Re: [PATCH RFC bpf-next 1/4] bpf: add struct largest member size in func model

["Alexis Lothoré" <alexis.lothore@bootlin.com>]

On Mon Apr 21, 2025 at 4:14 AM CEST, Xu Kuohai wrote:
> On 4/21/2025 12:02 AM, Alexis Lothoré wrote:
>> Hi Xu,
>> 
>> On Thu Apr 17, 2025 at 4:10 PM CEST, Xu Kuohai wrote:
>>> On 4/17/2025 3:14 PM, Alexis Lothoré wrote:
>>>> Hi Andrii,
>>>>
>>>> On Wed Apr 16, 2025 at 11:24 PM CEST, Andrii Nakryiko wrote:
>>>>> On Fri, Apr 11, 2025 at 1:32 PM Alexis Lothoré (eBPF Foundation)
>>>>> <alexis.lothore@bootlin.com> wrote:

[...]

>> Ah, thanks for those clear examples, I completely overlooked this
>> possibility. And now that you mention it, I feel a bit dumb because I now
>> remember that you mentioned this in Puranjay's series...
>> 
>> I took a quick look at the x86 JIT compiler for reference, and saw no code
>> related to this specific case neither. So I searched in the kernel for
>> actual functions taking struct arguments by value AND being declared with some
>> packed or aligned attribute. I only found a handful of those, and none
>> seems to take enough arguments to have the corresponding struct passed on the
>> stack. So rather than supporting this very specific case, I am tempted
>> to just return an error for now during trampoline creation if we detect such
>> structure (and then the JIT compiler can keep using data size to compute
>> alignment, now that it is sure not to receive custom alignments). Or am I
>> missing some actual cases involving those very specific alignments ?
>> 
>
> How can we reliably 'detect' the case? If a function has such a parameter
> but we fail to detect it, the BPF trampoline will pass an incorrect value
> to the function, which is also unacceptable.

That's a question I still have to answer :) I imagined being able to detect
it thanks to some info somewhere in BTF, but I have to dig further to find
how.


Alexis

-- 
Alexis Lothoré, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com