Re: [PATCH] s390/mm: Allow hugepage migration for s390

[Gerald Schaefer <gerald.schaefer@linux.ibm.com>]

On Wed, 11 Nov 2026 18:21:49 +0100
Gerald Schaefer <gerald.schaefer@linux.ibm.com> wrote:

> On Wed, 11 Nov 2026 17:09:45 +0100
> Alexander Gordeev <agordeev@linux.ibm.com> wrote:
> 
> [...]
> > > @@ -241,3 +241,15 @@ unsigned int __init arch_hugetlb_cma_order(void)
> > >  
> > >  	return 0;
> > >  }
> > > +
> > > +bool __init arch_hugetlb_valid_size(unsigned long size)
> > > +{
> > > +	return __hugetlb_valid_size(size);
> > > +}
> > > +
> > > +#ifdef CONFIG_ARCH_ENABLE_HUGEPAGE_MIGRATION
> > 
> > Could you please use IS_ENABLED() instead?
> 
> Does the benefit mentioned in Documentation/process/coding-style.rst
> for usage "within code" also apply in such a case, i.e. not within
> code?
> 
> Or are there other benefits or coding style requirements? Just wondering,
> because I never saw that in such a case where a whole function is behind
> such an #ifdef.

Ah ok, there is of course some code in the function, and it is not (yet)
referencing symbols that will not exist if the condition is not met. So
I guess it also makes some sense here. Will send a v2.

> 
> > 
> > > +bool arch_hugetlb_migration_supported(struct hstate *h)
> > > +{
> > > +	return __hugetlb_valid_size(huge_page_size(h));
> > > +}
> > > +#endif
> > 
> > Otherwise, LGTM.
> > 
> > Thanks!
> 
>