linux-kernel - Re: [PATCH] firmware: remove old CONFIG_FW_LOADER

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Yd3Twxj4FjYvBwuo@bombadil.infradead.org>
Date:   Tue, 11 Jan 2022 11:00:19 -0800
From:   Luis Chamberlain <mcgrof@...nel.org>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     linux-kernel@...r.kernel.org, Borislav Petkov <bp@...en8.de>
Subject: Re: [PATCH] firmware: remove old CONFIG_FW_LOADER_MODULE test

On Thu, Jan 06, 2022 at 03:44:08PM +0100, Greg Kroah-Hartman wrote:
> On Thu, Dec 30, 2021 at 10:39:32AM +0100, Greg Kroah-Hartman wrote:
> > The CONFIG_FW_LOADER_MODULE check in firmware.h is very obsolete given
> > that this symbol went away decades ago, so it should be removed.
> > 
> > Reported-by: Borislav Petkov <bp@...en8.de>
> > Cc: Luis Chamberlain <mcgrof@...nel.org>
> > Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
> > ---
> >  include/linux/firmware.h | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/include/linux/firmware.h b/include/linux/firmware.h
> > index 3b057dfc8284..0b1c4d9c5465 100644
> > --- a/include/linux/firmware.h
> > +++ b/include/linux/firmware.h
> > @@ -34,7 +34,7 @@ static inline bool firmware_request_builtin(struct firmware *fw,
> >  }
> >  #endif
> >  
> > -#if defined(CONFIG_FW_LOADER) || (defined(CONFIG_FW_LOADER_MODULE) && defined(MODULE))
> > +#if defined(CONFIG_FW_LOADER)
> >  int request_firmware(const struct firmware **fw, const char *name,
> >  		     struct device *device);
> >  int firmware_request_nowarn(const struct firmware **fw, const char *name,
> > -- 
> > 2.34.1
> > 
> 
> Odd, 0-day reports a bunch of warnings with this change:
> 
> drivers/base/firmware_loader/main.c:1054:1: error: redefinition of 'request_firmware_nowait'
> drivers/base/firmware_loader/main.c:811:1: error: redefinition of 'request_firmware'
> drivers/base/firmware_loader/main.c:838:5: error: redefinition of 'firmware_request_nowarn'
> drivers/base/firmware_loader/main.c:863:5: error: redefinition of 'request_firmware_direct'
> drivers/base/firmware_loader/main.c:887:5: error: redefinition of 'firmware_request_platform'
> drivers/base/firmware_loader/main.c:941:1: error: redefinition of 'request_firmware_into_buf'
> drivers/base/firmware_loader/main.c:970:1: error: redefinition of 'request_partial_firmware_into_buf'
> drivers/base/firmware_loader/main.c:992:6: error: redefinition of 'release_firmware'
> 
> 
> Luis, any ideas?

Yeah I think this comes from the fact that modules get a respective
_MODULE define in the generated file
include/generated/autoconf.h 

For example:

vagrant@...vops-dev /data/linux-next (git::20211203-umh-fix-exitcodes)$
grep CONFIG_CRC8 .config
CONFIG_CRC8=m
vagrant@...vops-dev /data/linux-next (git::20211203-umh-fix-exitcodes)$
grep CONFIG_CRC8 include/generated/autoconf.h
#define CONFIG_CRC8_MODULE 1 

So I think the above was put in place to ask if its built-in or a
module.

  Luis