Date: Sat, 18 Nov 2006 19:47:17 -0500
From: ROBERT HANCOCK <hancockr@shaw.ca>
Subject: Re: ata2: EH in ADMA mode, notifier 0x0 notifier_error 0x0 gen_ctl
To: linux-kernel@vger.kernel.org, christiand59@web.de
Message-id: <cb8795142da89.455f6345@shaw.ca>
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii
Content-language: en
Content-transfer-encoding: 7bit
Content-disposition: inline
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1417
Lines: 23

Christian wrote:
> During my I/O load test, after about half an hour of heavy I/O on three SATAII 
> disks the system suddenly hung for about 3 seconds. After that I checked 
> dmesg and found the following error output:
> 
> [ 4574.193809] ata2: EH in ADMA mode, notifier 0x0 notifier_error 0x0 gen_ctl 
> 0x1501000 status 0x400
> [ 4574.193826] ata2: CPB 0: ctl_flags 0x1f, resp_flags 0x1
> [ 4574.193835] ata2: CPB 1: ctl_flags 0x1f, resp_flags 0x2

All this output is from the debugging code I have in the error handler in sata_nv for ADMA mode.

> [ 4574.194366] ata2: Resetting port
> [ 4574.194411] ata2.00: exception Emask 0x0 SAct 0x2 SErr 0x0 action 0x2 
> frozen
> [ 4574.194453] ata2.00: tag 1 cmd 0x60 Emask 0x4 stat 0x40 err 0x0 (timeout)

Hmm, it looks like the controller thinks the command has been sent to the drive and has "released" the command for the drive to do its thing, and hasn't received a response back yet. (At least that's what I believe bit 1 in the response flags means..) This might not be the fault of the controller or driver, it might just be the drive not responding. Can you post some drive information (like full dmesg from bootup)?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/