Re: [PATCH v3 3/7] mm/sparse.c: introduce a new function clear_subsection_map()

[Baoquan He <bhe@redhat.com>]

On 03/09/20 at 02:38pm, David Hildenbrand wrote:
> On 09.03.20 14:32, Baoquan He wrote:
> > On 03/09/20 at 09:59am, David Hildenbrand wrote:
> >> On 07.03.20 09:42, Baoquan He wrote:
> >>> Factor out the code which clear subsection map of one memory region from
> >>> section_deactivate() into clear_subsection_map().
> >>>
> >>> Signed-off-by: Baoquan He <bhe@redhat.com>
> >>> ---
> >>>  mm/sparse.c | 31 ++++++++++++++++++++++++-------
> >>>  1 file changed, 24 insertions(+), 7 deletions(-)
> >>>
> >>> diff --git a/mm/sparse.c b/mm/sparse.c
> >>> index e37c0abcdc89..d9dcd58d5c1d 100644
> >>> --- a/mm/sparse.c
> >>> +++ b/mm/sparse.c
> >>> @@ -726,15 +726,11 @@ static void free_map_bootmem(struct page *memmap)
> >>>  }
> >>>  #endif /* CONFIG_SPARSEMEM_VMEMMAP */
> >>>  
> >>> -static void section_deactivate(unsigned long pfn, unsigned long nr_pages,
> >>> -		struct vmem_altmap *altmap)
> >>> +static int clear_subsection_map(unsigned long pfn, unsigned long nr_pages)
> >>>  {
> >>>  	DECLARE_BITMAP(map, SUBSECTIONS_PER_SECTION) = { 0 };
> >>>  	DECLARE_BITMAP(tmp, SUBSECTIONS_PER_SECTION) = { 0 };
> >>>  	struct mem_section *ms = __pfn_to_section(pfn);
> >>> -	bool section_is_early = early_section(ms);
> >>> -	struct page *memmap = NULL;
> >>> -	bool empty = false;
> >>>  	unsigned long *subsection_map = ms->usage
> >>>  		? &ms->usage->subsection_map[0] : NULL;
> >>>  
> >>> @@ -745,8 +741,31 @@ static void section_deactivate(unsigned long pfn, unsigned long nr_pages,
> >>>  	if (WARN(!subsection_map || !bitmap_equal(tmp, map, SUBSECTIONS_PER_SECTION),
> >>>  				"section already deactivated (%#lx + %ld)\n",
> >>>  				pfn, nr_pages))
> >>> +		return -EINVAL;
> >>> +
> >>> +	bitmap_xor(subsection_map, map, subsection_map, SUBSECTIONS_PER_SECTION);
> >>> +
> >>
> >> Nit: I'd drop this line.
> > 
> > It's fine to me. I usually keep one line for the returning. I will
> > remove it when update.
> > 
> >>
> >>> +	return 0;
> >>> +}
> >>> +
> >>> +static bool is_subsection_map_empty(struct mem_section *ms)
> >>> +{
> >>> +	return bitmap_empty(&ms->usage->subsection_map[0],
> >>> +			    SUBSECTIONS_PER_SECTION);
> >>> +}
> >>> +
> >>> +static void section_deactivate(unsigned long pfn, unsigned long nr_pages,
> >>> +		struct vmem_altmap *altmap)
> >>> +{
> >>> +	struct mem_section *ms = __pfn_to_section(pfn);
> >>> +	bool section_is_early = early_section(ms);
> >>> +	struct page *memmap = NULL;
> >>> +	bool empty = false;
> >>
> >> Nit: No need to initialize empty.
> > 
> > This is inherited from patch 1.
> > 
> >>
> >>> +
> >>> +	if (clear_subsection_map(pfn, nr_pages))
> >>>  		return;
> >>>  
> >>
> >> Nit: I'd drop this empty line.
> >>
> >>> +	empty = is_subsection_map_empty(ms);
> >>>  	/*
> >>>  	 * There are 3 cases to handle across two configurations
> >>>  	 * (SPARSEMEM_VMEMMAP={y,n}):
> >>> @@ -764,8 +783,6 @@ static void section_deactivate(unsigned long pfn, unsigned long nr_pages,
> >>>  	 *
> >>>  	 * For 2/ and 3/ the SPARSEMEM_VMEMMAP={y,n} cases are unified
> >>>  	 */
> >>> -	bitmap_xor(subsection_map, map, subsection_map, SUBSECTIONS_PER_SECTION);
> >>> -	empty = bitmap_empty(subsection_map, SUBSECTIONS_PER_SECTION);
> >>
> >> I do wonder why you moved this up the comment?
> > 
> > Since this empty will cover two places of handling, so moved it up,
> > seems this is what I was thinking. Can move it back here.
> 
> You're moving the whole comment later, was just wondering (makes it
> slightly harder to review).

I see, sorry for the confusion. I will move it back when repost.