Re: AMD A10: MCE Instruction Cache Error

From: Alexander Holler
Date: Tue Nov 06 2012 - 11:02:19 EST


Am 06.11.2012 15:47, schrieb Borislav Petkov:
On Tue, Nov 06, 2012 at 02:14:46PM +0100, Alexander Holler wrote:
Am 06.11.2012 12:44, schrieb Alexander Holler:
Am 06.11.2012 12:18, schrieb Alexander Holler:
I will now to tests with leaving fglrx off.

s/to/do/ ;)

That was gone fast. Disabled fglrx, started tests, full halt without any
visible on the serial (I needed to press the reset button):

One after another, now I've got this:

[ 5698.640830] [Hardware Error]: CPU:0
MC2_STATUS[Over|CE|MiscV|-|AddrV|-|-|CECC]: 0xdc2540c000040136
[ 5698.649866] [Hardware Error]: MC2_ADDR: 0x0000000002299678
[ 5698.655443] [Hardware Error]: Combined Unit Error: Fill ECC error
on data fills.
[ 5698.662849] [Hardware Error]: cache level: L2, tx: DATA, mem-tx: DRD

I think it's now really an RMA and I can stop doing further tests.

Are you sure the temperature conditions of the box are optimal? IOW,
there's nothing overheating in there?

Yes. At least if the boxed fan is enough, which I have to assume. Environment temperature is around 18Â C or even colder.

Regards,

Alexander

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