On Wed, May 17, 2017 at 10:01:29AM +0200, Jiri Slaby wrote:

On 05/15/2017, 03:16 PM, Will Deacon wrote:

quoted

Whilst I think this is a good idea, the code in question actually results
in undefined behaviour per the C spec and is reported by UBSAN.

Hi, yes, I know -- this patch was the 1st from the series of 3 which I
sent a long time ago to fix that up too. But I remember your patch, so I
sent only this one this time.

quoted

See my
patch fixing arm64 here (which I'd forgotten about):

https://www.spinics.net/lists/linux-arch/msg38564.html

But, as stated in the thread above, I think we should go a step further
and remove FUTEX_OP_{OR,ANDN,XOR,OPARG_SHIFT} altogether. They don't
appear to be used by userspace, and this whole thing is a total mess.

Any thoughts?

Ok, I am all for that. I think the only question is who is going to do
the work and submit it :)? Do I understand correctly to eliminate all
these functions and the path into the kernel? But won't this break API
-- are there really no users of this interface?

That's the million-dollar question, really. I don't know of any code using
it, and I couldn't find any when I looked (also nothing reported by Debian
Codesearch afaict), but I was hoping linux-arch might have some thoughts
on this too.

For now, I'll queue my arm64 patch before I forget about it again!

Will