[Bug debug/17799] New: Non-optimizing compile loses 'this'

[Bug debug/17799] New: Non-optimizing compile loses 'this'

[rth at gcc dot gnu dot org]

With the attached test case, compile with

  g++ -g z.cc -lpthread

(gdb) b 'Foo::bar()'
Breakpoint 1 at 0x804850d: file z.c, line 34.
(gdb) run
Starting program: /home/rth/x/a.out
[Thread debugging using libthread_db enabled]
[New Thread -1220950400 (LWP 31166)]
[New Thread -1220953168 (LWP 31169)]
[Switching to Thread -1220953168 (LWP 31169)]
 
Breakpoint 1, Foo::bar (this=0x80498c0) at z.c:34
34        if (_y!=MODEB)
Current language:  auto; currently c++
(gdb) p this
$1 = (Foo * const) 0x80498c0
(gdb) n
38        return true;
(gdb) p this
$2 = (Foo * const) 0x0


While it's completely understandable for variables to be optimized away
with optimization, it's uncool for it to happen when compiling at -O0.

I'm pretty sure this due to the fact that 'this' is marked DECL_ARTIFICIAL
by the C++ front end, and DECL_ARTIFICIAL variables are allocated to 
registers rather than on the stack.

This is a regression from gcc 3.4.

-- 
           Summary: Non-optimizing compile loses 'this'
           Product: gcc
           Version: 4.0.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: debug
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: rth at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17799

[Bug debug/17799] Non-optimizing compile loses 'this'

[rth at gcc dot gnu dot org]

------- Additional Comments From rth at gcc dot gnu dot org  2004-10-03 02:25 -------
Created an attachment (id=7261)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7261&action=view)
test case


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17799

[Bug debug/17799] [4.0 Regression] Non-optimizing compile loses 'this'

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |wrong-debug
            Summary|Non-optimizing compile loses|[4.0 Regression] Non-
                   |'this'                      |optimizing compile loses
                   |                            |'this'
   Target Milestone|---                         |4.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17799

[Bug debug/17799] [4.0 Regression] Non-optimizing compile loses 'this'

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-03 17:48 -------
Confirmed.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2004-10-03 17:48:28
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17799

[Bug c++/17799] [4.0 Regression] Non-optimizing compile loses 'this'

[mmitchel at gcc dot gnu dot org]

------- Additional Comments From mmitchel at gcc dot gnu dot org  2004-12-21 19:43 -------
Richard, do you consider this a C++ front-end bug?

It does not seem to be a front-end bug to me; the "this" parameter is in fact
compiler-generated.  It's been marked DECL_ARTIFICIAL since time immemorila. 
Maybe the middle end should be performing this optimization at -O0 only if
DECL_IGNORED_P is set.  If you agree, we should recategorize this defect.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17799

[Bug middle-end/17799] [4.0 Regression] Non-optimizing compile loses 'this'

[rth at gcc dot gnu dot org]

------- Additional Comments From rth at gcc dot gnu dot org  2004-12-22 05:27 -------
I think I agree about only putting stuff in registers for DECL_IGNORED_P.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |rth at gcc dot gnu dot org
                   |dot org                     |
             Status|NEW                         |ASSIGNED
          Component|c++                         |middle-end


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17799

[Bug middle-end/17799] [4.0 Regression] Non-optimizing compile loses 'this'

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-01-01 01:43 -------
Subject: Bug 17799

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	rth@gcc.gnu.org	2005-01-01 01:43:13

Modified files:
	gcc            : ChangeLog c-decl.c dwarf2asm.c function.c 
	                 gimplify.c tree-inline.c tree-outof-ssa.c 
	                 var-tracking.c 
	gcc/cp         : ChangeLog call.c class.c decl2.c init.c rtti.c 
	                 tree.c 

Log message:
	PR middle-end/17799
	* function.c (use_register_for_decl): Check DECL_IGNORED_P instead
	of DECL_ARTIFICIAL.
	(assign_parms_augmented_arg_list): Set DECL_IGNORED_P.
	* c-decl.c (build_compound_literal): Likewise.
	* dwarf2asm.c (dw2_force_const_mem): Likewise.
	* gimplify.c (create_artificial_label): Likewise.
	* tree-inline.c (expand_call_inline): Likewise.
	* var-tracking.c (vt_initialize): Likewise.
	* tree-outof-ssa.c (create_temp): Copy DECL_IGNORED_P.
	cp/
	* call.c (make_temporary_var_for_ref_to_temp): Set DECL_IGNORED_P.
	* class.c (build_vtable): Don't conditionallize setting it
	based on DWARF2_DEBUG.
	(layout_class_type): Set DECL_IGNORED_P.
	* decl2.c (get_guard): Likewise.
	* rtti.c (get_tinfo_decl, build_lang_decl): Likewise.
	* tree.c (build_local_temp): Likewise.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.6992&r2=2.6993
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/c-decl.c.diff?cvsroot=gcc&r1=1.619&r2=1.620
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/dwarf2asm.c.diff?cvsroot=gcc&r1=1.40&r2=1.41
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/function.c.diff?cvsroot=gcc&r1=1.597&r2=1.598
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/gimplify.c.diff?cvsroot=gcc&r1=2.100&r2=2.101
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/tree-inline.c.diff?cvsroot=gcc&r1=1.160&r2=1.161
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/tree-outof-ssa.c.diff?cvsroot=gcc&r1=2.38&r2=2.39
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/var-tracking.c.diff?cvsroot=gcc&r1=2.23&r2=2.24
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&r1=1.4564&r2=1.4565
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/call.c.diff?cvsroot=gcc&r1=1.525&r2=1.526
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/class.c.diff?cvsroot=gcc&r1=1.697&r2=1.698
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/decl2.c.diff?cvsroot=gcc&r1=1.762&r2=1.763
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/init.c.diff?cvsroot=gcc&r1=1.407&r2=1.408
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/rtti.c.diff?cvsroot=gcc&r1=1.207&r2=1.208
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/tree.c.diff?cvsroot=gcc&r1=1.421&r2=1.422



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17799

[Bug middle-end/17799] [4.0 Regression] Non-optimizing compile loses 'this'

[rth at gcc dot gnu dot org]

------- Additional Comments From rth at gcc dot gnu dot org  2005-01-01 01:45 -------
Fixed.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17799