Mirror of the gdb mailing list
 help / color / mirror / Atom feed
* simulator for sparc64?
@ 2009-10-15 14:50 Joel Sherrill
  2009-10-28 10:20 ` Doug Evans
  0 siblings, 1 reply; 10+ messages in thread
From: Joel Sherrill @ 2009-10-15 14:50 UTC (permalink / raw)
  To: gdb

Hi,

Is there a simulator for sparc64?  I found a
DejaGNU manual that used it as an example but
the sim code doesn't look like it is there.
sis only supports sparc32 AFAIK.

Thanks.

-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
   Support Available             (256) 722-9985



^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: simulator for sparc64?
  2009-10-15 14:50 simulator for sparc64? Joel Sherrill
@ 2009-10-28 10:20 ` Doug Evans
  2009-10-28 12:58   ` Joel Sherrill
  0 siblings, 1 reply; 10+ messages in thread
From: Doug Evans @ 2009-10-28 10:20 UTC (permalink / raw)
  To: Joel Sherrill; +Cc: gdb

On Thu, Oct 15, 2009 at 6:33 AM, Joel Sherrill
<joel.sherrill@oarcorp.com> wrote:
> Hi,
>
> Is there a simulator for sparc64?  I found a
> DejaGNU manual that used it as an example but
> the sim code doesn't look like it is there.
> sis only supports sparc32 AFAIK.
>
> Thanks.

Heh.

That entry either refers to Shade (a sparc64 simulator from Sun used
during the initial sparc64 gcc port), or to a sparc64 simulator I was
working on at the time (archived in a tarball somewhere, but never
completed).


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: simulator for sparc64?
  2009-10-28 10:20 ` Doug Evans
@ 2009-10-28 12:58   ` Joel Sherrill
  2009-10-28 13:59     ` Jakob Engblom
  0 siblings, 1 reply; 10+ messages in thread
From: Joel Sherrill @ 2009-10-28 12:58 UTC (permalink / raw)
  To: Doug Evans; +Cc: gdb

Doug Evans wrote:
> On Thu, Oct 15, 2009 at 6:33 AM, Joel Sherrill
> <joel.sherrill@oarcorp.com> wrote:
>   
>> Hi,
>>
>> Is there a simulator for sparc64?  I found a
>> DejaGNU manual that used it as an example but
>> the sim code doesn't look like it is there.
>> sis only supports sparc32 AFAIK.
>>
>> Thanks.
>>     
>
> Heh.
>
> That entry either refers to Shade (a sparc64 simulator from Sun used
> during the initial sparc64 gcc port), or to a sparc64 simulator I was
> working on at the time (archived in a tarball somewhere, but never
> completed).
>   
Can I infer that since many years have passed since
then, that it refers to nothing that is currently
available? :(

--joel


^ permalink raw reply	[flat|nested] 10+ messages in thread

* RE: simulator for sparc64?
  2009-10-28 12:58   ` Joel Sherrill
@ 2009-10-28 13:59     ` Jakob Engblom
  2009-10-29  0:28       ` Joel Sherrill
  2009-10-30 17:25       ` Python pretty printers and dynamic cast Elmenthaler, Jens
  0 siblings, 2 replies; 10+ messages in thread
From: Jakob Engblom @ 2009-10-28 13:59 UTC (permalink / raw)
  To: 'Joel Sherrill', 'Doug Evans'; +Cc: gdb

We have a commercial one if you want one.


Best regards,

/jakob

_______________________________________________________

Jakob Engblom, PhD, Technical Marketing Manager

Virtutech                   Direct: +46 8 690 07 47   
Drottningholmsvägen 22      Mobile: +46 709 242 646  
11243 Stockholm             Web:    www.virtutech.com 
Sweden
________________________________________________________
  





^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: simulator for sparc64?
  2009-10-28 13:59     ` Jakob Engblom
@ 2009-10-29  0:28       ` Joel Sherrill
  2009-10-30 17:25       ` Python pretty printers and dynamic cast Elmenthaler, Jens
  1 sibling, 0 replies; 10+ messages in thread
From: Joel Sherrill @ 2009-10-29  0:28 UTC (permalink / raw)
  To: Jakob Engblom; +Cc: 'Doug Evans', gdb

Jakob Engblom wrote:
> We have a commercial one if you want one.
>
>   
That's actually that the person porting RTEMS to
SPARC64 is using via a university license.  :)

I will email off list.

