status update for GDB 7.7?

status update for GDB 7.7?

[Joel Brobecker]

Hello all,

I thought it would be useful to do a quick status update for the 7.7
release. At this point, we have two items left on the TODO list:

 * [PhilippeW] "catch syscall" with gdbserver
   https://sourceware.org/ml/gdb-patches/2013-08/msg00922.html

 * [Jan/Doug] Record objfile->original_name as an absolute path
   https://sourceware.org/ml/gdb-patches/2013-10/msg00275.html

Hard to tell whether they are done or not. Can we have confirmation
either way?

For myself, strictly not necessary, but I'd like to have:

   [Joel] [GDB/MI] Help determine if GDB/MI command exists or not
   http://www.sourceware.org/ml/gdb-patches/2013-11/msg00476.html

   I've just sent a ping.

There may be:

   Fix PR 16201: internal error on a cygwin program linked against a DLL with no .data section
   http://www.sourceware.org/ml/gdb-patches/2013-11/msg00865.html

Anything else?

Thank you,
-- 
Joel

Re: status update for GDB 7.7?

[Jan Kratochvil]

On Mon, 02 Dec 2013 04:56:37 +0100, Joel Brobecker wrote:
>  * [Jan/Doug] Record objfile->original_name as an absolute path
>    https://sourceware.org/ml/gdb-patches/2013-10/msg00275.html
> 
> Hard to tell whether they are done or not. Can we have confirmation
> either way?

It is not checked in.  The patch is IMO done in the link above.

Just Doug required some whole new feature/patchset just because of the
testcase there and currently I work on other more urgent things:
	https://sourceware.org/ml/gdb-patches/2013-10/msg00287.html


Jan

Re: status update for GDB 7.7?

[Joel Brobecker]

> It is not checked in.  The patch is IMO done in the link above.
> 
> Just Doug required some whole new feature/patchset just because of the
> testcase there and currently I work on other more urgent things:
> 	https://sourceware.org/ml/gdb-patches/2013-10/msg00287.html

OK. So what do you guys want to do with respect to the release?
Some possible options:

  (1) Defer the patch for the next release, hoping there will be
      more time to look into the new ideas?

  (2) Or defer the release while waiting for someone to find that
      time?

  (3) Commit the patch as is, and defer the new improvement for
      later?

Any other suggestions? From my vantage point, it sounds like we have
let, in this thread, best be the enemy of good. IIUC, the patch is
OK, but blocked because we could do better in the testcase? Or did
a small issue in the testcase raise a bigger issue?

-- 
Joel

Re: status update for GDB 7.7?

[Jan Kratochvil]

On Mon, 02 Dec 2013 08:49:20 +0100, Joel Brobecker wrote:
>   (1) Defer the patch for the next release, hoping there will be
>       more time to look into the new ideas?
> 
>   (2) Or defer the release while waiting for someone to find that
>       time?
> 
>   (3) Commit the patch as is, and defer the new improvement for
>       later?

also possibly:
(4) Commit the fix and defer the testcase for improvement later.


> IIUC, the patch is
> OK, but blocked because we could do better in the testcase?

Yes.


> Or did a small issue in the testcase raise a bigger issue?

It raised a bigger issue but completely unrelated to the fix under test.


Regards,
Jan

Re: status update for GDB 7.7?

[Joel Brobecker]

> also possibly:
> (4) Commit the fix and defer the testcase for improvement later.

Works for me! I'll let you guys decide which option you would prefer.

-- 
Joel

Re: status update for GDB 7.7?

[Doug Evans]

Great.  gmail has started off my return from holidays by sending rich text.
Apologies for the resend.

