[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHk-=whyEQy771ixppPmMSYtPcFS5ZtqVWUYow8gWd=pMnATNA@mail.gmail.com>
Date: Tue, 25 Feb 2020 10:27:34 -0800
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Willy Tarreau <w@....eu>
Cc: Denis Efremov <efremov@...ux.com>, Jens Axboe <axboe@...nel.dk>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
linux-block <linux-block@...r.kernel.org>
Subject: Re: [PATCH 01/10] floppy: cleanup: expand macro FDCS
On Tue, Feb 25, 2020 at 10:15 AM Willy Tarreau <w@....eu> wrote:
>
> On Tue, Feb 25, 2020 at 10:08:51AM -0800, Linus Torvalds wrote:
> >
> > So we can remove at least the FD_IOPORT mess from the header file, I bet.
> >
> > Worst case - if somebody finds some case that uses them, we can put it back.
>
> I like that. And at least we'll know how they use it (likely without the
> dependency on fdc).
Note that the way uapi header files generally got created was by just
moving header files that user space used mechanically. See for example
commit 607ca46e97a1 ("UAPI: (Scripted) Disintegrate include/linux")
which created most of them.
There was no careful vetting of "this is the part that is used by user
space". It was just a "these are the files user space has used".
So it's not really a "the uapi files are set in stone and you can't
change them". Instead, you should think of the uapi files as a big red
blinking warning that says "sure, you can change them, but you need to
be very careful and think about the fact that user space may be
including this thing".
So it's a "think hard about it" rather than a "don't go there".
Of course, usually it's much _simpler_ to just "don't go there" ;)
Linus
Powered by blists - more mailing lists