[Gambas-user] Problems with interpreter optimization and CPU behaviour

[Benoit Minisini]

Le 08/10/2022 à 20:44, Benoit Minisini a écrit :
> Le 08/10/2022 à 20:15, Jussi Lahtinen a écrit :
>> Could be some limit in the speculative execution. Any correlations to 
>> the L1 - L4 cache sizes of your CPU?
>>
>> Jussi
>>
> 
> My Ryzen 5700G L1 cache size is 64Kb by core, according to this link: 
> https://www.techpowerup.com/cpu-specs/ryzen-7-5700g.c2472
> 
> The object file of the interpreter loop (gbx_exec_loop.c) has less than 
> 32Kb of machine code inside according to 'objdump'.
> 

OK, L1 cache is usually divided in two: one half for instructions, 
another half for the data. So the interpreter loop must be less than 32K.

What to put elsewhere? How to reduce the verbosity of the x86 
instructions? Difficult problems...

-- 
Benoît Minisini.