linux-kernel - Re: [GIT PULL] regulator updates for v4.7

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 17 May 2016 11:18:08 +0100
From:	Mark Brown <broonie@...nel.org>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
Cc:	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Liam Girdwood <lgirdwood@...il.com>
Subject: Re: [GIT PULL] regulator updates for v4.7

On Mon, May 16, 2016 at 07:38:23PM -0700, Linus Torvalds wrote:
> On Mon, May 16, 2016 at 5:27 AM, Mark Brown <broonie@...nel.org> wrote:

> > Mark Brown (26):
> >       regulator: Deprecate regulator_can_change_voltage()

> This needs to be fixed.

> It causes a build warning, and it's just not acceptable. We've had
> driver subsystems do this before, and then the warning stays around
> for years and nothing gets done.

> The warning is not just ugly, it ends up often hiding real problems
> when people get used to seeing warnings during the build.

I agree, it's not great, I wouldn't normally do it and it's certainly
not good to just shove the annotation on there with no other action.  In
this case a combination of this and pushing patches at people seems to
be working - I am expecting all the broken users to be fixed during the
merge window (-next looks OK) at which point I can send a patch to you
dropping the API.

> The whole "__deprecated" crap has been a bigger problem in itself than
> it has ever actually solved.

> So don't do it. Either remove bad interfaces or leave them alone. None
> of this "let's annoy everybody else with our deprecation messages
> whether they care about it or not".

I was also trying to remove the users via other means so I could delete
the function.  The reason I added the warning was that all the users
were in the embedded graphics drivers and for whatever reason it's
historically been suprisingly difficult to get any attention on issues
there that don't show up during the build so I hoped this would help
push things along faster and avoid new attempts to use it.  Like I say
something seems to have worked better this time around, though I don't
know how much of it is down to the warning of course.

I'll send a patch either deleting the interface or reverting this after
-rc1 (or earlier if the users are gone from your tree before then) if
that's OK?

Download attachment "signature.asc" of type "application/pgp-signature" (474 bytes)