linux-kernel - [PATCH] docs: Fix blk-iolatency peer throttling description

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>] [day] [month] [year] [list]

Message-ID: <20260114110837.84126-1-yizhou.tang@shopee.com>
Date: Wed, 14 Jan 2026 19:08:37 +0800
From: Tang Yizhou <yizhou.tang@...pee.com>
To: tj@...nel.org,
	corbet@....net,
	axboe@...nel.dk,
	hch@....de
Cc: cgroups@...r.kernel.org,
	linux-doc@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	Tang Yizhou <yizhou.tang@...pee.com>
Subject: [PATCH] docs: Fix blk-iolatency peer throttling description

From: Tang Yizhou <yizhou.tang@...pee.com>

The current text states that peers with a lower latency target are
throttled, which is the opposite of the actual behavior. In fact,
blk-iolatency throttles peer groups with a higher latency target in order
to protect the more latency-sensitive group.

In addition, peer groups without a configured latency target are also
throttled, as they are treated as lower priority compared to groups with
explicit latency requirements.

Update the documentation to reflect the correct throttling behavior.

Signed-off-by: Tang Yizhou <yizhou.tang@...pee.com>
---
 Documentation/admin-guide/cgroup-v2.rst | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
index 7f5b59d95fce..d6f7ef08b67d 100644
--- a/Documentation/admin-guide/cgroup-v2.rst
+++ b/Documentation/admin-guide/cgroup-v2.rst
@@ -2238,8 +2238,9 @@ IO Latency
 
 This is a cgroup v2 controller for IO workload protection.  You provide a group
 with a latency target, and if the average latency exceeds that target the
-controller will throttle any peers that have a lower latency target than the
-protected workload.
+controller will throttle any peers that have a higher latency target than the
+protected workload, as well as peers that do not have a latency target
+configured.
 
 The limits are only applied at the peer level in the hierarchy.  This means that
 in the diagram below, only groups A, B, and C will influence each other, and
@@ -2265,8 +2266,9 @@ How IO Latency Throttling Works
 
 io.latency is work conserving; so as long as everybody is meeting their latency
 target the controller doesn't do anything.  Once a group starts missing its
-target it begins throttling any peer group that has a higher target than itself.
-This throttling takes 2 forms:
+target it begins throttling any peer group that has a higher target than itself,
+as well as any peer group without a latency target. This throttling takes 2
+forms:
 
 - Queue depth throttling.  This is the number of outstanding IO's a group is
   allowed to have.  We will clamp down relatively quickly, starting at no limit
-- 
2.43.0