On Sun, 5 Dec 2004, Peter Arremann wrote:
> Thanks,
>
> so is not just me... The question is, is this a OO or a fedora problem?
> I remember reading somewhere that the openoffice.org version of OO uses ispell
> while the OO version from Fedora has been modified to use aspell. Is that
> correct? Just wanting to make sure that I report the bug at the right
> place...
Strace gives some stuff (when the memory usage jumps up)
******
open("/usr/lib/ooo-1.1/share/dict/ooo/en_AU.dic", O_RDONLY) = 31
fstat64(31, {st_mode=S_IFREG|0644, st_size=513822, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb0c08000
read(31, "45654\nAA\naardvark/SM\nAaron/M\naba"..., 4096) = 4096
mmap2(NULL, 733184, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xaac3d000
read(31, "ustic/SY\nacquaint/AGSD\nacquainta"..., 4096) = 4096
read(31, "fforest/nNGDS\naffray/MDGS\naffric"..., 4096) = 4096
read(31, "lison/M\nalliterate/vVNGSDn\nallit"..., 4096) = 4096
retc..
% grep dict strace.log
open("/home/balay/.rhopenoffice1.1/user/wordbook/dictionary.lst", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/lib/ooo-1.1/share/dict/ooo/dictionary.lst", O_RDONLY) = 31
open("/home/balay/.rhopenoffice1.1/user/wordbook/dictionary.lst", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/lib/ooo-1.1/share/dict/ooo/dictionary.lst", O_RDONLY) = 31
open("/home/balay/.rhopenoffice1.1/user/wordbook/dictionary.lst", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/lib/ooo-1.1/share/dict/ooo/dictionary.lst", O_RDONLY) = 31
open("/usr/lib/ooo-1.1/share/dict/ooo/en_US.dic", O_RDONLY) = 31
open("/usr/lib/ooo-1.1/share/dict/ooo/en_US.aff", O_RDONLY) = 31
open("/usr/lib/ooo-1.1/share/dict/ooo/en_US.dic", O_RDONLY) = 31
open("/usr/lib/ooo-1.1/share/dict/ooo/en_US.aff", O_RDONLY) = 31
open("/usr/lib/ooo-1.1/share/dict/ooo/bg_BG.dic", O_RDONLY) = 31
open("/usr/lib/ooo-1.1/share/dict/ooo/bg_BG.aff", O_RDONLY) = 31
open("/usr/lib/ooo-1.1/share/dict/ooo/bg_BG.dic", O_RDONLY) = 31
o
etc..
***
So OO is reading in all the dictionary files - which is consuming all
the memory. Maybe this operation needs to be done more efficiently..
Satish
