On Fri, Oct 9, 2026 at 3:29 PM [off-list ref] wrote:
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] hid: documentation incorrectly relaxes asynchronous requirement for output_report
--
commit 6d8fa62fc058c7f5630a13cc92c85b7867bc9869
Author: Jarrett Schultz [off-list ref]
Documentation: Correction in HID output_report callback description.
This commit updates the documentation for the HID output_report callback.
It changes the description to state that the callback might be asynchronous
rather than strictly requiring it, citing i2c-hid as an example.
quoted
diff --git a/Documentation/hid/hid-transport.rst b/Documentation/hid/hid-transport.rst
index 6f1692da296cf..2008cf432af12 100644
--- a/Documentation/hid/hid-transport.rst
+++ b/Documentation/hid/hid-transport.rst
@@ -327,8 +327,8 @@ The available HID callbacks are:
Send raw output report via intr channel. Used by some HID device drivers
which require high throughput for outgoing requests on the intr channel. This
- must not cause SET_REPORT calls! This must be implemented as asynchronous
- output report on the intr channel!
+ must not cause SET_REPORT calls! This call might be asynchronous, so the
+ caller should not expect an immediate response!
[Severity: Medium]
Does this documentation change contradict the strict architectural guarantees
of the HID subsystem?
The HID subsystem design rules mandate that the output_report callback must
be asynchronous to prevent latency spikes, performance degradation, or
deadlocks on the high-throughput interrupt channel.
By stating that the call "might be asynchronous", could this inadvertently
mislead future transport driver authors into implementing blocking,
synchronous callbacks where they are not permitted?
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261009-send-upstream-v5-0-384af01da3ee@chromium.org?part=1
The "must be asynchronous" rule doesn't match the in-tree transports.
usbhid and i2c-hid both block in `output_report()`, and hid-input calls it
from a workqueue as a "synchronous output report". This patch only
brings the documentation in line with existing behavior; the actual
requirements don't change.