Sign in

username:

password:



Not a member?

Search hc11



Search tips

Subscribe to hc11



Ads

Discussion Groups

Discussion Groups | M68HC11 | Code Auto Execution problem

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

Code Auto Execution problem - Israel Brewster - Nov 28 20:11:00 2004


I am having some difficulty getting my code to execute with BUFFALO disabled. I am using an
Adapt11C24DXEVB, and have my code loaded into the internal EEPROM starting at address
B600. If I power up the board with BUFFALO enabled, and issue the go B600 command,
everything works perfectly. However, if I then pull E0 high to disable buffalo, and reset the
board, my program only runs until it tries to do an A/D conversion on port E7, where it hangs
waiting for the conversion complete flag to be set.

I am using the branch on bit clear command to loop back to itself until the flag is set, and am
certain this command is correct both because I copied it from some code I have used before,
and because, as I mentioned already, it DOES run properly when called from BUFFALO. If I
replace the branch command with a delay loop that should be more than long enough for the
conversions to complete (Y index register is decremented from $0700) program execution
continues successfully past this point- except that the A/D conversion is never completed,
which obviously causes the branches based on the result of the A/D conversion to branch
improperly. Specifically, it always takes the branch for when the value is less than $41, while
the actual value should be (and is when run from BUFFALO) close to or equal to $FF.

I have already checked that I am setting the stack pointer ($CFFF) and enabling the interrupts
(CLI), with no change. Can anyone tell me what I am missing here? Thanks.






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


Re: Code Auto Execution problem - Jaap van Ganswijk - Nov 29 5:04:00 2004

At 2004-11-29 02:11, Israel Brewster wrote:
>I am having some difficulty getting my code to execute with BUFFALO disabled. I am using an
>Adapt11C24DXEVB, and have my code loaded into the internal EEPROM starting at address
>B600. If I power up the board with BUFFALO enabled, and issue the go B600 command,
>everything works perfectly. However, if I then pull E0 high to disable buffalo, and reset the
>board, my program only runs until it tries to do an A/D conversion on port E7, where it hangs
>waiting for the conversion complete flag to be set.
>
>I am using the branch on bit clear command to loop back to itself until the flag is set, and am
>certain this command is correct both because I copied it from some code I have used before,
>and because, as I mentioned already, it DOES run properly when called from BUFFALO. If I
>replace the branch command with a delay loop that should be more than long enough for the
>conversions to complete (Y index register is decremented from $0700) program execution
>continues successfully past this point- except that the A/D conversion is never completed,
>which obviously causes the branches based on the result of the A/D conversion to branch
>improperly. Specifically, it always takes the branch for when the value is less than $41, while
>the actual value should be (and is when run from BUFFALO) close to or equal to $FF.
>
>I have already checked that I am setting the stack pointer ($CFFF) and enabling the interrupts
>(CLI), with no change. Can anyone tell me what I am missing here? Thanks.

Beware that the bitwise instructions were added to the 6811
instruction set by aliens from outer space and that they
tend to use unexpected read-modify-write-cycles to get
things done. That may upset registers in peripheral chips.

Tip from a red blooded earthling (and compiler writer):
Don't use the bit instructions!

Because some hardware idiot thought it was doable to add
a certain instruction, that doesn't mean it makes sense
to use it!

(Don't get me started on the 6812...) Greetings,
Jaap

-- Chip Directory
-- http://www.chipdir.biz/
-- http://www.chipdir.info/
-- http://www.chipdir.net/
-- http://www.chipdir.nl/
-- http://www.chipdir.org/
-- And about 30 other mirror sites world-wide.
--
-- To subscribe to a free 'chip issues, questions and answers'
-- mailing list, send a message to with
-- in the body 'subscribe chipdir-L'. About 500 experts are
-- willing to think with you about electronics problems etc.




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

Re: Code Auto Execution problem - Israel Brewster - Nov 29 10:56:00 2004


I have used bit instructions quite frequently in my programing and found them to be a
valuable tool, making my code much simpler and more compact - frequently doing in one
line what would otherwise take four or five lines at least. Be that as it may, the problem
here has nothing to do with the bit instruction- as I mentioned in my original post, I have
tried replacing it with a simple delay loop. The problem is that with BUFFALO disabled, the
A/D conversion never takes place. Thus the conversion complete flag is never set, and
thus the bit instruction loops infinitely waiting for it to be set. This is shown by the
behavior experienced when the bit instruction is replaced with a delay loop, as described
in my original post. Thanks anyway.

--- In , Jaap van Ganswijk <ganswijk@x> wrote:
> At 2004-11-29 02:11, Israel Brewster wrote:
> >I am having some difficulty getting my code to execute with BUFFALO disabled. I am
using an
> >Adapt11C24DXEVB, and have my code loaded into the internal EEPROM starting at
address
> >B600. If I power up the board with BUFFALO enabled, and issue the go B600 command,
> >everything works perfectly. However, if I then pull E0 high to disable buffalo, and reset
the
> >board, my program only runs until it tries to do an A/D conversion on port E7, where it
hangs
> >waiting for the conversion complete flag to be set.
> >
> >I am using the branch on bit clear command to loop back to itself until the flag is set,
and am
> >certain this command is correct both because I copied it from some code I have used
before,
> >and because, as I mentioned already, it DOES run properly when called from BUFFALO.
If I
> >replace the branch command with a delay loop that should be more than long enough
for the
> >conversions to complete (Y index register is decremented from $0700) program
execution
> >continues successfully past this point- except that the A/D conversion is never
completed,
> >which obviously causes the branches based on the result of the A/D conversion to
branch
> >improperly. Specifically, it always takes the branch for when the value is less than $41,
while
> >the actual value should be (and is when run from BUFFALO) close to or equal to $FF.
> >
> >I have already checked that I am setting the stack pointer ($CFFF) and enabling the
interrupts
> >(CLI), with no change. Can anyone tell me what I am missing here? Thanks.
>
> Beware that the bitwise instructions were added to the 6811
> instruction set by aliens from outer space and that they
> tend to use unexpected read-modify-write-cycles to get
> things done. That may upset registers in peripheral chips.
>
> Tip from a red blooded earthling (and compiler writer):
> Don't use the bit instructions!
>
> Because some hardware idiot thought it was doable to add
> a certain instruction, that doesn't mean it makes sense
> to use it!
>
> (Don't get me started on the 6812...) > Greetings,
> Jaap
>
> -- Chip Directory
> -- http://www.chipdir.biz/
> -- http://www.chipdir.info/
> -- http://www.chipdir.net/
> -- http://www.chipdir.nl/
> -- http://www.chipdir.org/
> -- And about 30 other mirror sites world-wide.
> --
> -- To subscribe to a free 'chip issues, questions and answers'
> -- mailing list, send a message to listguru@f... with
> -- in the body 'subscribe chipdir-L'. About 500 experts are
> -- willing to think with you about electronics problems etc.






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

Re: Re: Code Auto Execution problem - Scott Grodevant - Nov 29 11:17:00 2004

Hi all and Israel,

My memory fails me for specifics about the HC11 (without going to the manual), but I am pretty sure there is a configuration register associated with the ADC that must be initialized before using it. Without BUFFALO running to do it for you, the ADC is probably disabled.

Good luck,
Scott Israel Brewster <> wrote:

I have used bit instructions quite frequently in my programing and found them to be a
valuable tool, making my code much simpler and more compact - frequently doing in one
line what would otherwise take four or five lines at least. Be that as it may, the problem
here has nothing to do with the bit instruction- as I mentioned in my original post, I have
tried replacing it with a simple delay loop. The problem is that with BUFFALO disabled, the
A/D conversion never takes place. Thus the conversion complete flag is never set, and
thus the bit instruction loops infinitely waiting for it to be set. This is shown by the
behavior experienced when the bit instruction is replaced with a delay loop, as described
in my original post. Thanks anyway.

--- In , Jaap van Ganswijk <ganswijk@x> wrote:
> At 2004-11-29 02:11, Israel Brewster wrote:
> >I am having some difficulty getting my code to execute with BUFFALO disabled. I am
using an
> >Adapt11C24DXEVB, and have my code loaded into the internal EEPROM starting at
address
> >B600. If I power up the board with BUFFALO enabled, and issue the go B600 command,
> >everything works perfectly. However, if I then pull E0 high to disable buffalo, and reset
the
> >board, my program only runs until it tries to do an A/D conversion on port E7, where it
hangs
> >waiting for the conversion complete flag to be set.
> >
> >I am using the branch on bit clear command to loop back to itself until the flag is set,
and am
> >certain this command is correct both because I copied it from some code I have used
before,
> >and because, as I mentioned already, it DOES run properly when called from BUFFALO.
If I
> >replace the branch command with a delay loop that should be more than long enough
for the
> >conversions to complete (Y index register is decremented from $0700) program
execution
> >continues successfully past this point- except that the A/D conversion is never
completed,
> >which obviously causes the branches based on the result of the A/D conversion to
branch
> >improperly. Specifically, it always takes the branch for when the value is less than $41,
while
> >the actual value should be (and is when run from BUFFALO) close to or equal to $FF.
> >
> >I have already checked that I am setting the stack pointer ($CFFF) and enabling the
interrupts
> >(CLI), with no change. Can anyone tell me what I am missing here? Thanks.
>
> Beware that the bitwise instructions were added to the 6811
> instruction set by aliens from outer space and that they
> tend to use unexpected read-modify-write-cycles to get
> things done. That may upset registers in peripheral chips.
>
> Tip from a red blooded earthling (and compiler writer):
> Don't use the bit instructions!
>
> Because some hardware idiot thought it was doable to add
> a certain instruction, that doesn't mean it makes sense
> to use it!
>
> (Don't get me started on the 6812...) > Greetings,
> Jaap
>
> -- Chip Directory
> -- http://www.chipdir.biz/
> -- http://www.chipdir.info/
> -- http://www.chipdir.net/
> -- http://www.chipdir.nl/
> -- http://www.chipdir.org/
> -- And about 30 other mirror sites world-wide.
> --
> -- To subscribe to a free 'chip issues, questions and answers'
> -- mailing list, send a message to listguru@f... with
> -- in the body 'subscribe chipdir-L'. About 500 experts are
> -- willing to think with you about electronics problems etc.

Yahoo! Groups Sponsor
Get unlimited calls to

U.S./Canada ---------------------------------
Yahoo! Groups Links

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





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

RE: Code Auto Execution problem - Nick Reitter - Nov 29 12:04:00 2004

Hi, Bit 7, ADPU,in the Option register, $1039, must be set.
Keep having fun.
Nick Reitter
Personal home page http://www.reitter.net/nick/
Where we got o church http://www.shepherdwlsv.org/
Where I used to work before I retired http://web.alfredstate.edu/eet/ -----Original Message-----
From: Jaap van Ganswijk [mailto:]
Sent: Monday, November 29, 2004 5:05 AM
To:
Subject: Re: [m68HC11] Code Auto Execution problem
At 2004-11-29 02:11, Israel Brewster wrote:
>I am having some difficulty getting my code to execute with BUFFALO
disabled. I am using an
>Adapt11C24DXEVB, and have my code loaded into the internal EEPROM starting
at address
>B600. If I power up the board with BUFFALO enabled, and issue the go B600
command,
>everything works perfectly. However, if I then pull E0 high to disable
buffalo, and reset the
>board, my program only runs until it tries to do an A/D conversion on port
E7, where it hangs
>waiting for the conversion complete flag to be set.
>
>I am using the branch on bit clear command to loop back to itself until the
flag is set, and am
>certain this command is correct both because I copied it from some code I
have used before,
>and because, as I mentioned already, it DOES run properly when called from
BUFFALO. If I
>replace the branch command with a delay loop that should be more than long
enough for the
>conversions to complete (Y index register is decremented from $0700)
program execution
>continues successfully past this point- except that the A/D conversion is
never completed,
>which obviously causes the branches based on the result of the A/D
conversion to branch
>improperly. Specifically, it always takes the branch for when the value is
less than $41, while
>the actual value should be (and is when run from BUFFALO) close to or equal
to $FF.
>
>I have already checked that I am setting the stack pointer ($CFFF) and
enabling the interrupts
>(CLI), with no change. Can anyone tell me what I am missing here? Thanks.

Beware that the bitwise instructions were added to the 6811
instruction set by aliens from outer space and that they
tend to use unexpected read-modify-write-cycles to get
things done. That may upset registers in peripheral chips.

Tip from a red blooded earthling (and compiler writer):
Don't use the bit instructions!

Because some hardware idiot thought it was doable to add
a certain instruction, that doesn't mean it makes sense
to use it!

(Don't get me started on the 6812...) Greetings,
Jaap

-- Chip Directory
-- http://www.chipdir.biz/
-- http://www.chipdir.info/
-- http://www.chipdir.net/
-- http://www.chipdir.nl/
-- http://www.chipdir.org/
-- And about 30 other mirror sites world-wide.
--
-- To subscribe to a free 'chip issues, questions and answers'
-- mailing list, send a message to with
-- in the body 'subscribe chipdir-L'. About 500 experts are
-- willing to think with you about electronics problems etc. Yahoo! Groups Links




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

Re: Re: Code Auto Execution problem - Tony Papadimitriou - Nov 29 12:40:00 2004

----- Original Message -----
From: "Israel Brewster" <>
To: < > I have used bit instructions quite frequently in my programing and found them
to be a
> valuable tool, making my code much simpler and more compact - frequently doing
in one
> line what would otherwise take four or five lines at least. Be that as it may,
the problem
> here has nothing to do with the bit instruction- as I mentioned in my original
post, I have
> tried replacing it with a simple delay loop. The problem is that with BUFFALO
disabled, the
> A/D conversion never takes place. Thus the conversion complete flag is never
set, and
> thus the bit instruction loops infinitely waiting for it to be set. This is
shown by the
> behavior experienced when the bit instruction is replaced with a delay loop,
as described
> in my original post. Thanks anyway.

If it helps, here's one routine I've used:

;**************************************************************
; Purpose: Read an A/D channel (4 readings) and return average
; Input : A = A/D channel to read
; Output : A = Averaged reading

ReadAD pshx
pshb
ldx #REGS
anda #%111 ;isolate channel number
sta [ADCTL,x ;convert channel in A (0-7)
clra ;D will sum up 4 readings
brclr [ADCTL,X,CCF.,* ;wait for conversion to finish
ldb [ADR1,x ;D = 1st reading
addb [ADR2,x
adca #0 ;D = 1st+2nd readings
addb [ADR3,x
adca #0 ;D = 1st+2nd+3rd readings
addb [ADR4,x
adca #0 ;D = sum of all 4 readings
lsrd:2 ;average (divide by 4)
tba ;Result now in A
pulb
pulx
rts
;**************************************************************

Undefined labels:
REGS is usually $1000
ADCTL is usually REGS+$30
ADR1..ADR4 are usually REGS+$31 .. REGS+$34
CCF. is $80

Using ASM11 v1.84d so syntax may require mods.





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

Re: Code Auto Execution problem - Israel Brewster - Nov 29 14:59:00 2004


Thank you, that certainly sounds promising. Would you be able to tell me where I could
find more information about this register and the functions of its various bits? It is not
mentioned in the HC11 book I have, aside from an acknowledgment of its existence in the
I/O registers table. I'll give it a try, see what happens. Thanks!

--- In , "Nick Reitter" <nick@r...> wrote:
> Hi, Bit 7, ADPU,in the Option register, $1039, must be set.
> Keep having fun.
> Nick Reitter
> Personal home page http://www.reitter.net/nick/
> Where we got o church http://www.shepherdwlsv.org/
> Where I used to work before I retired http://web.alfredstate.edu/eet/ > -----Original Message-----
> From: Jaap van Ganswijk [mailto:ganswijk@x...]
> Sent: Monday, November 29, 2004 5:05 AM
> To:
> Subject: Re: [m68HC11] Code Auto Execution problem >
> At 2004-11-29 02:11, Israel Brewster wrote:
> >I am having some difficulty getting my code to execute with BUFFALO
> disabled. I am using an
> >Adapt11C24DXEVB, and have my code loaded into the internal EEPROM starting
> at address
> >B600. If I power up the board with BUFFALO enabled, and issue the go B600
> command,
> >everything works perfectly. However, if I then pull E0 high to disable
> buffalo, and reset the
> >board, my program only runs until it tries to do an A/D conversion on port
> E7, where it hangs
> >waiting for the conversion complete flag to be set.
> >
> >I am using the branch on bit clear command to loop back to itself until the
> flag is set, and am
> >certain this command is correct both because I copied it from some code I
> have used before,
> >and because, as I mentioned already, it DOES run properly when called from
> BUFFALO. If I
> >replace the branch command with a delay loop that should be more than long
> enough for the
> >conversions to complete (Y index register is decremented from $0700)
> program execution
> >continues successfully past this point- except that the A/D conversion is
> never completed,
> >which obviously causes the branches based on the result of the A/D
> conversion to branch
> >improperly. Specifically, it always takes the branch for when the value is
> less than $41, while
> >the actual value should be (and is when run from BUFFALO) close to or equal
> to $FF.
> >
> >I have already checked that I am setting the stack pointer ($CFFF) and
> enabling the interrupts
> >(CLI), with no change. Can anyone tell me what I am missing here? Thanks.
>
> Beware that the bitwise instructions were added to the 6811
> instruction set by aliens from outer space and that they
> tend to use unexpected read-modify-write-cycles to get
> things done. That may upset registers in peripheral chips.
>
> Tip from a red blooded earthling (and compiler writer):
> Don't use the bit instructions!
>
> Because some hardware idiot thought it was doable to add
> a certain instruction, that doesn't mean it makes sense
> to use it!
>
> (Don't get me started on the 6812...) > Greetings,
> Jaap
>
> -- Chip Directory
> -- http://www.chipdir.biz/
> -- http://www.chipdir.info/
> -- http://www.chipdir.net/
> -- http://www.chipdir.nl/
> -- http://www.chipdir.org/
> -- And about 30 other mirror sites world-wide.
> --
> -- To subscribe to a free 'chip issues, questions and answers'
> -- mailing list, send a message to listguru@f... with
> -- in the body 'subscribe chipdir-L'. About 500 experts are
> -- willing to think with you about electronics problems etc. > Yahoo! Groups Links






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

Re: Code Auto Execution problem - Israel Brewster - Nov 29 15:52:00 2004


Yep, that did the trick. Thanks all for the suggestions!

--- In , "Israel Brewster" <fsijb@u...> wrote:
>
> Thank you, that certainly sounds promising. Would you be able to tell me where I could
> find more information about this register and the functions of its various bits? It is not
> mentioned in the HC11 book I have, aside from an acknowledgment of its existence in
the
> I/O registers table. I'll give it a try, see what happens. Thanks!
>
> --- In , "Nick Reitter" <nick@r...> wrote:
> > Hi, Bit 7, ADPU,in the Option register, $1039, must be set.
> > Keep having fun.
> > Nick Reitter
> > Personal home page http://www.reitter.net/nick/
> > Where we got o church http://www.shepherdwlsv.org/
> > Where I used to work before I retired http://web.alfredstate.edu/eet/
> >
> >
> > -----Original Message-----
> > From: Jaap van Ganswijk [mailto:ganswijk@x]
> > Sent: Monday, November 29, 2004 5:05 AM
> > To:
> > Subject: Re: [m68HC11] Code Auto Execution problem
> >
> >
> >
> > At 2004-11-29 02:11, Israel Brewster wrote:
> > >I am having some difficulty getting my code to execute with BUFFALO
> > disabled. I am using an
> > >Adapt11C24DXEVB, and have my code loaded into the internal EEPROM starting
> > at address
> > >B600. If I power up the board with BUFFALO enabled, and issue the go B600
> > command,
> > >everything works perfectly. However, if I then pull E0 high to disable
> > buffalo, and reset the
> > >board, my program only runs until it tries to do an A/D conversion on port
> > E7, where it hangs
> > >waiting for the conversion complete flag to be set.
> > >
> > >I am using the branch on bit clear command to loop back to itself until the
> > flag is set, and am
> > >certain this command is correct both because I copied it from some code I
> > have used before,
> > >and because, as I mentioned already, it DOES run properly when called from
> > BUFFALO. If I
> > >replace the branch command with a delay loop that should be more than long
> > enough for the
> > >conversions to complete (Y index register is decremented from $0700)
> > program execution
> > >continues successfully past this point- except that the A/D conversion is
> > never completed,
> > >which obviously causes the branches based on the result of the A/D
> > conversion to branch
> > >improperly. Specifically, it always takes the branch for when the value is
> > less than $41, while
> > >the actual value should be (and is when run from BUFFALO) close to or equal
> > to $FF.
> > >
> > >I have already checked that I am setting the stack pointer ($CFFF) and
> > enabling the interrupts
> > >(CLI), with no change. Can anyone tell me what I am missing here? Thanks.
> >
> > Beware that the bitwise instructions were added to the 6811
> > instruction set by aliens from outer space and that they
> > tend to use unexpected read-modify-write-cycles to get
> > things done. That may upset registers in peripheral chips.
> >
> > Tip from a red blooded earthling (and compiler writer):
> > Don't use the bit instructions!
> >
> > Because some hardware idiot thought it was doable to add
> > a certain instruction, that doesn't mean it makes sense
> > to use it!
> >
> > (Don't get me started on the 6812...)
> >
> >
> > Greetings,
> > Jaap
> >
> > -- Chip Directory
> > -- http://www.chipdir.biz/
> > -- http://www.chipdir.info/
> > -- http://www.chipdir.net/
> > -- http://www.chipdir.nl/
> > -- http://www.chipdir.org/
> > -- And about 30 other mirror sites world-wide.
> > --
> > -- To subscribe to a free 'chip issues, questions and answers'
> > -- mailing list, send a message to listguru@f... with
> > -- in the body 'subscribe chipdir-L'. About 500 experts are
> > -- willing to think with you about electronics problems etc.
> >
> >
> >
> >
> >
> > Yahoo! Groups Links





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

Re: Code Auto Execution problem - Mark Schultz - Nov 29 20:17:00 2004


--- In , Scott Grodevant <scott14468@y...>
wrote:
> Hi all and Israel,
>
> My memory fails me for specifics about the HC11 (without going to
> the manual), but I am pretty sure there is a configuration register
> associated with the ADC that must be initialized before using it.
> Without BUFFALO running to do it for you, the ADC is probably
> disabled.

What Scott says here comes close to the advice I will offer.

When you transition from a debug environment (e.g.
BUFFALO/PCBUG/JBug/etc.) to a 'standalone' runtime environment, you
must take care to ensure that your program includes ALL necessary
initialization code. This means things like: Setting the stack
pointer and status register properly, initializing all your RAM
variables to known values, setting values in time-protected registers
such as HPRIO, and initializing ALL subsystems you are using.

You can get away without doing much of the above because a typical
monitor program will place the MCU in a known, safe-operation state
as part of its own initialization. This is not true when you start
up out of reset directly into your own program. You are responsible
for initializing any subsystem/peripheral you choose to use, and are
also responsible for initializing the CPU registers and memory
appropriately.





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

Re: Code Auto Execution problem - Israel Brewster - Nov 30 12:16:00 2004


yeah, good advice- and absolutely true. The annoying thing about this situation is that, as
far as I can tell, the need to initialize the OPTION register is completely undocumented, at
least in the materials I have available. Its existence is not even mentioned in the book I
have on the subject. So while I am aware of the need to initialize everything, I don't know
what all needs to be initialized. This, of course, leads to very annoying problems such as
the one I just experienced. If anyone wants to suggest some good reference sources for
programing with the HC11/HC12 (thinking of moving up) that mention such things, I
would appreciate it. Thanks again everyone.

--- In , "Mark Schultz" <n9xmj@y...> wrote:
>
> --- In , Scott Grodevant <scott14468@y...>
> wrote:
> > Hi all and Israel,
> >
> > My memory fails me for specifics about the HC11 (without going to
> > the manual), but I am pretty sure there is a configuration register
> > associated with the ADC that must be initialized before using it.
> > Without BUFFALO running to do it for you, the ADC is probably
> > disabled.
>
> What Scott says here comes close to the advice I will offer.
>
> When you transition from a debug environment (e.g.
> BUFFALO/PCBUG/JBug/etc.) to a 'standalone' runtime environment, you
> must take care to ensure that your program includes ALL necessary
> initialization code. This means things like: Setting the stack
> pointer and status register properly, initializing all your RAM
> variables to known values, setting values in time-protected registers
> such as HPRIO, and initializing ALL subsystems you are using.
>
> You can get away without doing much of the above because a typical
> monitor program will place the MCU in a known, safe-operation state
> as part of its own initialization. This is not true when you start
> up out of reset directly into your own program. You are responsible
> for initializing any subsystem/peripheral you choose to use, and are
> also responsible for initializing the CPU registers and memory
> appropriately.






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

RE: Re: Code Auto Execution problem - Paul Hutchinson - Nov 30 13:13:00 2004

For the 68HC11 you need the Motorola reference manual literature#
M68HC11RM/AD and the Motorola technical data for the particular family of
HC11.

They are available as pdf files on the freescale web site.

I suspect there are equivalent manuals for the HC12 series also on the web
site.

Paul > -----Original Message-----
> From: Israel Brewster [mailto:]
> Sent: Tuesday, November 30, 2004 12:16 PM
>
> yeah, good advice- and absolutely true. The annoying thing about
> this situation is that, as
> far as I can tell, the need to initialize the OPTION register is
> completely undocumented, at
> least in the materials I have available. Its existence is not
> even mentioned in the book I
> have on the subject. So while I am aware of the need to
> initialize everything, I don't know
> what all needs to be initialized. This, of course, leads to very
> annoying problems such as
> the one I just experienced. If anyone wants to suggest some good
> reference sources for
> programing with the HC11/HC12 (thinking of moving up) that
> mention such things, I
> would appreciate it. Thanks again everyone.




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

Re: Re: Code Auto Execution problem - Scott Grodevant - Nov 30 13:35:00 2004

Hi all and Israel,

Well, the need to initialize the OPTION register isn't exactly undocumented. The manual states it's default settings. Aparently they differ from your application's requirements. In the world of electronic design, all manufacturers publish detailed specifications for their devices, it is up to the designer to obtain and read them. Yeah, we all make mistakes.

The manuals will not tell you what registers to initialize. In general, you must initialize any registers dealing with hardware you intend to use or any that are in a state that differs from your application's requirements. Further, if you don't initialize registers that are in a "correct" state, there is no guarantee (although it is rather unlikely) future mask sets won't be incompatible with your code. Good luck,
Scott
Israel Brewster <> wrote:

yeah, good advice- and absolutely true. The annoying thing about this situation is that, as
far as I can tell, the need to initialize the OPTION register is completely undocumented, at
least in the materials I have available. Its existence is not even mentioned in the book I
have on the subject. So while I am aware of the need to initialize everything, I don't know
what all needs to be initialized. This, of course, leads to very annoying problems such as
the one I just experienced. If anyone wants to suggest some good reference sources for
programing with the HC11/HC12 (thinking of moving up) that mention such things, I
would appreciate it. Thanks again everyone.

--- In , "Mark Schultz" <n9xmj@y...> wrote:
>
> --- In , Scott Grodevant <scott14468@y...>
> wrote:
> > Hi all and Israel,
> >
> > My memory fails me for specifics about the HC11 (without going to
> > the manual), but I am pretty sure there is a configuration register
> > associated with the ADC that must be initialized before using it.
> > Without BUFFALO running to do it for you, the ADC is probably
> > disabled.
>
> What Scott says here comes close to the advice I will offer.
>
> When you transition from a debug environment (e.g.
> BUFFALO/PCBUG/JBug/etc.) to a 'standalone' runtime environment, you
> must take care to ensure that your program includes ALL necessary
> initialization code. This means things like: Setting the stack
> pointer and status register properly, initializing all your RAM
> variables to known values, setting values in time-protected registers
> such as HPRIO, and initializing ALL subsystems you are using.
>
> You can get away without doing much of the above because a typical
> monitor program will place the MCU in a known, safe-operation state
> as part of its own initialization. This is not true when you start
> up out of reset directly into your own program. You are responsible
> for initializing any subsystem/peripheral you choose to use, and are
> also responsible for initializing the CPU registers and memory
> appropriately.

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

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





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

Re: Re: Code Auto Execution problem - bart homerson - Nov 30 17:23:00 2004

I would have to agree. In fact, whenever I write a program I start with a "shell" that contains a reference to every register, and a vector table to all of the interupts. If I don't need it, then I delete it. I have a terrible memory, if you haven't already figured out by reading my other postings, so doing such things helps to relieve my frustrations. Just a suggestion. If you want to see a bit of what I am talking about, then you can look at the Bit-Banging program that I uploaded to this group's file section. You can see the Interupt vector table at the bottom of the file. You will also notice many registers that are EQU'ed, in the begining of the file, are not being used. I really did not need to conserve space for this program so I left them there.

Take care.

LF

Israel Brewster <> wrote:

yeah, good advice- and absolutely true. The annoying thing about this situation is that, as
far as I can tell, the need to initialize the OPTION register is completely undocumented, at
least in the materials I have available. Its existence is not even mentioned in the book I
have on the subject. So while I am aware of the need to initialize everything, I don't know
what all needs to be initialized. This, of course, leads to very annoying problems such as
the one I just experienced. If anyone wants to suggest some good reference sources for
programing with the HC11/HC12 (thinking of moving up) that mention such things, I
would appreciate it. Thanks again everyone.

--- In , "Mark Schultz" <n9xmj@y...> wrote:
>
> --- In , Scott Grodevant <scott14468@y...>
> wrote:
> > Hi all and Israel,
> >
> > My memory fails me for specifics about the HC11 (without going to
> > the manual), but I am pretty sure there is a configuration register
> > associated with the ADC that must be initialized before using it.
> > Without BUFFALO running to do it for you, the ADC is probably
> > disabled.
>
> What Scott says here comes close to the advice I will offer.
>
> When you transition from a debug environment (e.g.
> BUFFALO/PCBUG/JBug/etc.) to a 'standalone' runtime environment, you
> must take care to ensure that your program includes ALL necessary
> initialization code. This means things like: Setting the stack
> pointer and status register properly, initializing all your RAM
> variables to known values, setting values in time-protected registers
> such as HPRIO, and initializing ALL subsystems you are using.
>
> You can get away without doing much of the above because a typical
> monitor program will place the MCU in a known, safe-operation state
> as part of its own initialization. This is not true when you start
> up out of reset directly into your own program. You are responsible
> for initializing any subsystem/peripheral you choose to use, and are
> also responsible for initializing the CPU registers and memory
> appropriately.

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

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




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

Re: Code Auto Execution problem - Israel Brewster - Nov 30 22:11:00 2004


Well, I'm sure the OPTION register is documented somewhere, just not in any of the
documentation I have looked at. I guess I just have all the wrong documentation :-) If I had
had some sort of a list giving all the registers, what their purpose was, and what they were
initialized to, I could have figured this out myself in about five minutes flat (or less,
depending on how long it took me to scan the list). Unfortunately, I have no such list.

--- In , "Nick Reitter" <nick@r...> wrote:
> Here are two texts that good:
> "Microcontroller Technology - The 68HC11"; Peter Saspov, Prentice Hall
> "Data Acquisition and Process Control with the M68HC11 Microcontroller";
> Driscoll etal; Prentice Hall
>
> Nick

Yeah, that second one is what I have. Good book, but I am finding it to be more and more
lacking as I learn more about the microcontroller and programing. Like the fact that it
doesn't mention the option register, at least that I have found. I'll have to check out that
first one sometime. Thanks again for the assistance and info

--- In , Scott Grodevant <scott14468@y...> wrote:
> Hi all and Israel,
>
> Well, the need to initialize the OPTION register isn't exactly undocumented. The manual
states it's default settings. Aparently they differ from your application's requirements.
In the world of electronic design, all manufacturers publish detailed specifications for their
devices, it is up to the designer to obtain and read them. Yeah, we all make mistakes.
>
> The manuals will not tell you what registers to initialize. In general, you must initialize
any registers dealing with hardware you intend to use or any that are in a state that differs
from your application's requirements. Further, if you don't initialize registers that are in a
"correct" state, there is no guarantee (although it is rather unlikely) future mask sets won't
be incompatible with your code. > Good luck,
> Scott






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

Re: Re: Code Auto Execution problem - bart homerson - Dec 1 10:44:00 2004

I think you need the MC68HC11 Reference GUIDE and not the reference MANUAL. The reference GUIDE has a complete listing of all of the registers and interupts. I will post it in the "Files" section of this group.

LF

Israel Brewster <> wrote:

Well, I'm sure the OPTION register is documented somewhere, just not in any of the
documentation I have looked at. I guess I just have all the wrong documentation :-) If I had
had some sort of a list giving all the registers, what their purpose was, and what they were
initialized to, I could have figured this out myself in about five minutes flat (or less,
depending on how long it took me to scan the list). Unfortunately, I have no such list.

--- In , "Nick Reitter" <nick@r...> wrote:
> Here are two texts that good:
> "Microcontroller Technology - The 68HC11"; Peter Saspov, Prentice Hall
> "Data Acquisition and Process Control with the M68HC11 Microcontroller";
> Driscoll etal; Prentice Hall
>
> Nick

Yeah, that second one is what I have. Good book, but I am finding it to be more and more
lacking as I learn more about the microcontroller and programing. Like the fact that it
doesn't mention the option register, at least that I have found. I'll have to check out that
first one sometime. Thanks again for the assistance and info

--- In , Scott Grodevant <scott14468@y...> wrote:
> Hi all and Israel,
>
> Well, the need to initialize the OPTION register isn't exactly undocumented. The manual
states it's default settings. Aparently they differ from your application's requirements.
In the world of electronic design, all manufacturers publish detailed specifications for their
devices, it is up to the designer to obtain and read them. Yeah, we all make mistakes.
>
> The manuals will not tell you what registers to initialize. In general, you must initialize
any registers dealing with hardware you intend to use or any that are in a state that differs
from your application's requirements. Further, if you don't initialize registers that are in a
"correct" state, there is no guarantee (although it is rather unlikely) future mask sets won't
be incompatible with your code. > Good luck,
> Scott

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

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





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

Re: Code Auto Execution problem - Israel Brewster - Dec 1 12:47:00 2004


Thanks! That looks very useful.

--- In , bart homerson <odddooo1@y...> wrote:
> I think you need the MC68HC11 Reference GUIDE and not the reference MANUAL. The
reference GUIDE has a complete listing of all of the registers and interupts. I will post it in
the "Files" section of this group.
>
> LF
>
> Israel Brewster <fsijb@u...> wrote:
>
> Well, I'm sure the OPTION register is documented somewhere, just not in any of the
> documentation I have looked at. I guess I just have all the wrong documentation :-) If I
had
> had some sort of a list giving all the registers, what their purpose was, and what they
were
> initialized to, I could have figured this out myself in about five minutes flat (or less,
> depending on how long it took me to scan the list). Unfortunately, I have no such list.
>
> --- In , "Nick Reitter" <nick@r...> wrote:
> > Here are two texts that good:
> > "Microcontroller Technology - The 68HC11"; Peter Saspov, Prentice Hall
> > "Data Acquisition and Process Control with the M68HC11 Microcontroller";
> > Driscoll etal; Prentice Hall
> >
> > Nick
>
> Yeah, that second one is what I have. Good book, but I am finding it to be more and
more
> lacking as I learn more about the microcontroller and programing. Like the fact that it
> doesn't mention the option register, at least that I have found. I'll have to check out that
> first one sometime. Thanks again for the assistance and info
>
> --- In , Scott Grodevant <scott14468@y...> wrote:
> > Hi all and Israel,
> >
> > Well, the need to initialize the OPTION register isn't exactly undocumented. The
manual
> states it's default settings. Aparently they differ from your application's requirements.
> In the world of electronic design, all manufacturers publish detailed specifications for
their
> devices, it is up to the designer to obtain and read them. Yeah, we all make mistakes.
> >
> > The manuals will not tell you what registers to initialize. In general, you must
initialize
> any registers dealing with hardware you intend to use or any that are in a state that
differs
> from your application's requirements. Further, if you don't initialize registers that are in
a
> "correct" state, there is no guarantee (although it is rather unlikely) future mask sets
won't
> be incompatible with your code.
> >
> >
> > Good luck,
> > Scott >
>
> Yahoo! Groups SponsorADVERTISEMENT > ---------------------------------
> Yahoo! Groups Links
>
> To >
> ---------------------------------




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

Re: Re: Code Auto Execution problem - bart homerson - Dec 1 16:54:00 2004

Israel,

You are quite welcome! I had the same problem when I was first starting out with these chips. Now, I always keep a copy on every computer that I might possibly write a program on. I used to have a hard copy of this guide that comes as a little book that fits in your pocket, but I lost it. I got my copy included with my starter kit from Technological arts, but I think you can order a hard copy from Freescale semiconductor.

LF

Israel Brewster <> wrote:

Thanks! That looks very useful.

--- In , bart homerson <odddooo1@y...> wrote:
> I think you need the MC68HC11 Reference GUIDE and not the reference MANUAL. The
reference GUIDE has a complete listing of all of the registers and interupts. I will post it in
the "Files" section of this group.
>
> LF
>
> Israel Brewster <fsijb@u...> wrote:
>
> Well, I'm sure the OPTION register is documented somewhere, just not in any of the
> documentation I have looked at. I guess I just have all the wrong documentation :-) If I
had
> had some sort of a list giving all the registers, what their purpose was, and what they
were
> initialized to, I could have figured this out myself in about five minutes flat (or less,
> depending on how long it took me to scan the list). Unfortunately, I have no such list.
>
> --- In , "Nick Reitter" <nick@r...> wrote:
> > Here are two texts that good:
> > "Microcontroller Technology - The 68HC11"; Peter Saspov, Prentice Hall
> > "Data Acquisition and Process Control with the M68HC11 Microcontroller";
> > Driscoll etal; Prentice Hall
> >
> > Nick
>
> Yeah, that second one is what I have. Good book, but I am finding it to be more and
more
> lacking as I learn more about the microcontroller and programing. Like the fact that it
> doesn't mention the option register, at least that I have found. I'll have to check out that
> first one sometime. Thanks again for the assistance and info
>
> --- In , Scott Grodevant <scott14468@y...> wrote:
> > Hi all and Israel,
> >
> > Well, the need to initialize the OPTION register isn't exactly undocumented. The
manual
> states it's default settings. Aparently they differ from your application's requirements.
> In the world of electronic design, all manufacturers publish detailed specifications for
their
> devices, it is up to the designer to obtain and read them. Yeah, we all make mistakes.
> >
> > The manuals will not tell you what registers to initialize. In general, you must
initialize
> any registers dealing with hardware you intend to use or any that are in a state that
differs
> from your application's requirements. Further, if you don't initialize registers that are in
a
> "correct" state, there is no guarantee (although it is rather unlikely) future mask sets
won't
> be incompatible with your code.
> >
> >
> > Good luck,
> > Scott >
>
> Yahoo! Groups SponsorADVERTISEMENT > ---------------------------------
> Yahoo! Groups Links
>
> To >
> ---------------------------------
> ">http://mail.yahoo.com

[Non-text portions of this message have been removed]



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