Summary: | sys-apps/util-linux-2.26 - multilib_check_headers fails: Header files have changed between ABIs. / Header checksum mismatch, aborting. | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | tman <cornicx> |
Component: | Current packages | Assignee: | Gentoo's Team for Core System packages <base-system> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | const, gmturner007, multilib+disabled, tetromino |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
/var/tmp/portage/sys-apps/util-linux-2.26/temp/build.log
multilib_wrap_libfdisk_h_header.patch Add install-nodist_fdiskincHEADERS to make install |
Description
tman
2015-02-19 20:33:10 UTC
Created attachment 396996 [details]
/var/tmp/portage/sys-apps/util-linux-2.26/temp/build.log
Created attachment 397020 [details, diff]
multilib_wrap_libfdisk_h_header.patch
Confirmed; easy-but-not-necessarily-correct ebuild patch attached will get ebuild merging... but it seems the header file is only generated for x86_64, which really seems spooky, and possibly flat-out-wrong. Perhaps there is a deeper underlying bug causing the header not to be deployed for abi_x86_32..., or perhaps that's "just how it works."
Really, someone should look into this before blindly applying the enclosed patch to gx86! I'm in a rush though atm -- maybe later tonight I'll look into it.
(In reply to Greg Turner from comment #2) > Created attachment 397020 [details, diff] [details, diff] > multilib_wrap_libfdisk_h_header.patch > > Confirmed; easy-but-not-necessarily-correct ebuild patch attached will get > ebuild merging... but it seems the header file is only generated for x86_64, > which really seems spooky, and possibly flat-out-wrong. Perhaps there is a > deeper underlying bug causing the header not to be deployed for > abi_x86_32..., or perhaps that's "just how it works." > > Really, someone should look into this before blindly applying the enclosed > patch to gx86! I'm in a rush though atm -- maybe later tonight I'll look > into it. Pseudo-off-topic matter: it seems this assymetry triggers what looks to me an awful lot like a bug in multilib-build.eclass... see below. /* This file is auto-generated by multilib-build.eclass * as a multilib-friendly wrapper. For the original content, * please see the files that are #included below. */ #if defined(__x86_64__) /* amd64 */ # if defined(__ILP32__) /* x32 ABI */ # error "abi_x86_x32 not supported by the package." # else /* 64-bit ABI */ # include <x86_64-pc-linux-gnu/libfdisk/libfdisk.h> # endif #elif defined(__i386__) /* plain x86 */ # include <x86_64-pc-linux-gnu/libfdisk/libfdisk.h> Surely, that can't be right, can it? Created attachment 397026 [details, diff]
Add install-nodist_fdiskincHEADERS to make install
I think this is a more correct patch, as it actually ensures that the same header is installed for all ABIs.
@jcallen, please apply that to 2.26 and 9999. <Poly-C_atwork> Can someone please close bug #540702 for me? I've fixed the bug but have no access to bugzilla currently. |