Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cdt-dev] Upcoming CDT Service Release and OS X Debugging.

     Unfortunately I don't think Apple is going to tell anyone their
"road map". The best thing to do would be for EVERYONE who is impacted
on this list to file a bug report at Apple asking for a newer version
of GDB. We have a perfectly valid reason for the request. Whether or
not Apple choses to do anything is anyone's guess. If someone happens
to go to Apple's World Wide Developers Conference (WWDC) this year
they could track down someone from the Xcode team and try to talk to
them about what is going on. Other than that, there is not a whole lot
that can be done. If CDT wants to support OS X debugging then CDT is
going to somehow have to support Apple's GDB debugger. Whether that is
through DSF or through CDI (for Helios at least) does not matter to
me. It would seem that there are just a few bugs that have patches
available and one could simply update CDT in the 6.0.2 SR with a note
saying that OS X Debugging has been worked on but may not be perfect.
I'll accept that.

   Again. I can support by testing new patches and trying out the
hopefully upcoming DSF-GDB-6.3 code.
_________________________________________________________
Mike Jackson                  mike.jackson@xxxxxxxxxxxxxx


On Tue, Jan 5, 2010 at 5:43 PM, Tobias Hahn <tobias.hahn@xxxxxxxxxxx> wrote:
> My personal experience when I last tried was that CDI had different but less issues than DSF with apple's gdb. The 'cause' for all the trouble on OS X is not CDI support rotting, but rather that apple decided about 5 years ago that it will stick with gdb 6.3.50-20050815 forever, neither merging bugfixes nor enhancements from fsf, and just making it good enough to support XCode. (Edit and continue anyone?)
>
> Otoh, fsf-gdb development has continued over the past 5 years, and so naturally these two gdbs will diverge more and more over time. If I am not mistaken, gdb-DSF support was designed with a current fsf-gdb in mind. So naturally, I would expect more issues with the more modern DSF, and less with the legacy CDI. Not to discourage anyone from starting to support apple-gdb-DSF. It is the right way to go IMHO, but it definitely won't be for free.
>
> Just my 2ct,
> Tobias
>
>
> Am 05.01.2010 um 22:20 schrieb Marc Khouzam:
>
>> Maybe, with a little luck, those issues won't be present when using DSF-GDB.
>> But you'll have to give it a try...
>>
>>> -----Original Message-----
>>> From: cdt-dev-bounces@xxxxxxxxxxx
>>> [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Michael Jackson
>>> Sent: January-05-10 12:59 PM
>>> To: CDT General developers list.
>>> Subject: Re: [cdt-dev] Upcoming CDT Service Release and OS X
>>> Debugging.
>>>
>>> So, How hard is it going to be to "port" a CDI patch to the
>>> DSF Framework? My initial thought is that all the hard work
>>> that went into creating those patches will have to be redone.
>>> It took long enough to get those in the first place.
>>>
>>>   I know no one is working on CDI but is it possible to get
>>> the patches applied for the 6.0.2 service release? I guess
>>> the other way is to temporarily "fork" until Helios is
>>> release with (hopefully) fixing debugging. I would really
>>> hate to do that but the OS X Users are really working with an
>>> inferior product at this point and the possibility of having
>>> something so close to working properly is just killing me.
>>>
>>> _________________________________________________________
>>> Mike Jackson                  mike.jackson@xxxxxxxxxxxxxx
>>>
>>>
>>> On Jan 5, 2010, at 12:40 PM, Marc Khouzam wrote:
>>>
>>>> Hi,
>>>>
>>>> Mike is right that we agreed on a way forward for debugging
>>> with OS X.
>>>> Here is what we agreed on.
>>>>
>>>> First, fixing things for CDI is tricky because there is not
>>> really any
>>>> active committers working on CDI.  This is not just for OS
>>> X, but for
>>>> CDI in general.
>>>>
>>>> So, to get OS X debugging working, we'd like to help the community
>>>> move away from CDI and towards DSF-GDB.  We hope that if
>>> patches are
>>>> written for DSF-GDB, they will have a better chance of being
>>>> committed.
>>>> To make this easier still, Pawel had the good idea of
>>> creating a set
>>>> of services specific to Apple's version of GDB.  This will allow to
>>>> commit patches specific to OS X without affecting other GDB
>>> versions.
>>>> I have opened http://bugs.eclipse.org/298883 to track that
>>> effort.
>>>> I will
>>>> try to have a patch ready this week.
>>>>
>>>> Once that is ready, it will be up to the interested members of the
>>>> community to try debugging with DSF-GDB and OS X and
>>> getting patches
>>>> written.
>>>>
>>>> Let's try to have DSF-GDB debugging properly on OS X for Helios.
>>>>
>>>> Marc
>>>>
>>>>
>>>>> -----Original Message-----
>>>>> From: cdt-dev-bounces@xxxxxxxxxxx
>>>>> [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Michael Jackson
>>>>> Sent: January-05-10 12:14 PM
>>>>> To: CDT General developers list.
>>>>> Subject: [cdt-dev] Upcoming CDT Service Release and OS X Debugging.
>>>>>
>>>>>
>>>>> From the CDT conference call today it seems that there is a plan
>>>>> going forward to shape up the OS X Debugging experience.
>>>>> I'll let someone else run through the details as my notes
>>> are shaky
>>>>> at best.
>>>>> But what I did take away from the call is that the
>>> implementation may
>>>>> take some time? With the next service release coming up I would be
>>>>> curious if the non-DSF patches could be committed so at
>>> least some of
>>>>> the major debugging "Bugs" are fixed for the next service release?
>>>>>
>>>>>  Thoughts and comments? The slow start up of the debugger and the
>>>>> thread-info seem like pretty low hanging fruit.
>>>>>
>>>>>  Also, from the call it sounds like some of the issues with
>>>>> supporting OS X is not lack of hardware (someone on the call
>>>>> mentioned they use CDT on their Mac) but the lack of
>>> Developer time
>>>>> to check patches, review code and/or actually debug
>>> issues? Is that a
>>>>> fair statement?
>>>>>
>>>>> _________________________________________________________
>>>>> Mike Jackson                  mike.jackson@xxxxxxxxxxxxxx
>>>>> BlueQuartz Software                    www.bluequartz.net
>>>>> Principal Software Engineer                  Dayton, Ohio
>>>>>
>>>>> _______________________________________________
>>>>> cdt-dev mailing list
>>>>> cdt-dev@xxxxxxxxxxx
>>>>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>>>>> _______________________________________________
>>>> cdt-dev mailing list
>>>> cdt-dev@xxxxxxxxxxx
>>>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>>>
>>> _______________________________________________
>>> cdt-dev mailing list
>>> cdt-dev@xxxxxxxxxxx
>>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>>> _______________________________________________
>> cdt-dev mailing list
>> cdt-dev@xxxxxxxxxxx
>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>
> _______________________________________________
> cdt-dev mailing list
> cdt-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>


Back to the top