qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: John Snow <jsnow@redhat.com>
To: Vladimir Sementsov-Ogievskiy <vsementsov@parallels.com>,
	qemu-devel@nongnu.org
Cc: kwolf@redhat.com, den@openvz.org, stefanha@redhat.com
Subject: Re: [Qemu-devel] [PATCH 5/9] block: BdrvDirtyBitmap store/restore interface
Date: Thu, 08 Jan 2015 16:22:09 -0500	[thread overview]
Message-ID: <54AEF501.4000407@redhat.com> (raw)
In-Reply-To: <1418307457-25996-6-git-send-email-vsementsov@parallels.com>



On 12/11/2014 09:17 AM, Vladimir Sementsov-Ogievskiy wrote:
> Several functions to provide necessary access to BdrvDirtyBitmap for
> block-migration.c
>
> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@parallels.com>
> ---
>   block.c               | 61 +++++++++++++++++++++++++++++++++++++++++++++++++--
>   include/block/block.h | 10 +++++++++
>   2 files changed, 69 insertions(+), 2 deletions(-)
>
> diff --git a/block.c b/block.c
> index 6edf1dc..7d42620 100644
> --- a/block.c
> +++ b/block.c
> @@ -5511,8 +5511,65 @@ void bdrv_reset_dirty_bitmap(BlockDriverState *bs, BdrvDirtyBitmap *bitmap,
>       hbitmap_reset(bitmap->bitmap, cur_sector, nr_sectors);
>   }
>
> -static void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector,
> -                           int nr_sectors)
> +const char *bdrv_dirty_bitmap_name(const BdrvDirtyBitmap *bitmap)
> +{
> +    return bitmap->name;
> +}
> +
> +uint64_t bdrv_dbm_data_size(const BdrvDirtyBitmap *bitmap, uint64_t count)
> +{
> +    return hbitmap_data_size(bitmap->bitmap, count);
> +}
> +

Stefan recommended to me for V10 that I avoid using different 
abbreviations such as "dbm" to avoid lengthy functions, so I removed any 
mention of "DBM" from my patches. We should coordinate and ensure any 
abbreviations we use for the BdrvDirtyBitmap functions are used 
consistently... or not at all.

> +void bdrv_dbm_store_data(const BdrvDirtyBitmap *bitmap, uint8_t *buf,
> +                         uint64_t start, uint64_t count)
> +{
> +    hbitmap_store_data(bitmap->bitmap, buf, start, count);
> +}
> +
> +void bdrv_dbm_restore_data(BdrvDirtyBitmap *bitmap, uint8_t *buf,
> +                           uint64_t start, uint64_t count)
> +{
> +    hbitmap_restore_data(bitmap->bitmap, buf, start, count);
> +}
> +
> +BdrvDirtyBitmap **bdrv_dbm_find_all_named(BlockDriverState *bs, int *count)
> +{
> +    BdrvDirtyBitmap *bm, **res, **iter;
> +    assert(count);
> +

Should force *count back to zero before usage.

> +    QLIST_FOREACH(bm, &bs->dirty_bitmaps, list) {
> +        if (bm->name != NULL) {
> +            (*count)++;
> +        }
> +    }
> +
> +    iter = res = g_malloc(sizeof(*res) * (*count));
> +    QLIST_FOREACH(bm, &bs->dirty_bitmaps, list) {
> +        if (bm->name != NULL) {
> +            *iter++ = bm;
> +        }
> +    }
> +
> +    return res;
> +}
> +
> +void bdrv_dbm_restore_finish(void)
> +{
> +    BlockDriverState *bs;
> +    BdrvDirtyBitmap *bm;
> +
> +    for (bs = bdrv_next(NULL); bs != NULL; bs = bdrv_next(bs)) {
> +        QLIST_FOREACH(bm, &bs->dirty_bitmaps, list) {
> +            if (bm->name != NULL) {
> +                hbitmap_restore_finish(bm->bitmap);
> +            }
> +        }
> +    }
> +}
> +
> +void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector,
> +                    int nr_sectors)

Is removing the static keyword here for set_dirty intentional?

