[Gmsh] Out of memory

[Mark van Doesburg]

Hello Christophe,

The pos file was generated with getdp version 2.0.0-cvs-20090525.
Note that it's a large simulation with 3.8e6 DoF (complex) and at nine
frequencies. The error occurs while reading the .pos file.

Here's the backtrace (note this is of version gmsh-2.3.1-cvs-20090512)

Breakpoint 6, Realloc (ptr=0x0, size=<value optimized out>)
    at MallocUtils.cpp:42
42          Msg::Fatal("Out of memory (buy some more RAM!)");
(gdb) bt
#0  Realloc (ptr=0x0, size=<value optimized out>) at MallocUtils.cpp:42
#1  0x00000000004b6f65 in List_Realloc (liste=0xf8eb00,
    n=<value optimized out>) at ListUtils.cpp:64
#2  0x00000000004b7102 in List_Add (liste=0x7f63340008b0, data=0x7fff4ef30f28)
    at ListUtils.cpp:72
#3  0x00000000006769e6 in gmsh_yyparse ()
#4  0x00000000004b1495 in ParseFile (fileName=
        {static npos = 18446744073709551615, _M_dataplus = {<std::allocator<char>> = {<__gnu_cxx::new_allocator<char>> = {<No data fields>}, <No data fields>}, _M_p = 0x7fff4ef34550 "8QÍ"}}, close=true, warnIfMissing=<value optimized out>)
    at OpenFile.cpp:174
#5  0x0000000000568670 in GModel::readGEO (this=0xccf550,
    name=<value optimized out>) at GModelIO_Geo.cpp:39
#6  0x00000000004b2c60 in MergeFile (fileName=
        {static npos = 18446744073709551615, _M_dataplus = {<std::allocator<char>> = {<__gnu_cxx::new_allocator<char>> = {<No data fields>}, <No data fields>}, _M_p = 0x7fff4ef34860 "8QÍ"}}, warnIfMissing=<value optimized out>)
    at OpenFile.cpp:351
#7  0x00000000004b2db9 in OpenProject (fileName=
        {static npos = 18446744073709551615, _M_dataplus = {<std::allocator<char>> = {<__gnu_cxx::new_allocator<char>> = {<No data fields>}, <No data fields>}, _M_p = 0x7fff4ef348d0 "8QÍ"}}) at OpenFile.cpp:427
#8  0x000000000041b0db in main (argc=2, argv=0x7fff4ef34a48) at Main.cpp:76

Mark.


Christophe Geuzaine <cgeuzaine at ulg.ac.be> wrote:

	Mark van Doesburg wrote:
	> Hello,
	> 
	> Gmsh was complaining that I should buy more RAM... that seemed highly
	> unlikely and it turns out it was lying ;-)

	Hi Mark - Can you tell us where such large old-style lists were 
	generated in the code? (Are you using the old post-pro format?)



	> 
	> Here's a patch.
	> 
	> regards,
	> 
	> Mark van Doesburg
	> 
	> 
	> ------------------------------------------------------------------------
	> 
	> _______________________________________________
	> gmsh mailing list
	> gmsh at geuz.org
	> http://www.geuz.org/mailman/listinfo/gmsh


	-- 
	Prof. Christophe Geuzaine
	University of Liege, Electrical Engineering and Computer Science
	http://www.montefiore.ulg.ac.be/~geuzaine