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;
Date:   Mon, 1 May 2023 15:08:03 -0700
From:   Jakub Kicinski <kuba@kernel.org>
To:     Kalle Valo <kvalo@kernel.org>
Cc:     Ping-Ke Shih <pkshih@realtek.com>, netdev@vger.kernel.org,
        linux-wireless@vger.kernel.org
Subject: Re: pull-request: wireless-next-2023-04-21
Message-ID: <20230501150803.6c4963ac@kernel.org>
In-Reply-To: <87cz3os2wr.fsf@kernel.org>
References: <20230421104726.800BCC433D2@smtp.kernel.org>
        <20230421075404.63c04bca@kernel.org>
        <e31dae6daa6640859d12bf4c4fc41599@realtek.com>
        <87leigr06u.fsf@kernel.org>
        <20230425071848.6156c0a0@kernel.org>
        <87cz3os2wr.fsf@kernel.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Fri, 28 Apr 2023 13:43:16 +0300 Kalle Valo wrote:
> > I don't think it's that much extra work, the driver requires FW
> > according to modinfo, anyway, so /lib/firmware is already required.
> > And on smaller systems with few hundred MB of RAM it'd be nice to not
> > hold all the stuff in kernel memory, I'd think.  
> 
> Later in this thread Ping explained pretty well the challenges here,
> that sums exactly what I'm worried about.
> 
> > We have a rule against putting FW as a static table in the driver
> > source, right? Or did we abandon that? Isn't this fundamentally similar?  
> 
> My understanding is that these are just initialisation values for
> hardware, not executable code. (Ping, please correct me if I
> misunderstood.) So that's why I thought these are ok to have in kernel.
> So I took practicality over elegance here.

Alright, I'll try to make someone else do this outside of wireless,
and come back with real life experience disproving the concerns :)