qemu-devel
[Top][All Lists]
Advanced

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

Re: [Qemu-devel] [RFC V1 12/14] qcow2: Add qcow2_dedup_update_metrics to


From: Eric Blake
Subject: Re: [Qemu-devel] [RFC V1 12/14] qcow2: Add qcow2_dedup_update_metrics to compute dedup RAM usage.
Date: Thu, 17 Jan 2013 08:40:19 -0700
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130110 Thunderbird/17.0.2

On 01/17/2013 04:11 AM, Benoît Canet wrote:
> Le Wednesday 16 Jan 2013 à 13:10:12 (-0700), Eric Blake a écrit :
>> On 01/16/2013 09:25 AM, Benoît Canet wrote:
>>> ---
>>>  block/qcow2-dedup.c |   13 +++++++++++++
>>>  block/qcow2.h       |    1 +
>>>  2 files changed, 14 insertions(+)
>>>
>>> diff --git a/block/qcow2-dedup.c b/block/qcow2-dedup.c
>>> index db23b71..4305746 100644
>>> --- a/block/qcow2-dedup.c
>>> +++ b/block/qcow2-dedup.c
>>> @@ -1311,3 +1311,16 @@ void qcow2_dedup_close(BlockDriverState *bs)
>>>  {
>>>      qcow2_dedup_free(bs);
>>>  }
>>> +
>>> +#define GTREE_NODE_SIZE sizeof(int) * 5
>>
>> Improperly parenthesized.  Also, this feels like a magic number, is
>> there an actual sizeof(struct) you could use instead of hand-computing
>> how much is used per node?
> 
> No the glib implementation totally hide it's structures.

Also, are you sure that this value is correct on both 32-bit and 64-bit
machines, or does a gtree node include a void* which changes how much
memory it uses based on platform?  Even adding a comment pointing to the
internal glib implementation that you were copying from would be
helpful, so that it doesn't feel quite so magic.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

Attachment: signature.asc
Description: OpenPGP digital signature


reply via email to

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