|
From: Susan in C. <sma...@uc...> - 2006-12-03 01:50:13
|
Hello! I working with an application that uses a HUGE amount of memory... but I am running on really fast machines that should be able to handle it. However, when I run valgrind, I see the following: margulie@fuzzy:~/gc_heuristic_11_18_06$ valgrind --tool=memcheck --leak- check=yes ./gc_heur l 0 3 k 9 2 4 .05 p <snip> ==29954== Warning: set address range perms: large range 748127592, a 0, v 1 Killed margulie@fuzzy:~/gc_heuristic_11_18_06$ When I run it under gdb (just as a point of information), I see the following: Program terminated with signal SIGKILL, Killed. The program no longer exists. (gdb) bt No stack. (gdb) Can anyone help? Since I know I am working with a lot of memory, is it possible that I am running into some system limitation? How would I check that? When I run my program under valgrind on smaller examples, it works perfectly and reports no errors or leaks of any kind. It's only as I increase the size of the examples, that it seems to become a problem. Thank you *very* much in advance for any comments! I have a screen shot of the true output, but I couldn't see how to attach it here (which is why I snipped) Best, Susan |
|
From: Julian S. <js...@ac...> - 2006-12-03 02:03:51
|
Sounds like your program was killed by the kernel's out-of-memory process-killer. Presumably shortly after you allocated an array of 748 M bytes. Are you sure the machine has enough available swap space to run your app? J On Sunday 03 December 2006 01:46, Susan in California wrote: > Hello! I working with an application that uses a HUGE amount of memory... > but I am running on really fast machines that should be able to handle it. > However, when I run valgrind, I see the following: > > margulie@fuzzy:~/gc_heuristic_11_18_06$ valgrind --tool=memcheck --leak- > check=yes ./gc_heur l 0 3 k 9 2 4 .05 p > <snip> > ==29954== Warning: set address range perms: large range 748127592, a 0, v 1 > Killed > margulie@fuzzy:~/gc_heuristic_11_18_06$ > > When I run it under gdb (just as a point of information), I see the > following: > > Program terminated with signal SIGKILL, Killed. > The program no longer exists. > (gdb) bt > No stack. > (gdb) > > Can anyone help? Since I know I am working with a lot of memory, is it > possible that I am running into some system limitation? How would I check > that? > > When I run my program under valgrind on smaller examples, it works > perfectly and reports no errors or leaks of any kind. It's only as I > increase the size of the examples, that it seems to become a problem. > > Thank you *very* much in advance for any comments! I have a screen shot of > the true output, but I couldn't see how to attach it here (which is why I > snipped) > > Best, > Susan > > > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Valgrind-users mailing list > Val...@li... > https://lists.sourceforge.net/lists/listinfo/valgrind-users |
|
From: Nicholas N. <nj...@cs...> - 2006-12-03 11:00:34
|
On Sun, 3 Dec 2006, Julian Seward wrote: > Sounds like your program was killed by the kernel's out-of-memory > process-killer. Presumably shortly after you allocated an array > of 748 M bytes. > > Are you sure the machine has enough available swap space to run your > app? Or maybe you hit a system ulimit? See "man ulimit" for details. Note that running your program under Valgrind tools it increases the memory usage a lot. Also, if you are running Memcheck and you are using a pre-3.2.0 version, upgrade to 3.2.0 as it is much more memory-efficient. Nick |
|
From: Julian S. <js...@ac...> - 2006-12-03 13:38:03
|
> On Sunday 03 December 2006 10:51, Nicholas Nethercote wrote: > > Are you sure the machine has enough available swap space to run your > > app? > > Or maybe you hit a system ulimit? See "man ulimit" for details. Come to think of it that's a more likely explaination than mine is. J |