[Rd] Debugging tools and practices in Windows?

[Andre Mikulec]

I have not tried this in several years, but inside the Makefiles of the R source code, the following will replace the (non)debugging flags with debugging flags.


This (still) may or may not work ...


find . -type f -name 'M*k*' -exec sed -i 's/-O3/-O0 -ggdb3/g' {} \;
find . -type f -name 'M*k*' -exec sed -i 's/-O2/-O0 -ggdb3/g' {} \;
find . -type f -name 'M*k*' -exec sed -i 's/-shared/-shared -O0 -ggdb3/g' {} \;

I was sort of fresh at the this debugging when I started gdb debugging and time has passed by, so the recommended (current) 'R on windows' compiler flags can be found in the g++ documentation for the correct g++ used to compile the current R on Windows.

Andre Mikulec
Andre_Mikulec at Hotmail.com


________________________________
From: R-devel <r-devel-bounces at r-project.org> on behalf of Ray Donnelly <rdonnelly at continuum.io>
Sent: Thursday, February 23, 2017 8:37 PM
To: Tomas Kalibera
Cc: r-devel


On Thu, Feb 23, 2017 at 7:49 PM, Tomas Kalibera <tomas.kalibera at gmail.com>
wrote:


> https://cran.r-project.org/bin/windows/base/rw-FAQ.html

cran.r-project.org

dows XP or later, including on 64-bit versions: See Can I use R on 64-bit Windows?.



>
> Tomas
>
>
>
> On 02/23/2017 08:10 PM, Javier Luraschi wrote:
>
>> Right, I'm talking about C code.
>>
>> Do you remember if you had to set specific CFLAGS or other settings to get
>> gdb working? I wasn't able to get gdb() working with the standard build
>> settings.
>>
>> Otherwise, Rprintf() would work for sure.
>>
>> Thank you!
>>
>>
>>
>> On Thu, Feb 23, 2017 at 10:55 AM, Duncan Murdoch <
>> murdoch.duncan at gmail.com>
>> wrote:
>>
>> On 23/02/2017 1:36 PM, Javier Luraschi wrote:
>>>
>>> Hello r-devel, could someone share the tools and practices they use to

>>>>
>>>> For instance, I would like to set a breakpoint in `gl_loadhistory` and
>>>> troubleshoot from there.
>>>>
>>>> You're talking about debugging the C code rather than the R code, I
>>> think.
>>>

t,
>>> I use gdb.  There used to be a front end for it (Insight) that worked in

>>
>> names lots of other front ends, but I haven't tried any of them in

>>>
>>>

(it allows you to debug externally-built executables).


> The other choice is the old fashioned method:  add lots of Rprintf()
>>> statements to the source and recompile.
>>>
>>> Duncan Murdoch
>>>
>>>
>>>         [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-devel at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-devel
R-devel Info Page - ETH Zurich<https://stat.ethz.ch/mailman/listinfo/r-devel>
stat.ethz.ch
Public discussion list on R development. This list is intended for questions and discussion about code development in R. Questions likely to prompt discussion ...



>>
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
R-devel Info Page - ETH Zurich<https://stat.ethz.ch/mailman/listinfo/r-devel>
stat.ethz.ch
Public discussion list on R development. This list is intended for questions and discussion about code development in R. Questions likely to prompt discussion ...



>

        [[alternative HTML version deleted]]

______________________________________________
R-devel at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
R-devel Info Page - ETH Zurich<https://stat.ethz.ch/mailman/listinfo/r-devel>
stat.ethz.ch
Public discussion list on R development. This list is intended for questions and discussion about code development in R. Questions likely to prompt discussion ...




	[[alternative HTML version deleted]]