You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@arrow.apache.org by "Neal Richardson (Jira)" <ji...@apache.org> on 2021/09/03 12:33:00 UTC
[jira] [Comment Edited] (ARROW-13870) [R] Should we proactively
disable multithreading on 32bit windows?
[ https://issues.apache.org/jira/browse/ARROW-13870?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17409478#comment-17409478 ]
Neal Richardson edited comment on ARROW-13870 at 9/3/21, 12:32 PM:
-------------------------------------------------------------------
It's the 32-bit version using rtools35 (gcc 4.9); the newer toolchain is fine I think. I think the issue is that we don't know why it's not working and we haven't been able in the past to find out why. And at this point I don't know how much effort it is worth spending trying to solve it, given that this toolchain is no longer required on CRAN. I'm content to degrade gracefully on older Windows machines.
was (Author: npr):
It's the 32-bit version using rtools35 (gcc 4.9); the newer toolchain is fine I think. I think the issue is that we don't know why it's not working and we haven't been able in the past to find out why. And at this point I don't know how much effort it is worth spending trying to solve it, given that this toolchain is no longer required on CRAN.
> [R] Should we proactively disable multithreading on 32bit windows?
> ------------------------------------------------------------------
>
> Key: ARROW-13870
> URL: https://issues.apache.org/jira/browse/ARROW-13870
> Project: Apache Arrow
> Issue Type: Improvement
> Components: R
> Reporter: Jonathan Keane
> Priority: Minor
>
> We have numerous test skips for 32bit windows + multithreading. Should we add something to {{onLoad}} that detects this and proactively sets arrow to single-threaded (using {{set_cpu_count()}} and / or {{options(arrow.use_threads)}}?).
> Should we display a message that we did this and if someone wants to live experimentally they can increase that again?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)