Also just did a full errata update and retried the compilation. It failed exactly as before.
-----Original Message----- From: Ivan Krylov <ikry...@disroot.org> Sent: 08 May 2024 16:09 To: Simon Andrews <simon.andr...@babraham.ac.uk> Cc: Simon Andrews via R-help <r-help@r-project.org>; Miguel Esteva <estev...@wehi.edu.au> Subject: Re: [R] Compilation problems with R4.4.0 В Wed, 8 May 2024 14:25:08 +0000 Simon Andrews <simon.andr...@babraham.ac.uk> пишет: > #8 0x00007ffff3f87c0b in dgesv ( > n=12884901891, nrhs=4294967299, > a=<error reading variable: value has been optimized out>, > lda=12884901891, ipiv=..., b=<error reading variable: value has been > optimized out>, ldb=12884901891, info=0 > ) at /usr/src/debug/lapack-3.9.0-10.el9.x86_64/SRC/dgesv.f:167 > #9 0x00007ffff50b2a17 in La_solve ( > tolin=<optimized out>, Bin=<optimized out>, A=0x18f2f68 > ) at Lapack.c:1246 > #10 mod_do_lapack ( > call=<optimized out>, op=<optimized out>, args=<optimized out>, > env=<optimized out> > ) at Lapack.c:1483 It looks like this R build is linked to a LAPACK. I now also notice LAPACK(generic) in the list of external libraries in your ./configure output. I've tried to link R to AlmaLinux's builds of reference BLAS and LAPACK using --with-blas='-l:libblas.so.3' --with-lapack='-l:liblapack.so.3' (where do I get libblas.so on AlmaLinux?), but that doesn't crash either. What does the config.log currently say about your BLAS and LAPACK situation? It's a number of lines between configure:47012: checking whether 'struct tm' includes tm_gmtoff and configure:48858: checking for iconv.h. If you'd like to link your R to custom BLAS and LAPACK, try specifying the necessary linker flags explicitly with --with-blas=-lwhatever, --with-lapack=-lwhatever. If you're fine with reference BLAS & LAPACK that come with R, try ./configure --without-blas --without-lapack. Normally, R's ./configure tries to prevent linking to an external LAPACK without linking to an external BLAS; I'm not sure how it got you LAPACK(generic) without BLAS(generic). -- Best regards, Ivan ------------------------------------ This email has been scanned for spam & viruses. If you believe this email should have been stopped by our filters, click the following link to report it (https://portal-uk.mailanyone.net/index.html#/outer/reportspam?token=dXNlcj1zaW1vbi5hbmRyZXdzQGJhYnJhaGFtLmFjLnVrO3RzPTE3MTUxODA5NjA7dXVpZD02NjNCOTVBMDY0RkEzNzNCMDM5QTkwQkZCQzJDRUJERjt0b2tlbj0yOTE5M2Q0NjFkZjQzYzhlMzk0ZDgxMTRkNjgyYTUwZjU3OWZlODgxOw%3D%3D). ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
