[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20220406121721.j6wlbzxher4xhtba@ti.com>
Date: Wed, 6 Apr 2022 17:47:21 +0530
From: Pratyush Yadav <p.yadav@...com>
To: <Tudor.Ambarus@...rochip.com>
CC: <michael@...le.cc>, <miquel.raynal@...tlin.com>, <richard@....at>,
<vigneshr@...com>, <linux-mtd@...ts.infradead.org>,
<linux-kernel@...r.kernel.org>, <Nicolas.Ferre@...rochip.com>
Subject: Re: [PATCH v2 4/4] mtd: spi-nor: sfdp: Keep SFDP definitions private
On 06/04/22 11:10AM, Tudor.Ambarus@...rochip.com wrote:
> On 4/6/22 13:31, Michael Walle wrote:
> >
> >>> That's correct, but I think exposing just the public defines in sfdp.h
> >>> is
> >>> the path to follow. We should keep private all the definitions that we
> >>> can
> >>> private in sfdp.c and expose publicly in sfdp.h just the ones that are
> >>> shared.
> >>> Flash collisions, and implicitly the need of public SFDP definitions,
> >>> should be
> >>> an exception, so I expect sfdp.h to be short in size.
> >>
> >> I disagree. I think we should keep everything in the same place. And
> >> since we need to expose this to manufacturer drivers, that place is
> >> sfdp.h. Who is going to cast the tiebreaking vote here? ;-)
> >
> > I'm leaning towards Pratyush opinion unless there is a clear advantage
> > to move the defines.
>
> Okay. Then we should move all the table definitions to sfdp.h, right?
We don't pass any other table to manufacturer drivers. Only BFPT. So
those can stay private. But I don't mind either in this case.
--
Regards,
Pratyush Yadav
Texas Instruments Inc.
Powered by blists - more mailing lists