linux-kernel - Re: [PATCH v2 3/7] mm/sparse.c: introduce a new function clear_subsection

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sun, 1 Mar 2020 13:20:28 +0800
From:   Baoquan He <bhe@...hat.com>
To:     David Hildenbrand <david@...hat.com>
Cc:     linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        akpm@...ux-foundation.org, richardw.yang@...ux.intel.com,
        osalvador@...e.de, dan.j.williams@...el.com, mhocko@...e.com,
        rppt@...ux.ibm.com, robin.murphy@....com
Subject: Re: [PATCH v2 3/7] mm/sparse.c: introduce a new function
 clear_subsection_map()

On 02/28/20 at 03:36pm, David Hildenbrand wrote:
> On 20.02.20 05:33, Baoquan He wrote:
> > Wrap the codes which clear subsection map of one memory region from
> > section_deactivate() into clear_subsection_map().
> > 
> > Signed-off-by: Baoquan He <bhe@...hat.com>
> > ---
> >  mm/sparse.c | 46 ++++++++++++++++++++++++++++++++++++++--------
> >  1 file changed, 38 insertions(+), 8 deletions(-)
> > 
> > diff --git a/mm/sparse.c b/mm/sparse.c
> > index 977b47acd38d..df857ee9330c 100644
> > --- a/mm/sparse.c
> > +++ b/mm/sparse.c
> > @@ -726,14 +726,25 @@ 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)
> > +/**
> > + * clear_subsection_map - Clear subsection map of one memory region
> > + *
> > + * @pfn - start pfn of the memory range
> > + * @nr_pages - number of pfns to add in the region
> > + *
> > + * This is only intended for hotplug, and clear the related subsection
> > + * map inside one section.
> > + *
> > + * Return:
> > + * * -EINVAL	- Section already deactived.
> > + * * 0		- Subsection map is emptied.
> > + * * 1		- Subsection map is not empty.
> > + */
> 
> Less verbose please (in my preference: none and simplify return handling)
> 
> > +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;
> >  	unsigned long *subsection_map = ms->usage
> >  		? &ms->usage->subsection_map[0] : NULL;
> >  
> > @@ -744,8 +755,28 @@ 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;
> > +		return -EINVAL;
> > +
> > +	bitmap_xor(subsection_map, map, subsection_map, SUBSECTIONS_PER_SECTION);
> >  
> > +	if (bitmap_empty(subsection_map, SUBSECTIONS_PER_SECTION))
> > +		return 0;
> > +
> 
> Can we please just have a
> 
> subsection_map_empty() instead and handle that in the caller?
> (you can then always return true in the !VMEMMAP variant)

I don't follow. Could you be more specific? or pseudo code please?

The old code has to handle below case in which subsection_map has been
cleared. And I introduce clear_subsection_map() to encapsulate all
subsection map realted code so that !VMEMMAP won't have to see it any
more.

        if (WARN(!subsection_map || !bitmap_equal(tmp, map, SUBSECTIONS_PER_SECTION),
                                "section already deactivated (%#lx + %ld)\n",
                                pfn, nr_pages))
                return;


> 
> I dislike the "rc" handling in the caller.
> 
> > +	return 1;
> > +}
> > +
> > +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;
> > +	int rc;
> > +
> > +
> 
> one superfluous empty line
> 
> > +	rc = clear_subsection_map(pfn, nr_pages);
> > +	if (IS_ERR_VALUE((unsigned long)rc))
> 
> huh? "if (rc < 0)" ? or am I missing something?

Both is fine to me, I have no preference, can change like this.

> 
> > +		return;
> >  	/*
> >  	 * There are 3 cases to handle across two configurations
> >  	 * (SPARSEMEM_VMEMMAP={y,n}):
> > @@ -763,8 +794,7 @@ 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);
> > -	if (bitmap_empty(subsection_map, SUBSECTIONS_PER_SECTION)) {
> > +	if (!rc) {
> >  		unsigned long section_nr = pfn_to_section_nr(pfn);
> >  
> >  		/*
> > @@ -786,7 +816,7 @@ static void section_deactivate(unsigned long pfn, unsigned long nr_pages,
> >  	else
> >  		depopulate_section_memmap(pfn, nr_pages, altmap);
> >  
> > -	if (bitmap_empty(subsection_map, SUBSECTIONS_PER_SECTION))
> > +	if (!rc)
> 
> I don't really like that handling.
> 
> Either
> 
> s/rc/section_empty/
> 
> or use a separate subsection_map_empty()
> 
> >  		ms->section_mem_map = (unsigned long)NULL;
> >  }
> >  
> > 
> 
> Thanks!
> 
> -- 
> Thanks,
> 
> David / dhildenb