Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <fcdb5238-5977-526a-4752-fa82893a580e@infradead.org>
Date:   Wed, 23 Nov 2022 08:35:44 -0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.5.0
Subject: Re: [PATCH -next] fbdev: offb: allow build when DRM_OFDRM=m
Content-Language: en-US
From:   Randy Dunlap <rdunlap@infradead.org>
To:     Arnd Bergmann <arnd@arndb.de>, linux-kernel@vger.kernel.org
Cc:     linux-fbdev@vger.kernel.org,
        Thomas Zimmermann <tzimmermann@suse.de>,
        Masahiro Yamada <masahiroy@kernel.org>,
        dri-devel@lists.freedesktop.org, Daniel Vetter <daniel@ffwll.ch>,
        Michal Suchanek <msuchanek@suse.de>,
        linuxppc-dev@lists.ozlabs.org, Helge Deller <deller@gmx.de>
References: <20221123031605.16680-1-rdunlap@infradead.org>
 <4b10b87d-f255-4839-8700-858d98ffb801@app.fastmail.com>
 <25c22370-b67a-33a6-f1e6-abf70760d866@infradead.org>
In-Reply-To: <25c22370-b67a-33a6-f1e6-abf70760d866@infradead.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 11/23/22 02:02, Randy Dunlap wrote:
> Hi Arnd,
> 
> On 11/23/22 01:08, Arnd Bergmann wrote:
>> On Wed, Nov 23, 2022, at 04:16, Randy Dunlap wrote:
>>> Fix build when CONFIG_FB_OF=y and CONFIG_DRM_OFDRM=m.
>>> When the latter symbol is =m, kconfig downgrades (limits) the 'select's
>>> under FB_OF to modular (=m). This causes undefined symbol references:
>>>
>>> powerpc64-linux-ld: drivers/video/fbdev/offb.o:(.data.rel.ro+0x58): 
>>> undefined reference to `cfb_fillrect'
>>> powerpc64-linux-ld: drivers/video/fbdev/offb.o:(.data.rel.ro+0x60): 
>>> undefined reference to `cfb_copyarea'
>>> powerpc64-linux-ld: drivers/video/fbdev/offb.o:(.data.rel.ro+0x68): 
>>> undefined reference to `cfb_imageblit'
>>>
>>> Fix this by allowing FB_OF any time that DRM_OFDRM != y so that the
>>> selected FB_CFB_* symbols will become =y instead of =m.
>>>
>>> In tristate logic (for DRM_OFDRM), this changes the dependency from
>>>     !DRM_OFDRM	== 2 - 1 == 1 => modular only (or disabled)
>>> to (boolean)
>>>     DRM_OFDRM != y == y, allowing the 'select's to cause the
>>> FB_CFB_* symbols to =y instead of =m.
>>>
>>
>> Is it actually a useful configuration to have OFDRM=m and
>> FB_OF=y though? I would expect in that case that the OFDRM
>> driver never binds to a device because it's already owned
>> by FB_OF.
>>
>>> diff -- a/drivers/video/fbdev/Kconfig b/drivers/video/fbdev/Kconfig
>>> --- a/drivers/video/fbdev/Kconfig
>>> +++ b/drivers/video/fbdev/Kconfig
>>> @@ -455,7 +455,7 @@ config FB_ATARI
>>>  config FB_OF
>>>  	bool "Open Firmware frame buffer device support"
>>>  	depends on (FB = y) && PPC && (!PPC_PSERIES || PCI)
>>> -	depends on !DRM_OFDRM
>>> +	depends on DRM_OFDRM != y
>>>  	select APERTURE_HELPERS
>>
>> I would instead make this 'depends on DRM_OFDRM=n', which
>> completely eliminates configs that have both driver enabled.
> 
> Yep, that works for me. Thanks.
> 
> Thomas, Michal, are you OK with that change?
> 
>> A nicer change would be to make FB_OF a tristate symbol,
>> which makes it possible to load one of the two modules if
>> both are enabled =m, while only allowing one of them to
>> be =y if the other is completely disabled. It looks like
>> offb was originally written to be usable as a loadable module,
>> but Kconfig has prevented this since at least the start of
>> the git history.

ISTM that a distro would prefer to have both DFM_OFDRM and
FB_OF as tristate symbols that could both be built as loadable
modules, as Arnd describes above.
I'll look into that.

-- 
~Randy