Re: [PATCH v4 2/2] arm64: mm: hugetlb: Enable HUGETLB_PAGE_FREE_VMEMMAP for arm64

[Muchun Song <songmuchun@bytedance.com>]

On Mon, Apr 11, 2022 at 03:42:52PM +0530, Anshuman Khandual wrote:
> 
> 
> On 3/31/22 12:26, Muchun Song wrote:
> > --- a/arch/arm64/mm/flush.c
> > +++ b/arch/arm64/mm/flush.c
> > @@ -68,6 +68,19 @@ EXPORT_SYMBOL_GPL(__sync_icache_dcache);
> >   */
> >  void flush_dcache_page(struct page *page)
> >  {
> > +	/*
> > +	 * Only the head page's flags of HugeTLB can be cleared since the tail
> > +	 * vmemmap pages associated with each HugeTLB page are mapped with
> > +	 * read-only when CONFIG_HUGETLB_PAGE_FREE_VMEMMAP is enabled (more
> > +	 * details can refer to vmemmap_remap_pte()).  Although
> 
> Did you see real permission fault when flush_dcache_page() accessed remapped
> tail pages, with readonly vmemmap ? OR this change is from code inspection ?
>

Not a real word issue. Just a code-inspection one.
 
> > +	 * __sync_icache_dcache() only set PG_dcache_clean flag on the head
> > +	 * page struct, some tail page structs still can be seen the flag is
> 
> Sentence here needs restructuring ....               ^^^^^^^^^^^^^^^^^^
> 

Will do.

> > +	 * set since the head vmemmap page frame is reused (more details can
> > +	 * refer to the comments above page_fixed_fake_head()).
> > +	 */
> > +	if (hugetlb_free_vmemmap_enabled() && PageHuge(page))
> > +		page = compound_head(page);
> 
> This should also be applicable to any other platform with both configs enabled
> i.e ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE and ARCH_WANT_HUGETLB_PAGE_FREE_VMEMMAP ?

Yes.

> If yes, then how to ensure that the platforms change flush_dcache_page() before
> subscribing into ARCH_WANT_HUGETLB_PAGE_FREE_VMEMMAP ?
> 

The one who enable this feature has to make sure there is no any issues
(e.g. He should fix flush_dcache_page() or other related issues) with this
feature enabled.

Thanks.