Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::3:3 as permitted sender) client-ip=2620:137:e000::3:3;
Feedback-ID: i56a14606:Fastmail
User-Agent: Cyrus-JMAP/3.9.0-alpha0-1019-ged83ad8595-fm-20231002.001-ged83ad85
MIME-Version: 1.0
Message-Id: <d081871c-977c-43e9-afa3-a3c3e5880fea@app.fastmail.com>
In-Reply-To: <87y1g94szz.fsf@kernel.org>
References: <20231010155444.858483-1-arnd@kernel.org>
 <2023101051-unmasked-cleaver-79b3@gregkh> <87y1g94szz.fsf@kernel.org>
Date:   Wed, 11 Oct 2023 09:24:55 +0200
From:   "Arnd Bergmann" <arnd@arndb.de>
To:     "Kalle Valo" <kvalo@kernel.org>,
        "Greg Kroah-Hartman" <gregkh@linuxfoundation.org>
Cc:     "Arnd Bergmann" <arnd@kernel.org>,
        "Alexandre Belloni" <alexandre.belloni@bootlin.com>,
        linux-kernel@vger.kernel.org, linux-staging@lists.linux.dev,
        linux-wireless@vger.kernel.org,
        "Claudiu Beznea" <claudiu.beznea@tuxon.dev>,
        "Pavel Machek" <pavel@ucw.cz>, "Jakub Kicinski" <kuba@kernel.org>,
        "David S . Miller" <davem@davemloft.net>,
        linux-arm-kernel@lists.infradead.org,
        "Larry Finger" <Larry.Finger@lwfinger.net>
Subject: Re: [PATCH] [RFC] wireless: move obsolete drivers to staging
Content-Type: text/plain
Precedence: bulk

On Wed, Oct 11, 2023, at 07:40, Kalle Valo wrote:
> Greg Kroah-Hartman <gregkh@linuxfoundation.org> writes:
>
> We (the wireless folks) have been talking about dropping legacy drivers
> on and off for several years now. The problem is that we don't know
> which of them work and which not, for example IIRC someone reported
> recently that wl3501 still works.
>
> Personally I would be extremly happy to remove all the ancient drivers
> as that reduces the amount of code for us to maintain but is that the
> right thing to do for the users? I don't have an answer to that,
> comments very welcome.

I had a look at what openwrt enables, to see if any of the drivers
in my RFC patch are actually enabled, if anything supports legacy
embedded devices with these it would be openwrt. The good news here
is that openwrt intentionally leaves WEXT disabled, and none of them
are still in use.

I also did the same thing for the kernel defconfig files, which
has a few hits:

arch/arm/configs/pxa_defconfig:CONFIG_HERMES=m
arch/arm/configs/pxa_defconfig:CONFIG_HOSTAP=m
arch/arm/configs/pxa_defconfig:CONFIG_HOSTAP_CS=m
arch/arm/configs/pxa_defconfig:CONFIG_PCMCIA_HERMES=m
arch/arm/configs/pxa_defconfig:CONFIG_PCMCIA_SPECTRUM=m
arch/loongarch/configs/loongson3_defconfig:CONFIG_HOSTAP=m
arch/loongarch/configs/loongson3_defconfig:CONFIG_USB_NET_RNDIS_WLAN=m
arch/mips/configs/ip22_defconfig:CONFIG_HOSTAP=m
arch/mips/configs/ip27_defconfig:CONFIG_ATMEL=m
arch/mips/configs/ip27_defconfig:CONFIG_HERMES=m
arch/mips/configs/ip27_defconfig:CONFIG_HOSTAP=m
arch/mips/configs/ip27_defconfig:CONFIG_HOSTAP_PCI=m
arch/mips/configs/ip27_defconfig:CONFIG_HOSTAP_PLX=m
arch/mips/configs/ip27_defconfig:CONFIG_NORTEL_HERMES=m
arch/mips/configs/ip27_defconfig:CONFIG_PCI_ATMEL=m
arch/mips/configs/ip27_defconfig:CONFIG_PLX_HERMES=m
arch/mips/configs/ip27_defconfig:CONFIG_TMD_HERMES=m
arch/mips/configs/loongson2k_defconfig:CONFIG_HOSTAP=m
arch/mips/configs/loongson3_defconfig:CONFIG_HOSTAP=m
arch/mips/configs/malta_defconfig:CONFIG_ATMEL=m
arch/mips/configs/malta_defconfig:CONFIG_HOSTAP=m
arch/mips/configs/malta_defconfig:CONFIG_HOSTAP_PCI=m
arch/mips/configs/malta_defconfig:CONFIG_HOSTAP_PLX=m
arch/mips/configs/malta_defconfig:CONFIG_PCI_ATMEL=m
arch/mips/configs/malta_kvm_defconfig:CONFIG_ATMEL=m
arch/mips/configs/malta_kvm_defconfig:CONFIG_HOSTAP=m
arch/mips/configs/malta_kvm_defconfig:CONFIG_HOSTAP_PCI=m
arch/mips/configs/malta_kvm_defconfig:CONFIG_HOSTAP_PLX=m
arch/mips/configs/malta_kvm_defconfig:CONFIG_PCI_ATMEL=m
arch/mips/configs/maltaup_xpa_defconfig:CONFIG_ATMEL=m
arch/mips/configs/maltaup_xpa_defconfig:CONFIG_HOSTAP=m
arch/mips/configs/maltaup_xpa_defconfig:CONFIG_HOSTAP_PCI=m
arch/mips/configs/maltaup_xpa_defconfig:CONFIG_HOSTAP_PLX=m
arch/mips/configs/maltaup_xpa_defconfig:CONFIG_PCI_ATMEL=m

My interpretation here is that these defconfigs just enabled
all drivers that were relevant at the time when the boards
were new. The loongarch defconfig is a bit of an outlier,
as this is a fairly new platform.

Debian on the other just enables every driver, so there
is no good way to know what they actually use.

      Arnd