This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
RE: Clock problem with eCos 1.3.1 using uITRON compatibility
- From: jecer at Rational dot Com
- To: jlarmour at redhat dot com
- Cc: ecos-discuss at sources dot redhat dot com
- Date: Tue, 29 Jan 2002 12:03:57 -0500
- Subject: RE: [ECOS] Clock problem with eCos 1.3.1 using uITRON compatibility
Hello,
I did try standerd eCos tests. The clock0 and clockcnv tests in the kernel
package complete fine (although the latter one complains about several
overflows), but the clock1 and tm_basic tests hang. I suspect that the
tm_basic test hangs in the call to cyg_thread_delay(2). The results are the
same when I run with eCos without uITRON compatibility - in other words the
default configuration. The only thing that I modified is the memory layout.
This was necessary to make eCos 1.3.1 to work with the redboot downloaded in
binary format from Red Hat. I haven't seen a mention of a known problem with
these symptoms. Is there such a problem or am I doing something wrong?
I also tried the eCos from anonymous CVS. I got a snapshot yesterday
(January 28, 2002). A compiled program would not produce any ouput at all (i
didn't see anythning after the "continue" command to gdb). I even tried
changing the memory mapping to one that worked with eCos 1.3.1, but the
results were the same. Just to be safe, I wanted to try the new eCos with
redboot built from the same source tree. However, the resulting redboot.bin
didn't boot at all. Is the current eCos fundementaly broken for the i386
target?
thank you
Jiri
-----Original Message-----
From: Jonathan Larmour [mailto:jlarmour@redhat.com]
Sent: January 25, 2002 3:56 PM
To: jecer@Rational.Com
Cc: ecos-discuss@sources.redhat.com
Subject: Re: [ECOS] Clock problem with eCos 1.3.1 using uITRON
compatibility
jecer@rational.com wrote:
>
> Thanks for your help. I tried your suggestion (incrementing a global
> variable in Cyg_RealTimeClock::isr() and Cyg_RealTimeClock::dsr() ) and
> determined that the two functions are not being called at all. That
> explainst why get_tim always returns 0.
>
> Is there anything that needs to be done explicitly to enable the clock
> interrupt?
It should automatically be enabled when the kernel starts. Have you tried
some of the "standard" kernel tests (in install/tests/kernel)?
But I'll mention anonymous CVS again :-).
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