< yevaud>
jeremyrubin: the absolute worst case is probably one where an attacking miner attempts to balance two competing forks. it is free for them to swap work between two nodes on competing forks, but incredibly expensive to re-wind and sync forward repeatedly.
< yevaud>
jeremyrubin: though this is a lot less likely with per-output UTXOs, it's harder to craft a situation where you have two competing forks with completely dissimilar, distinct transactions.
< bitcoin-git>
[bitcoin] fanquake opened pull request #18125: doc: remove PPA note from release-process.md (master...dont_bug_the_bluematt) https://github.com/bitcoin/bitcoin/pull/18125
< jeremyrubin>
aj: why not?
< aj>
jeremyrubin: because you do .find() on it which is linear for a vec but log for a set?
< jeremyrubin>
yevaud: I don't think what you're saying has anything to do with what we're talking about?
< jeremyrubin>
and then we add the entire cache line
< aj>
oh, i missed a level >:(
< jeremyrubin>
get on my level!
< fanquake>
I was hoping we might be able to take advantage of Microsoft's "new" version checking APIs, for enabling various process mitigations / security features.
< fanquake>
However, it turns out they've written functions, such as IsWindows10OrGreater(), that will return false, even if the OS is Windows 10..., unless you bundle an additional XML file with your executable.
< bitcoin-git>
[bitcoin] Empact opened pull request #18130: Replace uses of boost::trim* with non-locale-dependent alternatives (master...2020-02-boost-trim) https://github.com/bitcoin/bitcoin/pull/18130
< bitcoin-git>
[bitcoin] jonasschnelli opened pull request #18131: add macOS notarization info to the release process doc (master...2020/02/notarize) https://github.com/bitcoin/bitcoin/pull/18131
< bitcoin-git>
[bitcoin] practicalswift closed pull request #17760: util: Don't depend on locale when trimming strings. Add tests. (master...trimming-without-borders) https://github.com/bitcoin/bitcoin/pull/17760