Re: [Xen-devel] linux-3.9-rc0 regression from 3.8 SATA controller not detected under xen

[Sander Eikelenboom <linux@xxxxxxxxxxxxxx>]

Thursday, February 28, 2013, 8:51:52 AM, you wrote:

>>>> On 27.02.13 at 18:50, Sander Eikelenboom <linux@xxxxxxxxxxxxxx> wrote:
>> Wednesday, February 27, 2013, 1:54:31 PM, you wrote:
>>>>>> On 27.02.13 at 12:46, Sander Eikelenboom <linux@xxxxxxxxxxxxxx> wrote:
>>>>   [   89.338827] ahci: probe of 0000:00:11.0 failed with error -22
>> 
>>> Which is -EINVAL. With nothing else printed, I'm afraid you need to
>>> find the origin of this return value by instrumenting the involved
>>> call tree.
>> 
>> Just wondering, is multiple msi's per device actually supported by xen ?

> No, it isn't; only MSI-X is. But that should be irrelevant to your
> case, as iirc one of the quirk related messages your logs had said
> something about disabling MSI mode for this controller (or even a
> wider part of the system).

Ah you seem to refer to the line:   [   24.957481] pci 0000:00:11.0: calling 
quirk_msi_intx_disable_ati_bug+0x0/0x50
That one threw me off as well, but it seems it's only "Calling" the quirk, 
comments in the code make it clear it's only for the SB700, i have a SB850.

And the "no it isn't" seems to be quite relevant in my case and in general, see 
later mails and the associated commits .. (all merged very earlier in this 
merge window and the ahci changes not via the ahci tree ...)

--
Sander

> Jan




_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel