From: Jon Maloy <jon.maloy@ericsson.com>
To: davem@davemloft.net
Cc: Jon Maloy <jon.maloy@ericsson.com>,
netdev@vger.kernel.org,
Paul Gortmaker <paul.gortmaker@windriver.com>,
tipc-discussion@lists.sourceforge.net
Subject: [PATCH net-next v2 3/7] tipc: eliminate unnecessary linearization of incoming buffers
Date: Fri, 13 Mar 2015 16:08:07 -0400 [thread overview]
Message-ID: <1426277291-2427-4-git-send-email-jon.maloy@ericsson.com> (raw)
In-Reply-To: <1426277291-2427-1-git-send-email-jon.maloy@ericsson.com>
Currently, TIPC linearizes all incoming buffers directly at reception
before passing them upwards in the stack. This is clearly a waste of
CPU resources, and must be avoided.
In this commit, we eliminate this unnecessary linearization. We still
ensure that at least the message header is linear, and that the buffer
is linearized where this is still needed, i.e. when unbundling and when
reversing messages.
In addition, we ensure that fragmented messages are validated after
reassembly before delivering them upwards in the stack.
Reviewed-by: Erik Hugne <erik.hugne@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/link.c | 5 -----
net/tipc/msg.c | 14 ++++++++++----
2 files changed, 10 insertions(+), 9 deletions(-)
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 944c8c6..8c6639d 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1075,13 +1075,8 @@ void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b_ptr)
if (unlikely(!tipc_msg_validate(skb)))
goto discard;
- /* Ensure message data is a single contiguous unit */
- if (unlikely(skb_linearize(skb)))
- goto discard;
-
/* Handle arrival of a non-unicast link message */
msg = buf_msg(skb);
-
if (unlikely(msg_non_seq(msg))) {
if (msg_user(msg) == LINK_CONFIG)
tipc_disc_rcv(net, skb, b_ptr);
diff --git a/net/tipc/msg.c b/net/tipc/msg.c
index 4a64caf..ff8c64c 100644
--- a/net/tipc/msg.c
+++ b/net/tipc/msg.c
@@ -165,6 +165,9 @@ int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf)
}
if (fragid == LAST_FRAGMENT) {
+ TIPC_SKB_CB(head)->validated = false;
+ if (unlikely(!tipc_msg_validate(head)))
+ goto err;
*buf = head;
TIPC_SKB_CB(head)->tail = NULL;
*headbuf = NULL;
@@ -172,7 +175,6 @@ int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf)
}
*buf = NULL;
return 0;
-
err:
pr_warn_ratelimited("Unable to build fragment list\n");
kfree_skb(*buf);
@@ -378,10 +380,14 @@ bool tipc_msg_bundle(struct sk_buff_head *list, struct sk_buff *skb, u32 mtu)
*/
bool tipc_msg_extract(struct sk_buff *skb, struct sk_buff **iskb, int *pos)
{
- struct tipc_msg *msg = buf_msg(skb);
+ struct tipc_msg *msg;
int imsz;
- struct tipc_msg *imsg = (struct tipc_msg *)(msg_data(msg) + *pos);
+ struct tipc_msg *imsg;
+ if (unlikely(skb_linearize(skb)))
+ return false;
+ msg = buf_msg(skb);
+ imsg = (struct tipc_msg *)(msg_data(msg) + *pos);
/* Is there space left for shortest possible message? */
if (*pos > (msg_data_sz(msg) - SHORT_H_SIZE))
goto none;
@@ -463,11 +469,11 @@ bool tipc_msg_reverse(u32 own_addr, struct sk_buff *buf, u32 *dnode,
if (skb_linearize(buf))
goto exit;
+ msg = buf_msg(buf);
if (msg_dest_droppable(msg))
goto exit;
if (msg_errcode(msg))
goto exit;
-
memcpy(&ohdr, msg, msg_hdr_sz(msg));
imp = min_t(uint, imp + 1, TIPC_CRITICAL_IMPORTANCE);
if (msg_isdata(msg))
--
1.9.1
------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
next prev parent reply other threads:[~2015-03-13 20:08 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-13 20:08 [PATCH net-next v2 0/7] tipc: some optimizations and impovements Jon Maloy
2015-03-13 20:08 ` [PATCH net-next v2 1/7] tipc: add framework for node capabilities exchange Jon Maloy
2015-03-13 20:08 ` [PATCH net-next v2 2/7] tipc: move message validation function to msg.c Jon Maloy
2015-03-13 20:08 ` Jon Maloy [this message]
2015-03-13 20:08 ` [PATCH net-next v2 4/7] tipc: extract bundled buffers by cloning instead of copying Jon Maloy
2015-03-13 20:08 ` [PATCH net-next v2 5/7] tipc: eliminate unnecessary call to broadcast ack function Jon Maloy
2015-03-13 20:08 ` [PATCH net-next v2 6/7] tipc: split link outqueue Jon Maloy
2015-03-13 20:08 ` [PATCH net-next v2 7/7] tipc: clean up handling of message priorities Jon Maloy
2015-03-14 18:39 ` [PATCH net-next v2 0/7] tipc: some optimizations and impovements David Miller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1426277291-2427-4-git-send-email-jon.maloy@ericsson.com \
--to=jon.maloy@ericsson.com \
--cc=davem@davemloft.net \
--cc=netdev@vger.kernel.org \
--cc=paul.gortmaker@windriver.com \
--cc=tipc-discussion@lists.sourceforge.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).