Re: [PATCH 2/2] KVM: Create debugfs dir and stat files for each VM

["Alex Bennée" <alex.bennee@linaro.org>]

Janosch Frank <frankja@linux.vnet.ibm.com> writes:

> On 11/27/2015 09:42 PM, Tyler Baker wrote:
>> On 27 November 2015 at 10:53, Tyler Baker <tyler.baker@linaro.org> wrote:
>>> On 27 November 2015 at 09:08, Tyler Baker <tyler.baker@linaro.org> wrote:
>>>> On 27 November 2015 at 00:54, Christian Borntraeger
>>>> <borntraeger@de.ibm.com> wrote:
>>>>> On 11/26/2015 09:47 PM, Christian Borntraeger wrote:
>>>>>> On 11/26/2015 05:17 PM, Tyler Baker wrote:
>>>>>>> Hi Christian,
>>>>>>>
>>>>>>> The kernelci.org bot recently has been reporting kvm guest boot
>>>>>>> failures[1] on various arm64 platforms in next-20151126. The bot
>>>>>>> bisected[2] the failures to the commit in -next titled "KVM: Create
>>>>>>> debugfs dir and stat files for each VM". I confirmed by reverting this
>>>>>>> commit on top of next-20151126 it resolves the boot issue.
>>>>>>>
<snip>
>>
> After a quick look into qemu I guess I've found the problem:
> kvm_init creates a vm, does checking and self initialization and
> then calls kvm_arch_init. The arch initialization indirectly
> calls kvm_arm_create_scratch_host_vcpu and that's where the
> trouble begins, as it also creates a VM.
>
> My assumption was, that nobody would create multiple VMs under
> the same PID. Christian and I are working on a solution on kernel
> side.

Yeah ARM is a little weird in that respect as the scratch VM is used to
probe capabilities. There is nothing in the API that says you can't have
multiple VMs per PID so I guess a better unique identifier is needed.

--
Alex Bennée