Re: [Qemu-block] [PATCH v11 09/10] qcow2: skip writing zero buffers to empty COW areas

[Anton Nefedov]

On 21/12/2018 7:16 PM, Vladimir Sementsov-Ogievskiy wrote:
> 18.12.2018 10:57, Anton Nefedov wrote:
>> If COW areas of the newly allocated clusters are zeroes on the backing image,
>> efficient bdrv_write_zeroes(flags=BDRV_REQ_ALLOCATE) can be used on the whole
>> cluster instead of writing explicit zero buffers later in perform_cow().
>>
>> iotest 060:
>> write to the discarded cluster does not trigger COW anymore.
>> Use a backing image instead.
>>
>> Signed-off-by: Anton Nefedov <address@hidden>
>> ---
>>    qapi/block-core.json       |  4 +-
>>    block/qcow2.h              |  6 +++
>>    block/qcow2-cluster.c      |  2 +-
>>    block/qcow2.c              | 89 +++++++++++++++++++++++++++++++++++++-
>>    block/trace-events         |  1 +
>>    tests/qemu-iotests/060     |  7 ++-
>>    tests/qemu-iotests/060.out |  5 ++-
>>    7 files changed, 108 insertions(+), 6 deletions(-)
>>
>> diff --git a/qapi/block-core.json b/qapi/block-core.json
>> index 762000f31f..204528b3f6 100644
>> --- a/qapi/block-core.json
>> +++ b/qapi/block-core.json
>> @@ -3009,6 +3009,8 @@
>>    #
>>    # @cor_write: a write due to copy-on-read (since 2.11)
>>    #
>> +# @cluster_alloc_space: an allocation of file space for a cluster (since 
>> 4.0)
>> +#
>>    # Since: 2.9
>>    ##
>>    { 'enum': 'BlkdebugEvent', 'prefix': 'BLKDBG',
>> @@ -3027,7 +3029,7 @@
>>                'pwritev_rmw_tail', 'pwritev_rmw_after_tail', 'pwritev',
>>                'pwritev_zero', 'pwritev_done', 'empty_image_prepare',
>>                'l1_shrink_write_table', 'l1_shrink_free_l2_clusters',
>> -            'cor_write'] }
>> +            'cor_write', 'cluster_alloc_space'] }
>>    
>>    ##
>>    # @BlkdebugInjectErrorOptions:
>> diff --git a/block/qcow2.h b/block/qcow2.h
>> index a98d24500b..32d2c04bfa 100644
>> --- a/block/qcow2.h
>> +++ b/block/qcow2.h
>> @@ -386,6 +386,12 @@ typedef struct QCowL2Meta
>>         */
>>        Qcow2COWRegion cow_end;
>>    
>> +    /*
>> +     * Indicates that COW regions are already handled and do not require
>> +     * any more processing.
>> +     */
>> +    bool skip_cow;
>> +
>>        /**
> 
> hmm, around it, all comments starts from '/**', so, I think, yours should too.
> (this note doesn't touch your other comments)
> 

that triggers a warning in checkpatch.pl

   WARNING: Block comments use a leading /* on a separate line

>>         * The I/O vector with the data from the actual guest write request.
>>         * If non-NULL, this is meant to be merged together with the data
>> diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
>> index e2737429f5..23e0702027 100644
>> --- a/block/qcow2-cluster.c
>> +++ b/block/qcow2-cluster.c
>> @@ -806,7 +806,7 @@ static int perform_cow(BlockDriverState *bs, QCowL2Meta 
>> *m)
>>        assert(start->offset + start->nb_bytes <= end->offset);
>>        assert(!m->data_qiov || m->data_qiov->size == data_bytes);
>>    
>> -    if (start->nb_bytes == 0 && end->nb_bytes == 0) {
>> +    if ((start->nb_bytes == 0 && end->nb_bytes == 0) || m->skip_cow) {
>>            return 0;
>>        }
>>    
>> diff --git a/block/qcow2.c b/block/qcow2.c
>> index 4897abae5e..161b935962 100644
>> --- a/block/qcow2.c
>> +++ b/block/qcow2.c
>> @@ -2021,6 +2021,11 @@ static bool merge_cow(uint64_t offset, unsigned bytes,
>>                continue;
>>            }
>>    
>> +        /* If COW regions are handled already, skip this too */
>> +        if (m->skip_cow) {
>> +            continue;
>> +        }
>> +
>>            /* The data (middle) region must be immediately after the
>>             * start region */
>>            if (l2meta_cow_start(m) + m->cow_start.nb_bytes != offset) {
>> @@ -2046,6 +2051,77 @@ static bool merge_cow(uint64_t offset, unsigned bytes,
>>        return false;
>>    }
>>    
>> +static bool is_unallocated(BlockDriverState *bs, int64_t offset, int64_t 
>> bytes)
>> +{
>> +    int64_t nr;
>> +    return !bytes ||
>> +        (!bdrv_is_allocated_above(bs, NULL, offset, bytes, &nr) && nr == 
>> bytes);
>> +}
>> +
>> +static bool is_zero_cow(BlockDriverState *bs, QCowL2Meta *m)
>> +{
>> +    /*
>> +     * This check is designed for optimization shortcut so it must be
>> +     * efficient.
>> +     * Instead of is_zero(), use is_unallocated() as it is faster (but not
>> +     * as accurate and can result in false negatives).
>> +     */
>> +    return is_unallocated(bs, m->offset + m->cow_start.offset,
> 
> here you add m->cow_start.offset ...[1]
> 
>> +                          m->cow_start.nb_bytes) &&
>> +           is_unallocated(bs, m->offset + m->cow_end.offset,
>> +                          m->cow_end.nb_bytes);
>> +}
>> +
>> +static int handle_alloc_space(BlockDriverState *bs, QCowL2Meta *l2meta)
>> +{
>> +    BDRVQcow2State *s = bs->opaque;
>> +    QCowL2Meta *m;
>> +
>> +    if (!(bs->file->bs->supported_zero_flags & BDRV_REQ_ALLOCATE)) {
>> +        return 0;
>> +    }
>> +
>> +    if (bs->encrypted) {
>> +        return 0;
>> +    }
>> +
>> +    for (m = l2meta; m != NULL; m = m->next) {
>> +        int ret;
>> +
>> +        if (!m->cow_start.nb_bytes && !m->cow_end.nb_bytes) {
>> +            continue;
>> +        }
>> +
>> +        if (!is_zero_cow(bs, m)) {
>> +            continue;
>> +        }
>> +
>> +        /*
>> +         * Conventional place for qcow2_pre_write_overlap_check() but in 
>> this
>> +         * case it is already done for these clusters
>> +         */
> 
> to be honest, check is done for original request not, not for cow_regions, 
> and,
> on the other hand,
> do_perform_cow_write _does_ the check.
> 
> So, I see two consistent ways:
> 1. you just add check here, like in do_perform_cow_write
> 2. improve check in qcow2_co_pwritev to cover the whole area, and drop checks 
> from
>      perform_cow.
> 
> I prefer 1, as it more helpful in case of bugs in the code (just check 
> exactly what
> you are going to write in the next write).
> 

ok

diff --git a/block/qcow2.c b/block/qcow2.c
index 161b935962..05a7cbebbd 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -2096,16 +2096,18 @@ static int handle_alloc_space(BlockDriverState 
*bs, QCowL2Meta *l2meta)
              continue;
          }

-        /*
-         * Conventional place for qcow2_pre_write_overlap_check() but 
in this
-         * case it is already done for these clusters
-         */
-
-        BLKDBG_EVENT(bs->file, BLKDBG_CLUSTER_ALLOC_SPACE);
          /*
           * instead of writing zero COW buffers,
           * efficiently zero out the whole clusters
           */
+
+        ret = qcow2_pre_write_overlap_check(bs, 0, m->alloc_offset,
+                                            m->nb_clusters * 
s->cluster_size);
+        if (ret < 0) {
+            return ret;
+        }
+
+        BLKDBG_EVENT(bs->file, BLKDBG_CLUSTER_ALLOC_SPACE);
          ret = bdrv_co_pwrite_zeroes(bs->file, m->alloc_offset,
                                      m->nb_clusters * s->cluster_size,
                                      BDRV_REQ_ALLOCATE);

>> +
>> +        BLKDBG_EVENT(bs->file, BLKDBG_CLUSTER_ALLOC_SPACE);
>> +        /*
>> +         * instead of writing zero COW buffers,
>> +         * efficiently zero out the whole clusters
>> +         */
>> +        ret = bdrv_co_pwrite_zeroes(bs->file, m->alloc_offset,
> 
> [1]... and here - not.
> And perform_cow uses start too.
> hmm..
> 
> On the other hand, here we just want to zero all allocated clusters, it is 
> not related
> to cow regions. So, ok.
> 
>> +                                    m->nb_clusters * s->cluster_size,
>> +                                    BDRV_REQ_ALLOCATE);
>> +        if (ret < 0) {
>> +            if (ret != -ENOTSUP && ret != -EAGAIN) {
>> +                return ret;
>> +            }
>> +            continue;
>> +        }
>> +
>> +        trace_qcow2_skip_cow(qemu_coroutine_self(), m->offset, 
>> m->nb_clusters);
>> +        m->skip_cow = true;
>> +    }
>> +    return 0;
>> +}
>> +
> 
> [...]
> 
> 
> with pre-write-check added:
> Reviewed-by: Vladimir Sementsov-Ogievskiy <address@hidden>
> 
> 
>