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, 3 Nov 2022 17:02:02 +0100
From:   Vasily Gorbik <gor@linux.ibm.com>
To:     Muchun Song <muchun.song@linux.dev>,
        Andrew Morton <akpm@linux-foundation.org>
Cc:     Muchun Song <songmuchun@bytedance.com>,
        Mike Kravetz <mike.kravetz@oracle.com>,
        Gerald Schaefer <gerald.schaefer@linux.ibm.com>,
        Linux Memory Management List <linux-mm@kvack.org>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH] mm: hugetlb_vmemmap: include missing linux/moduleparam.h
Message-ID: <your-ad-here.call-01667491322-ext-5168@work.hours>
References: <patch.git-296b83ca939b.your-ad-here.call-01667411912-ext-5073@work.hours>
 <71BFC75C-483D-42C0-8248-59FC8DBE148C@linux.dev>
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <71BFC75C-483D-42C0-8248-59FC8DBE148C@linux.dev>
MIME-Version: 1.0
Precedence: bulk

On Thu, Nov 03, 2022 at 10:51:59AM +0800, Muchun Song wrote:
> 
> 
> > On Nov 3, 2022, at 02:09, Vasily Gorbik <gor@linux.ibm.com> wrote:
> > 
> > The kernel test robot reported build failures with a 'randconfig' on s390:
> >>> mm/hugetlb_vmemmap.c:421:11: error: a function declaration without a
> > prototype is deprecated in all versions of C [-Werror,-Wstrict-prototypes]
> >   core_param(hugetlb_free_vmemmap, vmemmap_optimize_enabled, bool, 0);
> >             ^
> > 
> > Link: https://lore.kernel.org/linux-mm/202210300751.rG3UDsuc-lkp@intel.com/
> > Reported-by: kernel test robot <lkp@intel.com>
> > Fixes: 30152245c63b ("mm: hugetlb_vmemmap: replace early_param() with core_param()")
> 
> Thanks for your fixing. However, the correct commit that should be fixed is
> 42deddf3ef5c ("s390: select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP").
> 
> Reviewed-by: Muchun Song <songmuchun@bytedance.com>

Even on x86 the inclusion chain looks like:
  CC      mm/hugetlb_vmemmap.o
In file included from ./include/linux/module.h:22,
                 from ./include/linux/device/driver.h:21,
                 from ./include/linux/device.h:32,
                 from ./include/linux/blk_types.h:11,
                 from ./include/linux/writeback.h:13,
                 from ./include/linux/memcontrol.h:22,
                 from ./include/linux/swap.h:9,
                 from ./include/asm-generic/hugetlb.h:5,
                 from ./arch/x86/include/asm/hugetlb.h:6,
                 from ./include/linux/hugetlb.h:802,
                 from mm/hugetlb_vmemmap.h:11,
                 from mm/hugetlb_vmemmap.c:18:
./include/linux/moduleparam.h:2:2: warning: #warning included from here

With s390 defconfig:
  CC      mm/hugetlb_vmemmap.o
In file included from ./include/linux/module.h:22,
                 from ./include/linux/device/driver.h:21,
                 from ./include/linux/device.h:32,
                 from ./include/linux/blk_types.h:11,
                 from ./include/linux/writeback.h:13,
                 from ./include/linux/memcontrol.h:22,
                 from ./include/linux/swap.h:9,
                 from ./include/linux/userfaultfd_k.h:18, <-- randconfig disabled this path (CONFIG_USERFAULTFD)
                 from ./include/linux/hugetlb.h:14,
                 from mm/hugetlb_vmemmap.h:11,
                 from mm/hugetlb_vmemmap.c:17:
./include/linux/moduleparam.h:2:2: warning: #warning included from here

42deddf3ef5c ("s390: select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP") is
just a one liner activating this feature and only present in linux-next
and s390 features branch targeting v6.2. So, if you don't want to have

Fixes: 30152245c63b ("mm: hugetlb_vmemmap: replace early_param() with core_param()")

fine with me. But then just dropping Fixes tag is probably a better
option. I hope that won't prevent it from going into some v6.1-rcX.

@Andrew
should I resend? or would you be so kind to drop Fixes tag and add Reviewed-by
without it? Thank you