Re: ata1: COMRESET failed

[Aaron Lu <aaron.lu@intel.com>]

On 01/08/2013 01:24 AM, Borislav Petkov wrote:
> Adding some more people.
> 
> Aaron, Shane, do you guys have any ideas? I've been seeing something
> similar on an AMD desktop box recently and Jiri sees it on a laptop with
> an SSD.

I've seen a similar problem on a bug report recently for a Crucial SSD,
and upgrading the disk's firmware fixed the problem:
https://bugzilla.kernel.org/show_bug.cgi?id=49701

If the controller has problem, it should have problems for all kinds of
disks attached to it. So possible to test with other disks, Jiri?

Hope this helps.

-Aaron

> 
> Here's Jiri's original mail with more info: http://lkml.org/lkml/2012/10/26/208
> 
> Thanks.
> 
> On Mon, Dec 03, 2012 at 12:57:34PM +0100, Borislav Petkov wrote:
>> On Mon, Dec 03, 2012 at 12:16:44PM +0100, Jiri Slaby wrote:
>>> On 10/26/2012 01:56 PM, Jiri Slaby wrote:
>>>> Hello,
>>>>
>>>> at every boot I'm seeing:
>>>>
>>>> At 5th second:
>>>> ata1: link is slow to respond, please be patient (ready=0)
>>>>
>>>> At 10th second:
>>>> ata1: COMRESET failed (errno=-16)
>>>> ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
>>>> ata1.00: ACPI cmd ef/02:00:00:00:00:a0 (SET FEATURES) succeeded
>>>> ata1.00: ACPI cmd f5/00:00:00:00:00:a0 (SECURITY FREEZE LOCK) filtered out
>>>> ata1.00: ATA-7: INTEL SSDSA2M080G2GC, 2CV102HD, max UDMA/133
>>>> ata1.00: 156301488 sectors, multi 16: LBA48 NCQ (depth 31/32)
>>>> ata1.00: ACPI cmd ef/02:00:00:00:00:a0 (SET FEATURES) succeeded
>>>> ata1.00: ACPI cmd f5/00:00:00:00:00:a0 (SECURITY FREEZE LOCK) filtered out
>>>> ata1.00: configured for UDMA/133
>>>> scsi 0:0:0:0: Direct-Access     ATA      INTEL SSDSA2M080 2CV1 PQ: 0 ANSI: 5
>>>>
>>>>
>>>> So that I have to wait 5 s for the disk to come up. I tried 3.4.15 and
>>>> 3.6.3, both with the same result.
>>>>
>>>> 00:1f.2 SATA controller [0106]: Intel Corporation 7 Series Chipset
>>>> Family 6-port SATA Controller [AHCI mode] [8086:1e03] (rev 04)
>>>>
>>>> Any ideas what is going on?
>>>
>>> Anybody? I also tried 3.0, it's there too. And libahci.skip_host_reset=1
>>> works around the issue.
>>
>> FWIW, I had the same issue on an AMD chipset at the old job. It needs to
>> reset SATA link for some reason, probably BIOS programs it wrongly...
>>
>> + Alan.
>