linux-kernel - Re: [PATCH] x86, smp: Fix topology checks on AMD MCM

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20120604144850.GF15433@aftab.osrc.amd.com>
Date:	Mon, 4 Jun 2012 16:48:50 +0200
From:	Borislav Petkov <borislav.petkov@....com>
To:	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	"H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...nel.org>
CC:	Borislav Petkov <bp@...64.org>, X86-ML <x86@...nel.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Andreas Herrmann <andreas.herrmann3@....com>
Subject: Re: [PATCH] x86, smp: Fix topology checks on AMD MCM

On Mon, Jun 04, 2012 at 03:38:37PM +0200, Peter Zijlstra wrote:
> On Mon, 2012-06-04 at 15:37 +0200, Borislav Petkov wrote:
> > This is still crap.
> 
> *poof* and the patch is gone.. I'll wait for an update ;-)

Thanks.

So I'm looking more into this and there's another issue IMHO:

processor       : 2
vendor_id       : AuthenticAMD
cpu family      : 16
model           : 9
stepping        : 1
microcode       : 0x10000d9
cpu MHz         : 800.000
cache size      : 512 KB
physical id     : 0
siblings        : 12
core id         : 2
cpu cores       : 2
^^^^^^^^^^^^^^^^^^^

This is cpuinfo_x86.booted_cores and here's how it progresses on this
MCM box (below).

On core 2 it becomes 2 and it doesn't change again. Until the next
physical socket comes (cpu 12) where it starts again from 1.

And, it is normally the number of booted cores on the socket, i.e. 12 in
this case.

Now, before we go babbling about what the right fix is, maybe we should
remove this thing completely - I mean, it looks like no one uses it, it
is only in the /proc/cpuinfo thing. Ingo, hpa?

$ grep -E "(cpu cores|processor)" /proc/cpuinfo
processor       : 0
cpu cores       : 1
processor       : 1
cpu cores       : 2
processor       : 2
cpu cores       : 2
processor       : 3
cpu cores       : 2
processor       : 4
cpu cores       : 2
processor       : 5
cpu cores       : 2
processor       : 6
cpu cores       : 2
processor       : 7
cpu cores       : 2
processor       : 8
cpu cores       : 2
processor       : 9
cpu cores       : 2
processor       : 10
cpu cores       : 2
processor       : 11
cpu cores       : 2
processor       : 12
cpu cores       : 1
processor       : 13
cpu cores       : 2
processor       : 14
cpu cores       : 2
processor       : 15
cpu cores       : 2
processor       : 16
cpu cores       : 2
processor       : 17
cpu cores       : 2
processor       : 18
cpu cores       : 2
processor       : 19
cpu cores       : 2
processor       : 20
cpu cores       : 2
processor       : 21
cpu cores       : 2
processor       : 22
cpu cores       : 2
processor       : 23
cpu cores       : 2


-- 
Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
GM: Alberto Bozzo
Reg: Dornach, Landkreis Muenchen
HRB Nr. 43632 WEEE Registernr: 129 19551

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/