Re: [PATCH 0/2] mm->def_flags cleanups (Was: Change khugepaged to respect MMF_THP_DISABLE flag)

[Oleg Nesterov <oleg@redhat.com>]

On 01/22, Alex Thorlton wrote:
>
> On Wed, Jan 22, 2014 at 08:43:27PM +0100, Oleg Nesterov wrote:
> > On 01/22, Oleg Nesterov wrote:
> > >
> > > On 01/22, Alex Thorlton wrote:
> > > > > +	case PR_SET_THP_DISABLE:
> > > > > +	case PR_GET_THP_DISABLE:
> > > > > +		down_write(&me->mm->mmap_sem);
> > > > > +		if (option == PR_SET_THP_DISABLE) {
> > > > > +			if (arg2)
> > > > > +				me->mm->def_flags |= VM_NOHUGEPAGE;
> > > > > +			else
> > > > > +				me->mm->def_flags &= ~VM_NOHUGEPAGE;
> > > > > +		} else {
> > > > > +			error = !!(me->mm->flags && VM_NOHUGEPAGE);
> > > >
> > > > Should be:
> > > >
> > > > error = !!(me->mm->def_flags && VM_NOHUGEPAGE);
> > >
> > > No, we need to return 1 if this bit is set ;)
> >
> > Damn, you are right of course, we need "&". I didn't notice "&&"
> > in the patch I sent and misunderstood your "&&" above ;) Sorry.
>
> Actually, I didn't catch that either!  Looking at it, though, we
> definitely do want bitwise AND here, not logical.
>
> However, what I was originally referring to is:  Shouldn't we be
> checking mm->***def_flags*** for the VM_NOHUGEPAGE bit, as opposed
> to mm->flags?  i.e. I think we want this:
>
> error = !!(me->mm->def_flags & VM_NOHUGEPAGE);

Damn, of course you are right. I misunderstood you twice.

But so far I'm afraid this idea can't work anyway, although lets wait
for reply from s390 maintainers.

Oleg.