[Qemu-devel] PING: [PATCH] backends/hostmem: Ignore ENOSYS while setting MPOL_DEFAULT

[Qemu-devel] PING: [PATCH] backends/hostmem: Ignore ENOSYS while setting MPOL_DEFAULT

[Pavel Fedin]

 Hello! No news for a long time, we are at RC stage. Could we get this in?

Kind regards,
Pavel Fedin
Expert Engineer
Samsung Electronics Research center Russia

> -----Original Message-----
> From: qemu-devel-bounces+p.fedin=samsung.com@nongnu.org [mailto:qemu-devel-
> bounces+p.fedin=samsung.com@nongnu.org] On Behalf Of Eduardo Habkost
> Sent: Tuesday, October 27, 2015 8:32 PM
> To: Pavel Fedin
> Cc: 'Paolo Bonzini'; qemu-devel@nongnu.org
> Subject: Re: [Qemu-devel] [PATCH] backends/hostmem: Ignore ENOSYS while setting MPOL_DEFAULT
> 
> On Tue, Oct 27, 2015 at 03:51:31PM +0300, Pavel Fedin wrote:
> > Currently hostmem backend fails if CONFIG_NUMA is enabled for the qemu
> > (default), but NUMA is not supported by the kernel. This makes it
> > impossible to use ivshmem in such configurations.
> >
> > This patch fixes the problem by ignoring ENOSYS error if policy is set to
> > MPOL_DEFAULT. This way the code behaves in the same way as if CONFIG_NUMA
> > was not defined. qemu will still fail if the user specifies some other
> > policy, so that the user knows it.
> >
> > Signed-off-by: Pavel Fedin <p.fedin@samsung.com>
> 
> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
> 
> Thanks. Applied to numa tree, with the following indentation fix:
> 
> diff --git a/backends/hostmem.c b/backends/hostmem.c
> index 94a4ac0..1b4eb45 100644
> --- a/backends/hostmem.c
> +++ b/backends/hostmem.c
> @@ -315,7 +315,7 @@ host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
>                    maxnode ? backend->host_nodes : NULL, maxnode + 1, flags)) {
>              if (backend->policy != MPOL_DEFAULT || errno != ENOSYS) {
>                  error_setg_errno(errp, errno,
> -                             "cannot bind memory to host NUMA nodes");
> +                                 "cannot bind memory to host NUMA nodes");
>                  return;
>              }
>          }
> 
> > ---
> >  backends/hostmem.c | 6 ++++--
> >  1 file changed, 4 insertions(+), 2 deletions(-)
> >
> > diff --git a/backends/hostmem.c b/backends/hostmem.c
> > index 41ba2af..94a4ac0 100644
> > --- a/backends/hostmem.c
> > +++ b/backends/hostmem.c
> > @@ -313,9 +313,11 @@ host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
> >          assert(maxnode <= MAX_NODES);
> >          if (mbind(ptr, sz, backend->policy,
> >                    maxnode ? backend->host_nodes : NULL, maxnode + 1, flags)) {
> > -            error_setg_errno(errp, errno,
> > +            if (backend->policy != MPOL_DEFAULT || errno != ENOSYS) {
> > +                error_setg_errno(errp, errno,
> >                               "cannot bind memory to host NUMA nodes");
> > -            return;
> > +                return;
> > +            }
> >          }
> >  #endif
> >          /* Preallocate memory after the NUMA policy has been instantiated.
> > --
> > 1.9.5.msysgit.0
> >
> >
> 
> --
> Eduardo

Re: [Qemu-devel] PING: [PATCH] backends/hostmem: Ignore ENOSYS while setting MPOL_DEFAULT

[Paolo Bonzini]

On 23/11/2015 07:41, Pavel Fedin wrote:
>  Hello! No news for a long time, we are at RC stage. Could we get this in?

Yes, queued for -rc2.

