Matthew Fluet

The issue with rightward drift makes sense. Agreed that a heuristic that avoids the awkward splitting and is consistent with both `val` and `fun` would be best. I think that...

I'm note sure which TeX template group volunteers @Craig-Rodkin is referring to, but I've advocated for options to set a fixed-size for and to suppress the copyright space: #8 and...

Seems to have changed at https://github.com/MLton/mlton/commit/a2606d088b6729eed9fcf404b98e1a499d17f806#diff-a2c6d1e6417a222aa27337f1bd49eed465324427c02fdd25f3363fe1c4f0f2c9R516. Should be easy enough to fix, but will need to wait until after the end of the semester.

@nrnrnr Could you explain: > and also, the full syntax appears as if it may be incompatible with SML/NJ's syntax for `#line` directives. I'm not seeing any meaningful difference that...

Ok; I was wondering if the "and also" was indicating additional difference with SML/NJ's line directives. But, I presume you simply meant that the full syntax is a subset of...

Could you give more details about how to build and run the program to produce the segfault? It looks like this is more of a data structure library rather than...

Thanks. With that, I was able to intermittently reproduce a segfault (on amd64-linux).

And compiling with `-debug true`, I consistently get: ``` mlvg_vector: gc/invariant.c:13: assertIsObjptrInFromSpaceOrImmutableMutableOrRootStaticHeap: Assertion `isObjptrInFromSpace (s, *opp) || isObjptrInImmutableMutableOrRootStaticHeap (s, *opp)' failed. ```

@hummy123 #560 and #559 are related in the sense that they are both triggering bugs resulting from aggressive optimization of micro-benchmarks. In particular, both of these bugs stem from MLton...

Confirmed, with both MLton 20210117 and current sources. The bug is actually with the `useless` pass, just that the miscompilation doesn't trigger another ICE until the `splitTypes2` pass. This can...