Re: [PATCH 36/37] Cavium OCTEON: pmd_none - use pmd_val() in pmd_val().

[Paul Gortmaker <paul.gortmaker@windriver.com>]

ddaney@caviumnetworks.com wrote:
> From: David Daney <ddaney@caviumnetworks.com>
>
> Before marking pmd_val as invalid_pte_table, factor in existing value
> for pmd_val.
>
> Signed-off-by: Paul Gortmaker <Paul.Gortmaker@windriver.com>
> Signed-off-by: David Daney <ddaney@caviumnetworks.com>
> ---
>  arch/mips/include/asm/pgtable-64.h |    5 +++++
>  1 files changed, 5 insertions(+), 0 deletions(-)
>   

Hi David,

I think you missed my feedback on this patch -- you should be the primary
S.O.B.  on this, and at a minimum, it needs a better description of just why
this is required (if it actually is) on Cavium CPUs, but not on other 
boards.

Thanks,
Paul.

> diff --git a/arch/mips/include/asm/pgtable-64.h b/arch/mips/include/asm/pgtable-64.h
> index 943515f..bb93bd5 100644
> --- a/arch/mips/include/asm/pgtable-64.h
> +++ b/arch/mips/include/asm/pgtable-64.h
> @@ -129,7 +129,12 @@ extern pmd_t empty_bad_pmd_table[PTRS_PER_PMD];
>   */
>  static inline int pmd_none(pmd_t pmd)
>  {
> +#ifdef CONFIG_CPU_CAVIUM_OCTEON
> +	return (pmd_val(pmd) == (unsigned long) invalid_pte_table) ||
> +				(!pmd_val(pmd));
> +#else
>  	return pmd_val(pmd) == (unsigned long) invalid_pte_table;
> +#endif
>  }
>  
>  #define pmd_bad(pmd)		(pmd_val(pmd) & ~PAGE_MASK)
>