Re: One problem in reassign pci bus number?

[Don Dutile <ddutile@redhat.com>]

On 04/23/2012 06:07 PM, Yinghai Lu wrote:
> On Mon, Apr 23, 2012 at 2:09 PM, Don Dutile<ddutile@redhat.com>  wrote:
>> On 04/23/2012 04:19 PM, Yinghai Lu wrote:
>>>
>>> On Mon, Apr 23, 2012 at 12:46 PM, Don Dutile<ddutile@redhat.com>    wrote:
>>>>
>>>> I have run into a similar problem recently when trying to use
>>>> pci=assign-busses
>>>> with an SRIOV device behind a non-ARI-capable PCIe switch.
>>>> In this scenario, the assign-busses code assigned the next bus number,
>>>> which conflicted with an existing one on the system, and hangs the
>>>> system -- two bridges responding to the same PCI bus num evidently
>>>> confuses the hw! ;-)
>>>
>>>
>>> can you post boot log and lspci -vvxxx?
>>>
>>> Yinghai
>>
>>
>> Attached requested logs of linux-3.4-rc2 booted on RHEL6.2 installation.
>>
>> I don't have a boot log of failing condition (pci=assign-devices)
>
> what is pci=assign-devices ? you have own local patches to handle that?
>
> or you mean pci=assign-busses?
>
yes, that's what I meant. sorry, my bad... :(

>> b/c it wedges at boot, and the serial line doesn't output anything
>> even though I've tried every magic trick known with grub&  kernel boot
>> params.
>> ... the PCI log btwn early boot and until the console is reconfigured is
>> 'lost'
>> on the serial line, and that's when the hang occurs.
>> It's been 'fun' to debug w/o that serial output...
>>
>> Let me know if you need something else. (lspci -t ?)
>
> can you check busn_alloc patchset fix the overlapping problem for you?
> It already split scan bus to two pass, also it  will double check not
> scanned peer bridges.
>
> git://git.kernel.org/pub/scm/linux/kernel/git/yinghai/linux-yinghai.git
> for-pci-busn-alloc
>
> you can merge them to your 3.4-rc2...
>
> Thanks
>
> Yinghai
I'll try as soon as I get back to the office;
I'm out the next few days, and don't have external access to my test system.