Sign in

username:

password:



Not a member?

Search hc11



Search tips

Subscribe to hc11



Ads

Discussion Groups

Discussion Groups | M68HC11 | SCI vector

Technical discussions about Freescale Microcontrollers: M68HC11. (Freescale Semiconductor is a Subsidiary of Motorola).

SCI vector - bal_gill21 - Feb 6 12:26:00 2005


Hi,
I am using the SCI to transmit from the micro to the PC via RS232.
Does the vector at FFD6 & FFD8 (SCI serial system & SPI serial
transfer complete) need to be initialised, I have not initialised it
on my development system and it works fine.

Thanks

Bal




Re: SCI vector


(You need to be a member of hc11 -- send a blank email to hc11-subscribe@yahoogroups.com )


Re: SCI vector - Bob Smith - Feb 6 13:17:00 2005

You are probably getting away with that because your development system has
already initialized the vector for you.

Your program will most likely fail when you try to run stand alone.

Bob

--- Avoid computer viruses, Practice safe hex ---

-- Specializing in small, cost effective
embedded control systems --

http://www.smithmachineworks.com/embedprod.html Robert L. (Bob) Smith
Smith Machine Works, Inc.
9900 Lumlay Road
Richmond, VA 23236 804/745-2608
----- Original Message -----
From: "bal_gill21" <>
To: <>
Sent: Sunday, February 06, 2005 11:26 AM
Subject: [m68HC11] SCI vector >
> Hi,
> I am using the SCI to transmit from the micro to the PC via RS232.
> Does the vector at FFD6 & FFD8 (SCI serial system & SPI serial
> transfer complete) need to be initialised, I have not initialised it
> on my development system and it works fine.
>
> Thanks
>
> Bal > --------------------------------------------------------------------------
------
> Yahoo! Groups Links
>
> a.. To



(You need to be a member of hc11 -- send a blank email to hc11-subscribe@yahoogroups.com )

Re: SCI vector - Mark Schultz - Feb 7 13:47:00 2005


--- In , "Bob Smith" <bobsmith5@v...> wrote:

> You are probably getting away with that because your development
> system has already initialized the vector for you.
>
> Your program will most likely fail when you try to run stand alone.

> ----- Original Message -----
> From: "bal_gill21" <bal_gill21@y... > I am using the SCI to transmit from the micro to the PC via RS232.
> Does the vector at FFD6 & FFD8 (SCI serial system & SPI serial
> transfer complete) need to be initialised, I have not initialised it
> on my development system and it works fine.

If the SCI is being used in strictly 'polled' mode, with SCI
interrupts disabled, it is not *mandatory* to initialize the SCI
vector, since the SCI interrupt (being disabled) will never be invoked.

I have always considered it 'good practice' to initialize ALL of the
interrupt and trap vectors supported by a given CPU. Interrupts that
you are not using should point to some sort of generic 'inimplemented
interrupt' routine. On my projects, this routine often consists of a
few instructions that "do something visible" that will let me know
that an unexpected interrupt has occured. This could be as simple as
asserting or toggling a otherwise unused port pin. After performing
the 'visible action', the routine ends with a branch instruction that
branches to itself, effectively stopping the program. I don't use
STOP or WAI because these instructions can be disabled or terminated
under certain conditions.

-- Mark





(You need to be a member of hc11 -- send a blank email to hc11-subscribe@yahoogroups.com )

Re: SCI vector - bal_gill21 - Feb 7 14:48:00 2005


I have sent what I wanted to over the RS232 and the correct length of
data has been sent but the data is meaningless. It could be (or is)
because I did not initialise the SCI. I would normally initialise it
in the normal manner ie give the appropriate location in code a label
(myLabel) and then initialise it at the end ie

ORG $ffd6 ;vector for SCI
FDB myLabel

The problem is that i'm unsure where to label it because there is no
place where the SCI actually starts. I have put some of my code
below, I hope its legible and makes sense. Thanks guys

Bal

circle LDAA PORTC ;Read PORT C, bit 4 will be high when user
has pressed button to download data
CMPA #%00000111
BHI trnsmit ;Ready to transmit data
JMP circle ;Signal not recieved to transmit from port
C, so keep looping
trnsmit LDAA #$30 ;Set bits SCP1 AND SCP0 to 1 in BAUD register
for selection of 9600
STAA BAUD ;$102b
LDAA #$0C ;Turn SCI on, TE=RE=1 (transmit & recieve) in
SCCR2 register
STAA SCCR2
LDX #DATASRT ;Initialise X register, start of first data
address
LDAA #$00 ;Set character length in the SCCR1,one START
bit,8 DATA bits & 1 STOP bit
STAA SCCR1

sndchr LDAA $00,x ;Load accumulator A with the next ASCII code
to be transmitted
LDAB SCSR ;Read SCSR register as first step in the
clearing sequence of the TDRE bit
STAA SCDR ;Send character to be transmitted into the
TDR and clear TDRE
INX ;Increment X
CPX #DATASTP ;Last memory location for ASCCI code
BEQ lstSrn ;If complete message transmitted, go to last
screen (code carries on)
chtdre LDAB SCSR ;Check if TDRE=1 to determine if the TDR is
ready for next character
BMI sndchr ;If tdre=1, transmit the next character
JMP chtdre ;If tdre !=1, check TDRE again > > You are probably getting away with that because your development
> > system has already initialized the vector for you.
> >
> > Your program will most likely fail when you try to run stand
alone.
>
> > ----- Original Message -----
> > From: "bal_gill21" <bal_gill21@y...>
> >
> >
> > I am using the SCI to transmit from the micro to the PC via RS232.
> > Does the vector at FFD6 & FFD8 (SCI serial system & SPI serial
> > transfer complete) need to be initialised, I have not initialised
it
> > on my development system and it works fine.
>
> If the SCI is being used in strictly 'polled' mode, with SCI
> interrupts disabled, it is not *mandatory* to initialize the SCI
> vector, since the SCI interrupt (being disabled) will never be
invoked.
>
> I have always considered it 'good practice' to initialize ALL of the
> interrupt and trap vectors supported by a given CPU. Interrupts
that
> you are not using should point to some sort of
generic 'inimplemented
> interrupt' routine. On my projects, this routine often consists of
a
> few instructions that "do something visible" that will let me know
> that an unexpected interrupt has occured. This could be as simple
as
> asserting or toggling a otherwise unused port pin. After performing
> the 'visible action', the routine ends with a branch instruction
that
> branches to itself, effectively stopping the program. I don't use
> STOP or WAI because these instructions can be disabled or terminated
> under certain conditions.
>
> -- Mark





(You need to be a member of hc11 -- send a blank email to hc11-subscribe@yahoogroups.com )

Re: Re: SCI vector - Tony Papadimitriou - Feb 7 15:53:00 2005

----- Original Message -----
From: "bal_gill21" <>
To: < > ORG $ffd6 ;vector for SCI
> FDB myLabel
>
> The problem is that i'm unsure where to label it because there is no
> place where the SCI actually starts. I have put some of my code
> below, I hope its legible and makes sense. Thanks guys

You need to organize your code differently to use interrupts for the SCI. (You
actually need to organize your code, regardless.)

When using interrupts, in effect you're multitasking. Your main code is one
task, the Interrupt Service Routine (ISR) is another. The two run in parallel
without interfering with each other.

Let's say you use interrupts only for RX, where they are mostly needed. (For
now, you'll still do TX the simple way by calling the regular 'PutChar'
routine.)

For RX, you need to have a routine (your ISR) that will only do a very simply
job: receive a character and place it in a queue. It shouldn't worry about
anything else. It shouldn't even bother to see if anyone is interested for the
received character at the moment. This will allow your program to receive
characters asynchronously from the main program's demands for them.

In your main program, every time you come to process another character from the
SCI, you'll call a routine to de-queue the next available character (so your
regular 'GetChar' routine will de-queue from the RX queue instead of polling
SCSR to see if a character has arrived).

Look inside http://www.aspisys.com/asm11d84.zip for the file
./SAMPLES/FRAME/OS11/SCI_INT.MOD

You can compare the interrupt-driven version with an equivalent polled-mode
version (./SAMPLES/FRAME/OS11/SCI.MOD).

Hope this helps.

> Bal





(You need to be a member of hc11 -- send a blank email to hc11-subscribe@yahoogroups.com )

Re: SCI vector - bal_gill21 - Feb 8 14:47:00 2005


My SCI is not interupt driven. I was using the ISR to log data every
15 minutes for 24 hours. The ISR is then turned off and the system
waits for an input on PORTC. When the signal is recieved on PORTC all
of the logged data is sent through the MAX232 chip and to the PC
using the SCI all in one go. Data is sent to the PC ie the right
length, but it does not display the numbers it is supposed to.

