Elliot Gorokhovsky

Would [top](https://man7.org/linux/man-pages/man1/top.1.html) or [htop](https://htop.dev/) work for your use-case?

> I have seen higher than 100% cpu usage without specifying multiple threads. @cgbur zstd CLI with default settings uses one thread for (de)compression and another thread (or threads?) for...

Thanks for the report! Could you share the exact code you're using to call `ZSTD_compress`? (You could create a GitHub Gist of that code and share that link with me).

> @terrelln , @embg , anything I need to update on this PR? Thanks. Sorry for the delay, I just need to run benchmarks. Will do it tomorrow. Edit: have...

> This would make this patch much leaner, essentially a 2-lines change, with no distance interaction, while delivering > 90% of the speed gains. A no brainer for merge. Good...

Thanks for the report! I will put up a fix before the next release.

Hi, I am still seeing this issue on 0.17.0: ``` >>> zstandard.decompress(zstandard.compress(b"foo") + zstandard.compress(b"bar")) b'foo' ``` This is inconsistent with the C API, which will decompress all frames. It is...

@Federico-Maresca Please reach out if you have any questions :-)

@kientzle This isn't a bug in zstd. We limit memory allocation in the decompressor to 2^27 bytes by default so that applications don't need to worry about unexpected inputs causing...

@electrum Pinging you to sign the CLA so we can merge :)