Received-SPF: pass (google.com: domain of linux-kernel+bounces-141524-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
From: Breno Leitao <leitao@debian.org>
To: kuba@kernel.org,
	davem@davemloft.net,
	pabeni@redhat.com,
	edumazet@google.com
Cc: linux-kernel@vger.kernel.org,
	netdev@vger.kernel.org
Subject: [PATCH net-next v3 0/4] net: dqs: optimize if stall threshold is not set
Date: Thu, 11 Apr 2024 12:22:28 -0700
Message-ID: <20240411192241.2498631-1-leitao@debian.org>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

Here are four patches aimed at enhancing the Dynamic Queue Limit (DQL)
subsystem within the networking stack.

The first two commits involve code refactoring, while the third patch
introduces the actual change. The fourth patch just improves the cache
locality.

Typically, when DQL is enabled, stall information is always populated
through dql_queue_stall(). However, this information is only necessary
if a stall threshold is set, which is stored in struct dql->stall_thrs.

Although dql_queue_stall() is relatively inexpensive, it is not entirely
free due to memory barriers and similar overheads.

To optimize performance, refrain from calling dql_queue_stall() when no
stall threshold is set, thus avoiding the processing of unnecessary
information.

Changelog:

v3:
	* Read stall_thrs earlier, at dql_completed(), to avoid any
	  performance penalty, as suggested by Jakub.
v2:
	* Moved the stall_thrs to the very first cache line, as a new
	  patch. Suggested by Eric Dumazet.
v1:
	* https://lore.kernel.org/all/20240404145939.3601097-1-leitao@debian.org/


Breno Leitao (4):
  net: dql: Avoid calling BUG() when WARN() is enough
  net: dql: Separate queue function responsibilities
  net: dql: Optimize stall information population
  net: dqs: make struct dql more cache efficient

 include/linux/dynamic_queue_limits.h | 50 +++++++++++++++++-----------
 lib/dynamic_queue_limits.c           | 13 +++++---
 2 files changed, 39 insertions(+), 24 deletions(-)

-- 
2.43.0