Re: [PATCH V2 1/2] mm: Update generic gup implementation to handle hugepage directory

[Steve Capper]

On Fri, Oct 17, 2014 at 10:08:06AM +0530, Aneesh Kumar K.V wrote:
> Update generic gup implementation with powerpc specific details.
> On powerpc at pmd level we can have hugepte, normal pmd pointer
> or a pointer to the hugepage directory.
> 
> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@xxxxxxxxxxxxxxxxxx>
> ---
> Changes from V1: 
> * Folded arm/arm64 related changes into the patch
> * Dropped pgd_huge from generic header
> 
>  arch/arm/include/asm/pgtable.h   |   2 +
>  arch/arm64/include/asm/pgtable.h |   2 +
>  include/linux/mm.h               |  26 +++++++++
>  mm/gup.c                         | 113 +++++++++++++++++++--------------------
>  4 files changed, 84 insertions(+), 59 deletions(-)
> 

Hi Aneesh,
Thanks for coding this up. I've tested this for arm (Arndale board) and
arm64 (Juno); it builds without any issues and passes my futex on THP
tail test.

Please add my:
Tested-by: Steve Capper <steve.capper@xxxxxxxxxx>

As this patch progresses through -mm, the arm maintainer:
Russell King <linux@xxxxxxxxxxxxxxxx>

and arm64 maintainers:
Catalin Marinas <catalin.marinas@xxxxxxx>
Will Deacon <will.deacon@xxxxxxx>

should also be on CC.

Cheers,
-- 
Steve

> diff --git a/arch/arm/include/asm/pgtable.h b/arch/arm/include/asm/pgtable.h
> index 90aa4583b308..46f81fbaa4a5 100644
> --- a/arch/arm/include/asm/pgtable.h
> +++ b/arch/arm/include/asm/pgtable.h
> @@ -181,6 +181,8 @@ extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
>  /* to find an entry in a kernel page-table-directory */
>  #define pgd_offset_k(addr)	pgd_offset(&init_mm, addr)
>  
> +#define pgd_huge(pgd)		(0)
> +
>  #define pmd_none(pmd)		(!pmd_val(pmd))
>  #define pmd_present(pmd)	(pmd_val(pmd))
>  
> diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
> index cefd3e825612..ed8f42497ac4 100644
> --- a/arch/arm64/include/asm/pgtable.h
> +++ b/arch/arm64/include/asm/pgtable.h
> @@ -464,6 +464,8 @@ static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)
>  extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
>  extern pgd_t idmap_pg_dir[PTRS_PER_PGD];
>  
> +#define pgd_huge(pgd)		(0)
> +
>  /*
>   * Encode and decode a swap entry:
>   *	bits 0-1:	present (must be zero)
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index 02d11ee7f19d..f97732412cb4 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1219,6 +1219,32 @@ long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
>  		    struct vm_area_struct **vmas);
>  int get_user_pages_fast(unsigned long start, int nr_pages, int write,
>  			struct page **pages);
> +
> +#ifdef CONFIG_HAVE_GENERIC_RCU_GUP
> +#ifndef is_hugepd
> +/*
> + * Some architectures support hugepage directory format that is
> + * required to support different hugetlbfs sizes.
> + */
> +typedef struct { unsigned long pd; } hugepd_t;
> +#define is_hugepd(hugepd) (0)
> +#define __hugepd(x) ((hugepd_t) { (x) })
> +static inline int gup_hugepd(hugepd_t hugepd, unsigned long addr,
> +			     unsigned pdshift, unsigned long end,
> +			     int write, struct page **pages, int *nr)
> +{
> +	return 0;
> +}
> +#else
> +extern int gup_hugepd(hugepd_t hugepd, unsigned long addr,
> +		      unsigned pdshift, unsigned long end,
> +		      int write, struct page **pages, int *nr);
> +#endif
> +extern int gup_huge_pte(pte_t orig, pte_t *ptep, unsigned long addr,
> +			unsigned long sz, unsigned long end, int write,
> +			struct page **pages, int *nr);
> +#endif
> +
>  struct kvec;
>  int get_kernel_pages(const struct kvec *iov, int nr_pages, int write,
>  			struct page **pages);
> diff --git a/mm/gup.c b/mm/gup.c
> index cd62c8c90d4a..13c560ef9ddf 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -786,65 +786,31 @@ static int gup_pte_range(pmd_t pmd, unsigned long addr, unsigned long end,
>  }
>  #endif /* __HAVE_ARCH_PTE_SPECIAL */
>  
> -static int gup_huge_pmd(pmd_t orig, pmd_t *pmdp, unsigned long addr,
> -		unsigned long end, int write, struct page **pages, int *nr)
> +int gup_huge_pte(pte_t orig, pte_t *ptep, unsigned long addr,
> +		 unsigned long sz, unsigned long end, int write,
> +		 struct page **pages, int *nr)
>  {
> -	struct page *head, *page, *tail;
>  	int refs;
> +	unsigned long pte_end;
> +	struct page *head, *page, *tail;
>  
> -	if (write && !pmd_write(orig))
> -		return 0;
> -
> -	refs = 0;
> -	head = pmd_page(orig);
> -	page = head + ((addr & ~PMD_MASK) >> PAGE_SHIFT);
> -	tail = page;
> -	do {
> -		VM_BUG_ON_PAGE(compound_head(page) != head, page);
> -		pages[*nr] = page;
> -		(*nr)++;
> -		page++;
> -		refs++;
> -	} while (addr += PAGE_SIZE, addr != end);
>  
> -	if (!page_cache_add_speculative(head, refs)) {
> -		*nr -= refs;
> +	if (write && !pte_write(orig))
>  		return 0;
> -	}
>  
> -	if (unlikely(pmd_val(orig) != pmd_val(*pmdp))) {
> -		*nr -= refs;
> -		while (refs--)
> -			put_page(head);
> +	if (!pte_present(orig))
>  		return 0;
> -	}
>  
> -	/*
> -	 * Any tail pages need their mapcount reference taken before we
> -	 * return. (This allows the THP code to bump their ref count when
> -	 * they are split into base pages).
> -	 */
> -	while (refs--) {
> -		if (PageTail(tail))
> -			get_huge_page_tail(tail);
> -		tail++;
> -	}
> -
> -	return 1;
> -}
> -
> -static int gup_huge_pud(pud_t orig, pud_t *pudp, unsigned long addr,
