From mboxrd@z Thu Jan 1 00:00:00 1970 From: Santosh Shilimkar Subject: RE: OMAP4 ES2.0 Unhandled fault: imprecise external abort Date: Fri, 14 Jan 2011 23:31:34 +0530 Message-ID: References: <7c601ad1e2d45e00240055319cfec2f9@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Return-path: Received: from na3sys009aog112.obsmtp.com ([74.125.149.207]:44054 "EHLO na3sys009aog112.obsmtp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751923Ab1ANSBk (ORCPT ); Fri, 14 Jan 2011 13:01:40 -0500 Received: by ewy8 with SMTP id 8so1532090ewy.17 for ; Fri, 14 Jan 2011 10:01:35 -0800 (PST) In-Reply-To: Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: Bryan Wu Cc: TAO HU , TAO HU , linux-omap > -----Original Message----- > From: cooloney@gmail.com [mailto:cooloney@gmail.com] On Behalf Of > Bryan Wu > Sent: Friday, January 14, 2011 11:24 PM > To: Santosh Shilimkar > Cc: TAO HU; TAO HU; linux-omap > Subject: Re: OMAP4 ES2.0 Unhandled fault: imprecise external abort > > On Fri, Jan 14, 2011 at 4:06 PM, Santosh Shilimkar > wrote: > >> -----Original Message----- > >> From: TAO HU [mailto:tghk48@motorola.com] > >> Sent: Friday, January 14, 2011 1:29 PM > >> To: Santosh Shilimkar > >> Cc: Bryan Wu; TAO HU; linux-omap > >> Subject: Re: OMAP4 ES2.0 Unhandled fault: imprecise external > abort > >> > >> Hi, Santosh > >> > >> I think it depends on whether IRQ is enabled. > >> It is a limitation for current L3 error logging mechanism. > >> > > L3 IRQ is always enabled except low power states. I guess > > we patched this part as well so that error is not lost and > > its logged. > > > > It's not a limitation L3 interrupt but behavior of CPU. > > Because scenario's where the I bit is cleared at CPU level > > (Interrupt disable), it's immune to interrupts. > > > > Note the interrupt is not lost but just got delayed a bit > > > > Is there any testcase I can try to make sure the L3 error interrupt > is working? > I guess with simple readmem to some non-initalised modules should trigger the error. Just read the change log of the patch. I guess there are couple of examples