Re: [PATCH 1/2] s390x: fix memleaks in cpu_finalize

Re: [PATCH 1/2] s390x: fix memleaks in cpu_finalize

[Pan Nengyuan]

On 2/15/2020 4:32 PM, pannengyuan@huawei.com wrote:

I'm sorry for the mail's subject, it's a single patch.
[PATCH 1/2]  --->  [PATCH]

> From: Pan Nengyuan <pannengyuan@huawei.com>
> 
> This patch fix memleaks when we call tests/qtest/cpu-plug-test on s390x. The leak stack is as follow:
> 
> Direct leak of 48 byte(s) in 1 object(s) allocated from:
>     #0 0x7fb43c7cd970 in __interceptor_calloc (/lib64/libasan.so.5+0xef970)
>     #1 0x7fb43be2149d in g_malloc0 (/lib64/libglib-2.0.so.0+0x5249d)
>     #2 0x558ba96da716 in timer_new_full /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:530
>     #3 0x558ba96da716 in timer_new /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:551
>     #4 0x558ba96da716 in timer_new_ns /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:569
>     #5 0x558ba96da716 in s390_cpu_initfn /mnt/sdb/qemu-new/qemu/target/s390x/cpu.c:285
>     #6 0x558ba9c969ab in object_init_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:372
>     #7 0x558ba9c9eb5f in object_initialize_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:516
>     #8 0x558ba9c9f053 in object_new_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:684
>     #9 0x558ba967ede6 in s390x_new_cpu /mnt/sdb/qemu-new/qemu/hw/s390x/s390-virtio-ccw.c:64
>     #10 0x558ba99764b3 in hmp_cpu_add /mnt/sdb/qemu-new/qemu/hw/core/machine-hmp-cmds.c:57
>     #11 0x558ba9b1c27f in handle_hmp_command /mnt/sdb/qemu-new/qemu/monitor/hmp.c:1082
>     #12 0x558ba96c1b02 in qmp_human_monitor_command /mnt/sdb/qemu-new/qemu/monitor/misc.c:142
> 
> Reported-by: Euler Robot <euler.robot@huawei.com>
> Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
> ---
>  target/s390x/cpu.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c
> index cf84d307c6..fff793a4eb 100644
> --- a/target/s390x/cpu.c
> +++ b/target/s390x/cpu.c
> @@ -294,6 +294,10 @@ static void s390_cpu_finalize(Object *obj)
>  
>      qemu_unregister_reset(s390_cpu_machine_reset_cb, cpu);
>      g_free(cpu->irqstate);
> +    timer_del(cpu->env.tod_timer);
> +    timer_del(cpu->env.cpu_timer);
> +    timer_free(cpu->env.tod_timer);
> +    timer_free(cpu->env.cpu_timer);
>  #endif
>  }
>  
> 

[PATCH 1/2] s390x: fix memleaks in cpu_finalize

[pannengyuan]

From: Pan Nengyuan <pannengyuan@huawei.com>

This patch fix memleaks when we call tests/qtest/cpu-plug-test on s390x. The leak stack is as follow:

Direct leak of 48 byte(s) in 1 object(s) allocated from:
    #0 0x7fb43c7cd970 in __interceptor_calloc (/lib64/libasan.so.5+0xef970)
    #1 0x7fb43be2149d in g_malloc0 (/lib64/libglib-2.0.so.0+0x5249d)
    #2 0x558ba96da716 in timer_new_full /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:530
    #3 0x558ba96da716 in timer_new /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:551
    #4 0x558ba96da716 in timer_new_ns /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:569
    #5 0x558ba96da716 in s390_cpu_initfn /mnt/sdb/qemu-new/qemu/target/s390x/cpu.c:285
    #6 0x558ba9c969ab in object_init_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:372
    #7 0x558ba9c9eb5f in object_initialize_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:516
    #8 0x558ba9c9f053 in object_new_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:684
    #9 0x558ba967ede6 in s390x_new_cpu /mnt/sdb/qemu-new/qemu/hw/s390x/s390-virtio-ccw.c:64
    #10 0x558ba99764b3 in hmp_cpu_add /mnt/sdb/qemu-new/qemu/hw/core/machine-hmp-cmds.c:57
    #11 0x558ba9b1c27f in handle_hmp_command /mnt/sdb/qemu-new/qemu/monitor/hmp.c:1082
    #12 0x558ba96c1b02 in qmp_human_monitor_command /mnt/sdb/qemu-new/qemu/monitor/misc.c:142

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
---
 target/s390x/cpu.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c
index cf84d307c6..fff793a4eb 100644
--- a/target/s390x/cpu.c
+++ b/target/s390x/cpu.c
@@ -294,6 +294,10 @@ static void s390_cpu_finalize(Object *obj)
 
     qemu_unregister_reset(s390_cpu_machine_reset_cb, cpu);
     g_free(cpu->irqstate);
+    timer_del(cpu->env.tod_timer);
+    timer_del(cpu->env.cpu_timer);
+    timer_free(cpu->env.tod_timer);
+    timer_free(cpu->env.cpu_timer);
 #endif
 }
 
-- 
2.18.2

Re: [PATCH 1/2] s390x: fix memleaks in cpu_finalize

[Philippe Mathieu-Daudé]

Hi Pan,

On 2/15/20 9:32 AM, pannengyuan@huawei.com wrote:
> From: Pan Nengyuan <pannengyuan@huawei.com>
> 
> This patch fix memleaks when we call tests/qtest/cpu-plug-test on s390x. The leak stack is as follow:
> 
> Direct leak of 48 byte(s) in 1 object(s) allocated from:
>      #0 0x7fb43c7cd970 in __interceptor_calloc (/lib64/libasan.so.5+0xef970)
>      #1 0x7fb43be2149d in g_malloc0 (/lib64/libglib-2.0.so.0+0x5249d)
>      #2 0x558ba96da716 in timer_new_full /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:530
>      #3 0x558ba96da716 in timer_new /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:551
>      #4 0x558ba96da716 in timer_new_ns /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:569
>      #5 0x558ba96da716 in s390_cpu_initfn /mnt/sdb/qemu-new/qemu/target/s390x/cpu.c:285
>      #6 0x558ba9c969ab in object_init_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:372
>      #7 0x558ba9c9eb5f in object_initialize_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:516
>      #8 0x558ba9c9f053 in object_new_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:684
>      #9 0x558ba967ede6 in s390x_new_cpu /mnt/sdb/qemu-new/qemu/hw/s390x/s390-virtio-ccw.c:64
>      #10 0x558ba99764b3 in hmp_cpu_add /mnt/sdb/qemu-new/qemu/hw/core/machine-hmp-cmds.c:57
>      #11 0x558ba9b1c27f in handle_hmp_command /mnt/sdb/qemu-new/qemu/monitor/hmp.c:1082
>      #12 0x558ba96c1b02 in qmp_human_monitor_command /mnt/sdb/qemu-new/qemu/monitor/misc.c:142
> 
> Reported-by: Euler Robot <euler.robot@huawei.com>
> Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
> ---
>   target/s390x/cpu.c | 4 ++++
>   1 file changed, 4 insertions(+)
> 
> diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c
> index cf84d307c6..fff793a4eb 100644
> --- a/target/s390x/cpu.c
> +++ b/target/s390x/cpu.c
> @@ -294,6 +294,10 @@ static void s390_cpu_finalize(Object *obj)
>   
>       qemu_unregister_reset(s390_cpu_machine_reset_cb, cpu);
>       g_free(cpu->irqstate);
> +    timer_del(cpu->env.tod_timer);
> +    timer_del(cpu->env.cpu_timer);

Similarly to your other cleanups, shouldn't we move the timer_new_ns() 
into a realize() function, then do the timer_del() in unrealize()?

> +    timer_free(cpu->env.tod_timer);
> +    timer_free(cpu->env.cpu_timer);

The timer_free() is correct.

>   #endif
>   }
>   
> 

Re: [PATCH 1/2] s390x: fix memleaks in cpu_finalize

[Pan Nengyuan]

On 2/15/2020 6:28 PM, Philippe Mathieu-Daudé wrote:
> Hi Pan,
> 
> On 2/15/20 9:32 AM, pannengyuan@huawei.com wrote:
>> From: Pan Nengyuan <pannengyuan@huawei.com>
>>
>> This patch fix memleaks when we call tests/qtest/cpu-plug-test on s390x. The leak stack is as follow:
>>
>> Direct leak of 48 byte(s) in 1 object(s) allocated from:
>>      #0 0x7fb43c7cd970 in __interceptor_calloc (/lib64/libasan.so.5+0xef970)
>>      #1 0x7fb43be2149d in g_malloc0 (/lib64/libglib-2.0.so.0+0x5249d)
>>      #2 0x558ba96da716 in timer_new_full /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:530
>>      #3 0x558ba96da716 in timer_new /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:551
>>      #4 0x558ba96da716 in timer_new_ns /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:569
>>      #5 0x558ba96da716 in s390_cpu_initfn /mnt/sdb/qemu-new/qemu/target/s390x/cpu.c:285
>>      #6 0x558ba9c969ab in object_init_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:372
>>      #7 0x558ba9c9eb5f in object_initialize_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:516
>>      #8 0x558ba9c9f053 in object_new_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:684
>>      #9 0x558ba967ede6 in s390x_new_cpu /mnt/sdb/qemu-new/qemu/hw/s390x/s390-virtio-ccw.c:64
>>      #10 0x558ba99764b3 in hmp_cpu_add /mnt/sdb/qemu-new/qemu/hw/core/machine-hmp-cmds.c:57
>>      #11 0x558ba9b1c27f in handle_hmp_command /mnt/sdb/qemu-new/qemu/monitor/hmp.c:1082
>>      #12 0x558ba96c1b02 in qmp_human_monitor_command /mnt/sdb/qemu-new/qemu/monitor/misc.c:142
>>
>> Reported-by: Euler Robot <euler.robot@huawei.com>
>> Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
>> ---
>>   target/s390x/cpu.c | 4 ++++
>>   1 file changed, 4 insertions(+)
>>
>> diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c
>> index cf84d307c6..fff793a4eb 100644
>> --- a/target/s390x/cpu.c
>> +++ b/target/s390x/cpu.c
>> @@ -294,6 +294,10 @@ static void s390_cpu_finalize(Object *obj)
>>         qemu_unregister_reset(s390_cpu_machine_reset_cb, cpu);
>>       g_free(cpu->irqstate);
>> +    timer_del(cpu->env.tod_timer);
>> +    timer_del(cpu->env.cpu_timer);
> 
> Similarly to your other cleanups, shouldn't we move the timer_new_ns() into a realize() function, then do the timer_del() in unrealize()?

