< bitcoin-git> [bitcoin] JeremyRubin closed pull request #17268: Epoch Mempool (master...mempool-experiments-2) https://github.com/bitcoin/bitcoin/pull/17268
< fanquake> Wrote notes up for anyone that would like to test some of the changes in #17916
< gribble> https://github.com/bitcoin/bitcoin/issues/17916 | windows: Enable heap terminate-on-corruption by fanquake . Pull Request #17916 . bitcoin/bitcoin . GitHub
< bitcoin-git> [bitcoin] fanquake pushed 3 commits to master: https://github.com/bitcoin/bitcoin/compare/ceb789cf3a90...af05bd9e1e36
< bitcoin-git> bitcoin/master 6cbe620 MarcoFalke: scripted-diff: Replace CCriticalSection with RecursiveMutex
< bitcoin-git> bitcoin/master e09c701 MarcoFalke: scripted-diff: Bump copyright of files changed in 2020
< bitcoin-git> bitcoin/master af05bd9 fanquake: Merge #17891: scripted-diff: Replace CCriticalSection with RecursiveMutex
< bitcoin-git> [bitcoin] fanquake merged pull request #17891: scripted-diff: Replace CCriticalSection with RecursiveMutex (master...2001-c2-CC) https://github.com/bitcoin/bitcoin/pull/17891
< bitcoin-git> [bitcoin] fanquake pushed 3 commits to master: https://github.com/bitcoin/bitcoin/compare/af05bd9e1e36...002f9e9b4041
< bitcoin-git> bitcoin/master 7f5db16 Luke Dashjr: GUI: Use PACKAGE_NAME in modal overlay
< bitcoin-git> bitcoin/master 5855cc5 Luke Dashjr: bitcoin-wallet: Use PACKAGE_NAME in usage help
< bitcoin-git> bitcoin/master 002f9e9 fanquake: Merge #17923: refactor: Use PACKAGE_NAME in GUI modal overlay and bitcoin-...
< bitcoin-git> [bitcoin] fanquake merged pull request #17923: refactor: Use PACKAGE_NAME in GUI modal overlay and bitcoin-wallet (master...fix_pkgname_202001) https://github.com/bitcoin/bitcoin/pull/17923
< fanquake> mehscollider: can you take a look at #17843
< gribble> https://github.com/bitcoin/bitcoin/issues/17843 | wallet: Reset reused transactions cache by fjahr . Pull Request #17843 . bitcoin/bitcoin . GitHub
< meshcollider> fanquake: Yep sure :)
< fanquake> meshcollider: cheers ?
< bitcoin-git> [bitcoin] meshcollider pushed 2 commits to master: https://github.com/bitcoin/bitcoin/compare/002f9e9b4041...ac61ec9da679
< bitcoin-git> bitcoin/master 6fc554f Fabian Jahr: wallet: Reset reused transactions cache
< bitcoin-git> bitcoin/master ac61ec9 Samuel Dobson: Merge #17843: wallet: Reset reused transactions cache
< bitcoin-git> [bitcoin] meshcollider merged pull request #17843: wallet: Reset reused transactions cache (master...getbalances) https://github.com/bitcoin/bitcoin/pull/17843
< fanquake> meshcollider: can you confirm #17603 is closed by just that PR, and isn't also waiting for the other one mentioned in the PR body? Or diregard if you are looking at the other one now.
< gribble> https://github.com/bitcoin/bitcoin/issues/17603 | partial spend avoidance makes partial spends and getbalances doesnt notice . Issue #17603 . bitcoin/bitcoin . GitHub
< fanquake> *disregard
< meshcollider> I'll reopen the issue, I think it should be closed after the second one
< fanquake> Thanks
< wumpus> "make check" seems to hang indefinitely here on "bench/bench_bitcoin -evals=1 -scaling=0 > /dev/null"
< fanquake> wumpus: using latest master?
< wumpus> (doesn't seem to be actually benchmarking, CPU use of the bench_bitcoin process is 1% or so...)
< wumpus> yes
< wumpus> might be a fluke, dunno, maybe I can attach a debugger
< fanquake> Taking a look. Haven't seen any issues today
< wumpus> oh it passed; that definitely took long
< wumpus> (debugging is great, by the time I figured out how to enable ptrace on linux again it had already passed :-) )
< fanquake> heh
< wumpus> still, it's expected to pass virtually instantly isn't it, with the evals=1 scaling=0
< wumpus> (still not entirely sure why "test" runs a benchmark)
< fanquake> yes, here it takes ~3.8s
< wumpus> sending the results to /dev/null, at that :-)
< wumpus> running it manually also runs very slow, might be a hw or local issue? trying to run it on another machine
< wumpus> real 1m44.629s user 0m3.896s sys 0m0.853s
< wumpus> quite different from 3.8s :)
< fanquake> indeed: real0m3.910s user0m2.194s sys0m0.817s
< wumpus> it spends its time sleeping or waiting for i/o
< wumpus> done some shotgun profiling: fdatasync is the culprit, everyt ime
< wumpus> in CBanDB::Write ?!?
< wumpus> why is the BanMan dumping the ban list between every benchmark
< promag> fanquake: I've included instagibbs fix in the backport
< fanquake> promag thanks, can you update the pr body
< promag> will do after ci
< promag> it that doesn't go well then we could split the backport and push the non controversial to 0.19
< promag> *if
< wumpus> removing the m_node.banman instantiation it runs in 4 seconds
< promag> quite a difference indeed
< wumpus> can't reproduce it on any other machine, so seems a local issue, still strange, I've noticed slowness with fdatasync in VMs before but not on bare hw
< bitcoin-git> [bitcoin] Bushstar opened pull request #17928: depends: Consistent use of package variable (master...patch-4) https://github.com/bitcoin/bitcoin/pull/17928
< bitcoin-git> [bitcoin] kallewoof closed pull request #17838: test: test the >10 UTXO case for output groups (master...191231-test-avoidreuse-many-utxos) https://github.com/bitcoin/bitcoin/pull/17838
< wumpus> going to test #16702
< gribble> https://github.com/bitcoin/bitcoin/issues/16702 | p2p: supplying and using asmap to improve IP bucketing in addrman by naumenkogs . Pull Request #16702 . bitcoin/bitcoin . GitHub
< elichai2> which python version do we require in the functional test?
< fanquake> 3.5
< fanquake> See the .python-version file
< elichai2> hmm `p2p_invalid_messages.py` fails for me in 3.8 because of a warning it prints
< elichai2> oh wait
< elichai2> it requires 3.5
< elichai2> i'll check and PR
< fanquake> We support >= 3.5
< elichai2> (I thought the "fix" requires python 3.7)
< bitcoin-git> [bitcoin] fanquake opened pull request #17929: build: pass optimization flags to the linker (master...pass_optimizations_to_linker) https://github.com/bitcoin/bitcoin/pull/17929
< promag> fanquake: updated pr body
< fanquake> promag thanks. Can you also add #17634
< gribble> https://github.com/bitcoin/bitcoin/issues/17634 | qt: Fix comparison function signature by hebasto . Pull Request #17634 . bitcoin/bitcoin . GitHub
< fanquake> I think it makes sense to backport that, given it fixes the qt build on our minimum supported centos.
< fanquake> Should at least be able to compile if we claim to support it.
< promag> fanquake: done
< fanquake> thanks
< promag> #15443
< gribble> https://github.com/bitcoin/bitcoin/issues/15443 | qa: Add getdescriptorinfo functional test by promag . Pull Request #15443 . bitcoin/bitcoin . GitHub
< promag> meshcollider: please look this one ^
< wumpus> python 3.8 definitely supposed to be supported, if it doesn't that's a bug
< promag> jamesob: will blockman have it's own mutex?
< promag> going to push the worst pr of 2020, sorry
< fanquake> "delete whitespace"
< fanquake> At least there's another 11 months for someone to steal the title
< bitcoin-git> [bitcoin] promag opened pull request #17930: rpc: Don't lock cs_main in getblock if verbosity=0 (master...2020-01-getblock-avoid-lock) https://github.com/bitcoin/bitcoin/pull/17930
< promag> fanquake: lol
< promag> I said it's stupid :| it's wrong
< bitcoin-git> [bitcoin] promag closed pull request #17930: rpc: Don't lock cs_main in getblock if verbosity=0 (master...2020-01-getblock-avoid-lock) https://github.com/bitcoin/bitcoin/pull/17930
< promag> I don't get it, does ReadBlockFromDisk requires the lock or not?
< bitcoin-git> [bitcoin] elichai opened pull request #17931: test: Fix p2p_invalid_messages failing in Python 3.8 because of warning (master...2020-01-coroutine) https://github.com/bitcoin/bitcoin/pull/17931
< jamesob> promag: it certainly could, though because I don't think there'd be a huge performance impact I was going to wait until after the assumeutxo functionality all gets merged to start worrying about splitting the locks up - which I do think is probably worth doing
< jonatack> PR #16115 has 4 acks, might be RFM
< gribble> https://github.com/bitcoin/bitcoin/issues/16115 | On bitcoind startup, write config args to debug.log by LarryRuane . Pull Request #16115 . bitcoin/bitcoin . GitHub
< bitcoin-git> [bitcoin] btc-coder opened pull request #17932: replace deprecated method of recursive_directory_iterator with the st... (master...master) https://github.com/bitcoin/bitcoin/pull/17932
< bitcoin-git> [bitcoin] dongcarl opened pull request #17933: guix: Pin Guix using `guix time-machine` (master...2020-01-guix-time-machine) https://github.com/bitcoin/bitcoin/pull/17933
< hebasto> promag: fanquake: ty
< bitcoin-git> [bitcoin] hebasto opened pull request #17934: doc: Use CONFIG_SITE variable instead of --prefix option (master...20200115-prefix-depends) https://github.com/bitcoin/bitcoin/pull/17934
< bitcoin-git> [bitcoin] brakmic opened pull request #17935: gui: hide HD icon when no wallet loaded (master...hdwallet-icon-handling) https://github.com/bitcoin/bitcoin/pull/17935
< bitcoin-git> [bitcoin] fanquake closed pull request #17932: replace deprecated method of recursive_directory_iterator with the st... (master...master) https://github.com/bitcoin/bitcoin/pull/17932
< luke-jr> fwiw, I found a new getbalance bug on v0.19.0.1; debugging now
< bitcoin-git> [bitcoin] promag opened pull request #17937: gui: Remove WalletView and BitcoinGUI circular dependency (master...2020-01-remove-walletview-bitcoingui) https://github.com/bitcoin/bitcoin/pull/17937