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;
MIME-Version: 1.0
References: <20210419112725.42145-1-wanjiabing@vivo.com> <20210419160411.GG1904484@iweiny-DESK2.sc.intel.com>
 <874kg1yt0o.fsf@fossix.org>
In-Reply-To: <874kg1yt0o.fsf@fossix.org>
From:   Dan Williams <dan.j.williams@intel.com>
Date:   Tue, 20 Apr 2021 12:32:43 -0700
Message-ID: <CAPcyv4hD8gGdT6LABSBHRG2Bb59Zp1MycdQjB-CF9QHY-VHepQ@mail.gmail.com>
Subject: Re: [PATCH] libnvdimm.h: Remove duplicate struct declaration
To:     Santosh Sivaraj <santosh@fossix.org>
Cc:     Ira Weiny <ira.weiny@intel.com>, Wan Jiabing <wanjiabing@vivo.com>,
        linux-nvdimm <linux-nvdimm@lists.01.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        kael_w@yeah.net
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Apr 20, 2021 at 6:39 AM Santosh Sivaraj <santosh@fossix.org> wrote:
>
> Hi Ira,
>
> Ira Weiny <ira.weiny@intel.com> writes:
>
> > On Mon, Apr 19, 2021 at 07:27:25PM +0800, Wan Jiabing wrote:
> >> struct device is declared at 133rd line.
> >> The declaration here is unnecessary. Remove it.
> >>
> >> Signed-off-by: Wan Jiabing <wanjiabing@vivo.com>
> >> ---
> >>  include/linux/libnvdimm.h | 1 -
> >>  1 file changed, 1 deletion(-)
> >>
> >> diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
> >> index 01f251b6e36c..89b69e645ac7 100644
> >> --- a/include/linux/libnvdimm.h
> >> +++ b/include/linux/libnvdimm.h
> >> @@ -141,7 +141,6 @@ static inline void __iomem *devm_nvdimm_ioremap(struct device *dev,
> >>
> >>  struct nvdimm_bus;
> >>  struct module;
> >> -struct device;
> >>  struct nd_blk_region;
> >
> > What is the coding style preference for pre-declarations like this?  Should
> > they be placed at the top of the file?
> >
> > The patch is reasonable but if the intent is to declare right before use for
> > clarity, both devm_nvdimm_memremap() and nd_blk_region_desc() use struct
> > device.  So perhaps this duplicate is on purpose?
>
> There are other struct device usage much later in the file, which doesn't have
> any pre-declarations for struct device. So I assume this might not be on
> purpose :-)

Yeah, I believe it was just code movement and the duplicate was
inadvertently introduced. Patch looks ok to me.

>
> On a side note, types.h can also be removed, since it's already included in
> kernel.h.

That I don't necessarily agree with, it just makes future header
reworks more fraught for not much benefit.