[Bug c++/23168] New: [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[Bug c++/23168] New: [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[pomac at vapor dot com]

Hi,

First of all this is my first bugreport when it comes to compilers and it's not
that detailed so bear with me.

If i compile a special version of VisualBoyAdvance on my box (fixed a 64 bit
problem, not really tested) using gcc 4.0.1 as supplied by gentoo i will get a
OOM on my machine, see memory info.

gcc 4.0.1 seemed to hover with a resident memory size of 800 mb while 3.4.3 did
the same thing at about 342 mb. 3.4.3 seemed to peak at 600mb resident size
(checked with top, so not reliable).

This happens while compiling GBA.cpp which is a rediculusly large cpp file:
100K    GBA.cpp, but only 4234 lines.

Is this intended behavior, or indeed a bug?
(Btw, i have been trying to get the gentoo 4.1.x snapshots to compile for
further verification but they never seem to work =P)

Mem info: 
Mem:       1023420
Swap:      1642932

gcc (GCC) 4.0.1 (Gentoo 4.0.1, pie-8.7.8)
gcc (GCC) 3.4.3 20041125 (Gentoo 3.4.3-r1, ssp-3.4.3-0, pie-8.7.7)

-- 
           Summary: [OOM] GCC 4.0.1 OOM's while compiling VBA (works with
                    3.4.x)
           Product: gcc
           Version: 4.0.1
            Status: UNCONFIRMED
          Severity: critical
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: pomac at vapor dot com
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: amd64-linux-gnu
  GCC host triplet: amd64-linux-gnu
GCC target triplet: amd64-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23168

[Bug c++/23168] [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[pomac at vapor dot com]

------- Additional Comments From pomac at vapor dot com  2005-08-01 00:03 -------
Sorry, totally forgot about the source file, must be to tired for this but here
it is:
http://pomac.netswarm.net/misc/vis.tar.bz2

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23168

[Bug c++/23168] [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[pomac at vapor dot com]

------- Additional Comments From pomac at vapor dot com  2005-08-01 00:54 -------
.ii file available from here, 1.1 mb uncompressed... 19575 lines...
get it here: http://pomac.netswarm.net/misc/GBA.ii.gz

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23168

[Bug c++/23168] [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|critical                    |normal


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23168

[Bug rtl-optimization/23168] [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-01 01:39 -------
At least on ppc-darwin, the scheduler takes most of the time and most of the memory usage.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|c++                         |rtl-optimization
           Keywords|                            |compile-time-hog


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23168

[Bug rtl-optimization/23168] [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-01 01:53 -------
we are trying to allocate 1.5GB on ppc-darwin, this is just wrong  This is on the mainline.  I have not 
looked where we were allocating it yet.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23168

[Bug rtl-optimization/23168] [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-01 02:11 -------
#0  xmalloc_failed (size=1859158144) at /Users/pinskia/src/local3/gcc/libiberty/xmalloc.c:123
#1  0x0042762c in xcalloc (nelem=232394768, elsize=8) at /Users/pinskia/src/local3/gcc/libiberty/
xmalloc.c:164
#2  0x0039b99c in global_alloc (file=0x0) at /Users/pinskia/src/local3/gcc/gcc/global.c:526
#3  0x0039c6dc in rest_of_handle_global_alloc () at /Users/pinskia/src/local3/gcc/gcc/global.c:2490
#4  0x001b6438 in execute_one_pass (pass=0x8) at /Users/pinskia/src/local3/gcc/gcc/passes.c:787


hmm, oh, why.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
  GCC build triplet|amd64-linux-gnu             |
   GCC host triplet|amd64-linux-gnu             |
 GCC target triplet|amd64-linux-gnu             |
   Last reconfirmed|0000-00-00 00:00:00         |2005-08-01 02:11:54
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23168

[Bug rtl-optimization/23168] [OOM] GCC 4.0.1 OOM's while compiling VBA (works with 3.4.x)

[rguenth at gcc dot gnu dot org]

------- Additional Comments From rguenth at gcc dot gnu dot org  2005-08-01 09:10 -------
Well, nelem=232394768 is surely a bug in gcc (which I cannot reproduce on i686 btw.)

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23168