Re: Data corruption on i.MX6 IPU in arm_copy_from_user()

[Krzysztof Hałasa]

"Russell King (Oracle)" <linux@xxxxxxxxxxxxxxx> writes:

>> LDM12: 1 3 4 5 6 7 8 9 A B C D
>
> That's rather sad, and does look very much like a hardware bug.
>
> The question is what to do about it... there's Linus' "do not break
> userspace" edict and that's exactly what this change has done. So I
> suppose we're going to have to revert the change and put up with
> everything being slightly slower on arm32 than it otherwise would
> have been. That probably means we'll end up with almost every kernel
> tree out there carrying a revert of the revert to work around the
> fact that seemingly NXP broke their hardware - which itself is not
> a good idea. I guess we're just going to have to put up with that.

For userspace, it's quite a corner case, basically development-only -
and I guess there are very few people who will do things like this.

The same problem can manifest itself without any kernel involvement -
it's enough to mmap /dev/mem and use LDM on in completely in userspace.
This can't be fixed - unless we disallow IPU mmap.

Perhaps making sure the bug is clearly documented is better than doing
a partial fix. Ideally NXP should document it in their papers, and we
should add notes to IPU driver code.

The last one - I guess I can do.
-- 
Krzysztof Hałasa

Sieć Badawcza Łukasiewicz
Przemysłowy Instytut Automatyki i Pomiarów PIAP
Al. Jerozolimskie 202, 02-486 Warszawa