Re: Potential fix for rdar://4658012

["Richard Guenther" <richard dot guenther at gmail dot com>]

On 8/26/06, Richard Guenther <richard.guenther@gmail.com> wrote:
> On 26 Aug 2006 11:12:03 -0700, Ian Lance Taylor <iant@google.com> wrote:
> > "Richard Guenther" <richard.guenther@gmail.com> writes:
> >
> > > Instead the following patch makes the C testcase above share the stack
> > > slot.  (completely untested)
> > >
> > > Richard.
> > >
> > > Index: calls.c
> > > ===================================================================
> > > *** calls.c     (revision 116273)
> > > --- calls.c     (working copy)
> > > *************** expand_call (tree exp, rtx target, int i
> > > *** 1985,1991 ****
> > >             /* For variable-sized objects, we must be called with a target
> > >                specified.  If we were to allocate space on the stack here,
> > >                we would have no way of knowing when to free it.  */
> > > !           rtx d = assign_temp (TREE_TYPE (exp), 1, 1, 1);
> > >
> > >             mark_temp_addr_taken (d);
> > >             structure_value_addr = XEXP (d, 0);
> > > --- 1985,1991 ----
> > >             /* For variable-sized objects, we must be called with a target
> > >                specified.  If we were to allocate space on the stack here,
> > >                we would have no way of knowing when to free it.  */
> > > !           rtx d = assign_temp (TREE_TYPE (exp), 0, 1, 1);
> > >
> > >             mark_temp_addr_taken (d);
> > >             structure_value_addr = XEXP (d, 0);
> >
> > I expect that would fail for something like
> >
> > typedef struct { char x[1000]; } S;
> > S bar (void);
> > void quux (S, S);
> > void foo (void)
> > {
> >   quux (bar(), bar());
> > }
>
> Not really.  It passed testing for C and C++ at least.  For the above we get
> the still bad (-Os to not inline the memcpys)

Instead it fails for the testcase below.  Committed as obvious.

Richard.

Index: testsuite/ChangeLog
===================================================================
*** testsuite/ChangeLog (revision 116476)
--- testsuite/ChangeLog (working copy)
***************
*** 1,3 ****
--- 1,7 ----
+ 2006-08-26  Richard Guenther  <rguenther@suse.de>
+
+       * gcc.c-torture/compile/20060826-1.c: New testcase.
+
 2006-08-26  Lee Millward  <lee.millward@codesourcery.com>

         PR c++/28736
Index: testsuite/gcc.c-torture/compile/20060826-1.c
===================================================================
*** testsuite/gcc.c-torture/compile/20060826-1.c        (revision 0)
--- testsuite/gcc.c-torture/compile/20060826-1.c        (revision 0)
***************
*** 0 ****
--- 1,7 ----
+ typedef _Complex double S;
+ S bar (void);
+ void quux (S, S);
+ void foo (void)
+ {
+  quux (bar(), bar());
+ }