] fix mark_single_exit_loops

[Mostafa Hagog]

Gerald Pfeifer <gerald@pfeifer.com> wrote on 29/03/2005 15:45:01:

> On Thu, 10 Mar 2005, Mostafa Hagog wrote:
> > I cannot provide a test case because this happens with SMS patch that
> > isn't yet in mainline.
>
> I guess you still could add that test case, even if it does not fail
> with current mainline -- it might help catch regressions for other
> changes or others working on similar changes.
>
> Adding testcases up front, instead of waiting for a merge of a branch
> seems preferrable (when compiling/running the test properly does not
> depend on new functionality added on the branch).

A test case is possible only if you change the code of GCC.  SMS was using
loop information without SIMPLE_PREHEADERS - I change SMS use
SIMPLE_PREHEADERS.
under the above assumption that code is correct - I suppose we should
comment this
fact.

>
> Gerald