[Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Li Qiang]

In set_netdev(), the peers[i] is initialized
qemu_find_net_clients_except() when i is in
0 between 'queues' it can't be NULL.

Signed-off-by: Li Qiang <liq3ea@gmail.com>
---
 hw/core/qdev-properties-system.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index 8b22fb5..b45a7ef 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
     }
 
     for (i = 0; i < queues; i++) {
-        if (peers[i] == NULL) {
-            err = -ENOENT;
-            goto out;
-        }
 
         if (peers[i]->peer) {
             err = -EEXIST;
-- 
1.8.3.1

Re: [Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Laurent Vivier]

On 09/11/2018 09:13, Li Qiang wrote:
> In set_netdev(), the peers[i] is initialized
> qemu_find_net_clients_except() when i is in
> 0 between 'queues' it can't be NULL.
> 
> Signed-off-by: Li Qiang <liq3ea@gmail.com>
> ---
>  hw/core/qdev-properties-system.c | 4 ----
>  1 file changed, 4 deletions(-)
> 
> diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
> index 8b22fb5..b45a7ef 100644
> --- a/hw/core/qdev-properties-system.c
> +++ b/hw/core/qdev-properties-system.c
> @@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
>      }
>  
>      for (i = 0; i < queues; i++) {
> -        if (peers[i] == NULL) {
> -            err = -ENOENT;
> -            goto out;
> -        }
>  
>          if (peers[i]->peer) {
>              err = -EEXIST;
> 

Reviewed-by: Laurent Vivier <lvivier@redhat.com>

Re: [Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Thomas Huth]

On 2018-11-09 09:21, Laurent Vivier wrote:
> On 09/11/2018 09:13, Li Qiang wrote:
>> In set_netdev(), the peers[i] is initialized
>> qemu_find_net_clients_except() when i is in
>> 0 between 'queues' it can't be NULL.
>>
>> Signed-off-by: Li Qiang <liq3ea@gmail.com>
>> ---
>>  hw/core/qdev-properties-system.c | 4 ----
>>  1 file changed, 4 deletions(-)
>>
>> diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
>> index 8b22fb5..b45a7ef 100644
>> --- a/hw/core/qdev-properties-system.c
>> +++ b/hw/core/qdev-properties-system.c
>> @@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
>>      }
>>  
>>      for (i = 0; i < queues; i++) {
>> -        if (peers[i] == NULL) {
>> -            err = -ENOENT;
>> -            goto out;
>> -        }
>>  
>>          if (peers[i]->peer) {
>>              err = -EEXIST;
>>
> 
> Reviewed-by: Laurent Vivier <lvivier@redhat.com>

So who can pick up such patches? We don't have a maintainer for the QDEV
subsystem anymore... should it go via qemu-trivial?

 Thomas

Re: [Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Laurent Vivier]

On 09/11/2018 09:26, Thomas Huth wrote:
> On 2018-11-09 09:21, Laurent Vivier wrote:
>> On 09/11/2018 09:13, Li Qiang wrote:
>>> In set_netdev(), the peers[i] is initialized
>>> qemu_find_net_clients_except() when i is in
>>> 0 between 'queues' it can't be NULL.
>>>
>>> Signed-off-by: Li Qiang <liq3ea@gmail.com>
>>> ---
>>>  hw/core/qdev-properties-system.c | 4 ----
>>>  1 file changed, 4 deletions(-)
>>>
>>> diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
>>> index 8b22fb5..b45a7ef 100644
>>> --- a/hw/core/qdev-properties-system.c
>>> +++ b/hw/core/qdev-properties-system.c
>>> @@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
>>>      }
>>>  
>>>      for (i = 0; i < queues; i++) {
>>> -        if (peers[i] == NULL) {
>>> -            err = -ENOENT;
>>> -            goto out;
>>> -        }
>>>  
>>>          if (peers[i]->peer) {
>>>              err = -EEXIST;
>>>
>>
>> Reviewed-by: Laurent Vivier <lvivier@redhat.com>
> 
> So who can pick up such patches? We don't have a maintainer for the QDEV
> subsystem anymore... should it go via qemu-trivial?