Sure, I have no problem with that. I will change and repost it.

Thanks.

> 
>> +    timer_free(cpu->env.tod_timer);
>> +    timer_free(cpu->env.cpu_timer);
> 
> The timer_free() is correct.
> 
>>   #endif
>>   }
>>  
> 
> .

Re: [PATCH 1/2] s390x: fix memleaks in cpu_finalize

[David Hildenbrand]

On 15.02.20 11:28, Philippe Mathieu-Daudé wrote:
> Hi Pan,
> 
> On 2/15/20 9:32 AM, pannengyuan@huawei.com wrote:
>> From: Pan Nengyuan <pannengyuan@huawei.com>
>>
>> This patch fix memleaks when we call tests/qtest/cpu-plug-test on s390x. The leak stack is as follow:
>>
>> Direct leak of 48 byte(s) in 1 object(s) allocated from:
>>      #0 0x7fb43c7cd970 in __interceptor_calloc (/lib64/libasan.so.5+0xef970)
>>      #1 0x7fb43be2149d in g_malloc0 (/lib64/libglib-2.0.so.0+0x5249d)
>>      #2 0x558ba96da716 in timer_new_full /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:530
>>      #3 0x558ba96da716 in timer_new /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:551
>>      #4 0x558ba96da716 in timer_new_ns /mnt/sdb/qemu-new/qemu/include/qemu/timer.h:569
>>      #5 0x558ba96da716 in s390_cpu_initfn /mnt/sdb/qemu-new/qemu/target/s390x/cpu.c:285
>>      #6 0x558ba9c969ab in object_init_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:372
>>      #7 0x558ba9c9eb5f in object_initialize_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:516
>>      #8 0x558ba9c9f053 in object_new_with_type /mnt/sdb/qemu-new/qemu/qom/object.c:684
>>      #9 0x558ba967ede6 in s390x_new_cpu /mnt/sdb/qemu-new/qemu/hw/s390x/s390-virtio-ccw.c:64
>>      #10 0x558ba99764b3 in hmp_cpu_add /mnt/sdb/qemu-new/qemu/hw/core/machine-hmp-cmds.c:57
>>      #11 0x558ba9b1c27f in handle_hmp_command /mnt/sdb/qemu-new/qemu/monitor/hmp.c:1082
>>      #12 0x558ba96c1b02 in qmp_human_monitor_command /mnt/sdb/qemu-new/qemu/monitor/misc.c:142
>>
>> Reported-by: Euler Robot <euler.robot@huawei.com>
>> Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
>> ---
>>   target/s390x/cpu.c | 4 ++++
>>   1 file changed, 4 insertions(+)
>>
>> diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c
>> index cf84d307c6..fff793a4eb 100644
>> --- a/target/s390x/cpu.c
>> +++ b/target/s390x/cpu.c
>> @@ -294,6 +294,10 @@ static void s390_cpu_finalize(Object *obj)
>>   
>>       qemu_unregister_reset(s390_cpu_machine_reset_cb, cpu);
>>       g_free(cpu->irqstate);
>> +    timer_del(cpu->env.tod_timer);
>> +    timer_del(cpu->env.cpu_timer);
> 
> Similarly to your other cleanups, shouldn't we move the timer_new_ns() 
> into a realize() function, then do the timer_del() in unrealize()?

include/hw/qdev-core.h

"Trivial field initializations should go into #TypeInfo.instance_init.
Operations depending on @props static properties should go into @realize."

IMHO, that here is the right thing to do. But not strong feelings.

Reviewed-by: David Hildenbrand <david@redhat.com>


-- 
Thanks,

David / dhildenb