mpl
mpl copied to clipboard
MPLLang
The MaPLe compiler for efficient and scalable parallel functional programming
Preliminary implementation, not optimized yet. Just a simple hash-set, with linear probing and tombstones for deletion. Allocation is backed by the block allocator, to avoid space+time overhead of `malloc`. At...
[Skyline benchmark](https://github.com/MPLLang/parallel-ml-bench/blob/main/mpl/bench/skyline/Skyline.sml) from [mpllang/parallel-ml-bench](https://github.com/MPLLang/parallel-ml-bench). The bug sometimes causes a segfault, although I've also seen it hang. It also appears to occur only on small core counts. To reproduce: ``` $...
Using default Makefile, settings on x64 Linux. ``` /usr/bin/ld: /tmp/filei7cguX.o: relocation R_X86_64_32S against `.data' can not be used when making a PIE object; recompile with -fPIC /usr/bin/ld: /tmp/fileMpJHu9.o: relocation R_X86_64_32S...
**Background**: To do concurrent collection, we need an efficient way of checking whether or not a chunk is in scope of the collection. Currently this is done by, at the...
The `smith-normal-form` regression test is causing a segfault. For example, on commit dfddbd5e4111c0df8a5a37405172cc5fd717459a: ``` [mpl/regression]$ ../build/bin/mpl smith-normal-form.sml [mpl/regression]$ ./smith-normal-form ... Segmentation fault (core dumped) ``` I looked into it a...
The terminology in the hierarchical heaps code needs some updates. Here are some terms that I think should change, and what I would change them to. I will add more...
MPL currently avoids GC at the "top level" (i.e., outside any calls to `par`) because these collections typically interfere with parallelism. So far, for parallel programs, this hasn't been too...
Something I've been considering is making a MPL-specific structure with, for example: - `MPL.Parallel` with fork/join primitives - `MPL.Concurrency` with atomics, synchronization prims, etc. - `MPL.Unsafe` with e.g. array allocation...
We need to implement better memory management statistics. Much of the [MLton GC library](http://www.mlton.org/MLtonGC) is MLton-specific and no longer relevant to MPL. At the moment to measure GC, I've been...
MLton.Parallel is currently a "bag of stuff". It's worth organizing this. I'll go section by section ### Reorganization I propose the following: * `MLton.Parallel`: contains `numberOfProcessors`, `processorNumber`, `registerProcessorFunction`, and `initializeProcessors`...