It seems trivial enough to go via qemu-trivial, but it would be good to
have a R-b from someone else than me to confirm.

Thanks,
Laurent

Re: [Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Thomas Huth]

On 2018-11-09 09:13, Li Qiang wrote:
> In set_netdev(), the peers[i] is initialized
> qemu_find_net_clients_except() when i is in
> 0 between 'queues' it can't be NULL.
> 
> Signed-off-by: Li Qiang <liq3ea@gmail.com>
> ---
>  hw/core/qdev-properties-system.c | 4 ----
>  1 file changed, 4 deletions(-)
> 
> diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
> index 8b22fb5..b45a7ef 100644
> --- a/hw/core/qdev-properties-system.c
> +++ b/hw/core/qdev-properties-system.c
> @@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
>      }
>  
>      for (i = 0; i < queues; i++) {
> -        if (peers[i] == NULL) {
> -            err = -ENOENT;
> -            goto out;
> -        }
>  
>          if (peers[i]->peer) {
>              err = -EEXIST;
> 

Reviewed-by: Thomas Huth <thuth@redhat.com>

Re: [Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Laurent Vivier]

On 09/11/2018 09:13, Li Qiang wrote:
> In set_netdev(), the peers[i] is initialized
> qemu_find_net_clients_except() when i is in
> 0 between 'queues' it can't be NULL.
> 
> Signed-off-by: Li Qiang <liq3ea@gmail.com>
> ---
>  hw/core/qdev-properties-system.c | 4 ----
>  1 file changed, 4 deletions(-)
> 
> diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
> index 8b22fb5..b45a7ef 100644
> --- a/hw/core/qdev-properties-system.c
> +++ b/hw/core/qdev-properties-system.c
> @@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
>      }
>  
>      for (i = 0; i < queues; i++) {
> -        if (peers[i] == NULL) {
> -            err = -ENOENT;
> -            goto out;
> -        }
>  
>          if (peers[i]->peer) {
>              err = -EEXIST;
> 


Applied to my trivial-patches branch.

Thanks,
Laurent

Re: [Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Paolo Bonzini]

On 09/11/2018 09:29, Laurent Vivier wrote:
> On 09/11/2018 09:26, Thomas Huth wrote:
>> On 2018-11-09 09:21, Laurent Vivier wrote:
>>> On 09/11/2018 09:13, Li Qiang wrote:
>>>> In set_netdev(), the peers[i] is initialized
>>>> qemu_find_net_clients_except() when i is in
>>>> 0 between 'queues' it can't be NULL.
>>>>
>>>> Signed-off-by: Li Qiang <liq3ea@gmail.com>
>>>> ---
>>>>  hw/core/qdev-properties-system.c | 4 ----
>>>>  1 file changed, 4 deletions(-)
>>>>
>>>> diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
>>>> index 8b22fb5..b45a7ef 100644
>>>> --- a/hw/core/qdev-properties-system.c
>>>> +++ b/hw/core/qdev-properties-system.c
>>>> @@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
>>>>      }
>>>>  
>>>>      for (i = 0; i < queues; i++) {
>>>> -        if (peers[i] == NULL) {
>>>> -            err = -ENOENT;
>>>> -            goto out;
>>>> -        }
>>>>  
>>>>          if (peers[i]->peer) {
>>>>              err = -EEXIST;
>>>>
>>>
>>> Reviewed-by: Laurent Vivier <lvivier@redhat.com>
>>
>> So who can pick up such patches? We don't have a maintainer for the QDEV
>> subsystem anymore... should it go via qemu-trivial?
> 
> It seems trivial enough to go via qemu-trivial, but it would be good to
> have a R-b from someone else than me to confirm.

