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;
Subject: Re: [PATCH] m68k: Remove redefinitions of `PMD_SIZE`, `PMD_MASK`,
 etc.
To:     Geert Uytterhoeven <geert@linux-m68k.org>
Cc:     linux-m68k <linux-m68k@lists.linux-m68k.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        kbuild test robot <lkp@intel.com>
References: <20210721104318.1070133-1-twd2.me@gmail.com>
 <CAMuHMdUWvnz4jCyPctZevq8MdFtCD8MCNSfDbCVR+__rP_=gvw@mail.gmail.com>
From:   twd2 <twd2.me@gmail.com>
Message-ID: <117080be-e587-a734-bc6f-8c57488db9dc@gmail.com>
Date:   Mon, 26 Jul 2021 22:14:05 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101
 Thunderbird/78.12.0
MIME-Version: 1.0
In-Reply-To: <CAMuHMdUWvnz4jCyPctZevq8MdFtCD8MCNSfDbCVR+__rP_=gvw@mail.gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2021/7/26 17:23, Geert Uytterhoeven wrote:
> Hi Wende,
>
> On Wed, Jul 21, 2021 at 12:44 PM Wende Tan <twd2.me@gmail.com> wrote:
>> Remove redefinitions of `PMD_SIZE`, `PMD_MASK`, `PTRS_PER_PMD`,
>> and `__PAGETABLE_PMD_FOLDED` when PGTABLE_LEVELS == 2 (i.e. SUN3 ||
>> COLDFIRE).  They have already been defined in
>> `asm-generic/pgtable-nopmd.h`.
> Thanks for your patch!
>
>> See also https://lore.kernel.org/lkml/202107201907.lqC6B3dF-lkp@intel.com/#t
> I'm a bit confused: is this a new issue introduced by the patch from
> the above thread, or is this a pre-existing issue?
> If the latter, do you know when it was introduced, and how to reproduce it?


I think this is a pre-existing issue, introduced in
commit 60e50f34b13e ("m68k: mm: use pgtable-nopXd instead of 4level-fixup") .
This commit replaced the usage of `4level-fixup.h` in `pgtable_mm.h` to
`pgtable-nop*d.h` and removed the redundent definition of `PMD_SHIFT` but
forgot to clean up other macros like `PMD_SIZE` (it should be moved into
the `#if` condition) [1].

And, GCC will not complain until two definitions of a macro with the same
name differ (as the above thread changes `1UL` to `_UL(1)`), hiding this
issue.

[1] https://github.com/torvalds/linux/commit/60e50f34b13e9e40763be12aa55f2144d8da514c#diff-9ec45948bad35c0e35d0c2d301a2a643336ff8fe2ee8ea3fff19d8f0bb948a46L34

Thanks.
Wende


>> Reported-by: kernel test robot <lkp@intel.com>
>> Signed-off-by: Wende Tan <twd2.me@gmail.com>
> Gr{oetje,eeting}s,
>
>                         Geert
>