Daira-Emma Hopwood
Daira-Emma Hopwood
Maybe a good heuristic here is that a cell that is only constrained by a single polynomial constraint is suspicious, and should have to be manually annotated as such?
Quote-replying just to preserve the original comment, which does not describe the current PR contents. > # Test and bench for eval_polynomial and compute_inner_product > Found more efficient solution and...
Please force-push this PR so that it only touches the test and benchmarking code.
I would rather not depend on `logzero`, `requests`, or `tqdm`. `fetch-params` has to Just Work, and the easiest way to reduce the set of things that can go wrong is...
Yes that would help. Also, which version of zcashd were you running? And can you reproduce the segfault reliably?
If you can reproduce it, then one option would be to have a video call (e.g. Google Meet) and debug it that way.
#4816 would make it more elegant to pass structured error information from Rust to C++.
Thanks for working on this. However, something I don't like about this approach is that you can't see directly in the code that tests a configuration option, what that option...
Upstream recently added, in https://github.com/bitcoin/bitcoin/pull/25218, a `BResult` type similar to what this issue asks for. Unfortunately, it fixes the result type to Bitcoin Core's `bilingual_str` (initially added in https://github.com/bitcoin/bitcoin/pull/16362), which...
I'm going cold on using `BResult` at all because of this (sample code from their PR): ``` BResult doSomething(arg1, arg2, arg3, arg4) { // do something... if (error) return "something...