bug report being ignored

bug report being ignored

[echancrure]

Hi all,
I don't want to be flippant or anything (I am just ignorant of the gcc/
gnat process/effort involved), but I filled a bug report for gnat on
bugzilla #35050 ( http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35050 )
on the first of February and so far it has remained unconfirmed, never
mind looked into in view to be fixed.

It's not always easy to ask question like this without offending gcc/
gnat people who are doing a tremendous job.

It is an annoying bug for me as it has an impact on gnatxref that I
use a lot.

Is there anything that I can do to try to get that bug fixed???
Am I doing something wrong?

regards,
Chris

Re: bug report being ignored

[Samuel Tardieu]

>>>>> "Chris" == echancrure  <echancrure@gmail.com> writes:

Chris> Hi all, I don't want to be flippant or anything (I am just
Chris> ignorant of the gcc/ gnat process/effort involved), but I
Chris> filled a bug report for gnat on bugzilla #35050 (
Chris> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35050 ) on the
Chris> first of February and so far it has remained unconfirmed, never
Chris> mind looked into in view to be fixed.

Chris> It's not always easy to ask question like this without
Chris> offending gcc/ gnat people who are doing a tremendous job.

Chris> It is an annoying bug for me as it has an impact on gnatxref
Chris> that I use a lot.

Chris> Is there anything that I can do to try to get that bug fixed???
Chris> Am I doing something wrong?

As far as I'm concerned, I take no offense, but I will try to explain
the current situation of bug reports against the FSF GNAT tree.

You're not doing anything wrong... except that you expect non-paid
volunteers to do the job in a short time window. As one of the (very
few) people who fix GNAT bugs in the FSF version, I choose the issues
to address according to:

  - the time I have (I do this on my free time, and I have a full-time
    job and many other computer-related and computer-unrelated
    activities);

  - the tools I use (I've never needed gnatxref for my Ada
    development);

  - the severity of the bugs according to my own ratings (an internal
    compiler error is high on the list, especially when it is
    triggered by valid real-world Ada code, a problem on a
    specially-crafted example concerning an obscure RM rule is low on
    my list).

The main issue is the first one I describe, i.e. the time to work on
GNAT. A few months ago, I started fixing the bugs one after another,
in the hope that more people would join the effort. Some of them (in
particular Ludovic Brenta and Duncan Sands) do a tremendous job for
the community, but we really lack manpower. AdaCore is also very
helpful with volunteers working on the public version of GCC; we
*need* to find more people to work on FSF GNAT.

So unless someone gets excited by your bug report, you would better
try to fix it yourself if you have the needed expertise and time, or
wait until it gets fixed by AdaCore because one of their clients has
reported it.

The main point is: when people work on Free Software as a hobby, they
will set their own agenda. And when there aren't many people, not many
topics are addressed at once.

Anyway, thanks for your bug report. Properly reporting issues is already
an important way to contribute back to Free Software you use.

  Sam
-- 
Samuel Tardieu -- sam@rfc1149.net -- http://www.rfc1149.net/

Re: bug report being ignored

[echancrure]

On Feb 25, 10:33 pm, Samuel Tardieu <s...@rfc1149.net> wrote:
> >>>>> "Chris" == echancrure  <echancr...@gmail.com> writes:
>
> Chris> Hi all, I don't want to be flippant or anything (I am just
> Chris> ignorant of the gcc/ gnat process/effort involved), but I
> Chris> filled a bug report for gnat on bugzilla #35050 (
> Chris>http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35050) on the
> Chris> first of February and so far it has remained unconfirmed, never
> Chris> mind looked into in view to be fixed.
>
> Chris> It's not always easy to ask question like this without
> Chris> offending gcc/ gnat people who are doing a tremendous job.
>
> Chris> It is an annoying bug for me as it has an impact on gnatxref
> Chris> that I use a lot.
>
> Chris> Is there anything that I can do to try to get that bug fixed???
> Chris> Am I doing something wrong?
>
> As far as I'm concerned, I take no offense, but I will try to explain
> the current situation of bug reports against the FSF GNAT tree.
>
> You're not doing anything wrong... except that you expect non-paid
> volunteers to do the job in a short time window. As one of the (very
> few) people who fix GNAT bugs in the FSF version, I choose the issues
> to address according to:
>
>   - the time I have (I do this on my free time, and I have a full-time
>     job and many other computer-related and computer-unrelated
>     activities);
>
>   - the tools I use (I've never needed gnatxref for my Ada
>     development);
>
>   - the severity of the bugs according to my own ratings (an internal
>     compiler error is high on the list, especially when it is
>     triggered by valid real-world Ada code, a problem on a
>     specially-crafted example concerning an obscure RM rule is low on
>     my list).
>
> The main issue is the first one I describe, i.e. the time to work on
> GNAT. A few months ago, I started fixing the bugs one after another,
> in the hope that more people would join the effort. Some of them (in
> particular Ludovic Brenta and Duncan Sands) do a tremendous job for
> the community, but we really lack manpower. AdaCore is also very
> helpful with volunteers working on the public version of GCC; we
> *need* to find more people to work on FSF GNAT.
>
> So unless someone gets excited by your bug report, you would better
> try to fix it yourself if you have the needed expertise and time, or
> wait until it gets fixed by AdaCore because one of their clients has
> reported it.
>
> The main point is: when people work on Free Software as a hobby, they
> will set their own agenda. And when there aren't many people, not many
> topics are addressed at once.
>
> Anyway, thanks for your bug report. Properly reporting issues is already
> an important way to contribute back to Free Software you use.
>
>   Sam
> --
> Samuel Tardieu -- s...@rfc1149.net --http://www.rfc1149.net/

Thank you Sam for your clarifications, I understand the process a
little better.
I appreciate your effort (I have seen it at work previously) and that
of others.

I unfortunately lack the expertise and the time to fix gnat; maybe
someday.

regards,
chris

Re: bug report being ignored

[Tero Koskinen]

On Mon, 25 Feb 2008 23:33:32 +0100 Samuel Tardieu wrote:

> >>>>> "Chris" == echancrure  <echancrure@gmail.com> writes:
> 
> Chris> Hi all, I don't want to be flippant or anything (I am just
> Chris> ignorant of the gcc/ gnat process/effort involved), but I
> Chris> filled a bug report for gnat on bugzilla #35050 (
> 
> Chris> Is there anything that I can do to try to get that bug fixed???
> Chris> Am I doing something wrong?
> 
> So unless someone gets excited by your bug report,

I looked at the problem and probably found the location of the bug from
the GCC/GNAT sources.

Could you try a patch at
http://iki.fi/tero.koskinen/gcc/gcc-ada-xref_lib_adb.diff
and report if the patch solves the problem for you (and doesn't introduce
any new problems)?

The patch is made against gcc-trunk revision 131693, but probably
works for almost any gcc 4.x.x version.

-- 
Tero Koskinen - http://iki.fi/tero.koskinen/

Re: bug report being ignored

[Samuel Tardieu]

>>>>> "Tero" == Tero Koskinen <tero.koskinen@iki.fi> writes:

Tero> I looked at the problem and probably found the location of the
Tero> bug from the GCC/GNAT sources.

Tero> Could you try a patch at
Tero> http://iki.fi/tero.koskinen/gcc/gcc-ada-xref_lib_adb.diff and
Tero> report if the patch solves the problem for you (and doesn't
Tero> introduce any new problems)?

I submitted an identical patch three days ago:

  http://thread.gmane.org/gmane.comp.gcc.patches/157295

I'm waiting for its review before checking it in GCC sources.

As a general rule, when a bug in the GCC bugzilla is marked as being
assigned to someone (as this one was), it is better to contact the
assignee to avoid work duplication.

But thanks anyway, I hope to see new patches from you in the future!

  Sam
-- 
Samuel Tardieu -- sam@rfc1149.net -- http://www.rfc1149.net/