Re: [PATCH 0/2] mm: thp: reduce unnecessary start_stop_khugepaged() calls

[Lorenzo Stoakes (Oracle)]

On Thu, Mar 05, 2026 at 12:45:56PM +0000, Lorenzo Stoakes (Oracle) wrote:
> On Thu, Mar 05, 2026 at 01:41:33PM +0100, David Hildenbrand (Arm) wrote:
> >
> > >> +
> > >> +	/* Check if anything has to be done */
> > >> +	if (hugepage_pmd_enabled() == !!khugepaged_thread)
> > >
> > > This is horrible, better to break out the latter expression as a small static
> > > function like:
> > >
> > > static bool is_khugepaged_thread_running(void)
> > > {
> > > 	if (khugepaged_thread)
> > > 		return true;
> > > 	return false;
> >
> > return khugepaged_thread;
> >
> > should do the trick, given taht the return type of the function is a bool :)
>
> Yeah I know :)
>
> I mean it's subjective, but this form makes it clear that what you're returning
> is _not_ a boolean.

Sorry obviously meant to say what you're TESTING not what you're RETURNING.

>
> I was a bit 'meh' when I first saw this pattern, but it's grown on me, sort of a
> 'make it so simple you can't get confused' kinda thing.
>
> >
> > > }
> >
> > --
> > Cheers,
> >
> > David
>
> Cheers, Lorenzo