nbd: Implement NBD_CMD_WRITE_ZEROES on client

Upstream NBD protocol recently added the ability to efficiently
write zeroes without having to send the zeroes over the wire,
along with a flag to control whether the client wants a hole.

The generic block code takes care of falling back to the obvious
write of lots of zeroes if we return -ENOTSUP because the server
does not have WRITE_ZEROES.

Ideally, since NBD_CMD_WRITE_ZEROES does not involve any data
over the wire, we want to support transactions that are much
larger than the normal 32M limit imposed on NBD_CMD_WRITE.  But
the server may still have a limit smaller than UINT_MAX, so
until experimental NBD protocol additions for advertising various
command sizes is finalized (see [1], [2]), for now we just stick to
the same limits as normal writes.

[1] https://github.com/yoe/nbd/blob/extension-info/doc/proto.md
[2] https://sourceforge.net/p/nbd/mailman/message/35081223/

Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <1476469998-28592-17-git-send-email-eblake@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This commit is contained in:
Eric Blake 2016-10-14 13:33:18 -05:00 committed by Paolo Bonzini
parent 1f4d6d18ed
commit fa778fffdf
3 changed files with 41 additions and 0 deletions

View file

@ -275,6 +275,41 @@ int nbd_client_co_pwritev(BlockDriverState *bs, uint64_t offset,
return -reply.error;
}
int nbd_client_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,
int count, BdrvRequestFlags flags)
{
ssize_t ret;
NBDClientSession *client = nbd_get_client_session(bs);
NBDRequest request = {
.type = NBD_CMD_WRITE_ZEROES,
.from = offset,
.len = count,
};
NBDReply reply;
if (!(client->nbdflags & NBD_FLAG_SEND_WRITE_ZEROES)) {
return -ENOTSUP;
}
if (flags & BDRV_REQ_FUA) {
assert(client->nbdflags & NBD_FLAG_SEND_FUA);
request.flags |= NBD_CMD_FLAG_FUA;
}
if (!(flags & BDRV_REQ_MAY_UNMAP)) {
request.flags |= NBD_CMD_FLAG_NO_HOLE;
}
nbd_coroutine_start(client, &request);
ret = nbd_co_send_request(bs, &request, NULL);
if (ret < 0) {
reply.error = -ret;
} else {
nbd_co_receive_reply(client, &request, &reply, NULL);
}
nbd_coroutine_end(client, &request);
return -reply.error;
}
int nbd_client_co_flush(BlockDriverState *bs)
{
NBDClientSession *client = nbd_get_client_session(bs);

View file

@ -48,6 +48,8 @@ int nbd_client_co_pdiscard(BlockDriverState *bs, int64_t offset, int count);
int nbd_client_co_flush(BlockDriverState *bs);
int nbd_client_co_pwritev(BlockDriverState *bs, uint64_t offset,
uint64_t bytes, QEMUIOVector *qiov, int flags);
int nbd_client_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,
int count, BdrvRequestFlags flags);
int nbd_client_co_preadv(BlockDriverState *bs, uint64_t offset,
uint64_t bytes, QEMUIOVector *qiov, int flags);

View file

@ -466,6 +466,7 @@ static int nbd_co_flush(BlockDriverState *bs)
static void nbd_refresh_limits(BlockDriverState *bs, Error **errp)
{
bs->bl.max_pdiscard = NBD_MAX_BUFFER_SIZE;
bs->bl.max_pwrite_zeroes = NBD_MAX_BUFFER_SIZE;
bs->bl.max_transfer = NBD_MAX_BUFFER_SIZE;
}
@ -558,6 +559,7 @@ static BlockDriver bdrv_nbd = {
.bdrv_file_open = nbd_open,
.bdrv_co_preadv = nbd_client_co_preadv,
.bdrv_co_pwritev = nbd_client_co_pwritev,
.bdrv_co_pwrite_zeroes = nbd_client_co_pwrite_zeroes,
.bdrv_close = nbd_close,
.bdrv_co_flush_to_os = nbd_co_flush,
.bdrv_co_pdiscard = nbd_client_co_pdiscard,
@ -576,6 +578,7 @@ static BlockDriver bdrv_nbd_tcp = {
.bdrv_file_open = nbd_open,
.bdrv_co_preadv = nbd_client_co_preadv,
.bdrv_co_pwritev = nbd_client_co_pwritev,
.bdrv_co_pwrite_zeroes = nbd_client_co_pwrite_zeroes,
.bdrv_close = nbd_close,
.bdrv_co_flush_to_os = nbd_co_flush,
.bdrv_co_pdiscard = nbd_client_co_pdiscard,
@ -594,6 +597,7 @@ static BlockDriver bdrv_nbd_unix = {
.bdrv_file_open = nbd_open,
.bdrv_co_preadv = nbd_client_co_preadv,
.bdrv_co_pwritev = nbd_client_co_pwritev,
.bdrv_co_pwrite_zeroes = nbd_client_co_pwrite_zeroes,
.bdrv_close = nbd_close,
.bdrv_co_flush_to_os = nbd_co_flush,
.bdrv_co_pdiscard = nbd_client_co_pdiscard,