Re: [PATCH v3 1/4] x86/livepatch: align functions to ensure minimal distance between entry points

[Jan Beulich <jbeulich@xxxxxxxx>]

On 14.12.2023 16:20, Roger Pau Monné wrote:
> On Thu, Dec 14, 2023 at 02:53:13PM +0100, Jan Beulich wrote:
>> On 14.12.2023 14:37, Roger Pau Monné wrote:
>>> On Thu, Dec 14, 2023 at 12:18:13PM +0100, Jan Beulich wrote:
>>>> On 14.12.2023 11:17, Roger Pau Monne wrote:
>>>>> The minimal function size requirements for livepatch are either 5 bytes 
>>>>> (for
>>>>> jmp) or 9 bytes (for endbr + jmp) on x86, and always 4 bytes on Arm.  
>>>>> Ensure
>>>>> that distance between functions entry points is always at least of the 
>>>>> minimal
>>>>> required size for livepatch instruction replacement to be successful.
>>>>>
>>>>> Add an additional align directive to the linker script, in order to 
>>>>> ensure that
>>>>> the next section placed after the .text.* (per-function sections) is also
>>>>> aligned to the required boundary, so that the distance of the last 
>>>>> function
>>>>> entry point with the next symbol is also of minimal size.
>>>>>
>>>>> Note that it's possible for the compiler to end up using a higher function
>>>>> alignment regardless of the passed value, so this change just make sure 
>>>>> that
>>>>> the minimum required for livepatch to work is present.
>>>>
>>>> That's a possibility which we don't need to be concerned about. Yet isn't 
>>>> it
>>>> also possible that we override a larger, deemed better (e.g. 
>>>> performance-wise)
>>>> value?
>>>
>>> I'm kind of confused, the compiler will always choose the higher
>>> alignment.
>>
>> Will it? Before writing the reply I went through gcc's respective doc
>> section, without finding such a guarantee.
> 
> Hm, yes, checked with godbolt now and GCC behaves the opposite of
> clang, and will always attempt to honor the alignment passed in
> falign-functions.
> 
> Maybe for release builds we should select a 16b alignment on x86?

Might make sense, yes. Iirc there was a time where 16-byte alignment was
the default for functions in gcc.

Jan