[R-pkg-devel] Managing RNG in C code
Hello, I am working on a package where I am passing some computations from R to C and back. The C code runs many stochastic trajectories of a population model (with the RNG from R through multiple calls to GetRNGstate and PutRNGstate). I would like that each trajectory has its own seed. My problem is that the guidelines to develop R packages write in the section '6.3 Random number generation' that "The random number generator is private to R; there is no way to select the kind of RNG or set the seed except by evaluating calls to the R functions." The only way I see to get around this is to set the seed in R, run 1 trajectory in C, get results back in R, increment the seed in R, run a 2nd trajectory in C with this new seed, add the new results to the previous one in R, and repeat this e.g. 1000 times. Does this sound fine or should I be concerned that calling the C library so many times may slow down the computation (even if the library is only loaded 1 time at the beginning), e.g. is there a substantial overhead in interfacing from R to C and back? Alternatively, is there a hack to get access by force to the private RNG in R from C and set its seed? This way, I could call the C code only 1 time and run the 1000 trajectories within C, updating the seed in R for each new trajectory (this is what I did when I previously used the GSL). Thanks for any advice on this! Guillaume __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
[R-pkg-devel] Multi-threaded C with R's RNG
Hello, Has anyone written a package with a C code that uses R's RNG but in a multi-threaded way (e.g. to parallelize Monte Carlo simulations)? I do not have the knowledge to figure out myself how to solved this and would need to see an example. Thanks! Guillaume -- Guillaume Chapron, PhD Associate Professor Grimsö Wildlife Research Station Swedish University of Agricultural Sciences SE - 73091 Riddarhyttan, Sweden http://www.carnivore.science/ Twitter @CarnivoreSci __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
[R-pkg-devel] Rprintf with mclapply
Hello, I have a R package with a compiled C executable to run simulations. Simulations can be very long so I use mclapply to launch several instances (each with a different R seed). I would need to follow the simulation progress and I used to do that with printf in the C code. But this is not allowed on CRAN and I have to use Rprintf. The problem is that with mclapply, Rprintf prints nothing in the R console! Any idea how can I fix this? Thanks! Guillaume __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
[R-pkg-devel] scripting R through lldb
Hello,
I am debugging a package that contains some compiled C code. The C code uses R
random number generator so I cannot (or have not figured out how to) debug it
without running it through R. What I do is to type the following in the
terminal:
R --debugger=lldb
then in the lldb prompt
r
and within R
source("~/Desktop/test.R”)
which is the file calling the C library. When the library crashes I can
navigate the C memory to see where is the problem.
My question is whether there is a way to have this in single line or script to
run from the terminal. I have looked on stackoverflow but could not find.
Many thanks
Guillaume
__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
Re: [R-pkg-devel] scripting R through lldb
Fantastic, it works perfectly, many thanks!
Guillaume
> On 15 Jun 2016, at 21:49, Kevin Ushey wrote:
>
> Hi Guillaume,
>
> I have a script that does this (for OS X) here, taking advantage of
> the new '--batch' argument to lldb:
>
>https://github.com/kevinushey/etc/blob/master/lang/r/mac/bin/r-lldb
>
> Note that, on recent versions of OS X due to system integrity
> protection, this may not work with the system lldb, as it appears
> there are restrictions on how certain environment variables, e.g.
> DYLD_FALLBACK_LIBRARY_PATH, can be set for system executables. (If you
> are indeed on OS X, you can try building your own version of lldb --
> see e.g.
> https://github.com/kevinushey/etc/blob/master/platform/mac/install-llvm.sh
> for that)
>
> In the end, you should be able to use this script as e.g.
>
>r-lldb -f test.R
>
> and this will launch lldb, set R as the target process, with '-f
> test.R' passed to the R executable.
>
> Depending on your OS / configuration, you might need to tweak the
> script, but it will hopefully either 'work out of the box' or at least
> serve as a starting point for your version.
>
> Best,
> Kevin
>
> On Wed, Jun 15, 2016 at 12:39 PM, Guillaume Chapron
> wrote:
>> Hello,
>>
>> I am debugging a package that contains some compiled C code. The C code uses
>> R random number generator so I cannot (or have not figured out how to) debug
>> it without running it through R. What I do is to type the following in the
>> terminal:
>>
>> R --debugger=lldb
>>
>> then in the lldb prompt
>>
>> r
>>
>> and within R
>>
>> source("~/Desktop/test.R”)
>>
>> which is the file calling the C library. When the library crashes I can
>> navigate the C memory to see where is the problem.
>>
>> My question is whether there is a way to have this in single line or script
>> to run from the terminal. I have looked on stackoverflow but could not find.
>>
>> Many thanks
>>
>> Guillaume
>> __
>> R-package-devel@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-package-devel
__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
[R-pkg-devel] Compiler optimization flags with R package
Hello, I am trying to debug some C code that is run only through a R package. I debug with lldb but I always get a message telling that the package "was compiled with optimization - stepping may behave oddly; variables may not be available.” And in fact, I cannot evaluate all variables and understand what is wrong. I have created a Makevars file in ~/.R/ that contains C=clang -O0 -g but it does not seem to change much. What should I do to make sure I can evaluate all variables in lldb? I am on a Mac and I wrote C=clang because I read that it produced much better error messages but I am happy to use gcc as well if this can be a solution. Many thanks Guillaume __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
Re: [R-pkg-devel] Compiler optimization flags with R package
Thanks!
In that file, I see:
CC = clang
CFLAGS = -Wall -mtune=core2 -g -O2 $(LTO)
will having -O0 instead give me all the variables or is there a flag for better
debugging (I guess -g does it a little already)?
Thanks again
> On 17 Jun 2017, at 21:57, Dirk Eddelbuettel wrote:
>
>
> On 17 June 2017 at 21:32, Guillaume Chapron wrote:
> | I am trying to debug some C code that is run only through a R package. I
> debug with lldb but I always get a message telling that the package "was
> compiled with optimization - stepping may behave oddly; variables may not be
> available.” And in fact, I cannot evaluate all variables and understand what
> is wrong. I have created a Makevars file in ~/.R/ that contains C=clang -O0
> -g but it does not seem to change much. What should I do to make sure I can
> evaluate all variables in lldb? I am on a Mac and I wrote C=clang because I
> read that it produced much better error messages but I am happy to use gcc as
> well if this can be a solution.
>
> Edit the file Makeconf in e.g.
>
>R> file.path(Sys.getenv("R_HOME"), "etc", "Makeconf")
>[1] "/usr/lib/R/etc/Makeconf"
>R>
>
> It has those settings hardwired from when R itself was compiled for you. You
> probably want to keep a copy of the original file to be able to revert.
>
> Dirk
>
> --
> http://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org
__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
[R-pkg-devel] R package with Swift code
Hello, Does anyone know of an example of a R package that contains Swift code (https://swift.org), like many packages use C code? Thanks Guillaume __ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel
