From: Crescent Hsieh <crescentcy.hsieh@moxa.com>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Johan Hovold <johan@kernel.org>,
linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1 4/4] usb: serial: mxuport: add sysfs control for UART FIFO
Date: Fri, 15 May 2026 17:08:09 +0800 [thread overview]
Message-ID: <agbieSXDX1wUkE4K@moxa-ThinkCentre-M90t> (raw)
In-Reply-To: <2026051321-apply-charting-150e@gregkh>
Hi Greg and Johan,
I agree that USB-to-serial cannot provide deterministic latency.
However, the embedded UART used by Moxa UPort has a 512-byte FIFO, so
its FIFO/trigger configuration still has a measurable effect on the
typical latency on this hardware.
RQ_VENDOR_SET_FIFO_DISABLE is a historical name. On the current
firmware, it does not literally turn the UART FIFO on or off, but
adjusts the UART RX/TX trigger levels instead. Its semantics are close
to the trigger-level control I am currently working on for the PCI
serial cards [1].
In a 1-byte loopback latency test under the same wiring and host setup,
with 10,000 samples for each configuration, I measured the following
results (in us):
| Enable | Disable |
--------------------------|
Min | 637 | 210 |
Median | 764 | 235 |
P95 | 811 | 238 |
P99 | 820 | 260 |
Max | 4,214 | 3,807 |
Avg | 744 | 235 |
--------------------------|
So while this does not make USB deterministic, it does show that this
setting has a significant effect on the typical latency on this hardware
and also matters for compatibility with some legacy serial equipment.
[1]
https://lore.kernel.org/all/20260504084900.22380-16-crescentcy.hsieh@moxa.com
---
Sincerely,
Crescent Hsieh
next prev parent reply other threads:[~2026-05-15 9:08 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-24 3:50 [PATCH v1 0/4] usb: serial: mxuport: extend MXU50U support and runtime controls Crescent Hsieh
2026-03-24 3:50 ` [PATCH v1 1/4] usb: serial: mxuport: add support for more MXU50U UART devices Crescent Hsieh
2026-05-07 14:13 ` Johan Hovold
2026-03-24 3:50 ` [PATCH v1 2/4] usb: serial: mxuport: handle SEND_NEXT tx flow control Crescent Hsieh
2026-05-07 14:40 ` Johan Hovold
2026-03-24 3:50 ` [PATCH v1 3/4] usb: serial: mxuport: support serial interface mode configuration Crescent Hsieh
2026-05-07 15:56 ` Johan Hovold
2026-03-24 3:50 ` [PATCH v1 4/4] usb: serial: mxuport: add sysfs control for UART FIFO Crescent Hsieh
2026-05-07 15:59 ` Johan Hovold
2026-05-12 11:28 ` Crescent Hsieh
2026-05-13 8:20 ` Johan Hovold
2026-05-13 8:48 ` Greg Kroah-Hartman
2026-05-15 9:08 ` Crescent Hsieh [this message]
2026-05-15 9:17 ` Greg Kroah-Hartman
2026-03-30 7:24 ` [PATCH v1 0/4] usb: serial: mxuport: extend MXU50U support and runtime controls Crescent Hsieh
2026-03-30 7:54 ` Johan Hovold
2026-04-27 3:37 ` Crescent Hsieh
2026-04-27 10:43 ` Johan Hovold
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=agbieSXDX1wUkE4K@moxa-ThinkCentre-M90t \
--to=crescentcy.hsieh@moxa.com \
--cc=gregkh@linuxfoundation.org \
--cc=johan@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
/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