Re: New capabilities for plugins

["Alex Bennée" <alex.bennee@linaro.org>]

Florian Hofhammer <florian.hofhammer@epfl.ch> writes:

> Hi Alex,
>
> Sorry for the late reply, I've been out of office and did not check my
> mail.
>
> On 05/08/2025 17:30, Alex Bennée wrote:
>> I think to read the PC we would just need to make sure we properly
>> resolve it - internally QEMU does this for faults with:
>>      tb = tcg_tb_lookup(retaddr);
>>      cpu_restore_state_from_tb(cpu, tb, retaddr);
>> where retaddr is the address of the translated code. We just need to
>> special case PC handling in the read path.
>> *snip* No need - we just need to exit the loop via
>> cpu_loop_exit_restore() and
>> the code will do the right thing. However we probably don't want to
>> trigger that via register write as we would surprise the plugin -
>> especially if there are other hooks still to run. So we would want an
>> explicit helper to do it.
>
> Is this something the QEMU maintainers would be interested in? If yes,
> I'm happy to dig into the codebase and submit some patches for review.
> But this of course depends on whether such a feature is even desirable
> in QEMU (cf. the parallel discussion thread).

I think writing the patches would be a useful exercise anyway. The way
the plugin code is structured should mean you can keep the changes
fairly localised which would reduce the burden of maintaining an
out-of-tree patch if it isn't accepted. This wasn't really possible
pre-plugins as instrumentation was often deep in the frontends which is
actively maintained code with constant changes making re-basing a
nightmare.

>
> Best regards,
> Florian

-- 
Alex Bennée
Virtualisation Tech Lead @ Linaro