Forums

CW You Did It Again. CWerked the Memory Window in V5.1 for 56800.

Started by Pete Becher May 29, 2003
CodeWarrior,

You did it again. The Memory window in the Debugger
shows the bytes in reverse order. Is there an easy fix for this
reCWerk. (must be the new improved memory window mentioned in the
spin-doctor drivel that you want us to read when installing v5.1)

Pete



Pete,
There is a new option on the 5.1 tools. When you download onto
your board there should be a data menu that gets created on the
toolbar. If you pull down the data menu there should be a option in
there called 'Swap Endianess'. This will take care of the issue
that you are seeing.

Regards,
Irene

--- In , "Pete Becher" <pbecher@d...>
wrote:
> CodeWarrior,
>
> You did it again. The Memory window in the Debugger
> shows the bytes in reverse order. Is there an easy fix for this
> reCWerk. (must be the new improved memory window mentioned in the
> spin-doctor drivel that you want us to read when installing v5.1)
>
> Pete





Hi Irene,

I CAN NOT believe it!! Do the Metrowerks CW programmers have any
concept of embedded programming??

The Memory window works like crap. All the "Swap Endian" does is
make the ASCII window readable. IT DOES NOT SWAP THE BYTES IN THE
HEX WINDOW. How the hell is someone supposed to troubleshoot data in
say, a peripheral, when the data in the registers is shown as swapped.
And not only that but selecting Swap Endian does not just swap
immediately. You have to jump through some hoops like closing
windows, exiting debugger, restarting debugger and opening window.
And then you cannot be sure it will remain that way.

I CAN NOT believe it!!
Pete

--- In , "isierra95" <isierra95@y...>
wrote:
> Pete,
> There is a new option on the 5.1 tools. When you download onto
> your board there should be a data menu that gets created on the
> toolbar. If you pull down the data menu there should be a option
in
> there called 'Swap Endianess'. This will take care of the issue
> that you are seeing.
>
> Regards,
> Irene
>
> --- In , "Pete Becher" <pbecher@d...>
> wrote:
> > CodeWarrior,
> >
> > You did it again. The Memory window in the Debugger
> > shows the bytes in reverse order. Is there an easy fix for this
> > reCWerk. (must be the new improved memory window mentioned in the
> > spin-doctor drivel that you want us to read when installing v5.1)
> >
> > Pete



Hello Pete,

The endian swap command is only applicable when word size is 16, which is
the correct word size for this target.

Some of the advantages for this window are:
-window is resizable
-better scrolling
-dockable
-you can view source for P memory view
-You can set BP's in the memory window
-More intutive modification to memory
-size/window position saved in workspace
-ability to swap endian (for 16-bit word)

I hope you understand that while the behavior is a bit different why we
believe that the advantages far outweigh the inconvenience of a few changes
in procedures that you have to make.

Ron

> Hi Irene,
>
> I CAN NOT believe it!! Do the Metrowerks CW programmers have any
> concept of embedded programming??
>
> The Memory window works like crap. All the "Swap Endian" does is
> make the ASCII window readable. IT DOES NOT SWAP THE BYTES IN THE
> HEX WINDOW. How the hell is someone supposed to troubleshoot data in
> say, a peripheral, when the data in the registers is shown as swapped.
> And not only that but selecting Swap Endian does not just swap
> immediately. You have to jump through some hoops like closing
> windows, exiting debugger, restarting debugger and opening window.
> And then you cannot be sure it will remain that way.
>
> I CAN NOT believe it!!
> Pete
>
> --- In , "isierra95" <isierra95@y...>
> wrote:
>> Pete,
>> There is a new option on the 5.1 tools. When you download onto
>> your board there should be a data menu that gets created on the
>> toolbar. If you pull down the data menu there should be a option
> in
>> there called 'Swap Endianess'. This will take care of the issue
>> that you are seeing.
>>
>> Regards,
>> Irene
>>
>> --- In , "Pete Becher" <pbecher@d...>
>> wrote:
>>> CodeWarrior,
>>>
>>> You did it again. The Memory window in the Debugger
>>> shows the bytes in reverse order. Is there an easy fix for this
>>> reCWerk. (must be the new improved memory window mentioned in the
>>> spin-doctor drivel that you want us to read when installing v5.1)
>>>
>>> Pete

--
Get printed CodeWarrior Documentation (limited selection)
And all the latest versions of CodeWarrior Manuals

http://www.metrowerks.com/MW/Support/dev_resources/default.htm

Ron Liechty - - http://www.metrowerks.com



Hi Ron,

We went through this with version 5.0. Someone there had enough
sense to change it back for version 5.01. This was posted (msg 350)
after that fix. "CW 5.0 used little endian format in its new memory
view. CW 5.01 and higher using the big endian format."

I think where the Metrowerks Programmers/Designers are not
understanding is that the DSP56F80X series of processors have no
concept of BYTES internally. This processor is 16 bit based and all
data is stored that way. If you store a CHAR it is put into a 16 bit
memory location in basically Big Endian format. If you store a 16
bit WORD it stored as a word, neither Big Endian nor Little Endian
format. If you store a LONG then it would be stored in Little Endian
format (16 bit Little Endian).

SO SHOW IT THAT WAY IN THE MEMORY WINDOW, GET IT, YOU ARE DISPLAYING
WHAT IS "IN MEMORY". Sincerely,
Pete--- In , MW Ron <mwron@m...> wrote:
> Hello Pete,
>
> The endian swap command is only applicable when word size is 16,
which is
> the correct word size for this target.
>
> Some of the advantages for this window are:
> -window is resizable
> -better scrolling
> -dockable
> -you can view source for P memory view
> -You can set BP's in the memory window
> -More intutive modification to memory
> -size/window position saved in workspace
> -ability to swap endian (for 16-bit word)
>
> I hope you understand that while the behavior is a bit different
why we
> believe that the advantages far outweigh the inconvenience of a few
changes
> in procedures that you have to make.
>
> Ron
>
> > Hi Irene,
> >
> > I CAN NOT believe it!! Do the Metrowerks CW programmers have any
> > concept of embedded programming??
> >
> > The Memory window works like crap. All the "Swap Endian" does is
> > make the ASCII window readable. IT DOES NOT SWAP THE BYTES IN THE
> > HEX WINDOW. How the hell is someone supposed to troubleshoot
data in
> > say, a peripheral, when the data in the registers is shown as
swapped.
> > And not only that but selecting Swap Endian does not just swap
> > immediately. You have to jump through some hoops like closing
> > windows, exiting debugger, restarting debugger and opening window.
> > And then you cannot be sure it will remain that way.
> >
> > I CAN NOT believe it!!
> > Pete
> >
> > --- In , "isierra95" <isierra95@y...>
> > wrote:
> >> Pete,
> >> There is a new option on the 5.1 tools. When you download
onto
> >> your board there should be a data menu that gets created on the
> >> toolbar. If you pull down the data menu there should be a option
> > in
> >> there called 'Swap Endianess'. This will take care of the issue
> >> that you are seeing.
> >>
> >> Regards,
> >> Irene
> >>
> >> --- In , "Pete Becher" <pbecher@d...>
> >> wrote:
> >>> CodeWarrior,
> >>>
> >>> You did it again. The Memory window in the Debugger
> >>> shows the bytes in reverse order. Is there an easy fix for this
> >>> reCWerk. (must be the new improved memory window mentioned in
the
> >>> spin-doctor drivel that you want us to read when installing
v5.1)
> >>>
> >>> Pete
>
> --
> Get printed CodeWarrior Documentation (limited selection)
> And all the latest versions of CodeWarrior Manuals
>
> http://www.metrowerks.com/MW/Support/dev_resources/default.htm
>
> Ron Liechty - MWRon@m... - http://www.metrowerks.com



Hi Ron;

I have to agree with Pete.  You may recall I reported the endian view in version 5 as a bug and was very pleased to see it fixed in 5.01.  I'm surprised that you decided to bring it back in your latest version.

Is there any realistic reason that you can think of that anyone would ever want to view memory of THIS processor using byte reversal? Any reason at all?

Steve Glow

Pete Becher wrote:
Hi Ron,
We went through this with version 5.0. Someone there had enough sense to change
it back for version 5.01. This was posted (msg 350) after that fix. "CW 5.0 used
little endian format in its new memory view. CW 5.01 and higher using the big
endian format."
I think where the Metrowerks Programmers/Designers are not understanding is that
the DSP56F80X series of processors have no concept of BYTES internally. This
processor is 16 bit based and all data is stored that way. If you store a CHAR
it is put into a 16 bit memory location in basically Big Endian format. If you
store a 16 bit WORD it stored as a word, neither Big Endian nor Little Endian
format. If you store a LONG then it would be stored in Little Endian format (16
bit Little Endian).
SO SHOW IT THAT WAY IN THE MEMORY WINDOW, GET IT, YOU ARE DISPLAYING WHAT IS "IN
MEMORY".
Sincerely,
Pete--- In m...@yahoogroups.com, MW Ron <mwron@m...> wrote:
Hello Pete,
The endian swap command is only applicable when word size is 16, 
which is
the correct word size for this target.
Some of the advantages for this window are:
-window is resizable
-better scrolling
-dockable
-you can view source for P memory view
-You can set BP's in the memory window
-More intutive modification to memory
-size/window position saved in workspace
-ability to swap endian (for 16-bit word)
I hope you understand that while the behavior is a bit different 
why we
believe that the advantages far outweigh the inconvenience of a few
changes
in procedures that you have to make.
Ron

Hi Irene,
I CAN NOT believe it!! Do the Metrowerks CW programmers have any
concept of embedded programming??
The Memory window works like crap. All the "Swap Endian" does is
make the ASCII window readable. IT DOES NOT SWAP THE BYTES IN THE
HEX WINDOW. How the hell is someone supposed to troubleshoot 
data in
say, a peripheral, when the data in the registers is shown as 
swapped.
And not only that but selecting Swap Endian does not just swap
immediately. You have to jump through some hoops like closing
windows, exiting debugger, restarting debugger and opening window.
And then you cannot be sure it will remain that way.
I CAN NOT believe it!!
Pete
--- In m...@yahoogroups.com, "isierra95" <isierra95@y...>
wrote:
Pete,
There is a new option on the 5.1 tools. When you download 
onto
your board there should be a data menu that gets created on the
toolbar. If you pull down the data menu there should be a option
in 
there called 'Swap Endianess'. This will take care of the issue
that you are seeing.
Regards,
Irene
--- In m...@yahoogroups.com, "Pete Becher" <pbecher@d...>
wrote:
CodeWarrior,
You did it again. The Memory window in the Debugger
shows the bytes in reverse order. Is there an easy fix for this
reCWerk. (must be the new improved memory window mentioned in 
the
spin-doctor drivel that you want us to read when installing 
v5.1)
Pete
-- Get printed CodeWarrior Documentation (limited selection)
And all the latest versions of CodeWarrior Manuals
http://www.metrowerks.com/MW/Support/dev_resources/default.htm
Ron Liechty - MWRon@m... - http://www.metrowerks.com
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get A Free Psychic Reading! Your Online Answer To Life's Important Questions.
http://us.click.yahoo.com/Lj3uPC/Me7FAA/ySSFAA/PNArlB/TM
---------------------------------~->
_____________________________________
Note: If you do a simple "reply" with your email client, only the author of this
message will receive your answer. You need to do a "reply all" if you want your
answer to be distributed to the entire group.
_____________________________________
About this discussion group:
To Join: m...@yahoogroups.com
To Post: m...@yahoogroups.com
To Leave: m...@yahoogroups.com
Archives: http://www.yahoogroups.com/group/motoroladsp
More Groups: http://www.dsprelated.com/groups.php3
">http://docs.yahoo.com/info/terms/ 



Hello Steve and Pete,

Originally we had the window showing the big endian view but we had customer
suggestions that this view be native and we (perhaps hastily) changed this
behavior. This change however brought out several complaints so there was a
quick patch to revert back to the older behavior.

For the latest release we were able to make this optional so we felt that
everyone could be satisfied with the behavior. However, it is apparent that
this still needs some tuning.

Our engineering team is looking into methods where this will be less
confusing and more user friendly.

Ron

> Hi Ron;
>
> I have to agree with Pete. You may recall I reported the endian view in
> version 5 as a bug and was very pleased to see it fixed in 5.01. I'm
> surprised that you decided to bring it back in your latest version.
>
> Is there any realistic reason that you can think of that anyone would
> ever want to view memory of THIS processor using byte reversal? Any
> reason at all?
>
> Steve Glow
>
> Pete Becher wrote:
>
>> Hi Ron,
>>
>> We went through this with version 5.0. Someone there had enough
>> sense to change it back for version 5.01. This was posted (msg 350)
>> after that fix. "CW 5.0 used little endian format in its new memory
>> view. CW 5.01 and higher using the big endian format."
>>
>> I think where the Metrowerks Programmers/Designers are not
>> understanding is that the DSP56F80X series of processors have no
>> concept of BYTES internally. This processor is 16 bit based and all
>> data is stored that way. If you store a CHAR it is put into a 16 bit
>> memory location in basically Big Endian format. If you store a 16
>> bit WORD it stored as a word, neither Big Endian nor Little Endian
>> format. If you store a LONG then it would be stored in Little Endian
>> format (16 bit Little Endian).
>>
>> SO SHOW IT THAT WAY IN THE MEMORY WINDOW, GET IT, YOU ARE DISPLAYING
>> WHAT IS "IN MEMORY".
>>
>>
>> Sincerely,
>> Pete--- In , MW Ron <mwron@m...> wrote:
>>
>>
>>> Hello Pete,
>>>
>>> The endian swap command is only applicable when word size is 16,
>>>
>>>
>> which is
>>
>>
>>> the correct word size for this target.
>>>
>>> Some of the advantages for this window are:
>>> -window is resizable
>>> -better scrolling
>>> -dockable
>>> -you can view source for P memory view
>>> -You can set BP's in the memory window
>>> -More intutive modification to memory
>>> -size/window position saved in workspace
>>> -ability to swap endian (for 16-bit word)
>>>
>>> I hope you understand that while the behavior is a bit different
>>>
>>>
>> why we
>>
>>
>>> believe that the advantages far outweigh the inconvenience of a few
>>>
>>>
>> changes
>>
>>
>>> in procedures that you have to make.
>>>
>>> Ron
>>>
>>>
>>>
>>
>>
>>
>>
>>
>>>> Hi Irene,
>>>>
>>>> I CAN NOT believe it!! Do the Metrowerks CW programmers have any
>>>> concept of embedded programming??
>>>>
>>>> The Memory window works like crap. All the "Swap Endian" does is
>>>> make the ASCII window readable. IT DOES NOT SWAP THE BYTES IN THE
>>>> HEX WINDOW. How the hell is someone supposed to troubleshoot
>>>>
>>>>
>> data in
>>
>>
>>>> say, a peripheral, when the data in the registers is shown as
>>>>
>>>>
>> swapped.
>>
>>
>>>> And not only that but selecting Swap Endian does not just swap
>>>> immediately. You have to jump through some hoops like closing
>>>> windows, exiting debugger, restarting debugger and opening window.
>>>> And then you cannot be sure it will remain that way.
>>>>
>>>> I CAN NOT believe it!!
>>>> Pete
>>>>
>>>> --- In , "isierra95" <isierra95@y...>
>>>> wrote:
>>>>
>>>>
>>>>> Pete,
>>>>> There is a new option on the 5.1 tools. When you download
>>>>>
>>>>>
>> onto
>>
>>
>>>>> your board there should be a data menu that gets created on the
>>>>> toolbar. If you pull down the data menu there should be a option
>>>>>
>>>>>
>>>> in
>>>>
>>>>
>>>>> there called 'Swap Endianess'. This will take care of the issue
>>>>> that you are seeing.
>>>>>
>>>>> Regards,
>>>>> Irene
>>>>>
>>>>> --- In , "Pete Becher" <pbecher@d...>
>>>>> wrote:
>>>>>
>>>>>
>>>>>> CodeWarrior,
>>>>>>
>>>>>> You did it again. The Memory window in the Debugger
>>>>>> shows the bytes in reverse order. Is there an easy fix for this
>>>>>> reCWerk. (must be the new improved memory window mentioned in
>>>>>>
>>>>>>
>> the
>>
>>
>>>>>> spin-doctor drivel that you want us to read when installing
>>>>>>
>>>>>>
>> v5.1)
>>
>>
>>>>>> Pete
>>>>>>
>>>>>>
>>> --
>>> Get printed CodeWarrior Documentation (limited selection)
>>> And all the latest versions of CodeWarrior Manuals
>>>
>>> http://www.metrowerks.com/MW/Support/dev_resources/default.htm
>>>
>>> Ron Liechty - MWRon@m... - http://www.metrowerks.com
>>>
>>>
>>
>>
>>
>>
>> _____________________________________
>> Note: If you do a simple "reply" with your email client, only the author of
>> this message will receive your answer. You need to do a "reply all" if you
>> want your answer to be distributed to the entire group.
>>
>> _____________________________________
>> About this discussion group:
>>
>> To Join:
>>
>> To Post:
>>
>> To Leave:
>>
>> Archives: http://www.yahoogroups.com/group/motoroladsp
>>
>> More Groups: http://www.dsprelated.com/groups.php3
>>
>>
>> ">http://docs.yahoo.com/info/terms/
>>
>>
>>
>

--
Get printed CodeWarrior Documentation (limited selection)
And all the latest versions of CodeWarrior Manuals

http://www.metrowerks.com/MW/Support/dev_resources/default.htm

Ron Liechty - - http://www.metrowerks.com



Hi Ron,

I have replied with the following to my BUG report.

A couple hours ago I was debugging and playing with the Memory window
and actually got it to work properly. AMAZING! It now will show a
check mark indicator beside the Swap Endian selection and change the
Memory window format immediately. The display, with the Swap Endian
checked, shows memory exactly the way I want it to (the HEX window
has the 16 bit memory location's bytes in the proper order and the
ASCII window shows the CHARs). And it seems to be remembering the
setting!!

To me the main BUG is that the window gets confused for some reason
and works like crap. I cannot say for sure how I got it to work but
seems to have something to do with starting but not running the
debugger and then selecting View Memory from the Data menu and then
immediately switching the Endian format. Maybe some upgrade
installation issue?

The other issue I see is that Little Endian swaps the BYTES instead
of the WORDS. And maybe some other issues being discussed with tech
support.

Pete

--- In , MW Ron <mwron@m...> wrote:
> Hello Steve and Pete,
>
> Originally we had the window showing the big endian view but we had
customer
> suggestions that this view be native and we (perhaps hastily)
changed this
> behavior. This change however brought out several complaints so
there was a
> quick patch to revert back to the older behavior.
>
> For the latest release we were able to make this optional so we
felt that
> everyone could be satisfied with the behavior. However, it is
apparent that
> this still needs some tuning.
>
> Our engineering team is looking into methods where this will be less
> confusing and more user friendly.
>
> Ron
>
> > Hi Ron;
> >
> > I have to agree with Pete. You may recall I reported the endian
view in
> > version 5 as a bug and was very pleased to see it fixed in 5.01.
I'm
> > surprised that you decided to bring it back in your latest
version.
> >
> > Is there any realistic reason that you can think of that anyone
would
> > ever want to view memory of THIS processor using byte reversal?
Any
> > reason at all?
> >
> > Steve Glow
> >
> > Pete Becher wrote:
> >
> >> Hi Ron,
> >>
> >> We went through this with version 5.0. Someone there had enough
> >> sense to change it back for version 5.01. This was posted (msg
350)
> >> after that fix. "CW 5.0 used little endian format in its new
memory
> >> view. CW 5.01 and higher using the big endian format."
> >>
> >> I think where the Metrowerks Programmers/Designers are not
> >> understanding is that the DSP56F80X series of processors have no
> >> concept of BYTES internally. This processor is 16 bit based and
all
> >> data is stored that way. If you store a CHAR it is put into a
16 bit
> >> memory location in basically Big Endian format. If you store a
16
> >> bit WORD it stored as a word, neither Big Endian nor Little
Endian
> >> format. If you store a LONG then it would be stored in Little
Endian
> >> format (16 bit Little Endian).
> >>
> >> SO SHOW IT THAT WAY IN THE MEMORY WINDOW, GET IT, YOU ARE
DISPLAYING
> >> WHAT IS "IN MEMORY".
> >>
> >>
> >> Sincerely,
> >> Pete--- In , MW Ron <mwron@m...>
wrote:
> >>
> >>
> >>> Hello Pete,
> >>>
> >>> The endian swap command is only applicable when word size is 16,
> >>>
> >>>
> >> which is
> >>
> >>
> >>> the correct word size for this target.
> >>>
> >>> Some of the advantages for this window are:
> >>> -window is resizable
> >>> -better scrolling
> >>> -dockable
> >>> -you can view source for P memory view
> >>> -You can set BP's in the memory window
> >>> -More intutive modification to memory
> >>> -size/window position saved in workspace
> >>> -ability to swap endian (for 16-bit word)
> >>>
> >>> I hope you understand that while the behavior is a bit different
> >>>
> >>>
> >> why we
> >>
> >>
> >>> believe that the advantages far outweigh the inconvenience of a
few
> >>>
> >>>
> >> changes
> >>
> >>
> >>> in procedures that you have to make.
> >>>
> >>> Ron
> >>>
> >>>
> >>>
> >>
> >>
> >>
> >>
> >>
> >>>> Hi Irene,
> >>>>
> >>>> I CAN NOT believe it!! Do the Metrowerks CW programmers have
any
> >>>> concept of embedded programming??
> >>>>
> >>>> The Memory window works like crap. All the "Swap Endian" does
is
> >>>> make the ASCII window readable. IT DOES NOT SWAP THE BYTES IN
THE
> >>>> HEX WINDOW. How the hell is someone supposed to troubleshoot
> >>>>
> >>>>
> >> data in
> >>
> >>
> >>>> say, a peripheral, when the data in the registers is shown as
> >>>>
> >>>>
> >> swapped.
> >>
> >>
> >>>> And not only that but selecting Swap Endian does not just swap
> >>>> immediately. You have to jump through some hoops like closing
> >>>> windows, exiting debugger, restarting debugger and opening
window.
> >>>> And then you cannot be sure it will remain that way.
> >>>>
> >>>> I CAN NOT believe it!!
> >>>> Pete
> >>>>
> >>>> --- In , "isierra95"
<isierra95@y...>
> >>>> wrote:
> >>>>
> >>>>
> >>>>> Pete,
> >>>>> There is a new option on the 5.1 tools. When you download
> >>>>>
> >>>>>
> >> onto
> >>
> >>
> >>>>> your board there should be a data menu that gets created on
the
> >>>>> toolbar. If you pull down the data menu there should be a
option
> >>>>>
> >>>>>
> >>>> in
> >>>>
> >>>>
> >>>>> there called 'Swap Endianess'. This will take care of the
issue
> >>>>> that you are seeing.
> >>>>>
> >>>>> Regards,
> >>>>> Irene
> >>>>>
> >>>>> --- In , "Pete Becher"
<pbecher@d...>
> >>>>> wrote:
> >>>>>
> >>>>>
> >>>>>> CodeWarrior,
> >>>>>>
> >>>>>> You did it again. The Memory window in the Debugger
> >>>>>> shows the bytes in reverse order. Is there an easy fix for
this
> >>>>>> reCWerk. (must be the new improved memory window mentioned in
> >>>>>>
> >>>>>>
> >> the
> >>
> >>
> >>>>>> spin-doctor drivel that you want us to read when installing
> >>>>>>
> >>>>>>
> >> v5.1)
> >>
> >>
> >>>>>> Pete
> >>>>>>
> >>>>>>
> >>> --
> >>> Get printed CodeWarrior Documentation (limited selection)
> >>> And all the latest versions of CodeWarrior Manuals
> >>>
> >>> http://www.metrowerks.com/MW/Support/dev_resources/default.htm
> >>>
> >>> Ron Liechty - MWRon@m... - http://www.metrowerks.com
> >>>
> >>>
> >>
> >>
> >>
> >>
> >> _____________________________________
> >> Note: If you do a simple "reply" with your email client, only
the author of
> >> this message will receive your answer. You need to do a "reply
all" if you
> >> want your answer to be distributed to the entire group.
> >>
> >> _____________________________________
> >> About this discussion group:
> >>
> >> To Join:
> >>
> >> To Post:
> >>
> >> To Leave:
> >>
> >> Archives: http://www.yahoogroups.com/group/motoroladsp
> >>
> >> More Groups: http://www.dsprelated.com/groups.php3
> >>
> >>
> >> ">http://docs.yahoo.com/info/terms/
> >>
> >>
> >>
> >
>
> --
> Get printed CodeWarrior Documentation (limited selection)
> And all the latest versions of CodeWarrior Manuals
>
> http://www.metrowerks.com/MW/Support/dev_resources/default.htm
>
> Ron Liechty - MWRon@m... - http://www.metrowerks.com







Pete Becher wrote:

> Hi Ron,
>
> I have replied with the following to my BUG report.
>
> A couple hours ago I was debugging and playing with the Memory window
> and actually got it to work properly. AMAZING! It now will show a
> check mark indicator beside the Swap Endian selection and change the
> Memory window format immediately. The display, with the Swap Endian
> checked, shows memory exactly the way I want it to (the HEX window
> has the 16 bit memory location's bytes in the proper order and the
> ASCII window shows the CHARs). And it seems to be remembering the
> setting!!

I imagine if it worked correctly to start with we could have saved you a
whole lot of frustration. I know the compiler and debugging engineers are
discussing this and I will bring it up with the QA department. I
appreciate your letting us all know your trials and tribulations.

> To me the main BUG is that the window gets confused for some reason
> and works like crap. I cannot say for sure how I got it to work but
> seems to have something to do with starting but not running the
> debugger and then selecting View Memory from the Data menu and then
> immediately switching the Endian format. Maybe some upgrade
> installation issue?
>
> The other issue I see is that Little Endian swaps the BYTES instead
> of the WORDS. And maybe some other issues being discussed with tech
> support.

I'll follow up with support on this.

Ron >
> --- In , MW Ron <mwron@m...> wrote:
>> Hello Steve and Pete,
>>
>> Originally we had the window showing the big endian view but we had
> customer
>> suggestions that this view be native and we (perhaps hastily)
> changed this
>> behavior. This change however brought out several complaints so
> there was a
>> quick patch to revert back to the older behavior.
>>
>> For the latest release we were able to make this optional so we
> felt that
>> everyone could be satisfied with the behavior. However, it is
> apparent that
>> this still needs some tuning.
>>
>> Our engineering team is looking into methods where this will be less
>> confusing and more user friendly.
>>
>> Ron
>>
>>> Hi Ron;
>>>
>>> I have to agree with Pete. You may recall I reported the endian
> view in
>>> version 5 as a bug and was very pleased to see it fixed in 5.01.
> I'm
>>> surprised that you decided to bring it back in your latest
> version.
>>>
>>> Is there any realistic reason that you can think of that anyone
> would
>>> ever want to view memory of THIS processor using byte reversal?
> Any
>>> reason at all?
>>>
>>> Steve Glow
>>>
>>> Pete Becher wrote:
>>>
>>>> Hi Ron,
>>>>
>>>> We went through this with version 5.0. Someone there had enough
>>>> sense to change it back for version 5.01. This was posted (msg
> 350)
>>>> after that fix. "CW 5.0 used little endian format in its new
> memory
>>>> view. CW 5.01 and higher using the big endian format."
>>>>
>>>> I think where the Metrowerks Programmers/Designers are not
>>>> understanding is that the DSP56F80X series of processors have no
>>>> concept of BYTES internally. This processor is 16 bit based and
> all
>>>> data is stored that way. If you store a CHAR it is put into a
> 16 bit
>>>> memory location in basically Big Endian format. If you store a
> 16
>>>> bit WORD it stored as a word, neither Big Endian nor Little
> Endian
>>>> format. If you store a LONG then it would be stored in Little
> Endian
>>>> format (16 bit Little Endian).
>>>>
>>>> SO SHOW IT THAT WAY IN THE MEMORY WINDOW, GET IT, YOU ARE
> DISPLAYING
>>>> WHAT IS "IN MEMORY".
>>>>
>>>>
>>>> Sincerely,
>>>> Pete--- In , MW Ron <mwron@m...>
> wrote:
>>>>
>>>>
>>>>> Hello Pete,
>>>>>
>>>>> The endian swap command is only applicable when word size is 16,
>>>>>
>>>>>
>>>> which is
>>>>
>>>>
>>>>> the correct word size for this target.
>>>>>
>>>>> Some of the advantages for this window are:
>>>>> -window is resizable
>>>>> -better scrolling
>>>>> -dockable
>>>>> -you can view source for P memory view
>>>>> -You can set BP's in the memory window
>>>>> -More intutive modification to memory
>>>>> -size/window position saved in workspace
>>>>> -ability to swap endian (for 16-bit word)
>>>>>
>>>>> I hope you understand that while the behavior is a bit different
>>>>>
>>>>>
>>>> why we
>>>>
>>>>
>>>>> believe that the advantages far outweigh the inconvenience of a
> few
>>>>>
>>>>>
>>>> changes
>>>>
>>>>
>>>>> in procedures that you have to make.
>>>>>
>>>>> Ron
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>> Hi Irene,
>>>>>>
>>>>>> I CAN NOT believe it!! Do the Metrowerks CW programmers have
> any
>>>>>> concept of embedded programming??
>>>>>>
>>>>>> The Memory window works like crap. All the "Swap Endian" does
> is
>>>>>> make the ASCII window readable. IT DOES NOT SWAP THE BYTES IN
> THE
>>>>>> HEX WINDOW. How the hell is someone supposed to troubleshoot
>>>>>>
>>>>>>
>>>> data in
>>>>
>>>>
>>>>>> say, a peripheral, when the data in the registers is shown as
>>>>>>
>>>>>>
>>>> swapped.
>>>>
>>>>
>>>>>> And not only that but selecting Swap Endian does not just swap
>>>>>> immediately. You have to jump through some hoops like closing
>>>>>> windows, exiting debugger, restarting debugger and opening
> window.
>>>>>> And then you cannot be sure it will remain that way.
>>>>>>
>>>>>> I CAN NOT believe it!!
>>>>>> Pete
>>>>>>
>>>>>> --- In , "isierra95"
> <isierra95@y...>
>>>>>> wrote:
>>>>>>
>>>>>>
>>>>>>> Pete,
>>>>>>> There is a new option on the 5.1 tools. When you download
>>>>>>>
>>>>>>>
>>>> onto
>>>>
>>>>
>>>>>>> your board there should be a data menu that gets created on
> the
>>>>>>> toolbar. If you pull down the data menu there should be a
> option
>>>>>>>
>>>>>>>
>>>>>> in
>>>>>>
>>>>>>
>>>>>>> there called 'Swap Endianess'. This will take care of the
> issue
>>>>>>> that you are seeing.
>>>>>>>
>>>>>>> Regards,
>>>>>>> Irene
>>>>>>>
>>>>>>> --- In , "Pete Becher"
> <pbecher@d...>
>>>>>>> wrote:
>>>>>>>
>>>>>>>
>>>>>>>> CodeWarrior,
>>>>>>>>
>>>>>>>> You did it again. The Memory window in the Debugger
>>>>>>>> shows the bytes in reverse order. Is there an easy fix for
> this
>>>>>>>> reCWerk. (must be the new improved memory window mentioned in
>>>>>>>>
>>>>>>>>
>>>> the
>>>>
>>>>
>>>>>>>> spin-doctor drivel that you want us to read when installing
>>>>>>>>
>>>>>>>>
>>>> v5.1)
>>>>
>>>>
>>>>>>>> Pete
>>>>>>>>
>>>>>>>>
>>>>> --
>>>>> Get printed CodeWarrior Documentation (limited selection)
>>>>> And all the latest versions of CodeWarrior Manuals
>>>>>
>>>>> http://www.metrowerks.com/MW/Support/dev_resources/default.htm
>>>>>
>>>>> Ron Liechty - MWRon@m... - http://www.metrowerks.com
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>> _____________________________________
>>>> Note: If you do a simple "reply" with your email client, only
> the author of
>>>> this message will receive your answer. You need to do a "reply
> all" if you
>>>> want your answer to be distributed to the entire group.
>>
--
Get printed CodeWarrior Documentation (limited selection)
And all the latest versions of CodeWarrior Manuals

http://www.metrowerks.com/MW/Support/dev_resources/default.htm

Ron Liechty - - http://www.metrowerks.com