Bal

--- In , "Tony Papadimitriou" <tonyp@m...>
wrote:
> ----- Original Message -----
> From: "bal_gill21" <bal_gill21@y...>
> To: < > > ORG $ffd6 ;vector for SCI
> > FDB myLabel
> >
> > The problem is that i'm unsure where to label it because there is
no
> > place where the SCI actually starts. I have put some of my code
> > below, I hope its legible and makes sense. Thanks guys
>
> You need to organize your code differently to use interrupts for
the SCI. (You
> actually need to organize your code, regardless.)
>
> When using interrupts, in effect you're multitasking. Your main
code is one
> task, the Interrupt Service Routine (ISR) is another. The two run
in parallel
> without interfering with each other.
>
> Let's say you use interrupts only for RX, where they are mostly
needed. (For
> now, you'll still do TX the simple way by calling the
regular 'PutChar'
> routine.)
>
> For RX, you need to have a routine (your ISR) that will only do a
very simply
> job: receive a character and place it in a queue. It shouldn't
worry about
> anything else. It shouldn't even bother to see if anyone is
interested for the
> received character at the moment. This will allow your program to
receive
> characters asynchronously from the main program's demands for them.
>
> In your main program, every time you come to process another
character from the
> SCI, you'll call a routine to de-queue the next available character
(so your
> regular 'GetChar' routine will de-queue from the RX queue instead
of polling
> SCSR to see if a character has arrived).
>
> Look inside http://www.aspisys.com/asm11d84.zip for the file
> ./SAMPLES/FRAME/OS11/SCI_INT.MOD
>
> You can compare the interrupt-driven version with an equivalent
polled-mode
> version (./SAMPLES/FRAME/OS11/SCI.MOD).
>
> Hope this helps.
>
> > Bal
>
> tonyp@a...





(You need to be a member of hc11 -- send a blank email to hc11-subscribe@yahoogroups.com )

Re: Re: SCI vector - bart homerson - Feb 9 1:44:00 2005

Bal,
You may need to recheck the settings for the Com port on your computer. Also, What program are you using to display the data? If you are not already doing so, then use Hyperteminal, or the equivalent, in order to see what is really coming across the Com Port. If you are certain that the Com port settings are correct for Hyperterm, then you will be able to see exactly what the MCU is sending to the computer, however, if the settings are incorrect, then you may see a bunch of gobldygook.

For a test, write a very simple program that sends perhaps a word to the computer with the same settings in your MCU code that you have now and see if this turns up correct. That's about all I can think of until we know how you are displaying the data on the Computer.

LF

bal_gill21 <> wrote:

My SCI is not interupt driven. I was using the ISR to log data every
15 minutes for 24 hours. The ISR is then turned off and the system
waits for an input on PORTC. When the signal is recieved on PORTC all
of the logged data is sent through the MAX232 chip and to the PC
using the SCI all in one go. Data is sent to the PC ie the right
length, but it does not display the numbers it is supposed to.

Bal

--- In , "Tony Papadimitriou" <tonyp@m...>
wrote:
> ----- Original Message -----
> From: "bal_gill21" <bal_gill21@y...>
> To: < > > ORG $ffd6 ;vector for SCI
> > FDB myLabel
> >
> > The problem is that i'm unsure where to label it because there is
no
> > place where the SCI actually starts. I have put some of my code
> > below, I hope its legible and makes sense. Thanks guys
>
> You need to organize your code differently to use interrupts for
the SCI. (You
> actually need to organize your code, regardless.)
>
> When using interrupts, in effect you're multitasking. Your main
code is one
> task, the Interrupt Service Routine (ISR) is another. The two run
in parallel
> without interfering with each other.
>
> Let's say you use interrupts only for RX, where they are mostly
needed. (For
> now, you'll still do TX the simple way by calling the
regular 'PutChar'
> routine.)
>
> For RX, you need to have a routine (your ISR) that will only do a
very simply
> job: receive a character and place it in a queue. It shouldn't
worry about
> anything else. It shouldn't even bother to see if anyone is
interested for the
> received character at the moment. This will allow your program to
receive
> characters asynchronously from the main program's demands for them.
>
> In your main program, every time you come to process another
character from the
> SCI, you'll call a routine to de-queue the next available character
(so your
> regular 'GetChar' routine will de-queue from the RX queue instead
of polling
> SCSR to see if a character has arrived).
>
> Look inside http://www.aspisys.com/asm11d84.zip for the file
> ./SAMPLES/FRAME/OS11/SCI_INT.MOD
>
> You can compare the interrupt-driven version with an equivalent
polled-mode
> version (./SAMPLES/FRAME/OS11/SCI.MOD).
>
> Hope this helps.
>
> > Bal
>
> tonyp@a...

Yahoo! Groups SponsorADVERTISEMENT
document.write('');

---------------------------------
Yahoo! Groups Links

To
---------------------------------




(You need to be a member of hc11 -- send a blank email to hc11-subscribe@yahoogroups.com )

Re: SCI vector - bal_gill21 - Feb 9 13:00:00 2005


Yes, I am actually using hyperterminal. The settings are exactly the
way they should be because I have sent data to it using the
development system and its fine. Using the actual hardware shows the
data in a triagle shape in hyperterminal but the actual length of the
data is being transmitted correctly.

Bal

--- In , bart homerson <odddooo1@y...> wrote:
> Bal,
> You may need to recheck the settings for the Com port on your
computer. Also, What program are you using to display the data? If
you are not already doing so, then use Hyperteminal, or the
equivalent, in order to see what is really coming across the Com
Port. If you are certain that the Com port settings are correct for
Hyperterm, then you will be able to see exactly what the MCU is
sending to the computer, however, if the settings are incorrect, then
you may see a bunch of gobldygook.
>
> For a test, write a very simple program that sends perhaps a word
to the computer with the same settings in your MCU code that you have
now and see if this turns up correct. That's about all I can think
of until we know how you are displaying the data on the Computer.
>
> LF
>
> bal_gill21 <bal_gill21@y...> wrote:
>
> My SCI is not interupt driven. I was using the ISR to log data
every
> 15 minutes for 24 hours. The ISR is then turned off and the system
> waits for an input on PORTC. When the signal is recieved on PORTC
all
> of the logged data is sent through the MAX232 chip and to the PC
> using the SCI all in one go. Data is sent to the PC ie the right
> length, but it does not display the numbers it is supposed to.
>
> Bal
>
> --- In , "Tony Papadimitriou" <tonyp@m...>
> wrote:
> > ----- Original Message -----
> > From: "bal_gill21" <bal_gill21@y...>
> > To: <>
> >
> >
> > > ORG $ffd6 ;vector for SCI
> > > FDB myLabel
> > >
> > > The problem is that i'm unsure where to label it because there
is
> no
> > > place where the SCI actually starts. I have put some of my code
> > > below, I hope its legible and makes sense. Thanks guys
> >
> > You need to organize your code differently to use interrupts for
> the SCI. (You
> > actually need to organize your code, regardless.)
> >
> > When using interrupts, in effect you're multitasking. Your main
> code is one
> > task, the Interrupt Service Routine (ISR) is another. The two
run
> in parallel
> > without interfering with each other.
> >
> > Let's say you use interrupts only for RX, where they are mostly
> needed. (For
> > now, you'll still do TX the simple way by calling the
> regular 'PutChar'
> > routine.)
> >
> > For RX, you need to have a routine (your ISR) that will only do a
> very simply
> > job: receive a character and place it in a queue. It shouldn't
> worry about
> > anything else. It shouldn't even bother to see if anyone is
> interested for the
> > received character at the moment. This will allow your program
to
> receive
> > characters asynchronously from the main program's demands for
them.
> >
> > In your main program, every time you come to process another
> character from the
> > SCI, you'll call a routine to de-queue the next available
character
> (so your
> > regular 'GetChar' routine will de-queue from the RX queue instead
> of polling
> > SCSR to see if a character has arrived).
> >
> > Look inside http://www.aspisys.com/asm11d84.zip for the file
> > ./SAMPLES/FRAME/OS11/SCI_INT.MOD
> >
> > You can compare the interrupt-driven version with an equivalent
> polled-mode
> > version (./SAMPLES/FRAME/OS11/SCI.MOD).
> >
> > Hope this helps.
> >
> > > Bal
> >
> > tonyp@a... >
>
> Yahoo! Groups SponsorADVERTISEMENT
> document.write('');
>
> ---------------------------------
> Yahoo! Groups Links
>
> To >
> ---------------------------------



(You need to be a member of hc11 -- send a blank email to hc11-subscribe@yahoogroups.com )