From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-it0-x242.google.com (mail-it0-x242.google.com [IPv6:2607:f8b0:4001:c0b::242]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 1388B3B2A0 for ; Thu, 13 Oct 2016 12:08:11 -0400 (EDT) Received: by mail-it0-x242.google.com with SMTP id l13so6109945itl.2 for ; Thu, 13 Oct 2016 09:08:11 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=subject:to:references:cc:from:message-id:date:user-agent :mime-version:in-reply-to; bh=0fG9CctKPR5oiXQ1hMZOQ4AxoXMBEkixsogg7jqcB0Y=; b=w2hUwLU7O69AesIYvoxcr2ceozOIHnhqeLUnupLNPwPqYyp7K+eXerv7FUqjISI2QP QY90aVav/OrCoTF5ZavLoILwAYD7bsk4qTAF6YX5kJVDSxqf0nrAToBFVkuAQTI8E6mx An4i1zMH8hZlJ/1tJR8E9XmLqNlGuJ/ROBfEDtavn6/gw6FiY1smBzHblPiE512qJJDu tJ6bQz+3N31y45EptHSNsmePq+NplUA2phAAQnyIaZBNAr4SmU7JI1dm/CVdz5jLYT87 milgoFwvFhTTMRPXv+/owUrOiyOitW78LdTBXktRmFbd0YvMME3x7QPBG5cIIYA1UpwG hPUw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:message-id:date :user-agent:mime-version:in-reply-to; bh=0fG9CctKPR5oiXQ1hMZOQ4AxoXMBEkixsogg7jqcB0Y=; b=C5c12TfDVUY071qN0PvwS+3l6T/6c+NflsnP7OIAake8C6v+VVwrz4BkauJAZh+MVe BawkQHRoMrvNbDLbX6949pIp69ZKpJiFMJ/Nks0uiZJtdVRnYNoCCV3oCj/0R5/CtxwN gIRXbPGr34IrBprovKP5Pu3Y/5DmRnipJ701f9upeETYAKCYSaiwgD0oGmKbD0cOdWpl YnfJf9+jj8vN20+hVa9+o24WzmgDok0a7QGMpnwmHsqBFX+4IqMgcLwu+uUGTFFPeD4K OXLX7HkZRonQNJiDOZAr7cUPHdJ8YsTt4TJtMhoIgY/3MhJ06BsGUoPgKtcOlV80FSH5 u9Bw== X-Gm-Message-State: AA6/9RmqJCq79bUIrXOSFbyOJSG4MBmpZ0pDoheOUEUkJDFkuGmd8Xvi4TlCRFHYRBpObQ== X-Received: by 10.36.65.144 with SMTP id b16mr7203659itd.38.1476374891285; Thu, 13 Oct 2016 09:08:11 -0700 (PDT) Received: from [172.21.63.50] (hades.kettering.edu. [192.138.137.97]) by smtp.gmail.com with ESMTPSA id j69sm5969940iod.8.2016.10.13.09.08.10 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 13 Oct 2016 09:08:10 -0700 (PDT) To: =?UTF-8?Q?Toke_H=c3=b8iland-J=c3=b8rgensen?= References: <87y41sxnlp.fsf@toke.dk> Cc: make-wifi-fast@lists.bufferbloat.net From: Noah Causin Message-ID: Date: Thu, 13 Oct 2016 12:08:12 -0400 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:45.0) Gecko/20100101 Thunderbird/45.4.0 MIME-Version: 1.0 In-Reply-To: Content-Type: multipart/alternative; boundary="------------15453993C12115C47336F683" Subject: Re: [Make-wifi-fast] ath10k fq_codel support X-BeenThere: make-wifi-fast@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 13 Oct 2016 16:08:12 -0000 This is a multi-part message in MIME format. --------------15453993C12115C47336F683 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit It seems like the patch in LEDE was removed 5 days ago. https://github.com/lede-project/source/commit/ad51e09fd1301484820a466a49447a34d7504882 Search the page for: 311-ath10k-disable-wake_tx_queue-for-older-devices.patch I'm going to compile a new build for my Archer C7 and see. On 10/13/2016 11:59 AM, Michal Kazior wrote: > On 13 Octsuppber 2016 at 17:41, Toke Høiland-Jørgensen wrote: >> Noah Causin writes: >> >>> Hi, >>> >>> I was looking into getting an Ubiquiti Unifi AC-Pro to flash LEDE onto for >>> testing the ath10k performance and latency with fq_codel. >>> >>> I already have an Archer C7 V2 which uses the ath10k-firmware-qca988x package, >>> which the Unifi AC-Pro and Lite seem to use. >>> >>> Would I have all I need if I flashed the latest git version of LEDE? >>> >>> Also, what firmware versions support the intermediate software queues, >>> and would my Archer C7 be fine for preliminary testing? >> You can check if it's active by seeing if you can read >> /sys/kernel/debug/ieee80211/phy0/aqm - if that exists and outputs >> anything when you 'cat' it, the intermediate queues are active. >> >> They are disabled for some hardware revisions, but not sure which ones. > wake_tx_queue is disabled if fw doesn't advertise flow control support > (which is supported in qca99xx and qca40xx chip fw blobs only so far). > > qca988x found in archer c7 will have it disabled. > > The patch that disables it is [1]. You can revert it or tinker around > the logic it introduced if you want to force fq_codel. > > [1]: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/drivers/net/wireless/ath/ath10k?id=4ca1807815aa6801aaced7fdefa9edacc2521767 > > > Michal --------------15453993C12115C47336F683 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 8bit

It seems like the patch in LEDE was removed 5 days ago.

https://github.com/lede-project/source/commit/ad51e09fd1301484820a466a49447a34d7504882

Search the page for:

311-ath10k-disable-wake_tx_queue-for-older-devices.patch

I'm going to compile a new build for my Archer C7 and see.


On 10/13/2016 11:59 AM, Michal Kazior wrote:
On 13 Octsuppber 2016 at 17:41, Toke Høiland-Jørgensen <toke@toke.dk> wrote:

Noah Causin <n0manletter@gmail.com> writes:


Hi,

I was looking into getting an Ubiquiti Unifi AC-Pro to flash LEDE onto for
testing the ath10k performance and latency with fq_codel.

I already have an Archer C7 V2 which uses the ath10k-firmware-qca988x package,
which the Unifi AC-Pro and Lite seem to use.

Would I have all I need if I flashed the latest git version of LEDE?

Also, what firmware versions support the intermediate software queues,
and would my Archer C7 be fine for preliminary testing?

You can check if it's active by seeing if you can read
/sys/kernel/debug/ieee80211/phy0/aqm - if that exists and outputs
anything when you 'cat' it, the intermediate queues are active.

They are disabled for some hardware revisions, but not sure which ones.

wake_tx_queue is disabled if fw doesn't advertise flow control support
(which is supported in qca99xx and qca40xx chip fw blobs only so far).

qca988x found in archer c7 will have it disabled.

The patch that disables it is [1]. You can revert it or tinker around
the logic it introduced if you want to force fq_codel.

[1]: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/drivers/net/wireless/ath/ath10k?id=4ca1807815aa6801aaced7fdefa9edacc2521767


Michal

--------------15453993C12115C47336F683--