Gentoo Archives: gentoo-user

From: Alan McKinnon <alan.mckinnon@×××××.com>
To: gentoo-user@l.g.o
Subject: Re: [gentoo-user] Gentoo segfaults on virtualbox-4.1.14 when running on AMD bulldozer
Date: Tue, 01 May 2012 17:44:00
Message-Id: 20120501193923.42d12018@khamul.example.com
In Reply to: [gentoo-user] Gentoo segfaults on virtualbox-4.1.14 when running on AMD bulldozer by Markos Chandras
1 On Fri, 27 Apr 2012 20:35:47 +0100
2 Markos Chandras <hwoarang@g.o> wrote:
3
4 > Hi all
5 >
6 > This is a copy of the thread I recently opened on Gentoo forums[1].
7 >
8 > I replaced my Phenom II cpu with a new 6-core AMD bulldozer. However,
9 > I noticed that all of my Gentoo virtual machines throw (compiler)
10 > segmentation faults when building or running any application. This
11 > happens on a Gentoo testing VM, on a Gentoo stable VM and on a brand
12 > new Gentoo installation. I have no idea where to look for that
13 > problem.
14 >
15 > System:
16 >
17 > Motherboard: Asus M5A78L-M/USB3
18 > CPU: AMD FX(tm)-6100 Six-Core Processor
19 > Host: Gentoo x86_64 testing (has not problem at all, so I presume HW
20 > problem is eliminated)
21 >
22 > Any ideas?
23 >
24
25 I ran into the same system here: VMs crash soon after being launched
26 when they first start doing some real work. I only tested Gentoo guests
27 though.
28
29 But, I have Intel i7 cpu.
30
31 This is an ~amd64 (unstable) host, running amd64 (stable) guests.
32
33 The only clue is in messages:
34
35 May 1 16:25:01 khamul kernel: [73366.634971] VBoxHeadless[2642]:
36 segfault at 10 ip 00007fa35e40599f sp 00007fa35c5b8b00 error 4 in
37 VBoxDD.so[7fa35e311000+144000]
38
39 I downgraded virtual box from 4.1.14 to 4.1.12 and initial
40 tests are good. A guest is doing a full emerge world and the first
41 package is glibc. Previously it wouldn't even manage to download
42 sources fully.
43
44 It's too early to draw any real conclusions though. I'm still rather
45 suspicious of glibc on the host (just as hunch, as that version seems
46 to be popping up in many more bug reports than usual)
47
48
49 --
50 Alan McKinnnon
51 alan.mckinnon@×××××.com