On 11/30/23 05:33, Peter Zijlstra wrote:
> On Wed, Nov 29, 2023 at 03:07:15PM -0600, Madhavan T. Venkataraman wrote:
>
>> Kernel Lockdown
>> ---------------
>>
>> But, we must provide at least some security in V2. Otherwise, it is useless.
>>
>> So, we have implemented what we call a kernel lockdown. At the end of kernel
>> boot, Heki establishes permissions in the extended page table as mentioned
>> before. Also, it adds an immutable attribute for kernel text and kernel RO
>> data.
>> Beyond that point, guest requests that attempt to modify permissions on any
>> of
>> the immutable pages will be denied.
>>
>> This means that features like FTrace and KProbes will not work on kernel text
>> in V2. This is a temporary limitation. Once authentication is in place, the
>> limitation will go away.
>
> So either you're saying your patch 17 / text_poke is broken (so why
> include it ?!?) or your statement above is incorrect. Pick one.
>
It has been included so that people can be aware of the changes.
I will remove the text_poke() changes from the patchset and send it later when
I have some authentication in place. It will make sense then.
>
>> __text_poke()
>> This function is called by various features to patch text.
>> This calls heki_text_poke_start() and heki_text_poke_end().
>>
>> heki_text_poke_start() is called to add write permissions to the
>> extended page table so that text can be patched. heki_text_poke_end()
>> is called to revert write permissions in the extended page table.
>
> This, if text_poke works, then static_call / jump_label / ftrace and
> everything else should work, they all rely on this.
>
>
>> Peter mentioned the following:
>>
>> "if you want to mirror the native PTEs why don't you hook into the
>> paravirt page-table muck and get all that for free?"
>>
>> We did consider using a shadow page table kind of approach so that guest
>> page table
>> modifications can be intercepted and reflected in the page table entry. We
>> did not
>> do this for two reasons:
>>
>> - there are bits in the page table entry that are not permission bits. We
>> would like
>> the guest kernel to be able to modify them directly.
>
> This statement makes no sense.
>
>> - we cannot tell a genuine request from an attack.
>
> Why not? How is an explicit call different from an explicit call in a
> paravirt hook?
>
>>From a maintenance pov we already hate paravirt with a passion, but it
> is ever so much better than sprinkling yet another pile of crap
> (heki_*) around.
I only said that the idea was considered.
We can resume the discussion on this topic when I send the text_poke() changes
in a later
version of the Heki patchset.
Madhavan
