Tom Birch
Tom Birch
b2h5py is mostly out of scope for me. To be clear: * I want SHUFFLE + BYTEDELTA blosc2 compression * I want to be able to store the compressed data...
Ok, I tried what I suggested above, but I get an error on decompression because `dparams.schunk = NULL`. If I set `dparams.schunk` to point to the schunk then it decompresses...
Is this at all useful? These guys seem to have implemented support for tensor_core. Seems like they branched off Halide somewhere between 8.0 and 10.0 https://github.com/TUE-EE-ES/HalideTCU https://www.es.ele.tue.nl/~tbasten/papers/Scopes_camera_ready.pdf
In my case I have multiple "root" modules, and I want to have a shared set of third-party modules that are easy to access (e.g. via bazel_dep). I don't see...
+1 I would love to see this in an 8.x release
The issue is the heuristic [here](https://github.com/wagoodman/dive/blob/main/dive/image/docker/image_archive.go#L105) is broken because newer oci layers do gzip small layers, so they are skipped because they don't fill the buffer. Commenting out that check...
I hacked around it, pushed my changes here: https://github.com/froody/dive/tree/crane quite a significant change, basically uses crane as a dest for docker to "push" to, much faster downloads, and now we...
@reneleonhardt ok I'll try
The performance difference vs zstd-1.5.7 is significant: - the C version is >3x faster at any compression level using a single thread - the C version gets a ~linear speedup...
Hmm how would I encode a per-chunk value in vlmeta? I think I would need to either pre-allocate a large enough buffer before compression, and then index into this using...