D GC theory

deadalnix via Digitalmars-d digitalmars-d at puremagic.com
Tue Feb 24 10:40:12 PST 2015


On Tuesday, 24 February 2015 at 15:14:03 UTC, Ola Fosheim Grøstad 
wrote:
> On Tuesday, 24 February 2015 at 14:45:17 UTC, Marc Schütz wrote:
>> It's possible to (ab)use the MMU as a write barrier.
>
> Uhm...
>
> The throughput for L/SFENCE is 5 cycles and SFENCE 33 cycles... 
> The cost of a page miss is 1000 cycles + overhead for copying.
>
> And that's assuming that you have enough memory and that the 
> TLB isn't affected...

That's why this is used for immutable, or mostly immutable data, 
but generally avoided in the case of mutable data.


More information about the Digitalmars-d mailing list