< gmaxwell>
Maybe we should consider now making GBT return an error if you don't send the segwit support flag (perhaps with a config option to bypass that yesiknowmyminingsoftwareisabsoluteshit=1)
< jim2524>
z ma
< jim2524>
在吗
< jim2524>
在吗
< jim2524>
在吗
< provoostenator>
gmaxwell: any particular block that triggered this idea?
< gmaxwell>
23:29:03 < mn3monic> hello, I'm a testnet miner with a 0.16.3, someone looking at my blocks pointed out that I'm not mining segwit transactions, so I
< gmaxwell>
investigated and I've seen that actually my getblocktemplate doesn't include all the available transactions I see in my
< gmaxwell>
getrawmempool, and all the missings are segwit transactions.
< meshcollider>
do runtime errors get printed to the log file too
< meshcollider>
or just shown to the user
< booyah>
provoostenator: damn
< provoostenator>
Assuming it's unsafe to switch the default behavior then indeed throwing an error could make sense. We could add an argument where a miner can opt-out of any softfork.
< booyah>
would be funny if it turned out real SW adoption was 80% but 30% of miners didn't care to change from doc :D
< booyah>
gmaxwell: better call it yes_I_want_to_earn_less_money_by_not_mining_segwit=1
< booyah>
in RPC too. It's half joke, but actually I can imagine a miner owner and some tech guy. Tech guy sets "segwit=false" and I do not care at all. He sets "earn_less_money" and Im interested wtf
< karelb>
maybe a stupid question.... when building bitcoind, I keep seeing something called `libbitcoin_common` etc.... but that is unrelated to this, right - https://github.com/libbitcoin/libbitcoin
< bitcoin-git>
[bitcoin] Sjors opened pull request #14472: [doc] getblocktemplate: use SegWit in example (master...2018/10/doc-getblocktemplate-segwit) https://github.com/bitcoin/bitcoin/pull/14472
< echeveria>
karelb: totally unrelated.
< karelb>
I hoped so
< karelb>
:)
< karelb>
thx
< jonasschnelli>
sipa: you mean #14424? Yes. Will test
< firelegend>
In the context of ecc, what is a base point G?
< gmaxwell>
firelegend: an arbitrarily selected element of the group.
< firelegend>
The group being numbers between 0 and p?
< gmaxwell>
No. The elliptic curve group is the set of points that satisify the curve equation.
< firelegend>
Thanks, I'll keep on reading then.
< bitcoin-git>
[bitcoin] alecalve opened pull request #14474: bitcoin-tx: Use constant for n pubkeys check (master...bitcoin_tx_use_constant) https://github.com/bitcoin/bitcoin/pull/14474
< sipa>
karelb: yes, unrelated to libbitcoin; those are just the names for our internal modules
< gmaxwell>
booyah: I don't think I've seen any evidence of mainnet miners mining that way, but it's an easy mistake to make. letting it mine without segwit was a compatiblity move that has mostly served its purpose.
< henry_>
123
< fanquake>
wumpus 14383 and #14403 should be mergeable, unless you wanted another ACK on 14403. In which case, could someone read through the discussion and ACK/NACK/comment etc
< Varunram>
I don't know if this is a bug or expected behaviour but my regtest node doesn't seem to adjust difficulty every 2016 blocks. Two questions - a. Is this expected and b. Is there a way to force an adjustment
< harding>
Varunram: I think it's expected: test/functional/test_framework/blocktools.py: block.nBits = 0x207fffff # difficulty retargeting is disabled in REGTEST chainparams
< Varunram>
thanks harding!
< harding>
Varunram: looks like you could probably change it here (and recompile): src/chainparams.cpp: consensus.fPowNoRetargeting = true;
< Varunram>
ah, I need to test some reorg code on testnet, hence my question, this would help a lot, thanks!
< bitcoin-git>
[bitcoin] practicalswift opened pull request #14475: serialize: Document integer width assumptions we are making when calculating compact sizes (master...integer-width-assumptions) https://github.com/bitcoin/bitcoin/pull/14475
< phantomcircuit>
Varunram, it doesn't retarget cause that would be annoying in regtest
< sipa>
phantomcircuit: it used to retarget
< Varunram>
I saw a couple old SO posts that told me it did
< phantomcircuit>
sipa, yeah and it was annoying
< Varunram>
phantomcircuit: I guess I found it different simply because _I_ was expecting regtest to behave like mainnet and not anything else
< * Varunram>
annoying as in? difficulty rocketing up after a couple retargets?
< sipa>
Varunram: right, for some tests you needed multiple retarget windows worth of blocks
< sipa>
which was expensive
< Varunram>
ah, I see, thanks!
< achow101>
Varunram: if you want to test some reorg stuff, you can use invalidblock and reconsiderblock
< Varunram>
achow101: yeah, that's what I'm doing now :)
< ken2812221>
MarcoFalke: It seems that Drahtbot is offline for days. Is it crash?
< bitcoin-git>
[bitcoin] sipa opened pull request #14477: Add ability to convert solvability info to descriptor (master...201810_inferdescript) https://github.com/bitcoin/bitcoin/pull/14477