I would have expected Jason to pick it up, after all it's networking,
but it's okay if you want to take it too.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Thanks,

Paolo

Re: [Qemu-devel] [PATCH] hw: set_netdev: remove useless code

[Laurent Vivier]

On 09/11/2018 11:00, Paolo Bonzini wrote:
> On 09/11/2018 09:29, Laurent Vivier wrote:
>> On 09/11/2018 09:26, Thomas Huth wrote:
>>> On 2018-11-09 09:21, Laurent Vivier wrote:
>>>> On 09/11/2018 09:13, Li Qiang wrote:
>>>>> In set_netdev(), the peers[i] is initialized
>>>>> qemu_find_net_clients_except() when i is in
>>>>> 0 between 'queues' it can't be NULL.
>>>>>
>>>>> Signed-off-by: Li Qiang <liq3ea@gmail.com>
>>>>> ---
>>>>>  hw/core/qdev-properties-system.c | 4 ----
>>>>>  1 file changed, 4 deletions(-)
>>>>>
>>>>> diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
>>>>> index 8b22fb5..b45a7ef 100644
>>>>> --- a/hw/core/qdev-properties-system.c
>>>>> +++ b/hw/core/qdev-properties-system.c
>>>>> @@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
>>>>>      }
>>>>>  
>>>>>      for (i = 0; i < queues; i++) {
>>>>> -        if (peers[i] == NULL) {
>>>>> -            err = -ENOENT;
>>>>> -            goto out;
>>>>> -        }
>>>>>  
>>>>>          if (peers[i]->peer) {
>>>>>              err = -EEXIST;
>>>>>
>>>>
>>>> Reviewed-by: Laurent Vivier <lvivier@redhat.com>
>>>
>>> So who can pick up such patches? We don't have a maintainer for the QDEV
>>> subsystem anymore... should it go via qemu-trivial?
>>
>> It seems trivial enough to go via qemu-trivial, but it would be good to
>> have a R-b from someone else than me to confirm.
> 
> I would have expected Jason to pick it up, after all it's networking,
> but it's okay if you want to take it too.

If Jason pushes it before I do, I will remove it from my queue.
[I have the bad habit to rebase by branch before a pull request]

Thanks,
Laurent

[Qemu-devel] qdev has no maintainer (was: [PATCH] hw: set_netdev: remove useless code)

[Markus Armbruster]

Thomas Huth <thuth@redhat.com> writes:

> On 2018-11-09 09:21, Laurent Vivier wrote:
>> On 09/11/2018 09:13, Li Qiang wrote:
>>> In set_netdev(), the peers[i] is initialized
>>> qemu_find_net_clients_except() when i is in
>>> 0 between 'queues' it can't be NULL.
>>>
>>> Signed-off-by: Li Qiang <liq3ea@gmail.com>
>>> ---
>>>  hw/core/qdev-properties-system.c | 4 ----
>>>  1 file changed, 4 deletions(-)
>>>
>>> diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
>>> index 8b22fb5..b45a7ef 100644
>>> --- a/hw/core/qdev-properties-system.c
>>> +++ b/hw/core/qdev-properties-system.c
>>> @@ -288,10 +288,6 @@ static void set_netdev(Object *obj, Visitor *v, const char *name,
>>>      }
>>>  
>>>      for (i = 0; i < queues; i++) {
>>> -        if (peers[i] == NULL) {
>>> -            err = -ENOENT;
>>> -            goto out;
>>> -        }
>>>  
>>>          if (peers[i]->peer) {
>>>              err = -EEXIST;
>>>
>> 
>> Reviewed-by: Laurent Vivier <lvivier@redhat.com>
>
> So who can pick up such patches? We don't have a maintainer for the QDEV
> subsystem anymore... should it go via qemu-trivial?

We've never had a formal qdev maintainer.

"qdev is too important to continue much longer without a maintainer."
Message-ID: <m37h8v85r7.fsf@blackfin.pond.sub.org>
https://lists.gnu.org/archive/html/qemu-devel/2011-06/msg01053.html

The closest we got to appointing one was probably here:
Message-ID: <56B33C2A.8030405@redhat.com>
https://lists.gnu.org/archive/html/qemu-devel/2016-02/msg00949.html

In my opinion, not having a maintainer push or at least direct the cart
is one of the reasons of the relatively sad state of qdev.

Same for QOM, where we had an active maintainer for a while, but no
more.