Re: [Qemu-devel] [PATCH v12 07/17] qmp: Add support of "dirty-bitmap" sync mode for drive-backup

[John Snow]

On 02/13/2015 12:33 PM, Vladimir Sementsov-Ogievskiy wrote:
> On 10.02.2015 04:35, John Snow wrote:
> > ......
> > @@ -278,28 +305,61 @@ static void coroutine_fn backup_run(void *opaque)
> >               qemu_coroutine_yield();
> >               job->common.busy = true;
> >           }
> > +    } else if (job->sync_mode == MIRROR_SYNC_MODE_DIRTY_BITMAP) {
> > +        /* Dirty Bitmap sync has a slightly different iteration
> > method */
> > +        HBitmapIter hbi;
> > +        int64_t sector;
> > +        int64_t cluster;
> > +        int64_t last_cluster = -1;
> > +        bool polyrhythmic;
> > +
> > +        bdrv_dirty_iter_init(bs, job->sync_bitmap, &hbi);
> > +        /* Does the granularity happen to match our backup cluster
> > size? */
> > +        polyrhythmic =
> > (bdrv_dirty_bitmap_granularity(job->sync_bitmap) !=
> > +                        BACKUP_CLUSTER_SIZE);
> let it be false, i.e. granularity == cluster
> > +
> > +        /* Find the next dirty /sector/ and copy that /cluster/ */
> > +        while ((sector = hbitmap_iter_next(&hbi)) != -1) {
> then, don't we skip here the very first cluster, if it is dirty?

I don't think so. The HBitmapIterator is still in its initial state 
here, and it can and will return 0 if the first sector is dirty.

The iotest submitted in v12 tests writes to the very first sector, and I 
just verified it quickly that this function *does* return 0 for the 
first go-around.

> > +            if (yield_and_check(job)) {
> > +                goto leave;
> > +            }
> > +            cluster = sector / BACKUP_SECTORS_PER_CLUSTER;
> > +
> > +            /* Play some catchup with the progress meter */
> > +            if (cluster != last_cluster + 1) {
> > +                job->common.offset += ((cluster - last_cluster - 1) *
> > +                                       BACKUP_CLUSTER_SIZE);
> > +            }
> > +
> > +            do {
> > +                ret = backup_do_cow(bs, cluster *
> > BACKUP_SECTORS_PER_CLUSTER,
> > +                                    BACKUP_SECTORS_PER_CLUSTER,
> > &error_is_read);
> > +                if ((ret < 0) &&
> > +                    backup_error_action(job, error_is_read, -ret) ==
> > +                    BLOCK_ERROR_ACTION_REPORT) {
> > +                    goto leave;
> > +                }
> > +            } while (ret < 0);
> > +
> > +            /* Advance (or rewind) our iterator if we need to. */
> > +            if (polyrhythmic) {
> > +                bdrv_set_dirty_iter(&hbi,
> > +                                    (cluster + 1) *
> > BACKUP_SECTORS_PER_CLUSTER);
> > +            }
> > +
> > +            last_cluster = cluster;
> > +        }
> > +
> > +        /* Play some final catchup with the progress meter */
> > +        if (last_cluster + 1 < end) {
> > +            job->common.offset += ((end - last_cluster - 1) *
> > +                                   BACKUP_CLUSTER_SIZE);
> > +        }
> > +
> >       } else {

--
—js