Re: pv_ops dom0 USB fixed

[Jeremy Fitzhardinge <jeremy@goop.org>]

Pasi Kärkkäinen wrote:
> On Thu, Dec 11, 2008 at 05:10:24PM +0000, Andrew Lyon wrote:
>   
>> On Wed, Dec 10, 2008 at 9:03 PM, Andrew Lyon <andrew.lyon@gmail.com> wrote:
>>     
>>> On Wed, Dec 10, 2008 at 8:48 PM, Jeremy Fitzhardinge <jeremy@goop.org> wrote:
>>>       
>>>> Pasi Kärkkäinen wrote:
>>>>         
>>>>> On Wed, Dec 10, 2008 at 12:05:32PM -0800, Jeremy Fitzhardinge wrote:
>>>>>
>>>>>           
>>>>>> Ian Campbell noticed a missing TLB flush which was causing the USB
>>>>>> crashes/failures when booting the pvops dom0 kernel.  With that fixed,
>>>>>> enabling USB boots reliably and seems to work.
>>>>>>
>>>>>>
>>>>>>             
>>>>> Nice!
>>>>>
>>>>>
>>>>>           
>>>>>> Its quite possible this will also improve matters with ATA/SATA
>>>>>> controllers, though I haven't tested it so far.
>>>>>>
>>>>>> Anyway, its a significant fix and its worth trying the current pvops
>>>>>> patch queue again.  Please tell me what you find.
>>>>>>
>>>>>>
>>>>>>             
>>> Excellent news, I've tried the pv_ops dom0 kernel several times
>>> recently and had failure with usb and sata drivers, so fingers crossed
>>> it will work now.
>>>
>>> I will test tomorrow and report my results ;-)
>>>
>>> Andy
>>>
>>>       
>>>>> Hmm.. against what kernel/tree are these patches?
>>>>>
>>>>>           
>>>> See the wiki ;)
>>>>
>>>>
>>>> Pull the kernel.org/hg/linux-2.6 tree, "hg update $(cat KERNEL_VERSION)",
>>>> then "hg qpush -a"
>>>>
>>>>   J
>>>>
>>>> _______________________________________________
>>>> Xen-devel mailing list
>>>> Xen-devel@lists.xensource.com
>>>> http://lists.xensource.com/xen-devel
>>>>
>>>>         
>> I downloaded the source a few mins ago and tried pv_ops dom0 on my
>> test system a dell optiplex 755, it doesnt get very far at all into
>> the boot process:
>>
>> (XEN) *** LOADING DOMAIN 0 ***
>> (XEN)  Xen  kernel: 64-bit, lsb, compat32
>> (XEN)  Dom0 kernel: 64-bit, PAE, lsb, paddr 0x200000 -> 0x8a0418
>> (XEN) PHYSICAL MEMORY ARRANGEMENT:
>> (XEN)  Dom0 alloc.:   0000000120000000->0000000122000000 (951655 pages
>> to be allocated)
>> (XEN) VIRTUAL MEMORY ARRANGEMENT:
>> (XEN)  Loaded kernel: ffffffff80200000->ffffffff808a0418
>> (XEN)  Init. ramdisk: ffffffff808a1000->ffffffff808a1000
>> (XEN)  Phys-Mach map: ffffffff808a1000->ffffffff80ff3b38
>> (XEN)  Start info:    ffffffff80ff4000->ffffffff80ff44a4
>> (XEN)  Page tables:   ffffffff80ff5000->ffffffff81002000
>> (XEN)  Boot stack:    ffffffff81002000->ffffffff81003000
>> (XEN)  TOTAL:         ffffffff80000000->ffffffff81400000
>> (XEN)  ENTRY ADDRESS: ffffffff80765200
>> (XEN) Dom0 has maximum 2 VCPUs
>> (XEN) Scrubbing Free RAM: .done.
>> (XEN) Xen trace buffers: disabled
>> (XEN) Std. Loglevel: Errors and warnings
>> (XEN) Guest Loglevel: Nothing (Rate-limited: Errors and warnings)
>> (XEN) Xen is relinquishing VGA console.
>> (XEN) *** Serial input -> DOM0 (type 'CTRL-a' three times to switch
>> input to Xen)
>> (XEN) Freed 108kB init memory.
>> mapping kernel into physical memory
>> Xen: setup ISA identity maps
>> about to get started...
>> (XEN) ioapic_guest_write: apic=0, pin=2, old_irq=0, new_irq=-1
>> (XEN) ioapic_guest_write: old_entry=000009f0, new_entry=00010900
>> (XEN) ioapic_guest_write: Attempt to remove IO-APIC pin of in-use IRQ!
>> (XEN) ioapic_guest_write: apic=0, pin=4, old_irq=4, new_irq=-1
>> (XEN) ioapic_guest_write: old_entry=000009f1, new_entry=00010900
>> (XEN) ioapic_guest_write: Attempt to remove IO-APIC pin of in-use IRQ!
>> (XEN) ioapic_guest_write: apic=0, pin=4, old_irq=4, new_irq=4
>> (XEN) ioapic_guest_write: old_entry=000009f1, new_entry=000189f1
>> (XEN) ioapic_guest_write: Attempt to modify IO-APIC pin for in-use IRQ!
>>
>> And then hangs.
>>
>> I tried adding pci=nomsi to the kernel arguments, didnt seem to make
>> much difference.
>>
>>     
>
> Did you try specifying both pci=nomsi and nosmp ? 
>   

No, I think "nosmp" is a problem.  Let it be SMP if it wants.

However, you should also have "console=hvc0 earlyprink=xen" to see the 
console output (and perhaps it only works on a serial console for some 
reason).

    J