Re: [PATCH] bisect.c: make show_list() build again

2018-09-02 Thread Christian Couder
On Sun, Sep 2, 2018 at 9:42 AM, Nguyễn Thái Ngọc Duy  wrote:

> In order to stop this from happening again, the function is now
> compiled unconditionally but exits early unless DEBUG_BISECT is
> non-zero.

Thanks for going the extra mile and doing this!

I wonder if we should also try to make the show_list() function part
of the trace_*() functions to make it even more regular. This can be a
separate patch or topic though.


Re: [PATCH] bisect.c: make show_list() build again

2018-09-03 Thread Duy Nguyen
On Sun, Sep 2, 2018 at 9:57 AM Christian Couder
 wrote:
>
> On Sun, Sep 2, 2018 at 9:42 AM, Nguyễn Thái Ngọc Duy  
> wrote:
>
> > In order to stop this from happening again, the function is now
> > compiled unconditionally but exits early unless DEBUG_BISECT is
> > non-zero.
>
> Thanks for going the extra mile and doing this!
>
> I wonder if we should also try to make the show_list() function part
> of the trace_*() functions to make it even more regular. This can be a
> separate patch or topic though.

Yeah that's probably a good idea (though I'm not familiar with
bisect.c enough to take that step).
-- 
Duy


Re: [PATCH] bisect.c: make show_list() build again

2018-09-04 Thread Christian Couder
On Mon, Sep 3, 2018 at 7:31 PM, Duy Nguyen  wrote:
> On Sun, Sep 2, 2018 at 9:57 AM Christian Couder
>  wrote:
>>
>> Thanks for going the extra mile and doing this!
>>
>> I wonder if we should also try to make the show_list() function part
>> of the trace_*() functions to make it even more regular. This can be a
>> separate patch or topic though.
>
> Yeah that's probably a good idea (though I'm not familiar with
> bisect.c enough to take that step).

Maybe this could be a GSoC micro project or a left over bit.