> -		unsigned long end, int write, struct page **pages, int *nr)
> -{
> -	struct page *head, *page, *tail;
> -	int refs;
> +	pte_end = (addr + sz) & ~(sz-1);
> +	if (pte_end < end)
> +		end = pte_end;
>  
> -	if (write && !pud_write(orig))
> -		return 0;
> +	/* hugepages are never "special" */
> +	VM_BUG_ON(!pfn_valid(pte_pfn(orig)));
>  
>  	refs = 0;
> -	head = pud_page(orig);
> -	page = head + ((addr & ~PUD_MASK) >> PAGE_SHIFT);
> +	head = pte_page(orig);
> +	page = head + ((addr & (sz-1)) >> PAGE_SHIFT);
>  	tail = page;
>  	do {
>  		VM_BUG_ON_PAGE(compound_head(page) != head, page);
> @@ -859,13 +825,18 @@ static int gup_huge_pud(pud_t orig, pud_t *pudp, unsigned long addr,
>  		return 0;
>  	}
>  
> -	if (unlikely(pud_val(orig) != pud_val(*pudp))) {
> +	if (unlikely(pte_val(orig) != pte_val(*ptep))) {
>  		*nr -= refs;
>  		while (refs--)
>  			put_page(head);
>  		return 0;
>  	}
>  
> +	/*
> +	 * Any tail pages need their mapcount reference taken before we
> +	 * return. (This allows the THP code to bump their ref count when
> +	 * they are split into base pages).
> +	 */
>  	while (refs--) {
>  		if (PageTail(tail))
>  			get_huge_page_tail(tail);
> @@ -898,10 +869,19 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end,
>  			if (pmd_numa(pmd))
>  				return 0;
>  
> -			if (!gup_huge_pmd(pmd, pmdp, addr, next, write,
> -				pages, nr))
> +			if (!gup_huge_pte(__pte(pmd_val(pmd)), (pte_t *)pmdp,
> +					  addr, PMD_SIZE, next,
> +					  write, pages, nr))
>  				return 0;
>  
> +		} else if (unlikely(is_hugepd(__hugepd(pmd_val(pmd))))) {
> +			/*
> +			 * architecture have different format for hugetlbfs
> +			 * pmd format and THP pmd format
> +			 */
> +			if (!gup_hugepd(__hugepd(pmd_val(pmd)), addr, PMD_SHIFT,
> +					next, write, pages, nr))
> +				return 0;
>  		} else if (!gup_pte_range(pmd, addr, next, write, pages, nr))
>  				return 0;
>  	} while (pmdp++, addr = next, addr != end);
> @@ -909,22 +889,27 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end,
>  	return 1;
>  }
>  
> -static int gup_pud_range(pgd_t *pgdp, unsigned long addr, unsigned long end,
> +static int gup_pud_range(pgd_t pgd, unsigned long addr, unsigned long end,
>  		int write, struct page **pages, int *nr)
>  {
>  	unsigned long next;
>  	pud_t *pudp;
>  
> -	pudp = pud_offset(pgdp, addr);
> +	pudp = pud_offset(&pgd, addr);
>  	do {
>  		pud_t pud = ACCESS_ONCE(*pudp);
>  
>  		next = pud_addr_end(addr, end);
>  		if (pud_none(pud))
>  			return 0;
> -		if (pud_huge(pud)) {
> -			if (!gup_huge_pud(pud, pudp, addr, next, write,
> -					pages, nr))
> +		if (unlikely(pud_huge(pud))) {
> +			if (!gup_huge_pte(__pte(pud_val(pud)), (pte_t *)pudp,
> +					  addr, PUD_SIZE, next,
> +					  write, pages, nr))
> +				return 0;
> +		} else if (unlikely(is_hugepd(__hugepd(pud_val(pud))))) {
> +			if (!gup_hugepd(__hugepd(pud_val(pud)), addr, PUD_SHIFT,
> +					next, write, pages, nr))
>  				return 0;
>  		} else if (!gup_pmd_range(pud, addr, next, write, pages, nr))
>  			return 0;
> @@ -970,10 +955,21 @@ int __get_user_pages_fast(unsigned long start, int nr_pages, int write,
>  	local_irq_save(flags);
>  	pgdp = pgd_offset(mm, addr);
>  	do {
> +		pgd_t pgd = ACCESS_ONCE(*pgdp);
> +
>  		next = pgd_addr_end(addr, end);
> -		if (pgd_none(*pgdp))
> +		if (pgd_none(pgd))
>  			break;
> -		else if (!gup_pud_range(pgdp, addr, next, write, pages, &nr))
> +		if (unlikely(pgd_huge(pgd))) {
> +			if (!gup_huge_pte(__pte(pgd_val(pgd)), (pte_t *)pgdp,
> +					  addr, PGDIR_SIZE, next,
> +					  write, pages, &nr))
> +				break;
> +		} else if (unlikely(is_hugepd(__hugepd(pgd_val(pgd))))) {
> +			if (!gup_hugepd(__hugepd(pgd_val(pgd)), addr, PGDIR_SHIFT,
> +					next, write, pages, &nr))
> +				break;
> +		} else if (!gup_pud_range(pgd, addr, next, write, pages, &nr))
>  			break;
>  	} while (pgdp++, addr = next, addr != end);
>  	local_irq_restore(flags);
> @@ -1028,5 +1024,4 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write,
>  
>  	return ret;
>  }
> -
>  #endif /* CONFIG_HAVE_GENERIC_RCU_GUP */
> -- 
> 1.9.1
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/