This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
Re: tm_basic (memory) problem
- From: Jonathan Larmour <jlarmour at redhat dot com>
- To: Andrea Acquaviva <aacquaviva at deis dot unibo dot it>
- Cc: "ecos-discuss at sources dot redhat dot com" <ecos-discuss at sources dot redhat dot com>
- Date: Fri, 01 Mar 2002 18:32:00 +0000
- Subject: Re: [ECOS]tm_basic (memory) problem
- Organization: Red Hat UK Ltd.
- References: <3C7C1E36.CB04910A@deis.unibo.it>
Andrea Acquaviva wrote:
>
> Hi all,
> I run into problems while running tm_basic tests. The tests hangs as
> soon as it reaches the assignment of nsamples in tm_basic_main. The
> other tests don't have problems. I found that the dimension of the
> executable is quite strange:
>
> text data bss dec
> hex filename
> 26924 5124 23368684 23400732 165111c
> install/tests/kernel/current/tests/tm_basic
>
> I have 64MB of sdram installed on my board, so I don't think that this
> should be a problem. However, the strange thing is that the dimension is
> so great even if I comment all the code starting from:
> new_thread(run_all_tests, 0).
>
> Summarizing, what I observed is that if I do a reference to a static
> defined variable like nsamples or nmutexes the size of the executable
> increases a lot and the program hangs.
tm_basic has some funky stuff in it that chooses the sizes on the basis of
the available memory. Perhaps given the amount of memory it chooses
something that is too much for the CPU to deal with, so it slows right
down?
Certainly, look at tweaking the values it uses by default (look for the
reference to CYGMEM_REGION_ram_SIZE).
Jifl
--
Red Hat, Rustat House, Clifton Road, Cambridge, UK. Tel: +44 (1223) 271062
Maybe this world is another planet's Hell -Aldous Huxley || Opinions==mine
--
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss