DMI Wiki for working on the Spec

On Wed, Jun 07, 2006 at 05:55:39PM +1200, Nick Roberts
wrote:
>  > Ok. Any idea when that will be?
> 
> Not really.  Eli usually reviews patches quite quickly
but I don't know
> how many iterations will be needed.
> 
> A lot of the MI commands described in the GDB manual
haven't been implemented
> and I plan to suggest removing some of them.  I wonder
if it might be a good
> idea to populate the DMI spec with the MI commands that
are considered
> essential and then add to them as the need arises,
rather than starting with
> all of them and then whittling them down.
> 
> As a minimal set, these are the ones that I plan to use
for Emacs after it
> gets released:
> 
> -inferior-tty-set
> -file-list-exec-source-files
> -file-list-exec-source-file
> -data-list-register-names
> -data-list-register-values
> -data-list-changed-registers
> -gdb-set
> -gdb-show
> -var-create
> -var-evaluate-expression
> -var-update
> -var-delete
> -var-assign
> -break-list

-break-insert ?

> -stack-list-frames
> -stack-list-locals
> -stack-info-frame

Bob Rossi

_______________________________________________
dmi-discuss mailing list
dmi-discusslists.freestandards.org
http://lists.freestandards.org/cgi-bin/mailma
n/listinfo/dmi-discuss

Bob Rossi wrote:
>On Wed, Jun 07, 2006 at 05:55:39PM +1200, Nick Roberts
wrote:
>>  > Ok. Any idea when that will be?
>>
>> Not really.  Eli usually reviews patches quite
quickly but I don't know
>> how many iterations will be needed.
>>
>> A lot of the MI commands described in the GDB
manual haven't been implemented
>> and I plan to suggest removing some of them.  I
wonder if it might be a good
>> idea to populate the DMI spec with the MI commands
that are considered
>> essential and then add to them as the need arises,
rather than starting with
>> all of them and then whittling them down.

This is a good idea - though I have no idea which haven't
been implemented.

I would also like to know what Apple has implemented in
addition (if anything) or 
what their variations are (Jim?).

>>
>> As a minimal set, these are the ones that I plan to
use for Emacs after it
>> gets released:
>>
>> -inferior-tty-set
>> -file-list-exec-source-files
>> -file-list-exec-source-file
>> -data-list-register-names
>> -data-list-register-values
>> -data-list-changed-registers
>> -gdb-set
>> -gdb-show
>> -var-create
>> -var-evaluate-expression
>> -var-update
>> -var-delete
>> -var-assign
>> -break-list
>
>-break-insert ?

Yes this one is pretty critical.  What about watchpoints,
signal handling?

>
>> -stack-list-frames
>> -stack-list-locals
>> -stack-info-frame






_______________________________________________
dmi-discuss mailing list
dmi-discusslists.freestandards.org
http://lists.freestandards.org/cgi-bin/mailma
n/listinfo/dmi-discuss

I think we need to include all the commands used by existing
front- 
ends, but I agree that there is no need to include
unimplemented  
commands yet (though these may become part of the spec).

Greg


On Jun 7, 2006, at 10:40 AM, <susansmacchia.net>  
<susansmacchia.net> wrote:

> Bob Rossi wrote:
>> On Wed, Jun 07, 2006 at 05:55:39PM +1200, Nick
Roberts wrote:
>>>> Ok. Any idea when that will be?
>>>
>>> Not really.  Eli usually reviews patches quite
quickly but I  
>>> don't know
>>> how many iterations will be needed.
>>>
>>> A lot of the MI commands described in the GDB
manual haven't been  
>>> implemented
>>> and I plan to suggest removing some of them.  I
wonder if it  
>>> might be a good
>>> idea to populate the DMI spec with the MI
commands that are  
>>> considered
>>> essential and then add to them as the need
arises, rather than  
>>> starting with
>>> all of them and then whittling them down.
>
> This is a good idea - though I have no idea which
haven't been  
> implemented.
>
> I would also like to know what Apple has implemented in
addition  
> (if anything) or
> what their variations are (Jim?).
>
>>>
>>> As a minimal set, these are the ones that I
plan to use for Emacs  
>>> after it
>>> gets released:
>>>
>>> -inferior-tty-set
>>> -file-list-exec-source-files
>>> -file-list-exec-source-file
>>> -data-list-register-names
>>> -data-list-register-values
>>> -data-list-changed-registers
>>> -gdb-set
>>> -gdb-show
>>> -var-create
>>> -var-evaluate-expression
>>> -var-update
>>> -var-delete
>>> -var-assign
>>> -break-list
>>
>> -break-insert ?
>
> Yes this one is pretty critical.  What about
watchpoints, signal  
> handling?
>
>>
>>> -stack-list-frames
>>> -stack-list-locals
>>> -stack-info-frame
>
>
>
>
>
>
> _______________________________________________
> dmi-discuss mailing list
> dmi-discusslists.freestandards.org
> http://lists.freestandards.org/cgi-bin/mailma
n/listinfo/dmi-discuss

_______________________________________________
dmi-discuss mailing list
dmi-discusslists.freestandards.org
http://lists.freestandards.org/cgi-bin/mailma
n/listinfo/dmi-discuss

 > I think we need to include all the commands used by
existing front- 
 > ends, but I agree that there is no need to include
unimplemented  
 > commands yet (though these may become part of the
spec).

Does anybody know which ones Eclipse uses?

-- 
Nick                                           http://www.inet.net.n
z/~nickrob

_______________________________________________
dmi-discuss mailing list
dmi-discusslists.freestandards.org
http://lists.freestandards.org/cgi-bin/mailma
n/listinfo/dmi-discuss

 > >> As a minimal set, these are the ones that I
plan to use for Emacs after it
 > >> gets released:
 > >>
 > >> -inferior-tty-set
 > >> -file-list-exec-source-files
 > >> -file-list-exec-source-file
 > >> -data-list-register-names
 > >> -data-list-register-values
 > >> -data-list-changed-registers
 > >> -gdb-set
 > >> -gdb-show
 > >> -var-create
 > >> -var-evaluate-expression
 > >> -var-update
 > >> -var-delete
 > >> -var-assign
 > >> -break-list
 > >
 > >-break-insert ?
 > 
 > Yes this one is pretty critical.  What about
watchpoints, signal handling?

I just use the CLI command break/watch to create the
break/watchpoint and
-break-list to update the breakpoints buffer.

 > >> -stack-list-frames
 > >> -stack-list-locals
 > >> -stack-info-frame


-- 
Nick                                           http://www.inet.net.n
z/~nickrob

_______________________________________________
dmi-discuss mailing list
dmi-discusslists.freestandards.org
http://lists.freestandards.org/cgi-bin/mailma
n/listinfo/dmi-discuss

A quick look at the source shows they provide an
implementation for  
the following commands. There may be a few I missed.

Greg

-break-after
-break-condition
-break-delete
-break-disable
-break-enable
-break-insert
-break-list
-break-watch
-data-disassemble
-data-evaluate-expression
-data-list-changed-registers
-data-list-register-names
-data-list-register-values
-data-read-memory
-data-write-memory
-data-write-register-values
-environment-cd
-environment-directory
-environment-path
-environment-pwd
-exec-arguments
-exec-continue
-exec-finish
-exec-interrupt
-exec-next
-exec-next-instruction
-exec-return-exec-run
-exec-step
-exec-step-instruction
-exec-until
-file-exec-file
-file-symbol-file
-gdb-exit
-gdb-set
-gdb-show
-interpreter-exec
-stack-info-depth
-stack-list-arguments
-stack-list-frames
-stack-list-locals
-stack-select-frame
-target-detach
-target-download
-target-select
-thread-list-ids
-thread-select
-var-assign
-var-create
-var-delete
-var-evaluate-expression
-var-info-expression
-var-info-num-children
-var-info-type
-var-list-children
-var-set-format
-var-show-attributes
-var-show-format
-var-update


On Jun 7, 2006, at 2:27 PM, Nick Roberts wrote:

>> I think we need to include all the commands used by
existing front-
>> ends, but I agree that there is no need to include
unimplemented
>> commands yet (though these may become part of the
spec).
>
> Does anybody know which ones Eclipse uses?
>
> -- 
> Nick                                           http:// 
> www.inet.net.nz/~nickrob

_______________________________________________
dmi-discuss mailing list
dmi-discusslists.freestandards.org
http://lists.freestandards.org/cgi-bin/mailma
n/listinfo/dmi-discuss

 > A quick look at the source shows they provide an
implementation for  
 > the following commands. There may be a few I missed.

These are nearly (but not quite) all the MI commands which
are implemented
in GDB.

 > -break-after
 > -break-condition
 > -break-delete
 > -break-disable
 > -break-enable
 > -break-insert
 > -break-list
 > -break-watch
 > -data-disassemble
 > -data-evaluate-expression
 > -data-list-changed-registers
 > -data-list-register-names
 > -data-list-register-values
 > -data-read-memory

 > -data-write-memory
 > -data-write-register-values

These two are implemented but not documented.

 > -environment-cd
 > -environment-directory
 > -environment-path
 > -environment-pwd
 > -exec-arguments
 > -exec-continue
 > -exec-finish
 > -exec-interrupt
 > -exec-next
 > -exec-next-instruction

 > -exec-return-exec-run

-exec-return
-exec-run

 > -exec-step
 > -exec-step-instruction
 > -exec-until
 > -file-exec-file
 > -file-symbol-file
 > -gdb-exit
 > -gdb-set
 > -gdb-show
 > -interpreter-exec
 > -stack-info-depth
 > -stack-list-arguments
 > -stack-list-frames
 > -stack-list-locals
 > -stack-select-frame
 > -target-detach
 > -target-download
 > -target-select
 > -thread-list-ids
 > -thread-select
 > -var-assign
 > -var-create
 > -var-delete
 > -var-evaluate-expression
 > -var-info-expression
 > -var-info-num-children
 > -var-info-type
 > -var-list-children
 > -var-set-format
 > -var-show-attributes
 > -var-show-format
 > -var-update

This would be a good list to start with then.  The others
could be pared out
of the DMI spec for the moment.

-- 
Nick                                           http://www.inet.net.n
z/~nickrob

_______________________________________________
dmi-discuss mailing list
dmi-discusslists.freestandards.org
http://lists.freestandards.org/cgi-bin/mailma
n/listinfo/dmi-discuss