[ECOS] Problem with HTTPD

[Manuel Borchers <manuel at matronix dot de>]

Hi all,

I got the network driver for the netX working quite fine. So I tried to
get something running with less effort to test the driver. I decided to
try CYGPKG_HTTPD, which implements to promising monitor.

When accessing http://myecosip/ I get the "Page not found" quite fine.
So I surfed to http://myecosip/monitor/index.html 

I had assertions and tracing on and was suddenly facing a corrupt stack!
So, first guess was, my stack could be too small. I raised
CYGNUM_HAL_COMMON_INTERRUPTS_STACK_SIZE to 32k (when this didn't help, I
even raised it to 128k) with no luck at all.

So, I went further down into the httpd thread and found out that for
some reason fprintf is called with a format string pointing to 0xa (my
SDRAM starts at 0x80000000, this area is for hardware vectors on ARM!)

Two #defines that are causing this are html_head and html_heading. There
seem to be even more, because fprintf is called even more with format
strings pointing to 0xa or 0x2 and the like. For debugging I just
returned from fprintf in these cases, which resulted in a somewhat
working index page in my browser.

Any hints why this is happening? I guess the stack is being corrupted,
because fprintf is called with 0xa or 0x2.

Thanks for any pointers what could be configured wrong here!


Cheers,
Manuel

-- 
Manuel Borchers

Web: http://www.matronix.de
eMail: manuel@matronix.de

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss