Re: [R-pkg-devel] current docker image for ASAN

2024-01-16 Thread Dirk Eddelbuettel


On 16 January 2024 at 15:54, Steven Scott wrote:
| Greetings everyone, though I expect this message is mainly for Dirk.
| 
| CRAN checks of my bsts/Boom package generate an ASAN error that the CRAN
| maintainers have asked me to look into.  I recall doing this before (this
| error has been there for several years now) via a docker image that Dirk
| had set up.  I have two questions.
| 
| 1) Which docker image should I use?  I imagine it has been updated since
| the last time I tried.
| 2) Is the image built with an asan-appropriate libc++?  I'm asking because
| the last time I tried tracking down this error, ASAN identified that there
| was an error, but pointed to an irrelevant section of code.  Brian thinks
| libc++ is the culprit.

Thanks -- maybe see prior messages. The container is on a scheduled weekly
rebuild, but for reasons that long escape me I also switched at some point to
relying on the 'sumo' container Winston builds by collating several such
containers and have used this myself for several debugging trips:

   https://github.com/wch/r-debug

You want RDsan and/or RDcsan therein.

Dirk

-- 
dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel


[R-pkg-devel] current docker image for ASAN

2024-01-16 Thread Steven Scott
Greetings everyone, though I expect this message is mainly for Dirk.

CRAN checks of my bsts/Boom package generate an ASAN error that the CRAN
maintainers have asked me to look into.  I recall doing this before (this
error has been there for several years now) via a docker image that Dirk
had set up.  I have two questions.

1) Which docker image should I use?  I imagine it has been updated since
the last time I tried.
2) Is the image built with an asan-appropriate libc++?  I'm asking because
the last time I tried tracking down this error, ASAN identified that there
was an error, but pointed to an irrelevant section of code.  Brian thinks
libc++ is the culprit.

[[alternative HTML version deleted]]

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel