-
LjTools
LuaJIT 2.0 bytecode parser, viewer, assembler and test VM. Lua 5.1 parser, IDE and debugger.
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
-
SaaSHub
SaaSHub - Software Alternatives and Reviews. SaaSHub helps you find the best software and product alternatives
Did you have a look at LuaJIT? See e.g. http://luajit.org/, http://wiki.luajit.org/Bytecode-2.0, http://wiki.luajit.org/SSA-IR-2.0, and https://github.com/MethodicalAcceleratorDesign/MADdocs/blob/master/luajit/luajit-doc.pdf
If you want to re-use LuaJIT as a backend, have e.g. a look at https://github.com/rochus-keller/ljtools
Have a look at these benchmarks of mine from last year, comparing my attempts at optimising code with gcc-O3 (see table about half-way down).
This may be entirely irrelevant to what you are looking for, but a good widely used finite register-based VM is the eBPF VM in the Linux Kernel. The IOVisor uBPF project (https://github.com/iovisor/ubpf) is a version of the VM in user space.
Of course, these streaming compilers are primarily there to improve startup performance. Their output only runs until a more optimizing compiler finishes its job. But they are also not that far behind the heavier optimizers in quality! V8's Liftoff produces code which (as of the above post) tends to be 30%-100% slower than Turbofan's; Lightbeam is closer to 15% slower than native and can even beat Cranelift or SpiderMonkey. (In part this is because distributed Wasm programs are already optimized.)