Shubhamkar Ayare

I myself am unopinionated and have no preference for the behavior. But in the absence of the test of time (= wait for 5-10 years), who gets to decide (and...

Thanks for getting back! Nope, `--no-timestamps` does not help, it produces the same output. I tried out [whisperX](https://github.com/m-bain/whisperX) today. For this particular audio, it worked amazingly well! Interestingly, merely using...

I think `make-hash-set` should take an optional or keyword arg (defaulted to `equal`) specifying the test function. (This should also require an additional helper function `hash-set-test` analogous to `hash-table-test`.)

The C world would be happy with untyped SIMD packs, but it wouldn't fit well with lisp. Will we generate a distinct `sap-ref-X.Y` for every unboxable X and pack-fitting Y?...

Even without masking, there are [gather/scatter instructions](https://en.m.wikipedia.org/wiki/Gather/scatter_(vector_addressing)) that allow loading from / storing to offsets from a base address. I'm not sure about the speedup numbers though.

> The X.Y would be for each element type abbreviation X (f32, f64, u8, ..., s64), and for each corresponding pack width Y. Ah, wait, so was the plan to...

Oh, okay! I think I misunderstood somethings above. Yup, okay, seems like the issue can be resolved by (i) having SB-SIMD functions that work on foreign pointers, and (ii) providing...

A more general idea on the wishlist is for SIMD support in CFFI itself.

Yes, one solution would indeed be to set up a `"my-system/config"` that loads only the configuration systems of other systems and sets them up. The main `"my-system"` will contain the...

So, a few solutions come to mind - #### 1. Load `py4cl2-cffi/config` and set the config variables as soon as you can after the lisp starts This way, whenever py4cl2-cffi...