Hey /prog/

improve my virtual machine

git clone https://code.google.com/p/gpvm/

Forgive my lack of documentation. This will be remidied shortly. Also, some instructions remain unimplemented, as I haven't found a use for them yet.

Also, note the absolute lack of references (int &a = b). Haskell has created the idea in my head they are not needed, but I can see them being useful and I will almost undoubtedly implement them at some point.

I look forward to sorting through your abuse to see if anyone says anything constructive.

>>79
You'd need a compiler. Right now, the instruction set and VM itself arr going through puberty in a massive way. If you're interested I'd recommended helping with the development over using it right now. Many features don't exist yet and many are changing, so a compiler could easily and quickly become obsolete.

>>81
I'll write a scheme compiler for your virtual machine in scheme.

>>82
Great! I'll need to spend a while on making comprehensive docs

>>82
scheme compiler
It can't even do tail calls or longjmp.

It is now possible to call object file functions with the lof (Load object file), cof (Call objectfile function) and cofr (call object file function that returns).

Currently only possibly on x86 as it requeres some asm.

>>83
take your time and don't worry. I don't mind reading undocumented code.

http://llvm.org/
Why would anyone choose this over LLVM?

>>87
Because llvm is a Compiler Infrastructure

>>40
It can now calculate the first 30 numbers in the fibonacci sequence in 48 seconds.
It's all an elaborate troll.

>>89
That kinda impressed my at first, considering how it would take about 2 minuites beforehand. Also, I was using the recursive method. Using the iterative algorithm is takes about 0.021s.

I included a tiny, Haskell written compiler for a typeless, functionless RPN language in the repo. it can be extended to be an interpreter by adding the line system $ "gpvm " ++ outFile to the end (before return () of course)

>>87
It's easier to study this because it has a much smaller scope. LLVM is certainly excellent for what it does but it's otherwise a massive beast of a codebase to learn from.

>>40
How the fuck did you manage for it take that long? Even Ruby isn't that fucking slow.

New assembler
I apologize for being a novice at Haskell

>>93
no experience with MREs
Python was slower than this when it was new.

>>93
Because he's (I hope) a troll. No way could someone code that badly that it runs out of memory after 28 Fibonacci numbers unless they malloc(100000000) for each integer, even with ``memory leaks''. No way could someone take 48 seconds to calculate the first 30 numbers in the Fibonacci sequence unless they added sleep(1) for each level of recursion. These are 32-bit integers, not bignums.
>>40
Yesterday, it would run for 1:3x then be killed due to excessive memory usage (It leaked, which is now fixed), and would usually make it to about the 28th number in the sequence.
For reference: 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1597, 2584, 4181, 6765, 10946, 17711, 28657, 46368, 75025, 121393, 196418, 317811
What machine are you running this interpreter on? Hopefully it's something like a Commodore 64, but even an interpreter written in C64 BASIC would be faster than this VM.
Also, IHBT.

>>95
go flip another patty you uneducated bitch.

>>96
Your VM is a slow piece of shit. My grandma can calculate Fibs faster than this with pen and paper.

>>97
To be honest that's not fair, your grandmother has a very advanced instruction set and implicit memoization so she doesn't recalculate the numbers.

C>>99

>>99
nice dubs bro