Re: [PATCH v2 09/12] mm,thp: reduce ifdef'ery for THP in generic code

[Kirill A. Shutemov]

On Tue, Sep 22, 2015 at 04:04:53PM +0530, Vineet Gupta wrote:
> - pgtable-generic.c: Fold individual #ifdef for each helper into a top
>   level #ifdef. Makes code more readable

Makes sense.

> - Per Andrew's suggestion removed the dummy implementations for !THP
>   in asm-generic/page-table.h to have build time failures vs. runtime.

I'm not sure it's a good idea. This can lead to unnecessary #ifdefs where
otherwise call to helper would be eliminated by compiler as dead code.

What about dummy helpers with BUILD_BUG()?

> Signed-off-by: Vineet Gupta 
> ---
>  include/asm-generic/pgtable.h | 49 
> ---
>  mm/pgtable-generic.c  | 24 +++--
>  2 files changed, 21 insertions(+), 52 deletions(-)
> 
> diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
> index 29c57b2cb344..2112f4147816 100644
> --- a/include/asm-generic/pgtable.h
> +++ b/include/asm-generic/pgtable.h
> @@ -30,9 +30,12 @@ extern int ptep_set_access_flags(struct vm_area_struct 
> *vma,
>  #endif
>  
>  #ifndef __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS
> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
>  extern int pmdp_set_access_flags(struct vm_area_struct *vma,
>unsigned long address, pmd_t *pmdp,
>pmd_t entry, int dirty);
> +
> +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>  #endif
>  
>  #ifndef __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
> @@ -64,14 +67,6 @@ static inline int pmdp_test_and_clear_young(struct 
> vm_area_struct *vma,
>   set_pmd_at(vma->vm_mm, address, pmdp, pmd_mkold(pmd));
>   return r;
>  }
> -#else /* CONFIG_TRANSPARENT_HUGEPAGE */
> -static inline int pmdp_test_and_clear_young(struct vm_area_struct *vma,
> - unsigned long address,
> - pmd_t *pmdp)
> -{
> - BUG();
> - return 0;
> -}
>  #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>  #endif
>  
> @@ -81,8 +76,21 @@ int ptep_clear_flush_young(struct vm_area_struct *vma,
>  #endif
>  
>  #ifndef __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH
> -int pmdp_clear_flush_young(struct vm_area_struct *vma,
> -unsigned long address, pmd_t *pmdp);
> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
> +extern int pmdp_clear_flush_young(struct vm_area_struct *vma,
> +   unsigned long address, pmd_t *pmdp);
> +#else
> +/*
> + * Despite relevant to THP only, this API is called from generic rmap code
> + * under PageTransHuge(), hence needs a dummy implementation for !THP
> + */

Looks like a case I described above. BUILD_BUG_ON() should work fine here.

> +static inline int pmdp_clear_flush_young(struct vm_area_struct *vma,
> +  unsigned long address, pmd_t *pmdp)
> +{
> + BUG();
> + return 0;
> +}
> +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>  #endif

-- 
 Kirill A. Shutemov
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH v2 09/12] mm,thp: reduce ifdef'ery for THP in generic code

[Kirill A. Shutemov]

On Tue, Sep 22, 2015 at 04:04:53PM +0530, Vineet Gupta wrote:
> - pgtable-generic.c: Fold individual #ifdef for each helper into a top
>   level #ifdef. Makes code more readable

Makes sense.

> - Per Andrew's suggestion removed the dummy implementations for !THP
>   in asm-generic/page-table.h to have build time failures vs. runtime.

I'm not sure it's a good idea. This can lead to unnecessary #ifdefs where
otherwise call to helper would be eliminated by compiler as dead code.

What about dummy helpers with BUILD_BUG()?

> Signed-off-by: Vineet Gupta 
> ---
>  include/asm-generic/pgtable.h | 49 
> ---
>  mm/pgtable-generic.c  | 24 +++--
>  2 files changed, 21 insertions(+), 52 deletions(-)
> 
> diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
> index 29c57b2cb344..2112f4147816 100644
> --- a/include/asm-generic/pgtable.h
> +++ b/include/asm-generic/pgtable.h
> @@ -30,9 +30,12 @@ extern int ptep_set_access_flags(struct vm_area_struct 
> *vma,
>  #endif
>  
>  #ifndef __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS
> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
>  extern int pmdp_set_access_flags(struct vm_area_struct *vma,
>unsigned long address, pmd_t *pmdp,
>pmd_t entry, int dirty);
> +
> +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>  #endif
>  
>  #ifndef __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
> @@ -64,14 +67,6 @@ static inline int pmdp_test_and_clear_young(struct 
> vm_area_struct *vma,
>   set_pmd_at(vma->vm_mm, address, pmdp, pmd_mkold(pmd));
>   return r;
>  }
> -#else /* CONFIG_TRANSPARENT_HUGEPAGE */
> -static inline int pmdp_test_and_clear_young(struct vm_area_struct *vma,
> - unsigned long address,
> - pmd_t *pmdp)
> -{
> - BUG();
> - return 0;
> -}
>  #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>  #endif
>  
> @@ -81,8 +76,21 @@ int ptep_clear_flush_young(struct vm_area_struct *vma,
>  #endif
>  
>  #ifndef __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH
> -int pmdp_clear_flush_young(struct vm_area_struct *vma,
> -unsigned long address, pmd_t *pmdp);
> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
> +extern int pmdp_clear_flush_young(struct vm_area_struct *vma,
> +   unsigned long address, pmd_t *pmdp);
> +#else
> +/*
> + * Despite relevant to THP only, this API is called from generic rmap code
> + * under PageTransHuge(), hence needs a dummy implementation for !THP
> + */

Looks like a case I described above. BUILD_BUG_ON() should work fine here.

> +static inline int pmdp_clear_flush_young(struct vm_area_struct *vma,
> +  unsigned long address, pmd_t *pmdp)
> +{
> + BUG();
> + return 0;
> +}
> +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>  #endif

-- 
 Kirill A. Shutemov
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/