On Sun, Dec 1, 2013 at 11:36 PM, Jan Kratochvil
<jan.kratochvil@redhat.com> wrote:
> On Mon, 02 Dec 2013 04:56:37 +0100, Joel Brobecker wrote:
>>  * [Jan/Doug] Record objfile->original_name as an absolute path
>>    https://sourceware.org/ml/gdb-patches/2013-10/msg00275.html
>>
>> Hard to tell whether they are done or not. Can we have confirmation
>> either way?
>
> It is not checked in.  The patch is IMO done in the link above.
>
> Just Doug required some whole new feature/patchset just because of the
> testcase there and currently I work on other more urgent things:
>         https://sourceware.org/ml/gdb-patches/2013-10/msg00287.html

Umm,    eh?

What whole new feature/patchset am I *requiring*?
[hint: there is none :-)]
The subsequent discussion was just discussion of approaching the
problem differently that could have a more general use.
I never argued it was a requirement.

If people are ok with that test, then it "works for me".
I just wanted confirmation.
[Joel did reply it was ok with him, modulo seeing if there was a way
to improve it,
but he was the only one to offer an opinion on the test itself.
It could be changed a bit I think, but I'm not sure it would remove
the cheezyness factor, to me anyway.]

Re: status update for GDB 7.7?

[Jan Kratochvil]

On Mon, 02 Dec 2013 20:43:43 +0100, Doug Evans wrote:
> What whole new feature/patchset am I *requiring*?
> [hint: there is none :-)]

The discussion started with:

On Thu, 10 Oct 2013 18:31:22 +0200, Doug Evans wrote:
# I was thinking,
# We have auto-solib-add and sharedlibrary.
# We just need something like that for symfile_objfile.


> The subsequent discussion was just discussion of approaching the
> problem differently that could have a more general use.
> I never argued it was a requirement.

OK, so it could be a misunderstanding from me.


> If people are ok with that test, then it "works for me".

It seems so to me, therefore checked it in now.


Regards,
Jan

Re: status update for GDB 7.7?

[Joel Brobecker]

> > If people are ok with that test, then it "works for me".
> 
> It seems so to me, therefore checked it in now.

Great to see the misunderstanding cleared. Thanks everyone!

-- 
Joel

Re: status update for GDB 7.7?

[Philippe Waroquiers]

On Mon, 2013-12-02 at 07:56 +0400, Joel Brobecker wrote:

> I thought it would be useful to do a quick status update for the 7.7
> release. At this point, we have two items left on the TODO list:
> 
>  * [PhilippeW] "catch syscall" with gdbserver
>    https://sourceware.org/ml/gdb-patches/2013-08/msg00922.html

This will not be done for 7.7.

Several comments were given on the last version, a.o.
a suggestion to switch from a specific QCatchSyscall packet to
a more general packet QInsertCatchPoint.
I last discussed this in 
  https://sourceware.org/ml/gdb-patches/2013-10/msg00321.html
but did not progress after that.

NB: feedback on my last message welcome :).

Philippe

Re: status update for GDB 7.7?

[Joel Brobecker]

> On Mon, 2013-12-02 at 07:56 +0400, Joel Brobecker wrote:
> 
> > I thought it would be useful to do a quick status update for the 7.7
> > release. At this point, we have two items left on the TODO list:
> > 
> >  * [PhilippeW] "catch syscall" with gdbserver
> >    https://sourceware.org/ml/gdb-patches/2013-08/msg00922.html
> 
> This will not be done for 7.7.
> 
> Several comments were given on the last version, a.o.
> a suggestion to switch from a specific QCatchSyscall packet to
> a more general packet QInsertCatchPoint.
> I last discussed this in 
>   https://sourceware.org/ml/gdb-patches/2013-10/msg00321.html
> but did not progress after that.

Thanks for the update, Philippe. I was very sorry to hear that.
I've moved your entry in the release wiki page to "Excluded",
with a link to this email to explain the reason.

> NB: feedback on my last message welcome :).

:-). Generally speaking, feel free to send a ping 2 weeks after
first submission, and then every week thereafter. I think it's
acceptable to ping a single week after first submission, but
I personally seem incapable of doing it - it feels like it's not
giving the maintainers enough time to breathe... In any case,
I am sure you are well past the 2 weeks delay ;-).

-- 
Joel