Change Details

Removing the need to allocate here and instead iterate over the tx directly improves performance slightly.

Removing the need to allocate here and instead iterate over the tx directly improves performance slightly. See D15552 for some benchmarks, I haven't run a dedicated benchmark just for this diff, but improvements should be in a similar range.

Removing the need to allocate here and instead iterate over the tx directly improves performance slightly. See D15552 for some benchmarks, I haven't run a dedicated benchmark just for this diff, but improvements should be in a similar range.