I’m encountering a new (to me) problem when trying to compile my Stan-dependent packages built with rstantools in RStudio. My workflow, following the vignette, has been:
pkgbuild::compile_dll() roxygen2::roxygenize() devtools::install(upgrade = FALSE)
compile_dll(), which in the past has worked quickly and mostly silently, has suddenly started puking out a massive stream of warnings to the console. After several minutes of this it terminates in the following fatal error:
Error in (function (command = NULL, args = character(), error_on_status = TRUE, : System command 'Rcmd.exe' failed, exit status: 1, stdout + stderr (last 10 lines): E> make: *** [C:/R/R-40~1.5/etc/x64/Makeconf:227: stanExports_IPM_ICchinook_pp.o] Error 1 E> make: *** Waiting for unfinished jobs.... E> C:/rtools40/mingw64/bin/../lib/gcc/x86_64-w64-mingw32/8.3.0/../../../../x86_64-w64-mingw32/bin/as.exe: stanExports_IPM_LCRchum_pp.o: too many sections (66611) E> C:\Users\Buhle\AppData\Local\Temp\ccuYkIrh.s: Assembler messages: E> C:\Users\Buhle\AppData\Local\Temp\ccuYkIrh.s: Fatal error: can't write 56 bytes to section .text of stanExports_IPM_LCRchum_pp.o: 'file too big' E> C:/rtools40/mingw64/bin/../lib/gcc/x86_64-w64-mingw32/8.3.0/../../../../x86_64-w64-mingw32/bin/as.exe: stanExports_IPM_LCRchum_pp.o: too many sections (66611) E> C:\Users\Buhle\AppData\Local\Temp\ccuYkIrh.s: Fatal error: can't close stanExports_IPM_LCRchum_pp.o: file too big E> make: *** [C:/R/R-40~1.5/etc/x64/Makeconf: [...]
What’s really baffling is that the package where this first started happening (the example shown above) had just been compiled without incident two days earlier, and literally nothing in the package code had changed, nor had I made any changes (that I’m aware of) to any relevant system files. I’ve since verified that the problem affects all my Stan / rstantools-dependent packages, including some I hadn’t touched in months. The previously compiled and installed “good” version of the package shown above still runs fine, as does rstanarm. Stan models called with
rstan() still compile and run fine.
Things I’ve tried that haven’t worked:
- Deleting the contents of the
- Deleting the file
C:\Users\Buhle\AppData\Local\Temp\ccuYkIrh.smentioned in the error message, and any other similar-looking temp files.
compile_dll(), as described here.
- Checking to see if the failed compile attempt had left one of those
00-LOCKfiles in my
librarydirectory (it hadn’t).
- Based on the cryptic “file too big” error and this thread, adding the
-g0flag to my
- Deleting the entire contents of
Makevars.winand just using defaults.
- Uninstalling and reinstalling pkgbuild.
I haven’t yet uninstalled and reinstalled rstan and StanHeaders as mentioned here, mostly because it’s not clear to me whether the versions on
mc-stan.org are the more recent ones that might resurrect a deal-breaking issue with Stan files in RStudio that I just recently, finally, resolved.
I’d be grateful for any suggestions. As always, thanks!
R version 4.0.5 (2021-03-31) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows 7 x64 (build 7601) Service Pack 1 Matrix products: default locale:  LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252  LC_MONETARY=English_United States.1252 LC_NUMERIC=C  LC_TIME=English_United States.1252 attached base packages:  stats graphics grDevices utils datasets methods base loaded via a namespace (and not attached):  Rcpp_1.0.6 compiler_4.0.5 pillar_1.5.1 prettyunits_1.1.1  tools_4.0.5 pkgbuild_1.2.0 jsonlite_1.7.2 lifecycle_1.0.0  tibble_3.1.0 gtable_0.3.0 pkgconfig_2.0.3 rlang_0.4.10  DBI_1.1.1 cli_2.4.0 parallel_4.0.5 curl_4.3  loo_2.4.1 gridExtra_2.3 withr_2.4.1 dplyr_1.0.5  generics_0.1.0 vctrs_0.3.7 tidyselect_1.1.0 stats4_4.0.5  grid_4.0.5 glue_1.4.2 inline_0.3.17 R6_2.5.0  processx_3.5.1 fansi_0.4.2 rstan_2.21.2 purrr_0.3.4  ggplot2_3.3.3 callr_3.6.0 magrittr_2.0.1 codetools_0.2-18  matrixStats_0.58.0 scales_1.1.1 ps_1.6.0 StanHeaders_2.21.0-7  ellipsis_0.3.1 assertthat_0.2.1 colorspace_2.0-0 V8_3.4.0  utf8_1.2.1 RcppParallel_5.0.3 munsell_0.5.0 crayon_1.4.1