Ben, it's most like what Peter says. I can confirm it works; I just installed https://cran.r-project.org/src/base-prerelease/R-latest.tar.gz on an up-to-date CentOS 7.9.2009 system using the vanilla gcc (GCC) 4.8.5 that comes with that version and R compiles just fine and it passes 'make check' too. Since R is trying to move toward C++14 support by default, I agree with I?aki, you might wanr to build and run R with a newer version of gcc. gcc 4.8.5 will only give you C++11 support. RedHat's Software Collections (SCL) devtoolset:s is the easiest way to do this. I've done this too and can confirm that gcc 7.3.1 that comes with SCL devtoolset/7 is sufficient to get C++14 support. I'm sharing my installation with lots of users, so I'm make it all transparent to the end-user with environment modules, i.e. 'module load r/4.1.0' is all the user needs to know. /Henrik On Thu, Apr 29, 2021 at 7:28 AM Peter Dalgaard <pdalgd at gmail.com> wrote:> > You may want to check out your checkout.... > > I see: > > Peter-Dalgaards-iMac:R pd$ grep newsock src/main/connections.c > con = R_newsock(host, port, server, serverfd, open, timeout, options); > > but your file seems to have lost the ", options" bit somehow. Also, mine is line 3488, not 3477. > > Maybe you have an old file getting in the way? > > - Peter > > > On 29 Apr 2021, at 15:58 , Ben Bolker <bbolker at gmail.com> wrote: > > > > I probably don't want to go down this rabbit hole very far, but if anyone has any *quick* ideas ... > > > > Attempting to build R from scratch with a fresh SVN checkout on a somewhat out-of-date CentOS system (for which I don't have root access, although I can bug people if I care enough). > > > > ../r-devel/configure; make > > > > ends with > > > > gcc -std=gnu99 -I../../../r-devel/trunk/src/extra -I. -I../../src/include -I../../../r-devel/trunk/src/include -I/usr/local/include -I../../../r-devel/trunk/src/nmath -DHAVE_CONFIG_H -fopenmp -g -O2 -c ../../../r-devel/trunk/src/main/connections.c -o connections.o > > ../../../r-devel/trunk/src/main/connections.c: In function ?do_sockconn?: > > ../../../r-devel/trunk/src/main/connections.c:3477:5: error: too few arguments to function ?R_newsock? > > con = R_newsock(host, port, server, serverfd, open, timeout); > > ^ > > In file included from ../../../r-devel/trunk/src/main/connections.c:80:0: > > ../../../r-devel/trunk/src/include/Rconnections.h:83:13: note: declared here > > Rconnection R_newsock(const char *host, int port, int server, int serverfd, const char * const mode, int timeout, int options); > > ^ > > make[3]: *** [connections.o] Error 1 > > > > Any suggestions for a quick fix/diagnosis? > > > > cheers > > Ben Bolker > > > > ===> > > > > > $ gcc --version > > gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39) > > > > $ lsb_release -a > > LSB Version: :core-4.1-amd64:core-4.1-noarch:cxx-4.1-amd64:cxx-4.1-noarch:desktop-4.1-amd64:desktop-4.1-noarch:languages-4.1-amd64:languages-4.1-noarch:printing-4.1-amd64:printing-4.1-noarch > > Distributor ID: CentOS > > Description: CentOS Linux release 7.8.2003 (Core) > > Release: 7.8.2003 > > Codename: Core > > > > ______________________________________________ > > R-devel at r-project.org mailing list > > https://stat.ethz.ch/mailman/listinfo/r-devel > > -- > Peter Dalgaard, Professor, > Center for Statistics, Copenhagen Business School > Solbjerg Plads 3, 2000 Frederiksberg, Denmark > Phone: (+45)38153501 > Office: A 4.23 > Email: pd.mes at cbs.dk Priv: PDalgd at gmail.com > > ______________________________________________ > R-devel at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel
On Sat, 1 May 2021 at 03:41, Henrik Bengtsson <henrik.bengtsson at gmail.com> wrote:> > Ben, it's most like what Peter says. I can confirm it works; I just > installed https://cran.r-project.org/src/base-prerelease/R-latest.tar.gz > on an up-to-date CentOS 7.9.2009 system using the vanilla gcc (GCC) > 4.8.5 that comes with that version and R compiles just fine and it > passes 'make check' too.It's not that you can't compile R with gcc 4.8.5, it's that you'll have a hard time installing many packages. And that's why EPEL 7 has R 3.6 and cannot be updated to 4.> Since R is trying to move toward C++14 support by default, I agree > with I?aki, you might wanr to build and run R with a newer version of > gcc. gcc 4.8.5 will only give you C++11 support. RedHat's Software > Collections (SCL) devtoolset:s is the easiest way to do this. I've > done this too and can confirm that gcc 7.3.1 that comes with SCL > devtoolset/7 is sufficient to get C++14 support. I'm sharing my > installation with lots of users, so I'm make it all transparent to the > end-user with environment modules, i.e. 'module load r/4.1.0' is all > the user needs to know.-- I?aki ?car
Thanks -- yes, I can confirm that it installs OK after erasing and checking out SVN from scratch. On 4/30/21 9:40 PM, Henrik Bengtsson wrote:> Ben, it's most like what Peter says. I can confirm it works; I just > installed https://cran.r-project.org/src/base-prerelease/R-latest.tar.gz > on an up-to-date CentOS 7.9.2009 system using the vanilla gcc (GCC) > 4.8.5 that comes with that version and R compiles just fine and it > passes 'make check' too. > > Since R is trying to move toward C++14 support by default, I agree > with I?aki, you might wanr to build and run R with a newer version of > gcc. gcc 4.8.5 will only give you C++11 support. RedHat's Software > Collections (SCL) devtoolset:s is the easiest way to do this. I've > done this too and can confirm that gcc 7.3.1 that comes with SCL > devtoolset/7 is sufficient to get C++14 support. I'm sharing my > installation with lots of users, so I'm make it all transparent to the > end-user with environment modules, i.e. 'module load r/4.1.0' is all > the user needs to know. > > /Henrik > > On Thu, Apr 29, 2021 at 7:28 AM Peter Dalgaard <pdalgd at gmail.com> wrote: >> >> You may want to check out your checkout.... >> >> I see: >> >> Peter-Dalgaards-iMac:R pd$ grep newsock src/main/connections.c >> con = R_newsock(host, port, server, serverfd, open, timeout, options); >> >> but your file seems to have lost the ", options" bit somehow. Also, mine is line 3488, not 3477. >> >> Maybe you have an old file getting in the way? >> >> - Peter >> >>> On 29 Apr 2021, at 15:58 , Ben Bolker <bbolker at gmail.com> wrote: >>> >>> I probably don't want to go down this rabbit hole very far, but if anyone has any *quick* ideas ... >>> >>> Attempting to build R from scratch with a fresh SVN checkout on a somewhat out-of-date CentOS system (for which I don't have root access, although I can bug people if I care enough). >>> >>> ../r-devel/configure; make >>> >>> ends with >>> >>> gcc -std=gnu99 -I../../../r-devel/trunk/src/extra -I. -I../../src/include -I../../../r-devel/trunk/src/include -I/usr/local/include -I../../../r-devel/trunk/src/nmath -DHAVE_CONFIG_H -fopenmp -g -O2 -c ../../../r-devel/trunk/src/main/connections.c -o connections.o >>> ../../../r-devel/trunk/src/main/connections.c: In function ?do_sockconn?: >>> ../../../r-devel/trunk/src/main/connections.c:3477:5: error: too few arguments to function ?R_newsock? >>> con = R_newsock(host, port, server, serverfd, open, timeout); >>> ^ >>> In file included from ../../../r-devel/trunk/src/main/connections.c:80:0: >>> ../../../r-devel/trunk/src/include/Rconnections.h:83:13: note: declared here >>> Rconnection R_newsock(const char *host, int port, int server, int serverfd, const char * const mode, int timeout, int options); >>> ^ >>> make[3]: *** [connections.o] Error 1 >>> >>> Any suggestions for a quick fix/diagnosis? >>> >>> cheers >>> Ben Bolker >>> >>> ===>>> >>> >>> $ gcc --version >>> gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39) >>> >>> $ lsb_release -a >>> LSB Version: :core-4.1-amd64:core-4.1-noarch:cxx-4.1-amd64:cxx-4.1-noarch:desktop-4.1-amd64:desktop-4.1-noarch:languages-4.1-amd64:languages-4.1-noarch:printing-4.1-amd64:printing-4.1-noarch >>> Distributor ID: CentOS >>> Description: CentOS Linux release 7.8.2003 (Core) >>> Release: 7.8.2003 >>> Codename: Core >>> >>> ______________________________________________ >>> R-devel at r-project.org mailing list >>> https://stat.ethz.ch/mailman/listinfo/r-devel >> >> -- >> Peter Dalgaard, Professor, >> Center for Statistics, Copenhagen Business School >> Solbjerg Plads 3, 2000 Frederiksberg, Denmark >> Phone: (+45)38153501 >> Office: A 4.23 >> Email: pd.mes at cbs.dk Priv: PDalgd at gmail.com >> >> ______________________________________________ >> R-devel at r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-devel