Re: [PATCH v3 3/3] arm64: mm: enable per pmd page table lock

[Yu Zhao]

On Mon, Mar 11, 2019 at 12:12:28PM +0000, Mark Rutland wrote:
> Hi,
> 
> On Sat, Mar 09, 2019 at 06:19:06PM -0700, Yu Zhao wrote:
> > Switch from per mm_struct to per pmd page table lock by enabling
> > ARCH_ENABLE_SPLIT_PMD_PTLOCK. This provides better granularity for
> > large system.
> > 
> > I'm not sure if there is contention on mm->page_table_lock. Given
> > the option comes at no cost (apart from initializing more spin
> > locks), why not enable it now.
> > 
> > We only do so when pmd is not folded, so we don't mistakenly call
> > pgtable_pmd_page_ctor() on pud or p4d in pgd_pgtable_alloc(). (We
> > check shift against PMD_SHIFT, which is same as PUD_SHIFT when pmd
> > is folded).
> 
> Just to check, I take it pgtable_pmd_page_ctor() is now a NOP when the
> PMD is folded, and this last paragraph is stale?

Yes, and will remove it.

> > Signed-off-by: Yu Zhao <yuzhao@xxxxxxxxxx>
> > ---
> >  arch/arm64/Kconfig               |  3 +++
> >  arch/arm64/include/asm/pgalloc.h | 12 +++++++++++-
> >  arch/arm64/include/asm/tlb.h     |  5 ++++-
> >  3 files changed, 18 insertions(+), 2 deletions(-)
> > 
> > diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> > index cfbf307d6dc4..a3b1b789f766 100644
> > --- a/arch/arm64/Kconfig
> > +++ b/arch/arm64/Kconfig
> > @@ -872,6 +872,9 @@ config ARCH_WANT_HUGE_PMD_SHARE
> >  config ARCH_HAS_CACHE_LINE_SIZE
> >  	def_bool y
> >  
> > +config ARCH_ENABLE_SPLIT_PMD_PTLOCK
> > +	def_bool y if PGTABLE_LEVELS > 2
> > +
> >  config SECCOMP
> >  	bool "Enable seccomp to safely compute untrusted bytecode"
> >  	---help---
> > diff --git a/arch/arm64/include/asm/pgalloc.h b/arch/arm64/include/asm/pgalloc.h
> > index 52fa47c73bf0..dabba4b2c61f 100644
> > --- a/arch/arm64/include/asm/pgalloc.h
> > +++ b/arch/arm64/include/asm/pgalloc.h
> > @@ -33,12 +33,22 @@
> >  
> >  static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr)
> >  {
> > -	return (pmd_t *)__get_free_page(PGALLOC_GFP);
> > +	struct page *page;
> > +
> > +	page = alloc_page(PGALLOC_GFP);
> > +	if (!page)
> > +		return NULL;
> > +	if (!pgtable_pmd_page_ctor(page)) {
> > +		__free_page(page);
> > +		return NULL;
> > +	}
> > +	return page_address(page);
> >  }
> >  
> >  static inline void pmd_free(struct mm_struct *mm, pmd_t *pmdp)
> >  {
> >  	BUG_ON((unsigned long)pmdp & (PAGE_SIZE-1));
> > +	pgtable_pmd_page_dtor(virt_to_page(pmdp));
> >  	free_page((unsigned long)pmdp);
> >  }
> 
> It looks like arm64's existing stage-2 code is inconsistent across
> alloc/free, and IIUC this change might turn that into a real problem.
> Currently we allocate all levels of stage-2 table with
> __get_free_page(), but free them with p?d_free(). We always miss the
> ctor and always use the dtor.
> 
> Other than that, this patch looks fine to me, but I'd feel more
> comfortable if we could first fix the stage-2 code to free those stage-2
> tables without invoking the dtor.
> 
> Anshuman, IIRC you had a patch to fix the stage-2 code to not invoke the
> dtors. If so, could you please post that so that we could take it as a
> preparatory patch for this series?

Will do.

> Thanks,
> Mark.
> 
> > diff --git a/arch/arm64/include/asm/tlb.h b/arch/arm64/include/asm/tlb.h
> > index 106fdc951b6e..4e3becfed387 100644
> > --- a/arch/arm64/include/asm/tlb.h
> > +++ b/arch/arm64/include/asm/tlb.h
> > @@ -62,7 +62,10 @@ static inline void __pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte,
> >  static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmdp,
> >  				  unsigned long addr)
> >  {
> > -	tlb_remove_table(tlb, virt_to_page(pmdp));
> > +	struct page *page = virt_to_page(pmdp);
> > +
> > +	pgtable_pmd_page_dtor(page);
> > +	tlb_remove_table(tlb, page);
> >  }
> >  #endif
> >  
> > -- 
> > 2.21.0.360.g471c308f928-goog
> >