public inbox for ecos-discuss@sourceware.org
 help / color / mirror / Atom feed
* [ECOS] virtual vectors problem
@ 2005-09-07 13:26 Stephane Royo
  2005-09-07 13:47 ` Gary Thomas
  0 siblings, 1 reply; 3+ messages in thread
From: Stephane Royo @ 2005-09-07 13:26 UTC (permalink / raw)
  To: ecos-discuss

Hello,

I updated eCos via CVS, then my application two months ago. Since, i noticed
that the comm_channels variable -used to contain the functions pointer on
serial communication channel- is no more in data section but it is in bss
section.

So, when my application is configurated to works in RAM in order to debug
it, i have problem. Indeed, in this configuration i use the option "Inherit
console settings from ROM monitor", but unfortunately the comm_channels
variable is put to zero during the bss initialization consequently, all
subsequent accesses to comm_channels variable cause a hardware reset.

I didn't find anything in eCos sources or in my configuration (ecc file)
which explains this change of location, but to solve the problem i
initialized the comm_channels at the time of declaration : static
hal_virtual_comm_table_t
comm_channels[CYGNUM_HAL_VIRTUAL_VECTOR_NUM_CHANNELS+1]={0};

Anybody, could help me to understand the problem ?

Thank you,
Stephane


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

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [ECOS] virtual vectors problem
  2005-09-07 13:26 [ECOS] virtual vectors problem Stephane Royo
@ 2005-09-07 13:47 ` Gary Thomas
  2005-09-07 14:10   ` Stephane Royo
  0 siblings, 1 reply; 3+ messages in thread
From: Gary Thomas @ 2005-09-07 13:47 UTC (permalink / raw)
  To: Stephane Royo; +Cc: ecos-discuss

On Wed, 2005-09-07 at 15:25 +0200, Stephane Royo wrote:
> Hello,
> 
> I updated eCos via CVS, then my application two months ago. Since, i noticed
> that the comm_channels variable -used to contain the functions pointer on
> serial communication channel- is no more in data section but it is in bss
> section.
> 
> So, when my application is configurated to works in RAM in order to debug
> it, i have problem. Indeed, in this configuration i use the option "Inherit
> console settings from ROM monitor", but unfortunately the comm_channels
> variable is put to zero during the bss initialization consequently, all
> subsequent accesses to comm_channels variable cause a hardware reset.
> 
> I didn't find anything in eCos sources or in my configuration (ecc file)
> which explains this change of location, but to solve the problem i
> initialized the comm_channels at the time of declaration : static
> hal_virtual_comm_table_t
> comm_channels[CYGNUM_HAL_VIRTUAL_VECTOR_NUM_CHANNELS+1]={0};
> 
> Anybody, could help me to understand the problem ?

We'd need to know what platform/target you are using.  This stuff
works (has worked for many years), so it's probably something "special"
in your target support.

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------


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

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [ECOS] virtual vectors problem
  2005-09-07 13:47 ` Gary Thomas
@ 2005-09-07 14:10   ` Stephane Royo
  0 siblings, 0 replies; 3+ messages in thread
From: Stephane Royo @ 2005-09-07 14:10 UTC (permalink / raw)
  To: Gary Thomas; +Cc: ecos-discuss

I use a specific platform based on ARM7TDMI on which i ported eCos.

Stephane

----- Original Message -----
From: "Gary Thomas" <gary@mlbassoc.com>
To: "Stephane Royo" <stephane.royo@europe-technologies.com>
Cc: <ecos-discuss@sources.redhat.com>
Sent: Wednesday, September 07, 2005 3:47 PM
Subject: Re: [ECOS] virtual vectors problem


> On Wed, 2005-09-07 at 15:25 +0200, Stephane Royo wrote:
> > Hello,
> >
> > I updated eCos via CVS, then my application two months ago. Since, i
noticed
> > that the comm_channels variable -used to contain the functions pointer
on
> > serial communication channel- is no more in data section but it is in
bss
> > section.
> >
> > So, when my application is configurated to works in RAM in order to
debug
> > it, i have problem. Indeed, in this configuration i use the option
"Inherit
> > console settings from ROM monitor", but unfortunately the comm_channels
> > variable is put to zero during the bss initialization consequently, all
> > subsequent accesses to comm_channels variable cause a hardware reset.
> >
> > I didn't find anything in eCos sources or in my configuration (ecc file)
> > which explains this change of location, but to solve the problem i
> > initialized the comm_channels at the time of declaration : static
> > hal_virtual_comm_table_t
> > comm_channels[CYGNUM_HAL_VIRTUAL_VECTOR_NUM_CHANNELS+1]={0};
> >
> > Anybody, could help me to understand the problem ?
>
> We'd need to know what platform/target you are using.  This stuff
> works (has worked for many years), so it's probably something "special"
> in your target support.
>
> --
> ------------------------------------------------------------
> Gary Thomas                 |  Consulting for the
> MLB Associates              |    Embedded world
> ------------------------------------------------------------


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

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2005-09-07 14:10 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-09-07 13:26 [ECOS] virtual vectors problem Stephane Royo
2005-09-07 13:47 ` Gary Thomas
2005-09-07 14:10   ` Stephane Royo

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).