Op 04-04-2021 om 15:43 schreef Florian Klämpfl via fpc-pascal:
Am 04.04.2021 um 15:36 schrieb Joost van der Sluis via fpc-pascal
<fpc-pascal@lists.freepascal.org>:
Op 04-04-2021 om 13:33 schreef Florian Klämpfl via fpc-pascal:
Am 04.04.2021 um 12:50 schrieb Joost van der Sluis via fpc-pascal
<fpc-pascal@lists.freepascal.org>:
Isn't it at least a good practice to store self at Y. So we have Z free for
other calculations and can access members directly using ldd (),y+().
But maybe that's difficult?
Using Y might be indeed difficult as the compiler knows only after register
allocation that it does not need Y for other purposes. It would basically
require the ability to redo code generation.
In my head I've been thinking a lot about another register-allocator:
During the code-generation the code-generation only asks the register-allocator
'I need a register now that cas capabilities X,Y and Z). Or: give me the same
register as I used last time.
And then, afterwards, once code has been generated for the whole 'block', the
register-allocator fills in the registers. And store/restores them when needed.
This can be done using an algorithm that uses a tree to 'peel-down' (is this
English?) all the solutions. Just like is done with a regular-expression parser.
Just dreaming. I don't have the time to work on it, and I don't even know how
it works at the moment. But that would seem to be the ideal solution to me.
But this is how it is basically currently done?
But why do you need to redo the code generation? At the moment the real
registers are assigned, you do know if you need the y register for some
specific task, no?
Regards,
Joost.
_______________________________________________
fpc-pascal maillist - fpc-pascal@lists.freepascal.org
https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-pascal
