linux-kernel - Oprofile [still] doesn't work on 2.6.28-rc4 on certain CPU

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

Message-ID: <alpine.LNX.1.10.0811132147030.19853@jikos.suse.cz>
Date:	Thu, 13 Nov 2008 21:57:32 +0100 (CET)
From:	Jiri Kosina <jkosina@...e.cz>
To:	Andi Kleen <andi@...stfloor.org>,
	Robert Richter <robert.richter@....com>
cc:	oprofile-list@...ts.sf.net, Jiri Benc <jbenc@...e.cz>,
	Vilem Marsik <vmarsik@...e.cz>,
	Eric Dumazet <dada1@...mosbay.com>,
	Ingo Molnar <mingo@...e.hu>,
	Pekka Enberg <penberg@...helsinki.fi>,
	linux-kernel@...r.kernel.org
Subject: Oprofile [still] doesn't work on 2.6.28-rc4 on certain CPU

Hi,

there has been a bug [1] recently which caused NMIs to be stuck on certain 
Intel CPUs, resultung in oprofile not working.

This has been fixed by 

	commit 7c64ade53a6f977d73f16243865c42ceae999aea
	Author: Andi Kleen <andi@...stfloor.org>
	Date:   Fri Nov 7 14:02:49 2008 +0100

		oprofile: Fix p6 counter overflow check

but we are seeing the very same symptoms still on the following CPU, even 
with 2.6.28-rc4 (which includes 7c64ade5):

	processor       : 0
	vendor_id       : AuthenticAMD
	cpu family      : 15
	model           : 47
	model name      : AMD Athlon(tm) 64 Processor 3000+
	stepping        : 0

After starting oprofile, no NMI is delivered at all:

	vulture:~ # opcontrol --start
	Using 2.6+ OProfile kernel interface.
	Reading module info.
	Using log file /var/lib/oprofile/samples/oprofiled.log
	Daemon started.
	Profiler running.
	vulture:~ # grep NMI /proc/interrupts
	NMI:          0   Non-maskable interrupts
	vulture:~ # grep NMI /proc/interrupts
	NMI:          0   Non-maskable interrupts

on a quite similar CPU, everything seems to be working smoothly though 
with the very same kernel -- NMIs are being delivered, and oprofile 
performs proper profiling:

	processor       : 0
	vendor_id       : AuthenticAMD
	cpu family      : 15
	model           : 4
	model name      : AMD Athlon(tm) 64 Processor 3000+
	stepping        : 8

I haven't yet found a time to start bisecting this.

[1] http://lkml.org/lkml/2008/10/30/319

-- 
Jiri Kosina
SUSE Labs
--
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/