From mboxrd@z Thu Jan 1 00:00:00 1970 From: linux@arm.linux.org.uk (Russell King - ARM Linux) Date: Thu, 24 Apr 2014 19:13:36 +0100 Subject: L2C boot crash on BCM Capri In-Reply-To: <20140424174456.GV26756@n2100.arm.linux.org.uk> References: <7hlhuuprkc.fsf@paris.lan> <20140424174456.GV26756@n2100.arm.linux.org.uk> Message-ID: <20140424181336.GW26756@n2100.arm.linux.org.uk> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Thu, Apr 24, 2014 at 06:44:56PM +0100, Russell King - ARM Linux wrote: > On Thu, Apr 24, 2014 at 10:26:11AM -0700, Kevin Hilman wrote: > > [ 0.100000] L2C-310 enabling early BRESP for Cortex-A9 > > [ 0.110000] L2C-310 full line of zeros enabled for Cortex-A9 > > [ 0.110000] L2C-310 dynamic clock gating enabled, standby mode enabled > > [ 0.120000] Unhandled fault: imprecise external abort (0xc06) at 0x44a42d67 > > The unfortunate thing about imprecise external aborts is that they don't > tell us what went wrong - we have to guess. > > My guess would be that something isn't wired up correctly between the > Cortex-A9 and L2 controller in Broadcom SoCs. You could try disabling > the clock gating/standby mode - I doubt it's the FLZ feature as that > also requires the Cortex-A9 configuration to be changed as well, which > wouldn't have happened at the point this fault occurred. > > Also, it's possible that BRESP could cause it. Try disabling each in > turn and see what effect they have... BTW, this is something _you_ will have to do, because _I_ don't have access to the failing hardware. -- FTTC broadband for 0.8mile line: now at 9.7Mbps down 460kbps up... slowly improving, and getting towards what was expected from it.