[Bug tree-optimization/35609] [4.3/4.4 Regression] "is used uninitialized in this function" should be may warning

[rguenth at gcc dot gnu dot org]

--- Comment #4 from rguenth at gcc dot gnu dot org  2008-03-19 10:45 ---
Subject: Bug 35609

Author: rguenth
Date: Wed Mar 19 10:44:52 2008
New Revision: 133341

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=133341
Log:
2008-03-19  Richard Guenther  <[EMAIL PROTECTED]>

PR middle-end/35609
* tree-ssa.c (always_executed): New global flag.
(warn_uninitialized_var): If !always_executed warn with "maybe"
instead of "is".
(execute_early_warn_uninitialized): Compute post-dominators.
Initialize always_executed before processing each basic block.

* gcc.dg/testsuite/uninit-15.c: New testcase.
* gcc.dg/testsuite/uninit-16.c: Likewise.

Added:
trunk/gcc/testsuite/gcc.dg/uninit-15.c
trunk/gcc/testsuite/gcc.dg/uninit-16.c
Modified:
trunk/gcc/ChangeLog
trunk/gcc/testsuite/ChangeLog
trunk/gcc/tree-ssa.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35609

[Bug tree-optimization/35609] [4.3/4.4 Regression] "is used uninitialized in this function" should be may warning

[rguenth at gcc dot gnu dot org]

--- Comment #3 from rguenth at gcc dot gnu dot org  2008-03-18 16:53 ---
While this is a regression the underlying problem was latent and thus this
should be an enhacement report.  But I'll have a look.


-- 

rguenth at gcc dot gnu dot org changed:

   What|Removed |Added

 AssignedTo|unassigned at gcc dot gnu   |rguenth at gcc dot gnu dot
   |dot org |org
   Severity|normal  |enhancement
 Status|NEW |ASSIGNED
   Priority|P3  |P4
   Last reconfirmed|2008-03-16 21:32:33 |2008-03-18 16:53:41
   date||


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35609

[Bug tree-optimization/35609] [4.3/4.4 Regression] "is used uninitialized in this function" should be may warning

[rguenth at gcc dot gnu dot org]

--- Comment #2 from rguenth at gcc dot gnu dot org  2008-03-16 21:32 ---
The warning works as designed, the PHI node in question that causes the
warning only has a single incoming edge:

testfunc ()
{
  int alt_reloc;
  int foo.0;

:
  foo.0_1 = foo;
  if (foo.0_1 > 19)
goto ;
  else
goto ;

:
  if (foo.0_1 > 9)
goto ;
  else
goto ;

:
  if (alt_reloc_4(D) != 0)
goto ;
  else
goto ;

:
  bar = 42;

:
  return;

}

we don't check whether the uninitialized use post-dominates the function
entry.  I don't know if we should.

This is just a case of 4.3 optimizing better than 4.2 which makes the
warning less precise.


-- 

rguenth at gcc dot gnu dot org changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
 Ever Confirmed|0   |1
   Last reconfirmed|-00-00 00:00:00 |2008-03-16 21:32:33
   date||


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35609

[Bug tree-optimization/35609] [4.3/4.4 Regression] "is used uninitialized in this function" should be may warning

[pinskia at gcc dot gnu dot org]

--- Comment #1 from pinskia at gcc dot gnu dot org  2008-03-16 21:30 ---
So Jump threading comes along and threads the jump for some reason makes the
PHI node go away.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35609

[Bug tree-optimization/35609] [4.3/4.4 Regression] "is used uninitialized in this function" should be may warning

[pinskia at gcc dot gnu dot org]

-- 

pinskia at gcc dot gnu dot org changed:

   What|Removed |Added

Summary|[4.3/4.4 Regression] bogus  |[4.3/4.4 Regression] "is
   |"is used uninitialized in   |used uninitialized in this
   |this function" warning  |function" should be may
   ||warning
   Target Milestone|--- |4.3.1


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35609