Paolo

> Kind regards,
> Pavel Fedin
> Expert Engineer
> Samsung Electronics Research center Russia
> 
>> -----Original Message-----
>> From: qemu-devel-bounces+p.fedin=samsung.com@nongnu.org [mailto:qemu-devel-
>> bounces+p.fedin=samsung.com@nongnu.org] On Behalf Of Eduardo Habkost
>> Sent: Tuesday, October 27, 2015 8:32 PM
>> To: Pavel Fedin
>> Cc: 'Paolo Bonzini'; qemu-devel@nongnu.org
>> Subject: Re: [Qemu-devel] [PATCH] backends/hostmem: Ignore ENOSYS while setting MPOL_DEFAULT
>>
>> On Tue, Oct 27, 2015 at 03:51:31PM +0300, Pavel Fedin wrote:
>>> Currently hostmem backend fails if CONFIG_NUMA is enabled for the qemu
>>> (default), but NUMA is not supported by the kernel. This makes it
>>> impossible to use ivshmem in such configurations.
>>>
>>> This patch fixes the problem by ignoring ENOSYS error if policy is set to
>>> MPOL_DEFAULT. This way the code behaves in the same way as if CONFIG_NUMA
>>> was not defined. qemu will still fail if the user specifies some other
>>> policy, so that the user knows it.
>>>
>>> Signed-off-by: Pavel Fedin <p.fedin@samsung.com>
>>
>> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
>>
>> Thanks. Applied to numa tree, with the following indentation fix:
>>
>> diff --git a/backends/hostmem.c b/backends/hostmem.c
>> index 94a4ac0..1b4eb45 100644
>> --- a/backends/hostmem.c
>> +++ b/backends/hostmem.c
>> @@ -315,7 +315,7 @@ host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
>>                    maxnode ? backend->host_nodes : NULL, maxnode + 1, flags)) {
>>              if (backend->policy != MPOL_DEFAULT || errno != ENOSYS) {
>>                  error_setg_errno(errp, errno,
>> -                             "cannot bind memory to host NUMA nodes");
>> +                                 "cannot bind memory to host NUMA nodes");
>>                  return;
>>              }
>>          }
>>
>>> ---
>>>  backends/hostmem.c | 6 ++++--
>>>  1 file changed, 4 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/backends/hostmem.c b/backends/hostmem.c
>>> index 41ba2af..94a4ac0 100644
>>> --- a/backends/hostmem.c
>>> +++ b/backends/hostmem.c
>>> @@ -313,9 +313,11 @@ host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
>>>          assert(maxnode <= MAX_NODES);
>>>          if (mbind(ptr, sz, backend->policy,
>>>                    maxnode ? backend->host_nodes : NULL, maxnode + 1, flags)) {
>>> -            error_setg_errno(errp, errno,
>>> +            if (backend->policy != MPOL_DEFAULT || errno != ENOSYS) {
>>> +                error_setg_errno(errp, errno,
>>>                               "cannot bind memory to host NUMA nodes");
>>> -            return;
>>> +                return;
>>> +            }
>>>          }
>>>  #endif
>>>          /* Preallocate memory after the NUMA policy has been instantiated.
>>> --
>>> 1.9.5.msysgit.0
>>>
>>>
>>
>> --
>> Eduardo
> 
> 
> 

Re: [Qemu-devel] PING: [PATCH] backends/hostmem: Ignore ENOSYS while setting MPOL_DEFAULT

[Paolo Bonzini]

On 23/11/2015 11:09, Paolo Bonzini wrote:
> On 23/11/2015 07:41, Pavel Fedin wrote:
>>  Hello! No news for a long time, we are at RC stage. Could we get this in?
> 
> Yes, queued for -rc2.

... doh, Eduardo applied it to the NUMA tree already.  I missed that
backends/hostmem* is under NUMA and not memory.

Paolo

> Paolo
> 
>> Kind regards,
>> Pavel Fedin
>> Expert Engineer
>> Samsung Electronics Research center Russia
>>
>>> -----Original Message-----
>>> From: qemu-devel-bounces+p.fedin=samsung.com@nongnu.org [mailto:qemu-devel-
>>> bounces+p.fedin=samsung.com@nongnu.org] On Behalf Of Eduardo Habkost
>>> Sent: Tuesday, October 27, 2015 8:32 PM
>>> To: Pavel Fedin
>>> Cc: 'Paolo Bonzini'; qemu-devel@nongnu.org
>>> Subject: Re: [Qemu-devel] [PATCH] backends/hostmem: Ignore ENOSYS while setting MPOL_DEFAULT
>>>
>>> On Tue, Oct 27, 2015 at 03:51:31PM +0300, Pavel Fedin wrote:
>>>> Currently hostmem backend fails if CONFIG_NUMA is enabled for the qemu
>>>> (default), but NUMA is not supported by the kernel. This makes it
>>>> impossible to use ivshmem in such configurations.
>>>>
>>>> This patch fixes the problem by ignoring ENOSYS error if policy is set to
>>>> MPOL_DEFAULT. This way the code behaves in the same way as if CONFIG_NUMA
>>>> was not defined. qemu will still fail if the user specifies some other
>>>> policy, so that the user knows it.
>>>>
>>>> Signed-off-by: Pavel Fedin <p.fedin@samsung.com>
>>>
>>> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
>>>
>>> Thanks. Applied to numa tree, with the following indentation fix:
>>>
>>> diff --git a/backends/hostmem.c b/backends/hostmem.c
>>> index 94a4ac0..1b4eb45 100644
>>> --- a/backends/hostmem.c
>>> +++ b/backends/hostmem.c
>>> @@ -315,7 +315,7 @@ host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
>>>                    maxnode ? backend->host_nodes : NULL, maxnode + 1, flags)) {
>>>              if (backend->policy != MPOL_DEFAULT || errno != ENOSYS) {
>>>                  error_setg_errno(errp, errno,
>>> -                             "cannot bind memory to host NUMA nodes");
>>> +                                 "cannot bind memory to host NUMA nodes");
>>>                  return;
>>>              }
>>>          }
>>>
>>>> ---
>>>>  backends/hostmem.c | 6 ++++--
>>>>  1 file changed, 4 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/backends/hostmem.c b/backends/hostmem.c
>>>> index 41ba2af..94a4ac0 100644
>>>> --- a/backends/hostmem.c
>>>> +++ b/backends/hostmem.c
>>>> @@ -313,9 +313,11 @@ host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
>>>>          assert(maxnode <= MAX_NODES);
>>>>          if (mbind(ptr, sz, backend->policy,
>>>>                    maxnode ? backend->host_nodes : NULL, maxnode + 1, flags)) {
>>>> -            error_setg_errno(errp, errno,
>>>> +            if (backend->policy != MPOL_DEFAULT || errno != ENOSYS) {
>>>> +                error_setg_errno(errp, errno,
>>>>                               "cannot bind memory to host NUMA nodes");
>>>> -            return;
>>>> +                return;
>>>> +            }
>>>>          }
>>>>  #endif
>>>>          /* Preallocate memory after the NUMA policy has been instantiated.
>>>> --
>>>> 1.9.5.msysgit.0
>>>>
>>>>
>>>
>>> --
>>> Eduardo
>>
>>
>>
> 
> 

Re: [Qemu-devel] PING: [PATCH] backends/hostmem: Ignore ENOSYS while setting MPOL_DEFAULT

[Eduardo Habkost]

On Mon, Nov 23, 2015 at 11:11:28AM +0100, Paolo Bonzini wrote:
> 
> 
> On 23/11/2015 11:09, Paolo Bonzini wrote:
> > On 23/11/2015 07:41, Pavel Fedin wrote:
> >>  Hello! No news for a long time, we are at RC stage. Could we get this in?
> > 
> > Yes, queued for -rc2.
> 
> ... doh, Eduardo applied it to the NUMA tree already.  I missed that
> backends/hostmem* is under NUMA and not memory.

I had applied it to the numa tree but forgot to send a pull
request, sorry. I will submit a pull request in a moment.

-- 
Eduardo