[Cbe-oss-dev] [PATCH] powerpc: Introduce address space "slices"

[Benjamin Herrenschmidt]

On Tue, 2007-02-20 at 13:45 -0600, Adam Litke wrote:
> Your patch drops the pgoff check that prepare_hugepage_range used to
> check.  The misaligned_offset test in libhugetlbfs identified the
> problem.  The following patch (applied on top of yours) makes the
> problem go away.  I am not necessarily suggesting it's the correct
> fix... just concisely describing the problem.

Ok, I'll fold that into the patch. Ultimately, when I finally do the
generic changes, prepare_hugepage_range() will be going away. I will
either pass pgoff along to slice_g_u_a for it to validate the pgoff, or
I will let f_ops->mmap() be responsible of checking it. For SPEs, I do
the pgoff check there. Any reason tht wouldn't work for huge pages ?

Ben.

> commit 95bcfa9c7b086de320cd9a1ff9c7281f7f16b15f
> Author: Adam Litke <agl at us.ibm.com>
> Date:   Tue Feb 20 11:44:46 2007 -0800
> 
>     Restore the pgoff check for prepare_hugepage_range()
> 
> diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
> index cbb8c52..f38ab78 100644
> --- a/arch/powerpc/mm/hugetlbpage.c
> +++ b/arch/powerpc/mm/hugetlbpage.c
> @@ -349,6 +349,9 @@ int prepare_hugepage_range(unsigned long addr, unsigned long len, pgoff_t pgoff)
>  
>  	printk("prepare_hugepage_range(addr=0x%lx, len=0x%lx\n", addr, len);
>  
> +	if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT))
> +		return -EINVAL;
> +
>  	/* This is only useful for MAP_FIXED so we turn it into that */
>  	gua_addr = slice_get_unmapped_area(addr, len, MAP_FIXED,
>  					   mmu_huge_psize, 1, 0);
>