netdev - [PATCH net-next] dql: dql_queued() should write first to reduce bus transactions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1411711496.16953.92.camel@edumazet-glaptop2.roam.corp.google.com>
Date:	Thu, 25 Sep 2014 23:04:56 -0700
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Tom Herbert <therbert@...gle.com>
Cc:	Jesper Dangaard Brouer <brouer@...hat.com>,
	Linux Netdev List <netdev@...r.kernel.org>,
	"David S. Miller" <davem@...emloft.net>,
	Alexander Duyck <alexander.h.duyck@...el.com>,
	Toke Høiland-Jørgensen <toke@...e.dk>,
	Florian Westphal <fw@...len.de>,
	Jamal Hadi Salim <jhs@...atatu.com>,
	Dave Taht <dave.taht@...il.com>,
	John Fastabend <john.r.fastabend@...el.com>,
	Daniel Borkmann <dborkman@...hat.com>,
	Hannes Frederic Sowa <hannes@...essinduktion.org>
Subject: [PATCH net-next] dql: dql_queued() should write first to reduce bus
 transactions

From: Eric Dumazet <edumazet@...gle.com>

While doing high throughput test on a BQL enabled NIC,
I found a very high cost in ndo_start_xmit() when accessing BQL data.

It turned out the problem was caused by compiler trying to be
smart, but involving a bad MESI transaction :

  0.05 │  mov    0xc0(%rax),%edi    // LOAD dql->num_queued
  0.48 │  mov    %edx,0xc8(%rax)    // STORE dql->last_obj_cnt = count
 58.23 │  add    %edx,%edi
  0.58 │  cmp    %edi,0xc4(%rax)
  0.76 │  mov    %edi,0xc0(%rax)    // STORE dql->num_queued += count
  0.72 │  js     bd8


I got an incredible 10 % gain [1] by making sure cpu do not attempt
to get the cache line in Shared mode, but directly requests for
ownership.

New code :
	mov    %edx,0xc8(%rax)  // STORE dql->last_obj_cnt = count
	add    %edx,0xc0(%rax)  // RMW   dql->num_queued += count
	mov    0xc4(%rax),%ecx  // LOAD dql->adj_limit
	mov    0xc0(%rax),%edx  // LOAD dql->num_queued
	cmp    %edx,%ecx

The TX completion was running from another cpu, with high interrupts
rate.

Note that I am using barrier() as a soft hint, as mb() here could be
too heavy cost.

[1] This was a netperf TCP_STREAM with TSO disabled, but GSO enabled.

Signed-off-by: Eric Dumazet <edumazet@...gle.com>
---
 include/linux/dynamic_queue_limits.h |   12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/include/linux/dynamic_queue_limits.h b/include/linux/dynamic_queue_limits.h
index 5621547d631b..a4be70398ce1 100644
--- a/include/linux/dynamic_queue_limits.h
+++ b/include/linux/dynamic_queue_limits.h
@@ -73,14 +73,22 @@ static inline void dql_queued(struct dql *dql, unsigned int count)
 {
 	BUG_ON(count > DQL_MAX_OBJECT);
 
-	dql->num_queued += count;
 	dql->last_obj_cnt = count;
+
+	/* We want to force a write first, so that cpu do not attempt
+	 * to get cache line containing last_obj_cnt, num_queued, adj_limit
+	 * in Shared state, but directly does a Request For Ownership
+	 * It is only a hint, we use barrier() only.
+	 */
+	barrier();
+
+	dql->num_queued += count;
 }
 
 /* Returns how many objects can be queued, < 0 indicates over limit. */
 static inline int dql_avail(const struct dql *dql)
 {
-	return dql->adj_limit - dql->num_queued;
+	return ACCESS_ONCE(dql->adj_limit) - ACCESS_ONCE(dql->num_queued);
 }
 
 /* Record number of completed objects and recalculate the limit. */


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html