Re: [Qemu-devel] [PATCH 5/7] qcow2: refactor qcow2_co_pwritev: split out qcow2_co_do_pwritev

[Max Reitz <mreitz@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 3281 bytes --]

On 01.10.18 17:43, Vladimir Sementsov-Ogievskiy wrote:
> 28.09.2018 17:23, Max Reitz wrote:
>> On 07.08.18 19:43, Vladimir Sementsov-Ogievskiy wrote:
>>> Split out block which will be reused in async scheme.
>>>
>>> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
>>> ---
>>>   block/qcow2.c | 138
>>> ++++++++++++++++++++++++++++++++++++----------------------
>>>   1 file changed, 86 insertions(+), 52 deletions(-)
>>>
>>> diff --git a/block/qcow2.c b/block/qcow2.c
>>> index a0df8d4e50..4d669432d1 100644
>>> --- a/block/qcow2.c
>>> +++ b/block/qcow2.c
>>> @@ -2210,6 +2210,85 @@ static bool merge_cow(uint64_t offset,
>>> unsigned bytes,
>>>       return false;
>>>   }
>>>   +/* qcow2_co_do_pwritev
>>> + * Called without s->lock unlocked
>>> + * hd_qiov - temp qiov for any use. It is initialized so it is empty
>>> and
>>> + *           support adding up to qiov->niov + 2 elements
>>> + * l2meta  - if not NULL, qcow2_co_do_pwritev() will consume it.
>>> Caller must not
>>> + *           use it somehow after qcow2_co_do_pwritev() call
>>> + */
>>> +static coroutine_fn int qcow2_co_do_pwritev(BlockDriverState *bs,
>>> +                                            uint64_t
>>> file_cluster_offset,
>>> +                                            uint64_t offset,
>>> +                                            uint64_t bytes,
>>> +                                            QEMUIOVector *qiov,
>>> +                                            uint64_t qiov_offset,
>>> +                                            QCowL2Meta *l2meta)
>>> +{
>>> +    int ret;
>>> +    BDRVQcow2State *s = bs->opaque;
>>> +    void *crypt_buf = NULL;
>>> +    QEMUIOVector hd_qiov;
>>> +    int offset_in_cluster = offset_into_cluster(s, offset);
>>> +
>>> +    qemu_iovec_reset(&hd_qiov);
>> This shouldn't be here.
>>
>>> +    qemu_iovec_init(&hd_qiov, qiov->niov);
>>> +
>>> +    if (bs->encrypted) {
>>> +        assert(s->crypto);
>>> +        assert(bytes <= QCOW_MAX_CRYPT_CLUSTERS * s->cluster_size);
>>> +        crypt_buf = qemu_try_blockalign(bs->file->bs, bytes);
>>> +        qemu_iovec_to_buf(qiov, qiov_offset, crypt_buf, bytes);
>>> +
>>> +        if (qcrypto_block_encrypt(s->crypto,
>>> +                                  (s->crypt_physical_offset ?
>>> +                                   file_cluster_offset +
>>> offset_in_cluster :
>>> +                                   offset),
>>> +                                  crypt_buf,
>>> +                                  bytes, NULL) < 0) {
>> Same question as in the read case: Can't we make do without the bounce
>> buffer?
> 
> I think, we should not modify guest buffers..

Hmm, yes, agreed. O:-)

Max


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]