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 <anton.nefe...@virtuozzo.com> > --- > 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) > * 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). > + > + 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 <vsement...@virtuozzo.com> -- Best regards, Vladimir