Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20220226045047.643695-1-mikezackles@gmail.com>
 <c09a353b-8e52-6d9d-a3c6-743e4e6f444b@mail.uni-paderborn.de>
 <871qtyj6b6.fsf@kernel.org> <CAN+PumF3JfmwVh_7BffLza+v7iD58QsxanCvbCrS1ztSt=XjNw@mail.gmail.com>
 <d9dc0c00d9ea8bb5b9341a7bcd4a75fc369a7d8f.camel@intel.com>
In-Reply-To: <d9dc0c00d9ea8bb5b9341a7bcd4a75fc369a7d8f.camel@intel.com>
From:   Zachary Michaels <mikezackles@gmail.com>
Date:   Sun, 14 Aug 2022 09:01:24 -0700
Message-ID: <CAN+PumF73tYv0tFT3nGdbgpLCFEB3PgobYBgLisTVCw+q_5K6g@mail.gmail.com>
Subject: Re: [PATCH] iwlwifi: Make missed beacon timeout configurable
To:     "Greenman, Gregory" <gregory.greenman@intel.com>
Cc:     "kvalo@kernel.org" <kvalo@kernel.org>,
        "linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>,
        "tmvolin@gmail.com" <tmvolin@gmail.com>,
        "dennisba@mail.uni-paderborn.de" <dennisba@mail.uni-paderborn.de>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sun, Aug 14, 2022 at 6:05 AM Greenman, Gregory
<gregory.greenman@intel.com> wrote:
>
> On Wed, 2022-08-03 at 08:50 -0700, Zachary Michaels wrote:
> > > Why not just increase the default beacon timeout value?
> >
> > I made it a parameter because I was under the vague impression that a
> > larger timeout wouldn't be ideal on a well-behaved network, but this
> > isn't my area of expertise. I just thought this would be the most
> > conservative change since it would support identical default
> > operation.
> >
> > Thanks for taking a look.
>
> I see that this issue https://bugzilla.kernel.org/show_bug.cgi?id=203709
> has a long history from 2019. But, anyway, what value for
> beacon_timeout solves the problem?

It seems that most people have had success with 256, but that number
may be somewhat arbitrary.

> May I ask you to perform two experiments that could help us find
> the root cause for this issue? The first is to run with a default driver and
> to collect a firmware dump (here is a wiki with directions how to do that,
> https://wireless.wiki.kernel.org/en/users/drivers/iwlwifi/debugging#firmware_debugging)
> and the second is to load the defaut driver with modparam power_scheme=1. This will
> completely disable power save and will let us know if the issue is somehow caused by
> the NIC going to sleep and missing beacons.
>
> Thanks,
> Gregory

Unfortunately I am no longer experiencing this issue, but someone else
on this chain may be able to help. (Someone having problems requested
that I submit the patch.)

There may be some relevant information in the linked bug report if you
care to sift through it. There are reports that turning off power save
did not work, and there are also some firmware dumps.

Best,
-Zach