FORTIFY_SOURCE breaks ARM compilation in -next -- was Re: ARM compile failure in Re: linux-next: Tree for Apr 4

[Pavel Machek]

On Wed 2018-04-04 12:59:51, Tony Lindgren wrote:
> * Pavel Machek <pavel at ucw.cz> [180404 18:47]:
> > Hi!
> > 
> > > > When trying to build kernel for N900, I get:
> > > > 
> > > >   CC      lib/timerqueue.o
> > > >     CC      lib/vsprintf.o
> > > >     lib/string.c: In function 'strstr':
> > > >     lib/string.c:478:8: error: inlining failed in call to
> > > >     always_inline 'strlen': function not inlinable
> > > >     lib/string.c:903:5: error: called from here
> > > >     lib/string.c:478:8: error: inlining failed in call to
> > > >     always_inline 'strlen': function not inlinable
> > > >     lib/string.c:906:5: error: called from here
> > > >     lib/string.c:855:15: error: inlining failed in call to
> > > >     always_inline 'memcmp': function not inlinable
> > > >     ...
> > > > pavel at duo:/data/l/linux-next-n900$ git branch -l
> > > > * (detached from next-20180403)
> > > > 
> > > > I'm using
> > > > 
> > > > eval ` eldk-switch.sh -r 5.4 armv7a`
> > > > 
> > > > for cross-compilation.
> > > 
> > > Hmm I'm not seeing build issues with next-20180404, did this somehow
> > > already get sorted out?
> > 
> > It seems it is config-dependend. omap2plus_defconfig compiles ok for
> > me.
> > 
> > Can you try with attached .config?
> 
> That still builds for me. This with plain gcc-7.3.0 binutils-2.30
> built with buildall script.

Thanks.

Ok, let me try to bisect it. Compile-problem should be easy...

Hmm. And as it is compile-problem in single file, it should even be
reasonably fast. I did not realize how easy it would be:

#!/bin/bash
set -e
cp config.ok .config
yes '' | ARCH=arm make lib/string.o

And the winner is:

ee333554fed55555a986a90bb097ac7f9d6f05bf is the first bad commit
commit ee333554fed55555a986a90bb097ac7f9d6f05bf
Author: Jinbum Park <jinb.park7 at gmail.com>
Date:   Tue Mar 6 01:39:24 2018 +0100

    ARM: 8749/1: Kconfig: Add ARCH_HAS_FORTIFY_SOURCE

    CONFIG_FORTIFY_SOURCE detects various overflows at compile-time.
        (6974f0c4555e ("include/linux/string.h:
    add the option of fortified string.h functions)
    
    ARCH_HAS_FORTIFY_SOURCE means that the architecture can be built
    and
    run with CONFIG_FORTIFY_SOURCE.
    
    Since ARM can be built and run with that flag like other
    architectures,
    select ARCH_HAS_FORTIFY_SOURCE as default.
    
    Acked-by: Kees Cook <keescook at chromium.org>
    Signed-off-by: Jinbum Park <jinb.park7 at gmail.com>
    Signed-off-by: Russell King <rmk+kernel at armlinux.org.uk>

:040000 040000 2bff2f75ba36dd838a8ded3028b745d9f13f4c4a
									e61289e33a1fb71b1a6c05c442d45ea40bd5aba0
									M
									arch
bisect run success
pavel at duo:/data/l/linux-next-n900$ 


									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20180404/27e9066c/attachment.sig>