Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210727050447.7339-1-rdunlap@infradead.org>
In-Reply-To: <20210727050447.7339-1-rdunlap@infradead.org>
From:   Geert Uytterhoeven <geert@linux-m68k.org>
Date:   Wed, 28 Jul 2021 10:39:00 +0200
Message-ID: <CAMuHMdXrDLvPRNMy_pEtu1u=7ML8GT3mpKLba8vRnHU3iF8U+w@mail.gmail.com>
Subject: Re: [PATCH] efi: sysfb_efi: fix build when EFI is not set
To:     Randy Dunlap <rdunlap@infradead.org>
Cc:     Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Ard Biesheuvel <ardb@kernel.org>,
        linux-efi <linux-efi@vger.kernel.org>,
        DRI Development <dri-devel@lists.freedesktop.org>,
        Javier Martinez Canillas <javierm@redhat.com>,
        Thomas Zimmermann <tzimmermann@suse.de>,
        Mark Brown <broonie@kernel.org>,
        Linux-Next <linux-next@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8BIT
Precedence: bulk

Hi Randy,

On Tue, Jul 27, 2021 at 7:06 AM Randy Dunlap <rdunlap@infradead.org> wrote:
> When # CONFIG_EFI is not set, there are 2 definitions of
> sysfb_apply_efi_quirks(). The stub from sysfb.h should be used
> and the __init function from sysfb_efi.c should not be used.
>
> ../drivers/firmware/efi/sysfb_efi.c:337:13: error: redefinition of ‘sysfb_apply_efi_quirks’
>  __init void sysfb_apply_efi_quirks(struct platform_device *pd)
>              ^~~~~~~~~~~~~~~~~~~~~~
> In file included from ../drivers/firmware/efi/sysfb_efi.c:26:0:
> ../include/linux/sysfb.h:65:20: note: previous definition of ‘sysfb_apply_efi_quirks’ was here
>  static inline void sysfb_apply_efi_quirks(struct platform_device *pd)
>                     ^~~~~~~~~~~~~~~~~~~~~~
>
> Signed-off-by: Randy Dunlap <rdunlap@infradead.org>

Thanks for your patch!

> --- linext-20210726.orig/drivers/firmware/efi/sysfb_efi.c
> +++ linext-20210726/drivers/firmware/efi/sysfb_efi.c
> @@ -332,6 +332,7 @@ static const struct fwnode_operations ef
>         .add_links = efifb_add_links,
>  };
>
> +#ifdef CONFIG_EFI
>  static struct fwnode_handle efifb_fwnode;
>
>  __init void sysfb_apply_efi_quirks(struct platform_device *pd)
> @@ -354,3 +355,4 @@ __init void sysfb_apply_efi_quirks(struc
>                 pd->dev.fwnode = &efifb_fwnode;
>         }
>  }
> +#endif

How come you enter drivers/firmware/efi/ without CONFIG_EFI?

Oh:

drivers/firmware/Makefile:obj-$(CONFIG_EFI)             += efi/
drivers/firmware/Makefile:obj-$(CONFIG_UEFI_CPER)               += efi/

Looks like UEFI_CPER is not related to EFI at all (it's not set in
my arm64 config which has CONFIG_EFI=y). Perhaps it should be moved
to drivers/acpi/ instead?

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds