[Pull Request] [Natty] module RO/NX take 2, with fixes

Tim Gardner tim.gardner at canonical.com
Thu Dec 9 19:45:03 UTC 2010


On 12/09/2010 12:03 PM, Kees Cook wrote:
> Hi Tim,
>
> On Thu, Dec 09, 2010 at 08:08:31AM -0700, Tim Gardner wrote:
>>> Kees Cook (4):
>>>        Revert "Revert "x86: Add NX protection for kernel data""
>>>        Revert "Revert "x86: Add RO/NX protection for loadable kernel modules""
>>>        Revert "Revert "UBUNTU: [Config] update config for CONFIG_DEBUG_SET_MODULE_RONX""
>>>        x86: RO/NX protection for loadable kernel, jump_table fix
>>>
>>> Lin Ming (1):
>>>        x86: Resume trampoline must be executable
>>>
>>>   arch/x86/Kconfig.debug                    |   11 ++
>>>   arch/x86/include/asm/jump_label.h         |    2 +-
>>>   arch/x86/include/asm/pci.h                |    1 +
>>>   arch/x86/kernel/ftrace.c                  |    3 +
>>>   arch/x86/kernel/vmlinux.lds.S             |    8 +-
>>>   arch/x86/mm/init.c                        |    3 +-
>>>   arch/x86/mm/init_32.c                     |   20 +++-
>>>   arch/x86/mm/pageattr.c                    |    5 +-
>>>   arch/x86/pci/pcbios.c                     |   23 ++++
>>>   debian.master/config/config.common.ubuntu |    1 +
>>>   debian.master/config/enforce              |    1 +
>>>   include/linux/module.h                    |   11 ++-
>>>   kernel/module.c                           |  171 ++++++++++++++++++++++++++++-
>>>   13 files changed, 251 insertions(+), 9 deletions(-)
>>>
>>
>> Applied, though I'm curious how you've constructed your tree. The
>> first two reverts had already been applied in master by
>> Ubuntu-2.6.37-6.17. Anyways, pushed to master-next.
>
> I was reverting the reverts. There are 7 patches in total for the feature.
> 4 original, 2 fixes, and 1 config update. 2 of the original were reverted,
> and the 1 config was reverted. I reapplied those, and then added the 2
> fixes.
>
> I figure it'll all make sense on the next rebase.
>
> -Kees
>

I'm only mildly dyslexic :) Andy seems to have figured it out and pushed 
correctly.

-- 
Tim Gardner tim.gardner at canonical.com




More information about the kernel-team mailing list