[Tremor] Lomem / No-byte memory usage figures / compilation with LCC?

[Neil Stainton]

Hello,

I'm looking to implement the Tremor decoder on an FPGA platform, but
have a couple of questions I'd like to ask before comitting too much
time to the project.

I have available a 32-bit 'soft-core' processor, which I think is most
similar to a very basic Sparc in architecture, and a quickly hacked
version of the LCC compiler to work with. The first question is, has
anyone tried porting to this type of processor, or is anyone using LCC
in their development?

Secondly, I have a total of 64k of RAM available, which I can freely
divide between code space and data memory at compile time. (This
processor has a Harvard architecture). I took a look at Ripinder Singh's
memory usage figures posted a couple of months ago, and see that I could
probably do with a little more. Has anyone else any data to report? I
plan to implement alloca().

Thanks for any advice or info,

Neil