This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix PR47559
On Tue, 1 Feb 2011, Zdenek Dvorak wrote:
> Hi,
>
> > Committed as rev. 169487. That leaves us with the correctness issues
> > that we eliminate stores to global memory from endless loops. w/o
> > non-call exceptions we optimize the testcase to while (1);
>
> what correctness issues? If the loop is endless, eliminating the stores
> preserves the semantics,
I'm not sure. Consider this loop done in a thread, a lame way
of initializing a global and another thread using the data.
One might argue there are volatiles missing, but it at least looks
suspicious to me (we'd better move that store before the loop
instead of trying to move it after it).
I've opened PR47575 to track this possible issue.
Richard.