qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v2 2/6] hostmem: AddressSpace has its own map an


From: liu ping fan
Subject: Re: [Qemu-devel] [PATCH v2 2/6] hostmem: AddressSpace has its own map and maintained by RCU prepared style
Date: Mon, 6 May 2013 09:44:01 +0800

On Fri, May 3, 2013 at 5:10 PM, Stefan Hajnoczi <address@hidden> wrote:
> On Fri, May 03, 2013 at 10:45:18AM +0800, Liu Ping Fan wrote:
>> +/**
>> + * Install new regions list
>> + */
>> +static void hostmem_listener_commit(MemoryListener *listener)
>> +{
>> +    HostMem *tmp;
>> +    AddrSpaceMem *as_mem = container_of(listener, AddrSpaceMem, listener);
>> +
>> +    /* writer of cur_hostmem &next_hostmem is serialed by biglock
>
> s/serialed/serialized/
>
Will fix,
>> +         * in hotplug path. So only take care of r/w on cur_hostmem
>> +         */
>
> Indentation.
>
Will fix,
>> @@ -164,18 +203,30 @@ void hostmem_init(void)
>>          .coalesced_mmio_del = hostmem_listener_coalesced_mmio_dummy,
>>          .priority = 10,
>>      };
>> +    as_mem->cur_hostmem = g_new0(HostMem, 1);
>> +    as_mem->cur_hostmem->ref = 1;
>> +    memory_listener_register(&as_mem->listener, as);
>>
>> -    memory_listener_register(&system_mem->listener, &address_space_memory);
>> -    if (system_mem->num_new_regions > 0) {
>> -        hostmem_listener_commit(&system_mem->listener);
>> -    }
>
> The point of this if statement was to make the newly added regions
> visible.  I guess it is not necessary because exec.c is calling us
> before any memory gets initialized now?

Yes, before any memory_region added into system.

Regards,
Pingfan



reply via email to

[Prev in Thread] Current Thread [Next in Thread]