[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-05-12 17:33 -------
*** Bug 15400 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |leblanc at skycomputers dot
                   |                            |com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[leblanc at skycomputers dot com]

------- Additional Comments From leblanc at skycomputers dot com  2004-05-12 17:44 -------
Would someone please assign this?  This bug can severely impact systems with limited memory.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[bangerth at dealii dot org]

------- Additional Comments From bangerth at dealii dot org  2004-05-12 18:11 -------
Rather than looking just at the scopes of variables, this should properly  
be fixed by looking at the live ranges of variables.  
  
W.  

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[wilson at gcc dot gnu dot org]

------- Additional Comments From wilson at gcc dot gnu dot org  2004-05-12 22:54 -------
The only issue with the patch from Daniel Jacobowitz was a problem with
TARGET_EXPR, and I wasn't sure how to fix that.  Meanwhile, Ulrich Weigand
checked in a patch May 1 to fix an aliasing problem with TARGET_EXPRs that was
apparently related to the problem I identified last year, as his patch is a
superset of mine.  Therefore, it may be the case that Daniel's patch is OK now.
 It needs retesting.

The SAVE_EXPR/var_temp_slot_level stuff is still broken.  Someone needs to look
into that.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[wilson at specifixinc dot com]

------- Additional Comments From wilson at specifixinc dot com  2004-05-12 23:00 -------
Subject: Re:  Coelesce stack slots for disjoint scopes.

On Wed, 2004-05-12 at 10:44, leblanc at skycomputers dot com wrote:
> Would someone please assign this?  This bug can severely impact systems with limited memory.

See the patch in comment #6.  It should work, but this stuff is tricky,
as there is no way to prove that this patch is safe.  We may just have
to try it and see what happens.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[drow at false dot org]

------- Additional Comments From drow at false dot org  2004-05-12 23:47 -------
Subject: Re:  Coelesce stack slots for disjoint scopes.

On Wed, May 12, 2004 at 10:55:00PM -0000, wilson at gcc dot gnu dot org wrote:
> 
> ------- Additional Comments From wilson at gcc dot gnu dot org  2004-05-12 22:54 -------
> The only issue with the patch from Daniel Jacobowitz was a problem with
> TARGET_EXPR, and I wasn't sure how to fix that.  Meanwhile, Ulrich Weigand
> checked in a patch May 1 to fix an aliasing problem with TARGET_EXPRs that was
> apparently related to the problem I identified last year, as his patch is a
> superset of mine.  Therefore, it may be the case that Daniel's patch is OK now.
>  It needs retesting.
> 
> The SAVE_EXPR/var_temp_slot_level stuff is still broken.  Someone needs to look
> into that.

I'll try to take a look at these after the merge.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|3.5.0                       |---


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[naveens at noida dot hcltech dot com]

------- Additional Comments From naveens at noida dot hcltech dot com  2004-07-07 16:09 -------
 > we should allocate only one structure on the stack.
 > Here's some interesting numbers I got for this testcase.  HEAD+stack is
 > the -fstack-reorg patch posted some time ago.
 
A small clarification.
-fstack-reorg doesn't affect stack height significantly. It had a slightly 
different objective (i.e. reorganizing frame to reduce address arithemtic
for processors with limited offsets). Stack height reduction,
if any, occurs due only due alignment holes elimination. (not in this
case anyway)

I realize that work can be used to reduce stack height as well.
Myself and Rakesh Kumar plan to work on this problem shortly.

If the proposed patch (for this PR) works, it is fine, otherwise we can look 
into this. However, it seems it will not be a bug-fixing exercise, 
because this is actually a design deficiency.

Cannot give a commitment on timeline though.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[wilson at gcc dot gnu dot org]

------- Additional Comments From wilson at gcc dot gnu dot org  2004-07-08 02:05 -------
A related but slightly different problem was exposed by a testcase from Ian
Taylor.  The message with the testcase is here:
    http://gcc.gnu.org/ml/gcc/2004-06/msg01827.html
My analysis is here:
    http://gcc.gnu.org/ml/gcc/2004-06/msg01913.html

Briefly, expand_builtin returns a value even when the ignore parameter is true.
 This confuses other code.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[rth at gcc dot gnu dot org]

------- Additional Comments From rth at gcc dot gnu dot org  2004-09-02 07:15 -------
Working on it.  Finally.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |rth at gcc dot gnu dot org
                   |dot org                     |
             Status|NEW                         |ASSIGNED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-09-03 23:50 -------
Subject: Bug 9997

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	rth@gcc.gnu.org	2004-09-03 23:50:11

Modified files:
	gcc            : ChangeLog Makefile.in cfgexpand.c 

Log message:
	PR middle-end/9997
	* cfgexpand.c (LOCAL_ALIGNMENT): Provide default.
	(STACK_ALIGNMENT_NEEDED, FRAME_GROWS_DOWNWARD): Likewise.
	(struct stack_var, EOC, stack_vars, stack_vars_alloc, stack_vars_num,
	stack_vars_sorted, stack_vars_conflict, stack_vars_conflict_alloc,
	frame_phase, get_decl_align_unit, add_stack_var, triangular_index,
	resize_stack_vars_conflict, add_stack_var_conflict,
	stack_var_conflict_p, add_alias_set_conflicts, stack_var_size_cmp,
	union_stack_vars, partition_stack_vars, dump_stack_var_partition,
	expand_one_stack_var_at, expand_stack_vars, expand_one_stack_var,
	expand_one_static_var, expand_one_hard_reg_var,
	expand_one_register_var, expand_one_error_var, defer_stack_allocation,
	expand_one_var, expand_used_vars_for_block, clear_tree_used): New.
	(expand_used_vars): Rewrite.
	* Makefile.in (cfgexpand.o): Update dependencies.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.5243&r2=2.5244
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/Makefile.in.diff?cvsroot=gcc&r1=1.1367&r2=1.1368
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cfgexpand.c.diff?cvsroot=gcc&r1=2.18&r2=2.19



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[rth at gcc dot gnu dot org]

------- Additional Comments From rth at gcc dot gnu dot org  2004-09-03 23:52 -------
http://gcc.gnu.org/ml/gcc-patches/2004-09/msg00440.html

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |3.5.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2003-12-11 02:24 -------
For some reason this is no longer fixed on the tree-ssa but it could be and should be able 
to be fixed.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |pessimizes-code
   Last reconfirmed|0000-00-00 00:00:00         |2003-12-11 02:24:29
               date|                            |
   Target Milestone|---                         |tree-ssa


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997

[Bug middle-end/9997] Coelesce stack slots for disjoint scopes.

[pinskia at physics dot uc dot edu]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9997



------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-10 02:54 -------
This looks fixed on the tree-ssa branch.