Re: 3.2.11: PCI Express card cannot be re-detected withing cca 60sec timeframe

[Yinghai Lu]

On Tue, Apr 17, 2012 at 2:34 AM, Martin Mokrejs
<mmokrejs@xxxxxxxxxxxxxxxxxx> wrote:
> Yinghai Lu wrote:
>> On Tue, Apr 17, 2012 at 1:46 AM, Yinghai Lu <yinghai@xxxxxxxxxx> wrote:
>>> On Tue, Apr 17, 2012 at 12:27 AM, Martin Mokrejs
>>> <mmokrejs@xxxxxxxxxxxxxxxxxx> wrote:
>>>>>
>>>>> can you send out lspci -vvxxxx -s 00:1c.7 after removing USB 3.0 express card?
>>>>
>>>> No, that did not help (tried on 3.4-rc3). First I plugged in the NEC uPD720200 USB3.0 card
>>>> and unplugged it (card removal not reported :(), saved the lspci output.
>>>> Then I inserted the FireWire card, unplugged, again saved the lspci output.
>>>> Finally, saved the dmesg. All three files are attached.
>>>
>>> for USB 3.0:
>>>                SltSta: Status: AttnBtn- PowerFlt- MRL- CmdCplt- PresDet+ Interlock-
>>>                        Changed: MRL- PresDet- LinkState+
>>
>> LnkSta: Speed 2.5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive+
>> BWMgmt+ ABWMgmt-
>>
>> link still active even that card get removed, that looks like chipset
>> have problem.
>
> Yinghai,
>  so what should I ask Dell for? To fix BIOS of the laptop? What exactly? Or do you think
> this is a hardware issue with the chipset?

after the expresscard get removed, that root port socket status bits
still report Data Link Active, and
the card still present.

Better check with Dell at first, and ask them to talk with intel if
there is silicon bug with that chipset that report wrong status.

Let me check if i could rebase one my local patch do reset pcie link
to prove that problem it silicon related.

Yinghai
--
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/