4 messages · Simon Frost, Brian Ripley
Hi Folks, I compiled and installed Assist-0.1.0; I'm using R-2.0.0 running on SuSe Linux, on a dual AMD64 machine. However, I (apparently randomly) get segmentation faults when I use cubic2. Any ideas? Thanks Simon
Simon D.W. Frost, MA DPhil Adjunct Assistant Professor Department of Pathology University of California, San Diego UCSD Antiviral Research Center 150 W. Washington St. San Diego, CA 92103 USA Tel: +1 619 543 8080 Ext 275 Fax: +1 619 298 0177 Email: sdfrost at ucsd.edu
What is Assist-0.1.0? It's not part of R and it is not a CRAN nor Bioconductor package. There is a CRAN package `assist', and if you mean that please follow the advice in the posting guide and contact the maintainer. Now, that package says Title: A Suite of S-Plus Functions Implementing Smoothing Splines and S-PLUS functions don't run on 64-bit versions of R. I suggest you and the maintainer look for the use of `long *' in .C calls: S-PLUS uses `long *' where R uses `int *'.
I compiled and installed Assist-0.1.0; I'm using R-2.0.0 running on SuSe Linux, on a dual AMD64 machine. However, I (apparently randomly) get segmentation faults when I use cubic2. Any ideas?
Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595
Dear Brian, Apologies;I do mean 'assist' rather than 'Assist'; there are versions for R and for SPlus. I doubt whether it's to do with calling conventions; the functions sometimes work, sometimes they don't, and cause a segmentation fault. Thanks Simon
Simon D.W. Frost, MA DPhil Adjunct Assistant Professor Department of Pathology University of California, San Diego UCSD Antiviral Research Center 150 W. Washington St. San Diego, CA 92103 USA Tel: +1 619 543 8080 Ext 275 Fax: +1 619 298 0177 Email: sdfrost at ucsd.edu
R CMD check assist segfaults on AMD64 unless long is replaced by int in ker.c. If it is, it passes R CMD check. Next time please do not dismiss informed advice out of hand, but try it for yourself. Chunlei Ke: please update your package.
Dear Brian, Apologies;I do mean 'assist' rather than 'Assist'; there are versions for R and for SPlus. I doubt whether it's to do with calling conventions; the functions sometimes work, sometimes they don't, and cause a segmentation fault. Thanks Simon
Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595