aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorEric Blake <eblake@redhat.com>2017-07-07 15:30:48 -0500
committerPaolo Bonzini <pbonzini@redhat.com>2017-07-14 12:04:42 +0200
commit0c1d50bda7ae03146b4bf7bc3a12389897ea059c (patch)
tree4561bd187c795158a2ce5a0e901ed75596e6b4d8
parent8ecaeae82241b6e317d19e5558e52f9e2f979f74 (diff)
downloadqemu-0c1d50bda7ae03146b4bf7bc3a12389897ea059c.zip
qemu-0c1d50bda7ae03146b4bf7bc3a12389897ea059c.tar.gz
qemu-0c1d50bda7ae03146b4bf7bc3a12389897ea059c.tar.bz2
nbd: Implement NBD_INFO_BLOCK_SIZE on server
The upstream NBD Protocol has defined a new extension to allow the server to advertise block sizes to the client, as well as a way for the client to inform the server that it intends to obey block sizes. Thanks to a recent fix (commit df7b97ff), our real minimum transfer size is always 1 (the block layer takes care of read-modify-write on our behalf), but we're still more efficient if we advertise 512 when the client supports it, as follows: - OPT_INFO, but no NBD_INFO_BLOCK_SIZE: advertise 512, then fail with NBD_REP_ERR_BLOCK_SIZE_REQD; client is free to try something else since we don't disconnect - OPT_INFO with NBD_INFO_BLOCK_SIZE: advertise 512 - OPT_GO, but no NBD_INFO_BLOCK_SIZE: advertise 1 - OPT_GO with NBD_INFO_BLOCK_SIZE: advertise 512 We can also advertise the optimum block size (presumably the cluster size, when exporting a qcow2 file), and our absolute maximum transfer size of 32M, to help newer clients avoid EINVAL failures or abrupt disconnects on oversize requests. We do not reject clients for using the older NBD_OPT_EXPORT_NAME; we are no worse off for those clients than we used to be. Signed-off-by: Eric Blake <eblake@redhat.com> Message-Id: <20170707203049.534-9-eblake@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-rw-r--r--nbd/server.c48
-rw-r--r--nbd/trace-events1
2 files changed, 45 insertions, 4 deletions
diff --git a/nbd/server.c b/nbd/server.c
index e84d012..49ed574 100644
--- a/nbd/server.c
+++ b/nbd/server.c
@@ -365,6 +365,8 @@ static int nbd_negotiate_handle_info(NBDClient *client, uint32_t length,
uint16_t request;
uint32_t namelen;
bool sendname = false;
+ bool blocksize = false;
+ uint32_t sizes[3];
char buf[sizeof(uint64_t) + sizeof(uint16_t)];
const char *msg;
@@ -412,11 +414,16 @@ static int nbd_negotiate_handle_info(NBDClient *client, uint32_t length,
length -= sizeof(request);
trace_nbd_negotiate_handle_info_request(request,
nbd_info_lookup(request));
- /* For now, we only care about NBD_INFO_NAME; everything else
- * is either a request we don't know or something we send
- * regardless of request. */
- if (request == NBD_INFO_NAME) {
+ /* We care about NBD_INFO_NAME and NBD_INFO_BLOCK_SIZE;
+ * everything else is either a request we don't know or
+ * something we send regardless of request */
+ switch (request) {
+ case NBD_INFO_NAME:
sendname = true;
+ break;
+ case NBD_INFO_BLOCK_SIZE:
+ blocksize = true;
+ break;
}
}
@@ -448,6 +455,27 @@ static int nbd_negotiate_handle_info(NBDClient *client, uint32_t length,
}
}
+ /* Send NBD_INFO_BLOCK_SIZE always, but tweak the minimum size
+ * according to whether the client requested it, and according to
+ * whether this is OPT_INFO or OPT_GO. */
+ /* minimum - 1 for back-compat, or 512 if client is new enough.
+ * TODO: consult blk_bs(blk)->bl.request_alignment? */
+ sizes[0] = (opt == NBD_OPT_INFO || blocksize) ? BDRV_SECTOR_SIZE : 1;
+ /* preferred - Hard-code to 4096 for now.
+ * TODO: is blk_bs(blk)->bl.opt_transfer appropriate? */
+ sizes[1] = 4096;
+ /* maximum - At most 32M, but smaller as appropriate. */
+ sizes[2] = MIN(blk_get_max_transfer(exp->blk), NBD_MAX_BUFFER_SIZE);
+ trace_nbd_negotiate_handle_info_block_size(sizes[0], sizes[1], sizes[2]);
+ cpu_to_be32s(&sizes[0]);
+ cpu_to_be32s(&sizes[1]);
+ cpu_to_be32s(&sizes[2]);
+ rc = nbd_negotiate_send_info(client, opt, NBD_INFO_BLOCK_SIZE,
+ sizeof(sizes), sizes, errp);
+ if (rc < 0) {
+ return rc;
+ }
+
/* Send NBD_INFO_EXPORT always */
trace_nbd_negotiate_new_style_size_flags(exp->size,
exp->nbdflags | myflags);
@@ -459,6 +487,18 @@ static int nbd_negotiate_handle_info(NBDClient *client, uint32_t length,
return rc;
}
+ /* If the client is just asking for NBD_OPT_INFO, but forgot to
+ * request block sizes, return an error.
+ * TODO: consult blk_bs(blk)->request_align, and only error if it
+ * is not 1? */
+ if (opt == NBD_OPT_INFO && !blocksize) {
+ return nbd_negotiate_send_rep_err(client->ioc,
+ NBD_REP_ERR_BLOCK_SIZE_REQD, opt,
+ errp,
+ "request NBD_INFO_BLOCK_SIZE to "
+ "use this export");
+ }
+
/* Final reply */
rc = nbd_negotiate_send_rep(client->ioc, NBD_REP_ACK, opt, errp);
if (rc < 0) {
diff --git a/nbd/trace-events b/nbd/trace-events
index bd09fab..be3dce7 100644
--- a/nbd/trace-events
+++ b/nbd/trace-events
@@ -39,6 +39,7 @@ nbd_negotiate_handle_export_name_request(const char *name) "Client requested exp
nbd_negotiate_send_info(int info, const char *name, uint32_t length) "Sending NBD_REP_INFO type %d (%s) with remaining length %" PRIu32
nbd_negotiate_handle_info_requests(int requests) "Client requested %d items of info"
nbd_negotiate_handle_info_request(int request, const char *name) "Client requested info %d (%s)"
+nbd_negotiate_handle_info_block_size(uint32_t minimum, uint32_t preferred, uint32_t maximum) "advertising minimum 0x%" PRIx32 ", preferred 0x%" PRIx32 ", maximum 0x%" PRIx32
nbd_negotiate_handle_starttls(void) "Setting up TLS"
nbd_negotiate_handle_starttls_handshake(void) "Starting TLS handshake"
nbd_negotiate_options_flags(uint32_t flags) "Received client flags 0x%" PRIx32