[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v11 09/10] qcow2: skip writing zero buffers to e
From: |
Anton Nefedov |
Subject: |
Re: [Qemu-devel] [PATCH v11 09/10] qcow2: skip writing zero buffers to empty COW areas |
Date: |
Mon, 24 Dec 2018 08:21:37 +0000 |
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>
>
>
>
- [Qemu-devel] [PATCH v11 02/10] blkverify: set supported write/zero flags, (continued)
- [Qemu-devel] [PATCH v11 02/10] blkverify: set supported write/zero flags, Anton Nefedov, 2018/12/18
- [Qemu-devel] [PATCH v11 03/10] quorum: set supported write flags, Anton Nefedov, 2018/12/18
- [Qemu-devel] [PATCH v11 06/10] file-posix: reset fallocate-related flags without CONFIG_FALLOCATE*, Anton Nefedov, 2018/12/18
- [Qemu-devel] [PATCH v11 05/10] block: treat BDRV_REQ_ALLOCATE as serialising, Anton Nefedov, 2018/12/18
- [Qemu-devel] [PATCH v11 04/10] block: introduce BDRV_REQ_ALLOCATE flag, Anton Nefedov, 2018/12/18
- [Qemu-devel] [PATCH v11 08/10] block: support BDRV_REQ_ALLOCATE in passthrough drivers, Anton Nefedov, 2018/12/18
- [Qemu-devel] [PATCH v11 09/10] qcow2: skip writing zero buffers to empty COW areas, Anton Nefedov, 2018/12/18
- [Qemu-devel] [PATCH v11 07/10] file-posix: support BDRV_REQ_ALLOCATE, Anton Nefedov, 2018/12/18
- [Qemu-devel] [PATCH v11 10/10] iotest 134: test cluster-misaligned encrypted write, Anton Nefedov, 2018/12/18