>   {
>       BdrvDirtyBitmap *bitmap;
>       QLIST_FOREACH(bitmap, &bs->dirty_bitmaps, list) {
> diff --git a/include/block/block.h b/include/block/block.h
> index b21233c..09eff80 100644
> --- a/include/block/block.h
> +++ b/include/block/block.h
> @@ -459,6 +459,16 @@ void bdrv_dirty_iter_init(BlockDriverState *bs,
>   void bdrv_dirty_iter_set(struct HBitmapIter *hbi, int64_t offset);
>   int64_t bdrv_get_dirty_count(BlockDriverState *bs, BdrvDirtyBitmap *bitmap);
>
> +uint64_t bdrv_dbm_data_size(const BdrvDirtyBitmap *bitmap, uint64_t count);
> +void bdrv_dbm_store_data(const BdrvDirtyBitmap *bitmap, uint8_t *buf,
> +                         uint64_t start, uint64_t count);
> +void bdrv_dbm_restore_data(BdrvDirtyBitmap *bitmap, uint8_t *buf,
> +                           uint64_t start, uint64_t count);
> +bool bdrv_dbm_is_named(BdrvDirtyBitmap *bitmap);
> +const char *bdrv_dirty_bitmap_name(const BdrvDirtyBitmap *bitmap);
> +BdrvDirtyBitmap **bdrv_dbm_find_all_named(BlockDriverState *bs, int *count);
> +void bdrv_dbm_restore_finish(void);
> +
>   void bdrv_enable_copy_on_read(BlockDriverState *bs);
>   void bdrv_disable_copy_on_read(BlockDriverState *bs);
>
>

  reply	other threads:[~2015-01-08 21:22 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-12-11 14:17 [Qemu-devel] [PATCH 0/8] Dirty bitmaps migration Vladimir Sementsov-Ogievskiy
2014-12-11 14:17 ` [Qemu-devel] [PATCH 1/9] block: rename bdrv_reset_dirty_bitmap Vladimir Sementsov-Ogievskiy
2015-01-08 21:19   ` John Snow
2014-12-11 14:17 ` [Qemu-devel] [PATCH 2/9] block-migration: fix pending() return value Vladimir Sementsov-Ogievskiy
2015-01-08 21:20   ` John Snow
2015-01-09 19:01     ` Dr. David Alan Gilbert
2014-12-11 14:17 ` [Qemu-devel] [PATCH 3/9] block: fix spoiling all dirty bitmaps by mirror and migration Vladimir Sementsov-Ogievskiy
2015-01-08 21:20   ` John Snow
2014-12-11 14:17 ` [Qemu-devel] [PATCH 4/9] hbitmap: store / restore Vladimir Sementsov-Ogievskiy
2015-01-08 21:21   ` John Snow
2015-01-08 21:37     ` Paolo Bonzini
2015-01-13 12:59     ` Vladimir Sementsov-Ogievskiy
2015-01-13 17:08       ` John Snow
2015-01-14 10:29         ` Vladimir Sementsov-Ogievskiy
2014-12-11 14:17 ` [Qemu-devel] [PATCH 5/9] block: BdrvDirtyBitmap store/restore interface Vladimir Sementsov-Ogievskiy
2015-01-08 21:22   ` John Snow [this message]
2015-01-14 11:27   ` Vladimir Sementsov-Ogievskiy
2014-12-11 14:17 ` [Qemu-devel] [PATCH 6/9] block-migration: tiny refactoring Vladimir Sementsov-Ogievskiy
2015-01-08 21:23   ` John Snow
2015-01-14 12:26     ` Vladimir Sementsov-Ogievskiy
2015-01-14 16:53       ` John Snow
2014-12-11 14:17 ` [Qemu-devel] [PATCH 7/9] block-migration: remove not needed iothread lock Vladimir Sementsov-Ogievskiy
2015-01-08 21:24   ` John Snow
2015-01-16 12:54     ` Vladimir Sementsov-Ogievskiy
2015-01-08 22:28   ` Paolo Bonzini
2015-01-16 13:03     ` Vladimir Sementsov-Ogievskiy
2014-12-11 14:17 ` [Qemu-devel] [PATCH 8/9] migration: add dirty parameter Vladimir Sementsov-Ogievskiy
2014-12-11 15:18   ` Eric Blake
2014-12-15  8:33     ` Vladimir Sementsov-Ogievskiy
2015-01-08 21:51   ` John Snow
2015-01-08 22:29     ` Eric Blake
2015-01-08 22:31       ` John Snow
2015-01-08 22:37     ` Paolo Bonzini
2014-12-11 14:17 ` [Qemu-devel] [PATCH 9/9] block-migration: add named dirty bitmaps migration Vladimir Sementsov-Ogievskiy
2015-01-08 22:05   ` John Snow
2015-01-17 17:17     ` Vladimir Sementsov-Ogievskiy
2015-01-20 17:25       ` John Snow
2015-01-08 22:36   ` Paolo Bonzini
2015-01-08 22:45     ` Eric Blake
2015-01-08 22:49       ` Paolo Bonzini
2015-01-12 14:20     ` Vladimir Sementsov-Ogievskiy
2015-01-12 14:42       ` Paolo Bonzini
2015-01-12 16:48   ` Paolo Bonzini
2015-01-12 17:31     ` John Snow
2015-01-12 19:09       ` Paolo Bonzini
2015-01-13  9:16         ` Vladimir Sementsov-Ogievskiy
2015-01-13 16:35           ` John Snow

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=54AEF501.4000407@redhat.com \
    --to=jsnow@redhat.com \
    --cc=den@openvz.org \
    --cc=kwolf@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    --cc=vsementsov@parallels.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).