qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Zhi Hui Li <zhihuili@linux.vnet.ibm.com>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: zhihuili@cn.ibm.com, QEMU-devel <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] About the snapshot
Date: Tue, 06 Dec 2011 18:01:08 +0800	[thread overview]
Message-ID: <4EDDE7E4.1020503@linux.vnet.ibm.com> (raw)
In-Reply-To: <CAJSP0QWnt2At2-wcXCi82ugZmry4wHe=ZpSYtJbRwh09B5bjUw@mail.gmail.com>

On 2011年12月06日 17:40, Stefan Hajnoczi wrote:
> On Tue, Dec 6, 2011 at 9:07 AM, Zhi Hui Li<zhihuili@linux.vnet.ibm.com>  wrote:
>>
>> 1) :
>>
>> for example:
>>
>> BDRVQcowState *s = bs->opaque;
>>
>> s->snapshots
>> s->nb_snapshots
>>
>>
>> 1:use the command:   qemu-img snapshot ./test.qcow2  -c aa
>> the memory of the s->snapshot don't free,
>> if the s->nb_snapshots is large, Does it have some problems.
>>
>> 2: use the command:  qemu-system-x86_64  ./test.qcow2 -snapshot
>> when the program ends, Does it need to free the s->snapshots ?
>
> These two commands are unrelated.  QEMU uses the term "snapshot" for
> several different features:
>
> 1. qemu-img snapshot refers to "internal snapshots" that are contained
> within qcow2 image files.  The savevm/loadvm/delvm monitor commands
> operate on internal snapshots.
>
> 2. qemu -snapshot refers to a temporary qcow2 image file created to
> buffer any data that the guest writes.  When QEMU exits your disk
> image is not modified and the temporary qcow2 file is deleted.  You
> can also apply the buffer to the disk image using the "commit" monitor
> command.

yes, I understand what you say, but the qemu-img and savevm both call 
the function of qcow2_snapshot_create, when I use the command qemu-img 
snapshot ./test.qcow2  -c aa,
but the memory of the s->snapshot don't free.

>
>> 2):
>> in the function of
>> qcow2_update_snapshot_refcount
>>
>> it has some "  goto fail  ";
>> if the function runs some times, then something makes it goto fail,
>> I am not sure whether it will make the refcount  incorrect.
>
> When an error occurs its possible that refcount leaks are introduced
> (the refcount was increment but will never be used), but we should
> never decrement a reference that is still in use.
>
> Please be more specific about the problem so that Kevin or I can take a look.
>
> Stefan
>
Ok, I got it.

thank you very much !

  reply	other threads:[~2011-12-06 10:01 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-12-06  9:07 [Qemu-devel] About the snapshot Zhi Hui Li
2011-12-06  9:40 ` Stefan Hajnoczi
2011-12-06 10:01   ` Zhi Hui Li [this message]
2011-12-06 10:33     ` Stefan Hajnoczi
2011-12-06 14:55       ` Zhi Hui Li
2011-12-06 15:35         ` Stefan Hajnoczi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4EDDE7E4.1020503@linux.vnet.ibm.com \
    --to=zhihuili@linux.vnet.ibm.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@gmail.com \
    --cc=zhihuili@cn.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).