Summary: | www-apache/passenger-5.0.27 : src/.../fpclassify.hpp:137:75: error: ‘::isnan’ has not been declared | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Toralf Förster <toralf> |
Component: | Current packages | Assignee: | Hans de Graaff <graaff> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | ahodgson, ruby |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | |||
Bug Blocks: | 579962 | ||
Attachments: |
emerge-history.txt
environment www-apache:passenger-5.0.27:20160415-102025.log patch to fix building with glibc 2.23 |
Description
Toralf Förster
![]() Created attachment 430616 [details]
emerge-history.txt
Created attachment 430618 [details]
environment
Created attachment 430620 [details]
www-apache:passenger-5.0.27:20160415-102025.log
I can't seem to reproduce this so I'm wondering if you still can, or if this was perhaps related to the patched glibc version that was introduced for a short period of time? (In reply to Hans de Graaff from comment #4) that's why the blocker is set ;) (In reply to Hans de Graaff from comment #4) that's why the blocker is set - but I added #575232 too but FWIW www-apache/passenger-5.0.28 emerges fine i don't think this is related to sysmacros. the failure is not major/minor/makedev but math related. *** Bug 586354 has been marked as a duplicate of this bug. *** *** Bug 586958 has been marked as a duplicate of this bug. *** Reopening this since Toralf filed a couple of duplicates so I guess the problem is still present for him. I'm having the same problem; is there any workaround/patch available? (In reply to Arnim Eijkhoudt from comment #12) > I'm having the same problem; is there any workaround/patch available? I haven't been able to reproduce this so this bug is low-priority for me (compared to all other ruby-related issues). It might be useful to file this upstream instead and get it fixed there. Bump... This still happens with Passenger 5.0.30, Boost 1.61, GCC 5.4.0 and GLIBC 2.23-r2. I honestly don't know how to otherwise answer your question on reproduction; it happens consistently for me at the same point in the build, with the same reference (error) to Boost. Could you help me help you better in reproducing this? This is caused by the update to glibc 2.23; ::isnan must now be std::isnan, apparently. This is also now a problem in stable amd64, btw, the current stable www-apache/passenger no longer builds. I'll attach a patch to passenger 5.1.1 that fixes it; it should probably also work with 5.0.30. It's not extensively tested, but I have confirmed 5.1.1 builds with it and runs a couple of ruby apps. Created attachment 459822 [details, diff]
patch to fix building with glibc 2.23
This broke upgrades on Infra machines, so I verified & applied the fix. [master e27754fd8b] www-apache/passenger: bug#580078 isnan compile fix. Date: Fri Feb 3 22:43:23 2017 -0800 5 files changed, 19 insertions(+), 1 deletion(-) create mode 100644 www-apache/passenger/files/passenger-5.1.1-isnan.patch |