Lpc2000

--- In lpc2000@yahoogroups.com, "gerhard_uttenthaler"
<gerhard.uttenthaler@g...> wrote:
> --- In lpc2000@yahoogroups.com, "genie_23432" <pascal_poirier@h...>
> wrote:
> > --- In lpc2000@yahoogroups.com, "gerhard_uttenthaler" 
> > <gerhard.uttenthaler@g...> wrote:
> > > Hello!
> > > 
> > > I am working on a project using the LPC2119.
> > > 
> > > My main loop observes, whether a message has been received by
> > > the 
> > > CAN
> > > interrupt and if so, sends it on the other CAN channel.
> > > 
> > > Now, if i put a:
> > > 
> > > // wait for free UART
> > > while( (UART1_LSR & 0x60) == 0x00){;}
> > > // send a character
> > > UART1_THR = 0x32;
> > > 
> > > into this main loop and have a heavy CAN load on one channel the
> > > programm gets locked up after some period of time and the
> > > (external)
> > > watchdog triggers a reboot.
> > > 
> > > Without the UART accesses the programm works just fine.
> > > 
> > > Does anyone has any idea about that?
> > > 
> > > Thank you!
> > > 
> > > Gerhard
> > 
> > I am having an almost identical type of problem using the
> > LPC2119.  
> > Its in previous posts and I have since been told by Phillips that
> > if 
> > I was using IAR to contact IAR for this issue.  Not much in the
> > way 
> > of details other then to the finger to the IAR compiler/debugger.
> > 
> > I am waiting to hear from IAR now but its a big issue since I
> > cannot 
> > seem to predict this erratic behavior.  Are you currently using
> > IAR?  
> > Might help rule out the compiler or not...
> > 
> > PS If you do find the reason I wouldn't mind knowing since I am
> > going 
> > around in circles with this issue myself.
> 
> No, I am not using IAR, but the gnuarm toolchain (arm-elf-gcc
> version
> 3.3.3).
> 
> If I find out something helpful I will post it here.

Sincw I declared my buffer variables, where the received messages are
stored in the CAN interrupt routine as volatile and put a
disableIRQ/restoreIRQ around the buffer decrement in my CAN send
routine the reboots have gone away.

Sorry foe the noise.