qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Xu <peterx@redhat.com>
To: guangrong.xiao@gmail.com
Cc: pbonzini@redhat.com, mst@redhat.com, mtosatti@redhat.com,
	qemu-devel@nongnu.org, kvm@vger.kernel.org, dgilbert@redhat.com,
	jiang.biao2@zte.com.cn, wei.w.wang@intel.com,
	Xiao Guangrong <xiaoguangrong@tencent.com>
Subject: Re: [Qemu-devel] [PATCH v2 02/10] migration: stop compression to allocate and free memory frequently
Date: Wed, 28 Mar 2018 17:25:44 +0800	[thread overview]
Message-ID: <20180328092544.GE29554@xz-mi> (raw)
In-Reply-To: <20180327091043.30220-3-xiaoguangrong@tencent.com>

On Tue, Mar 27, 2018 at 05:10:35PM +0800, guangrong.xiao@gmail.com wrote:

[...]

> @@ -357,10 +358,20 @@ static void compress_threads_save_cleanup(void)
>      terminate_compression_threads();
>      thread_count = migrate_compress_threads();
>      for (i = 0; i < thread_count; i++) {
> +        /*
> +         * stream.opaque can be used to store private data, we use it
> +         * as a indicator which shows if the thread is properly init'd
> +         * or not
> +         */
> +        if (!comp_param[i].stream.opaque) {
> +            break;
> +        }

How about using comp_param[i].file?  The opaque seems to be hiding
deeper, and...

>          qemu_thread_join(compress_threads + i);
>          qemu_fclose(comp_param[i].file);
>          qemu_mutex_destroy(&comp_param[i].mutex);
>          qemu_cond_destroy(&comp_param[i].cond);
> +        deflateEnd(&comp_param[i].stream);
> +        comp_param[i].stream.opaque = NULL;
>      }
>      qemu_mutex_destroy(&comp_done_lock);
>      qemu_cond_destroy(&comp_done_cond);
> @@ -370,12 +381,12 @@ static void compress_threads_save_cleanup(void)
>      comp_param = NULL;
>  }
>  
> -static void compress_threads_save_setup(void)
> +static int compress_threads_save_setup(void)
>  {
>      int i, thread_count;
>  
>      if (!migrate_use_compression()) {
> -        return;
> +        return 0;
>      }
>      thread_count = migrate_compress_threads();
>      compress_threads = g_new0(QemuThread, thread_count);
> @@ -383,6 +394,12 @@ static void compress_threads_save_setup(void)
>      qemu_cond_init(&comp_done_cond);
>      qemu_mutex_init(&comp_done_lock);
>      for (i = 0; i < thread_count; i++) {
> +        if (deflateInit(&comp_param[i].stream,
> +                           migrate_compress_level()) != Z_OK) {

(indent issue)

> +            goto exit;
> +        }
> +        comp_param[i].stream.opaque = &comp_param[i];

...here from document:

        ZEXTERN int ZEXPORT deflateInit OF((z_streamp strm, int level));

        Initializes the internal stream state for compression. The
        fields zalloc, zfree and opaque must be initialized before by
        the caller. If zalloc and zfree are set to Z_NULL, deflateInit
        updates them to use default allocation functions.

So shall we init opaque first?  Otherwise looks good to me.

> +
>          /* comp_param[i].file is just used as a dummy buffer to save data,
>           * set its ops to empty.
>           */

Thanks,

-- 
Peter Xu

  reply	other threads:[~2018-03-28  9:26 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-27  9:10 [Qemu-devel] [PATCH v2 00/10] migration: improve and cleanup compression guangrong.xiao
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 01/10] migration: stop compressing page in migration thread guangrong.xiao
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 02/10] migration: stop compression to allocate and free memory frequently guangrong.xiao
2018-03-28  9:25   ` Peter Xu [this message]
2018-03-29  3:41     ` Xiao Guangrong
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 03/10] migration: stop decompression " guangrong.xiao
2018-03-28  9:42   ` Peter Xu
2018-03-29  3:43     ` Xiao Guangrong
2018-03-29  4:14       ` Peter Xu
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 04/10] migration: detect compression and decompression errors guangrong.xiao
2018-03-28  9:59   ` Peter Xu
2018-03-29  3:51     ` Xiao Guangrong
2018-03-29  4:25       ` Peter Xu
2018-03-30  3:11         ` Xiao Guangrong
2018-04-02  4:26           ` Peter Xu
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 05/10] migration: introduce control_save_page() guangrong.xiao
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 06/10] migration: move some code ram_save_host_page guangrong.xiao
2018-03-28 10:05   ` Peter Xu
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 07/10] migration: move calling control_save_page to the common place guangrong.xiao
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 08/10] migration: move calling save_zero_page " guangrong.xiao
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 09/10] migration: introduce save_normal_page() guangrong.xiao
2018-03-27  9:10 ` [Qemu-devel] [PATCH v2 10/10] migration: remove ram_save_compressed_page() guangrong.xiao

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=20180328092544.GE29554@xz-mi \
    --to=peterx@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=guangrong.xiao@gmail.com \
    --cc=jiang.biao2@zte.com.cn \
    --cc=kvm@vger.kernel.org \
    --cc=mst@redhat.com \
    --cc=mtosatti@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=wei.w.wang@intel.com \
    --cc=xiaoguangrong@tencent.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).