[<prev] [next>] [day] [month] [year] [list]
Message-Id: <200611171805.kAHI5S3C025845@harpo.it.uu.se>
Date: Fri, 17 Nov 2006 19:05:28 +0100 (MET)
From: Mikael Pettersson <mikpe@...uu.se>
To: ak@...e.de, mikpe@...uu.se
Cc: akpm@...l.org, arvidjaar@...l.ru, bunk@...sta.de,
discuss@...-64.org, earny@...4u.de, ebiederm@...ssion.com,
gregkh@...e.de, jens.axboe@...cle.com, komurojun-mbn@...ty.com,
len.brown@...el.com, linux-kernel@...r.kernel.org,
linux-pci@...ey.karlin.mff.cuni.cz,
linux-usb-devel@...ts.sourceforge.net, maan@...temlinux.org,
mingo@...hat.com, oprofile-list@...ts.sourceforge.net,
phil.el@...adoo.fr, prakash@...noor.de, romosan@...orax.lbl.gov,
shemminger@...l.org, stern@...land.harvard.edu, torvalds@...l.org,
wcohen@...hat.com
Subject: Re: [discuss] Re: 2.6.19-rc5: known regressions (v3)
On Fri, 17 Nov 2006 11:29:25 +0100, Andi Kleen wrote:
>On Friday 17 November 2006 10:59, Mikael Pettersson wrote:
>
>> It certainly worked when I originally implemented it.
>
>I don't think so. NMI watchdog never recovered no matter if oprofile
>used the counter or not.
If so then that's a bug in oprofile or the x86-64 kernel.
I just checked the 2.6.18 i386 kernel + the perfctr kernel
extension, and the NMI watchdog did start ticking again when
perfctr called release_lapic_nmi().
Before the {reserve,release}_lapic_nmi() API went into the
kernel the NMI watchdog might not have resumed, but that was
ages ago (the 2.6.6 kernel).
/Mikael
-
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/
Powered by blists - more mailing lists