Re: Fastcode MM Rule

Hello Pierre,
> I just don't need a dword marker between small blocks. I just need a byte
> to allow me to store whether the block is allocated, etc. Assuming 8-byte
> alignment of blocks, I can satisfy a request for 15 bytes out of a
> internal chunk of 16 bytes. If I had a dword boundary marker I would have
> to use a 24-byte chunk to satisfy the same request. Since more than 90% of
> requests in my appications are for less than 64 bytes, it makes a
> significant difference.

Looks interesting.
Just out of curiosity, have you ever verified what part of the total memory
consumption the blocks smaller than 32 bytes are responsible for?
For instance, in the Reservation System replay benchmark, are we talking
about 80%, 20%, 5%...?
In your tests with your new MM, what kind of memory reduction do you
observe?

Robert


.