Thanks for popping up.
> Best regards,
>
> /jakob
>
> _______________________________________________________
>
> Jakob Engblom, PhD, Technical Marketing Manager
>
> Virtutech                   Direct: +46 8 690 07 47   
> Drottningholmsvägen 22      Mobile: +46 709 242 646  
> 11243 Stockholm             Web:    www.virtutech.com 
> Sweden
> ________________________________________________________
>   
>
>
>
>
>   


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
   Support Available             (256) 722-9985



^ permalink raw reply	[flat|nested] 10+ messages in thread

* Python pretty printers and dynamic cast
  2009-10-28 13:59     ` Jakob Engblom
  2009-10-29  0:28       ` Joel Sherrill
@ 2009-10-30 17:25       ` Elmenthaler, Jens
  2009-10-30 17:51         ` Tom Tromey
  1 sibling, 1 reply; 10+ messages in thread
From: Elmenthaler, Jens @ 2009-10-30 17:25 UTC (permalink / raw)
  To: gdb


Hi all,

I have the following situation: I have an interface type (a class with pure virtual methods only, no data members), and an actual implementation type (having the actual data members). The interface type is used in all public interfaces, but the only implementation of it is the one implementation type.

Is there any way for a python pretty printer to do (or 'mimic') a dynamic cast, such that I can show the members of the implementation type where ever I have the interface type?
 
Jens.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Python pretty printers and dynamic cast
  2009-10-30 17:25       ` Python pretty printers and dynamic cast Elmenthaler, Jens
@ 2009-10-30 17:51         ` Tom Tromey
  2009-12-22  8:15           ` Elmenthaler, Jens
  0 siblings, 1 reply; 10+ messages in thread
From: Tom Tromey @ 2009-10-30 17:51 UTC (permalink / raw)
  To: Elmenthaler, Jens; +Cc: gdb

>>>>> "Jens" == Elmenthaler, Jens <jens.elmenthaler@verigy.com> writes:

Jens> Is there any way for a python pretty printer to do (or 'mimic') a
Jens> dynamic cast, such that I can show the members of the
Jens> implementation type where ever I have the interface type?

There is nothing built-in yet.  We've had a few requests for this,
though, and there was a patch posted to the archer list.  If you give it
a try, let us know how it turns out.

Tom


^ permalink raw reply	[flat|nested] 10+ messages in thread

* RE: Python pretty printers and dynamic cast
  2009-10-30 17:51         ` Tom Tromey
@ 2009-12-22  8:15           ` Elmenthaler, Jens
  2010-01-05 15:34             ` Tom Tromey
  0 siblings, 1 reply; 10+ messages in thread
From: Elmenthaler, Jens @ 2009-12-22  8:15 UTC (permalink / raw)
  To: tromey; +Cc: gdb

Tom> There is nothing built-in yet.  We've had a few requests for this,
Tom> though, and there was a patch posted to the archer list.  If you give it
Tom> a try, let us know how it turns out.

I gave it a try. Unfortunately, the patch only provided the C++ style casts on the gdb command line, not the python API. I got stuck figuring out how to get the gdb command line called from within python.

Nevertheless, another patch from the Archer mailing list made me really happy (and this applies to the python pretty printing facility in general!): http://sourceware.org/ml/archer/2009-q4/msg00013.html.
This patch adds a rtti_type method to gdb.Value.

In my case, I have a smart-pointer like class, i.e. it has just one member, which is a pointer to some base interface. We have many derived interfaces of that base interface, and we have at least one implementation of each of those derived interfaces.

I realized to find out the real implementation type of this pointer member is the real value for debugging our code. The "set print object on" has no effect on the MI interface, and thus out Eclipse CDT frontend does not show the implementation type members when expanding this pointer member.
I ended up having just one pretty printer with a children method figuring out the implementation type of the pointer member via the rtti_type method. I then use the normal cast method of the python API to (static) cast the value to its implementation type and return this value. This isn't clean, but didn't fail so far.

So, my vote as a user looks something like this:
Prio 1: commit this patch for the rtti_type method (I applied it to the current gdb 7.0 sources and was happy with it). ]
Prio 2: also add the C++ style casts to improve gdb command line experience for C++.
Prio 3: provide the C++ style casts in the Python API to make pretty printers like the one described above clean.

To those of you who celebrate Christmas - merry Christmas and enjoy your holidays. Jens.
 
