Re: Remove testsuite/gcc.c-torture/execute/20000603-1.c

[Diego Novillo <dnovillo at redhat dot com>]

On Thu, 2004-02-05 at 21:02, Richard Henderson wrote:
> On Thu, Feb 05, 2004 at 06:04:15PM -0500, Diego Novillo wrote:
> > a, UID 2, type memory tag: TMT.8, is written to, is dereferenced to store
> > b, UID 3, type memory tag: TMT.7, is written to, is dereferenced to load
> > TMT.7, UID 10
> > TMT.8, UID 11, is written to, may aliases: { a }
> [...]
> > With your modifications, we get:
> > 
> > a, UID 3, type memory tag: TMT.10, is written to, is dereferenced to store
> > b, UID 4, type memory tag: TMT.9, is written to, is dereferenced to load
> > TMT.9, UID 12, may aliases: { a }
> > TMT.10, UID 13, is written to, may aliases: { a }
> 
> Is this with or without inlining?
> 
With inlining, of course.  Without inlining DCE won't dare touch the
store to '*a' because 'a' is a PARM_DECLs, which equates to a global in
its view.

> The test clearly wants the function not to be inlined.
> Perhaps add __attribute__((noinline)) and see...
> 
Without inlining the test is guaranteed to pass.  But how does noinline
make the test any more valid?  Isn't it still invoking undefined
behaviour?


Thanks.  Diego.