Dawid Ciężarkiewicz

Yeap. When I change `8` to `2` it suddenly works again. And I think it used to work in the past here as well.

Unfortunately the number of lanes do change the final hash, so I can't just use lower value.

I've added `.configure_threads(1)` and now it works. But it is really not great that it crashes when number of lanes used is higher than number of detected cores(?).

The code in question: https://github.com/dpc/crev/blob/5a956deeddba6ac682c13e20533a0cbb2a7ce97a/crev-lib/src/id.rs#L139 ``` > cat ~/.config/crev/ids/FYlr8YoYGVvDwHQxqEIs89reKKDy-oWisoO0qXXEfHE.yaml --- version: -1 url: "https://github.com/dpc/crev-proofs" public-key: FYlr8YoYGVvDwHQxqEIs89reKKDy-oWisoO0qXXEfHE sealed-secret-key: seal-nonce: pass: version: 19 variant: argon2id iterations: 192 memory-size: 4096 lanes: 8 salt:...

The problem is that about lanes to thread relationship. It seems to be about threads to actual cpu cores. Setting lanes just makes number of threads go up too. I've...

`Job 3, 'gdb /home/dpc/lab/crev/target/d…' has stopped` :/

`SIGSEGV` never happens without this new test.

Even when I only execute this one new test it sometimes segfaults. ``` Finished release [optimized] target(s) in 0.26s Running target/release/deps/cargo_crev-211cbdc418c06591 running 0 tests test result: ok. 0 passed; 0...

Oh. Actually in `--release` when I run in gdb it sometimes prints `L3` and even `L4`.

Damn, rust made my segfault debugging skills so rusty... ``` warning: core file may not match specified executable file. [New LWP 19258] [New LWP 18988] [New LWP 18987] [Thread debugging...