-----Original Message-----
From: Tom Tromey [mailto:tromey@redhat.com] 
Sent: Freitag, 30. Oktober 2009 18:25
To: Elmenthaler, Jens
Cc: gdb@sourceware.org
Subject: Re: Python pretty printers and dynamic cast

>>>>> "Jens" == Elmenthaler, Jens <jens.elmenthaler@verigy.com> writes:

Jens> Is there any way for a python pretty printer to do (or 'mimic') a
Jens> dynamic cast, such that I can show the members of the
Jens> implementation type where ever I have the interface type?

There is nothing built-in yet.  We've had a few requests for this,
though, and there was a patch posted to the archer list.  If you give it
a try, let us know how it turns out.

Tom


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Python pretty printers and dynamic cast
  2009-12-22  8:15           ` Elmenthaler, Jens
@ 2010-01-05 15:34             ` Tom Tromey
  2010-01-07  8:55               ` Elmenthaler, Jens
  0 siblings, 1 reply; 10+ messages in thread
From: Tom Tromey @ 2010-01-05 15:34 UTC (permalink / raw)
  To: Elmenthaler, Jens; +Cc: gdb

>>>>> "Jens" == Elmenthaler, Jens <jens.elmenthaler@verigy.com> writes:

Tom> There is nothing built-in yet.  We've had a few requests for this,
Tom> though, and there was a patch posted to the archer list.  If you give it
Tom> a try, let us know how it turns out.

[...]
Jens> So, my vote as a user looks something like this:
Jens> Prio 1: commit this patch for the rtti_type method (I applied it
Jens> to the current gdb 7.0 sources and was happy with it). ]

I'm hoping we'll merge a bunch of python stuff "soon".
I'll make sure this is on our list.

Jens> Prio 2: also add the C++ style casts to improve gdb command line
Jens> experience for C++.

FWIW we have a patch to add these casts on the expr-cumulative branch.
I will look into merging that.

Jens> Prio 3: provide the C++ style casts in the Python API to make
Jens> pretty printers like the one described above clean.

Could you file a feature request for this in bugzilla?

Tom


^ permalink raw reply	[flat|nested] 10+ messages in thread

* RE: Python pretty printers and dynamic cast
  2010-01-05 15:34             ` Tom Tromey
@ 2010-01-07  8:55               ` Elmenthaler, Jens
  0 siblings, 0 replies; 10+ messages in thread
From: Elmenthaler, Jens @ 2010-01-07  8:55 UTC (permalink / raw)
  To: Tom Tromey; +Cc: gdb

Jens> Prio 3: provide the C++ style casts in the Python API to make
Jens> pretty printers like the one described above clean.

Tom> Could you file a feature request for this in bugzilla?

Done: Bug 11145

Jens.

-----Original Message-----
From: Tom Tromey [mailto:tromey@redhat.com] 
Sent: Dienstag, 5. Januar 2010 16:34
To: Elmenthaler, Jens
Cc: gdb@sourceware.org
Subject: Re: Python pretty printers and dynamic cast

>>>>> "Jens" == Elmenthaler, Jens <jens.elmenthaler@verigy.com> writes:

Tom> There is nothing built-in yet.  We've had a few requests for this,
Tom> though, and there was a patch posted to the archer list.  If you give it
Tom> a try, let us know how it turns out.

[...]
Jens> So, my vote as a user looks something like this:
Jens> Prio 1: commit this patch for the rtti_type method (I applied it
Jens> to the current gdb 7.0 sources and was happy with it). ]

I'm hoping we'll merge a bunch of python stuff "soon".
I'll make sure this is on our list.

Jens> Prio 2: also add the C++ style casts to improve gdb command line
Jens> experience for C++.

FWIW we have a patch to add these casts on the expr-cumulative branch.
I will look into merging that.

Jens> Prio 3: provide the C++ style casts in the Python API to make
Jens> pretty printers like the one described above clean.

Could you file a feature request for this in bugzilla?

Tom


^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2010-01-07  8:55 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-10-15 14:50 simulator for sparc64? Joel Sherrill
2009-10-28 10:20 ` Doug Evans
2009-10-28 12:58   ` Joel Sherrill
2009-10-28 13:59     ` Jakob Engblom
2009-10-29  0:28       ` Joel Sherrill
2009-10-30 17:25       ` Python pretty printers and dynamic cast Elmenthaler, Jens
2009-10-30 17:51         ` Tom Tromey
2009-12-22  8:15           ` Elmenthaler, Jens
2010-01-05 15:34             ` Tom Tromey
2010-01-07  8:55               ` Elmenthaler, Jens

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox