Pieter Wuille comments

Results 803 comments of


                                            Pieter Wuille

[IBD] coins: reduce lookups in dbcache layer propagation

Why is this Draft?

doc: improvements to doc/descriptors.md

@Sjors Done.

Optimized SFL cluster linearization

I have moved the changes to TxGraph here to a separate PR, #34085, as it's not actually dependent on the optimizations here.

Optimized SFL cluster linearization

Rebased after merge of #32545, but not marking as Ready for review yet, as I'm investigating another data structure layout. If that turns out to be fruitful, that may replace...

refactor: std::span compat fixes

utACK fa494a1d53f3f030fafe7b533d72b2200428a0fd

miniscript: account for all `StringType` variants in `Miniscriptdescriptor::ToString()`

Code review ACK 28a4fcb03c0fb1cd5112eca1eb36dcb13e0b4ff2 Conceptually seems fine too, but I'd want to see @achow101's opinion on that.

Performance regression in `scalar_mul` with Clang and `x86_64` assembly enabled

Can you post benchmarks with GCC on the same machine?

Copy the 's' parameter of _gej_rescale to avoid aliasing issues

I think we should decide whether to allow aliasing, or not, of this argument: * If we allow aliasing, the function should be changed like this PR does, but I...

bench: replace wall-clock timer with per-process CPU timer

If we're going to rework this, I'd suggest using the stabilized quartiles approach from https://cr.yp.to/papers/rsrst-20250727.pdf: * StQ1: the average of all samples between 1st and 3rd octile * StQ2: the...

bench: replace wall-clock timer with per-process CPU timer

Concept NACK on removing any ability to observe variance in timing. The current min/avg/max are far from perfect, but they work fairly well in practice. Improving is welcome, but removing...