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;
Date:   Thu, 31 Mar 2022 12:42:04 +0300
From:   Mike Rapoport <rppt@linux.ibm.com>
To:     Roman Gushchin <roman.gushchin@linux.dev>
Cc:     kernel test robot <lkp@intel.com>, Roman Gushchin <guro@fb.com>,
        llvm@lists.linux.dev, kbuild-all@lists.01.org,
        GNU/Weeb Mailing List <gwml@vger.gnuweeb.org>,
        linux-kernel@vger.kernel.org,
        Suren Baghdasaryan <surenb@google.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Linux Memory Management List <linux-mm@kvack.org>,
        Minchan Kim <minchan@google.com>,
        Nathan Chancellor <nathan@kernel.org>,
        Nick Desaulniers <ndesaulniers@google.com>
Subject: Re:
 [ammarfaizi2-block:google/android/kernel/common/android12-trusty-5.10
 4036/5872] WARNING: modpost: vmlinux.o(.text+0x4111c4): Section mismatch in
 reference from the function memblock_bottom_up() to the variable
 .meminit.data:memblock
Message-ID: <YkV3bCqtLaVH+mC7@linux.ibm.com>
References: <202203301412.MZ7wQvQz-lkp@intel.com>
 <YkTP/+jhhAX6xlAQ@carbon.dhcp.thefacebook.com>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <YkTP/+jhhAX6xlAQ@carbon.dhcp.thefacebook.com>
MIME-Version: 1.0
Precedence: bulk

(added llvm folks)

Hi,

On Wed, Mar 30, 2022 at 02:47:43PM -0700, Roman Gushchin wrote:
> On Wed, Mar 30, 2022 at 02:53:14PM +0800, kbuild test robot wrote:
> > Hi Roman,
> > 
> > FYI, the error/warning still remains.
> > 
> > tree:   https://github.com/ammarfaizi2/linux-block google/android/kernel/common/android12-trusty-5.10
> > head:   07055bfd3d810d41a38354693dfaa55a6f8c0025
> > commit: 0e0bfc41fdf4d79d39ebe929844cdee44f97366d [4036/5872] UPSTREAM: mm: cma: allocate cma areas bottom-up
> > config: x86_64-randconfig-a005 (https://download.01.org/0day-ci/archive/20220330/202203301412.MZ7wQvQz-lkp@intel.com/config)
> > compiler: clang version 15.0.0 (https://github.com/llvm/llvm-project 0f6d9501cf49ce02937099350d08f20c4af86f3d)
> > reproduce (this is a W=1 build):
> >         wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
> >         chmod +x ~/bin/make.cross
> >         # https://github.com/ammarfaizi2/linux-block/commit/0e0bfc41fdf4d79d39ebe929844cdee44f97366d
> >         git remote add ammarfaizi2-block https://github.com/ammarfaizi2/linux-block
> >         git fetch --no-tags ammarfaizi2-block google/android/kernel/common/android12-trusty-5.10
> >         git checkout 0e0bfc41fdf4d79d39ebe929844cdee44f97366d
> >         # save the config file to linux build tree
> >         mkdir build_dir
> >         COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=x86_64 SHELL=/bin/bash
> > 
> > If you fix the issue, kindly add following tag as appropriate
> > Reported-by: kernel test robot <lkp@intel.com>
> > 
> > All warnings (new ones prefixed by >>, old ones prefixed by <<):
> > 
> > >> WARNING: modpost: vmlinux.o(.text+0x4111c4): Section mismatch in reference from the function memblock_bottom_up() to the variable .meminit.data:memblock
> > The function memblock_bottom_up() references
> > the variable __meminitdata memblock.
> > This is often because memblock_bottom_up lacks a __meminitdata
> > annotation or the annotation of memblock is wrong.
> 
> I guess this patch should fix it, however I fail to reproduce the original issue.
> Maybe it's up to the specific compiler version.
> 
> --
> 
> From b55a8dd19f4156d7e24ec39b18ede06965ce1c4f Mon Sep 17 00:00:00 2001
> From: Roman Gushchin <roman.gushchin@linux.dev>
> Date: Wed, 30 Mar 2022 14:42:12 -0700
> Subject: [PATCH] memblock: fix memblock_bottom_up() and
>  memblock_set_bottom_up() annotations
> 
> memblock_bottom_up() and memblock_set_bottom_up() lack __meminitdata
> annotations causing compiler warnings like:
>   WARNING: modpost: vmlinux.o(.text+0x4111c4): Section mismatch in reference from the function memblock_bottom_up() to the
>   variable .meminit.data:memblock
> 
> Fix it by adding the missing annotation and removing the wrong
> __meminit annotation.
> 
> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: Roman Gushchin <roman.gushchin@linux.dev>
> ---
>  include/linux/memblock.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/memblock.h b/include/linux/memblock.h
> index 50ad19662a32..536bc2fc31e6 100644
> --- a/include/linux/memblock.h
> +++ b/include/linux/memblock.h
> @@ -460,7 +460,7 @@ static inline void *memblock_alloc_node(phys_addr_t size,
>  /*
>   * Set the allocation direction to bottom-up or top-down.
>   */
> -static inline __init_memblock void memblock_set_bottom_up(bool enable)
> +static inline __initdata_memblock void memblock_set_bottom_up(bool enable)

I think putting __initdata_memlock won't help here, because there should be
nothing wrong with __meminit function accessing __meminitdata data.

My guesstimate would be that the compiler decided not to inline this and
still dropped section attribute because of 'inline'.

If this is the case we I think we should 

	s/inline __init_memblock/__always_inline/

>  {
>  	memblock.bottom_up = enable;
>  }
> @@ -470,7 +470,7 @@ static inline __init_memblock void memblock_set_bottom_up(bool enable)
>   * if this is true, that said, memblock will allocate memory
>   * in bottom-up direction.
>   */
> -static inline __init_memblock bool memblock_bottom_up(void)
> +static inline __initdata_memblock bool memblock_bottom_up(void)
>  {
>  	return memblock.bottom_up;
>  }
> -- 
> 2.30.2
> 

-- 
Sincerely yours,
Mike.