New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Installation error #8
Comments
This is the error? |
make: *** [glm.o] Error 1 This is the only one that comes up |
Same error |
Hello, first of all thanks a lot for the package I was always frustrated how some very simple function was so unoptimised in R. I have the same issue on centos 7 with r version 3.6.
Rcpp version is 1.03 I actually suspect that it is due to the gcc compiler version (4.8.5 in my case) |
The same error when I use gcc 5.2.0. Then I installed another gcc version, ie gcc 6.4.0. The error had gone. |
To narrow in on the gcc requirements: I can confirm that Rfast requires something newer than gcc 5.3.1. I got it to work gcc 6.3.1 (see next comment). Those are the only versions I could test with. Maybe you could add a note on this in the DetailsFWIW, with gcc 4.8.3 (comes with CentOS 7) and gcc 5.3.1 (comes with SCL 'devtoolset-4'), the essential compilation error is: g++ -std=gnu++11 -I"/wynton/home/cbi/shared/software/CBI/R-3.6.2/lib64/R/include" -DNDEBUG -I../inst/include -I"/wyn
ton/home/cbi/hb/R/x86_64-pc-linux-gnu-library/3.6-CBI-revdepcheck/Rcpp/include" -I"/wynton/home/cbi/hb/R/x86_64-pc-l
inux-gnu-library/3.6-CBI-revdepcheck/RcppArmadillo/include" -I/usr/local/include -fopenmp -fpic -g -O2 -c glm.cpp
-o glm.o
glm.cpp: In function ‘Rcpp::List glm_logistic(Rcpp::NumericMatrix, Rcpp::NumericVector, double, int)’:
glm.cpp:60:21: error: call of overloaded ‘isinf(double&)’ is ambiguous
if ( isinf(negLL) || lambda[0]*ta/2 < tol || prevNegLL - negLL < lltol) {
In file included from /usr/include/features.h:375:0,
from /opt/rh/devtoolset-4/root/usr/include/c++/5.3.1/x86_64-redhat-linux/bits/os_defines.h:39,
from /opt/rh/devtoolset-4/root/usr/include/c++/5.3.1/x86_64-redhat-linux/bits/c++config.h:2255,
from /opt/rh/devtoolset-4/root/usr/include/c++/5.3.1/cmath:41,
from /wynton/home/cbi/hb/R/x86_64-pc-linux-gnu-library/3.6-CBI-revdepcheck/Rcpp/include/Rcpp/platfo
rm/compiler.h:100,
from /wynton/home/cbi/hb/R/x86_64-pc-linux-gnu-library/3.6-CBI-revdepcheck/Rcpp/include/Rcpp/r/head
ers.h:67,
from /wynton/home/cbi/hb/R/x86_64-pc-linux-gnu-library/3.6-CBI-revdepcheck/Rcpp/include/RcppCommon.
h:29,
from /wynton/home/cbi/hb/R/x86_64-pc-linux-gnu-library/3.6-CBI-revdepcheck/RcppArmadillo/include/Rc
ppArmadilloForward.h:26,
from /wynton/home/cbi/hb/R/x86_64-pc-linux-gnu-library/3.6-CBI-revdepcheck/RcppArmadillo/include/Rc
ppArmadillo.h:31,
from glm.cpp:3:
/usr/include/bits/mathcalls.h:202:1: note: candidate: int isinf(double)
__MATHDECL_1 (int,isinf,, (_Mdouble_ __value)) __attribute__ ((__const__));
... |
RedHat/CentOS users: If you're on RHEL/CentOS and have CentOS Software Collections (SCL) installed - check with ## Too old
$ gcc --version | head -1
gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39)
## Use a less old gcc version
$ scl enable devtoolset-6 bash
$ gcc --version |head -1
gcc (GCC) 6.3.1 20170216 (Red Hat 6.3.1-3)
## Now we can compile Rfast
$ Rscript -e "install.packages('Rfast')"
* installing *source* package ‘Rfast’ ...
** package ‘Rfast’ successfully unpacked and MD5 sums checked
** using staged installation
...
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (Rfast)
## Done; disable the SCL again
$ exit
## Back too old setup, ...
$ gcc --version | head -1
gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39)
## ... but it doesn't matter when it comes using Rfast
$ Rscript -e "packageVersion('Rfast')" -e "loadNamespace('Rfast')"
[1] ‘1.9.8’
<environment: namespace:Rfast>
$ |
The error has been fixed. |
Unable to install the package:
make: *** [glm.o] Error 1
ERROR: compilation failed for package ‘Rfast’
The text was updated successfully, but these errors were encountered: