From: Andrey Derkach <gabin278@gmail.com>
To: David Heidelberg <david@ixit.cz>
Cc: Jeff Johnson <jjohnson@kernel.org>,
Johannes Berg <johannes.berg@intel.com>,
Kalle Valo <kvalo@qca.qualcomm.com>,
Michal Kazior <michal.kazior@tieto.com>,
Richard Acayan <mailingradian@gmail.com>,
linux-wireless@vger.kernel.org, ath10k@lists.infradead.org,
linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org,
phone-devel@vger.kernel.org
Subject: Re: [PATCH v4] wifi: ath10k: make in-order rx amsdu buffers persistent
Date: Tue, 20 Oct 2026 22:50:20 +0300 [thread overview]
Message-ID: <20261020195021.1133728-1-gabin278@gmail.com> (raw)
In-Reply-To: <20261020-ath10k-a-msdu-v4-1-670b69d39059@ixit.cz>
On Tue, 20 Oct 2026 20:52:07 +0200, David Heidelberg wrote:
> The WCN3990 firmware may split an A-MSDU across multiple HTT "in-order"
> indications. Mainline drops these ("failed to extract amsdu: -11",
> rx_confused, recovery), collapsing throughput under load on affected
> devices. Stash the MSDUs of an incomplete MPDU in the driver state so
> it can be completed by the next indication instead of being dropped.
Tested on a Lenovo S5 Pro (SDM636) with a 7.2.3-based tree, on top of
"wifi: ath10k: trigger hardware recovery upon rx failures", replacing
the older out-of-tree version of this patch we carried before.
During TCP downloads the firmware split an A-MSDU across two in-order
indications (once in about 39,000 indications so far, counted with
kprobes on the stash and resume paths). The patch stashed the MSDUs and
the next indication completed the MPDU, with no warning and no
recovery. Throughput is the same as with the old version, about
60 Mbit/s here.
Tested-on: WCN3990 hw1.0 SNOC WLAN.HL.1.0.1.c2-00455-QCAHLSWMTPLZ-1
Tested-by: Andrey Derkach <gabin278@gmail.com>
next prev parent reply other threads:[~2026-10-20 19:50 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-20 18:52 [PATCH v4] wifi: ath10k: make in-order rx amsdu buffers persistent David Heidelberg via B4 Relay
2026-10-20 19:50 ` Andrey Derkach [this message]
2026-10-20 20:28 ` Andrey Derkach
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=20261020195021.1133728-1-gabin278@gmail.com \
--to=gabin278@gmail.com \
--cc=ath10k@lists.infradead.org \
--cc=david@ixit.cz \
--cc=jjohnson@kernel.org \
--cc=johannes.berg@intel.com \
--cc=kvalo@qca.qualcomm.com \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-wireless@vger.kernel.org \
--cc=mailingradian@gmail.com \
--cc=michal.kazior@tieto.com \
--cc=phone-devel@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