linux-kernel - Re: [GIT PULL] fbdev late fixes for v6.12-rc5

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]

Message-ID: <e3debf69-0507-41bd-999c-b3de79c809b5@gmx.de>
Date: Fri, 25 Oct 2024 21:35:57 +0200
From: Helge Deller <deller@....de>
To: Linus Torvalds <torvalds@...ux-foundation.org>,
 Helge Deller <deller@...nel.org>
Cc: linux-kernel@...r.kernel.org, linux-fbdev@...r.kernel.org,
 dri-devel@...ts.freedesktop.org
Subject: Re: [GIT PULL] fbdev late fixes for v6.12-rc5

Hi Linus,

On 10/25/24 20:31, Linus Torvalds wrote:
> On Fri, 25 Oct 2024 at 09:04, Helge Deller <deller@...nel.org> wrote:
>>
>> It's mostly about build warning fixes with cornercase CONFIG settings
>> and one big patch which removes the now unused da8xx fbdev driver.
>
> So I pulled this, but only later noticed that some of the Kconfig
> "fixes" are anything but.
>
> At least commit 447794e44744 ("fbdev: sstfb: Make CONFIG_FB_DEVICE
> optional") is not fixing anything, and very questionable.
>
> For no reason at all does it seem to enable 30-year old hardware in a
> new configuration.
>
> There were no build issues before, the build issues that existed were
> *introduced* by broken early versions of this patch.

That patch was the one I meant with "fixes [for] cornercase CONFIG settings".
But you are right that there aren't any issues fixed by this patch.

> Does anybody even *have* that hardware?

I do have a few of those (Voodoo2). Actually one is built-into one of my parisc
machines.

> Why were those pointless changes made?

When I accepted this patch I did not find it useless.
Maybe there are people who really enables Voodoo driver although
they prefer DRM. Maybe they don't even know the difference.
I applied it because I don't want compilation to fail at all
(which I see I was wrong in).

> Sure, the Voodoo1 was the bomb back in 1996 if you wanted to run
> hw-accelerated Quake, but in 2024, this change should have had more
> explanation for why anybody would care about the CONFIG_FB_DEVICE
> dependency.

Ok.

Btw, you will be astonished if you check the prices of those cards
on ebay nowadays.

> And in no case should it have been marked as a "fix".

Ok.
Do you want me to send a revert for this specific patch?

FWIW, just a few hours before I sent the pull request I did complain
about a similar patch (which I did not apply):
https://lore.kernel.org/linux-fbdev/7aabca78-dd34-4819-8a63-105d1a4cb4ba@gmx.de/T/#m070c6ba1047d26b856b0d6ac43592fc7b6f95518

Helge