linux-kernel - v4.9-rc6 - failure to load i386 modules

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <fa668f3c-d909-59d0-2800-c6dc539588a9@canonical.com>
Date:   Tue, 22 Nov 2016 08:36:05 -0700
From:   Tim Gardner <tim.gardner@...onical.com>
To:     rostedt@...dmis.org
Cc:     Al Viro <viro@...iv.linux.org.uk>,
        LKML <linux-kernel@...r.kernel.org>
Subject: v4.9-rc6 - failure to load i386 modules

Hi Steven - I recently started testing i386 QEMU and found that it won't
load most modules, e.g., 'floppy: disagrees about version of symbol
mcount' ad nauseam for almost all module load attempts. Your commit
5de0a8c0c240338cb5b73363b0673c6aa804bb1c ('x86: Fix export for mcount
and __fentry__') does not appear to fix every issue introduced with
commit 784d5699eddc55878627da20d3fe0c8542e2f1a2 ('x86: move exports to
actual definitions'). Interestingly, reverting commit 784d5699 _does_
fix my i386 issues, though I have not tested other arches for
regression. Any thoughts ? I've attached my config and kernel log.

Note that 'mcount' isn't the only symbol with problems.

drm: disagrees about version of symbol __sw_hweight32
drm: disagrees about version of symbol __put_user_4
drm: disagrees about version of symbol __put_user_8
etc

rtg
-- 
Tim Gardner tim.gardner@...onical.com

View attachment "config" of type "text/plain" (203553 bytes)

View attachment "dmesg.txt" of type "text/plain" (50607 bytes)