Re: [BUG] KCSAN: data-race in xas_clear_mark / xas_find_marked

[Mirsad Todorovac]

On 8/18/23 16:19, Matthew Wilcox wrote:
> On Fri, Aug 18, 2023 at 03:37:10PM +0200, Mirsad Todorovac wrote:
> > I am new to KCSAN. I was not aware of KCSAN false positives thus far, so my best bet was to report them.
> >
> > I thought that maybe READ_ONCE() was required, but I will trust your judgment.
> >
> > I hope I can find this resolved.
>
> I haven't looked into KCSAN in any detail, I don't know what the right
> way is to resolve this.

That's OK, I suppose.

I don't feel clarity about it either.

I am certain that the original developers have the big picture. :-)

Best regards,
Mirsad Todorovac