On Tue, 10 Jun 2008 18:01:07 -0400
Dave Kleikamp <[EMAIL PROTECTED]> wrote:
> mm: Allow architectures to define additional protection bits
>
> This patch allows architectures to define functions to deal with
> additional protections bits for mmap() and mprotect().
>
> arch_calc_vm_prot_bits() maps additonal protection bits to vm_flags
> arch_vm_get_page_prot() maps additional vm_flags to the vma's vm_page_prot
> arch_validate_prot() checks for valid values of the protection bits
>
> Note: vm_get_page_prot() is now pretty ugly. Suggestions?
It didn't get any better, no ;)
I wonder if we can do the ORing after doing the protection_map[]
lookup. I guess that's illogical even if it happens to work.
> Signed-off-by: Dave Kleikamp <[EMAIL PROTECTED]>
> ---
>
> include/linux/mman.h | 23 ++++++++++++++++++++++-
> mm/mmap.c | 5 +++--
> mm/mprotect.c | 2 +-
> 3 files changed, 26 insertions(+), 4 deletions(-)
>
> diff -Nurp linux001/include/linux/mman.h linux002/include/linux/mman.h
> --- linux001/include/linux/mman.h 2008-06-05 10:08:01.000000000 -0500
> +++ linux002/include/linux/mman.h 2008-06-10 16:48:59.000000000 -0500
> @@ -34,6 +34,26 @@ static inline void vm_unacct_memory(long
> }
>
> /*
> + * Allow architectures to handle additional protection bits
> + */
> +
> +#ifndef HAVE_ARCH_PROT_BITS
> +#define arch_calc_vm_prot_bits(prot) 0
> +#define arch_vm_get_page_prot(vm_flags) __pgprot(0)
> +
> +/*
> + * This is called from mprotect(). PROT_GROWSDOWN and PROT_GROWSUP have
> + * already been masked out.
> + *
> + * Returns true if the prot flags are valid
> + */
> +static inline int arch_validate_prot(unsigned long prot)
> +{
> + return (prot & ~(PROT_READ | PROT_WRITE | PROT_EXEC | PROT_SEM)) == 0;
> +}
> +#endif /* HAVE_ARCH_PROT_BITS */
argh, another HAVE_ARCH_foo.
A good (but verbose) way of doing this is to nuke the ifdefs and just
go and define these three things for each architecture. That can be
done via copy-n-paste into include/asm-*/mman.h or #include
<asm-generic/arch-mman.h>(?) within each asm/mman.h.
Another way would be
#ifndef arch_calc_vm_prot_bits
#define arch_calc_vm_prot_bits(prot) ...
> +/*
> * Optimisation macro. It is equivalent to:
> * (x & bit1) ? bit2 : 0
> * but this version is faster.
> @@ -51,7 +71,8 @@ calc_vm_prot_bits(unsigned long prot)
> {
> return _calc_vm_trans(prot, PROT_READ, VM_READ ) |
> _calc_vm_trans(prot, PROT_WRITE, VM_WRITE) |
> - _calc_vm_trans(prot, PROT_EXEC, VM_EXEC );
> + _calc_vm_trans(prot, PROT_EXEC, VM_EXEC) |
> + arch_calc_vm_prot_bits(prot);
> }
>
> /*
> diff -Nurp linux001/mm/mmap.c linux002/mm/mmap.c
> --- linux001/mm/mmap.c 2008-06-05 10:08:03.000000000 -0500
> +++ linux002/mm/mmap.c 2008-06-10 16:48:59.000000000 -0500
> @@ -72,8 +72,9 @@ pgprot_t protection_map[16] = {
>
> pgprot_t vm_get_page_prot(unsigned long vm_flags)
> {
> - return protection_map[vm_flags &
> - (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)];
> + return __pgprot(pgprot_val(protection_map[vm_flags &
> + (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
> + pgprot_val(arch_vm_get_page_prot(vm_flags)));
> }
> EXPORT_SYMBOL(vm_get_page_prot);
>
> diff -Nurp linux001/mm/mprotect.c linux002/mm/mprotect.c
> --- linux001/mm/mprotect.c 2008-06-05 10:08:03.000000000 -0500
> +++ linux002/mm/mprotect.c 2008-06-10 16:48:59.000000000 -0500
> @@ -239,7 +239,7 @@ sys_mprotect(unsigned long start, size_t
> end = start + len;
> if (end <= start)
> return -ENOMEM;
> - if (prot & ~(PROT_READ | PROT_WRITE | PROT_EXEC | PROT_SEM))
> + if (!arch_validate_prot(prot))
> return -EINVAL;
>
> reqprot = prot;
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-dev
