< luke-jr>
michaelfolkson: 9946 looks completely unrelated; 5872 is still a clean merge
< michaelfolkson>
Hmm appears so. 9946 referred to your 5872 to address Issue 9933
< * michaelfolkson>
shrugs
< bitcoin-git>
[bitcoin] luke-jr opened pull request #20121: configure: Allow users to explicitly enable libsecp256k1's GMP bignum support (master...secp256k1_allow_bignum) https://github.com/bitcoin/bitcoin/pull/20121
< luke-jr>
michaelfolkson: afaict, it's simply because 5872 fixed a (completely different) issue with spaces in paths
< pinheadmz>
aw. 9946... one of my first contirbutions ever :-)
< michaelfolkson>
Still causing problems in 2020 ;)
< luke-jr>
eh?
< michaelfolkson>
Sorry, just a joke (ignore me)
< sipa>
luke-jr: fwiw, after secp256k1 PR 767 we'll probably remove bignum support entirely
< luke-jr>
sipa: why does that PR talk about ARM specifically? it looks generic?
< sipa>
it is
< sipa>
luke-jr: i think ARM benchmarking may appear disproportionally due to the fact that the author doesn'thave any nativd 32-bit hardware and had to ask others for numbers
< luke-jr>
I see
< dburkett>
michaelfolkson: That presentation is exactly what I was looking for. Thanks!
< michaelfolkson>
No problem
< vasild>
ariard: I think it is ok to just relay even if the receiver may end up ditching the address (not further gossip it)
< wumpus>
vasild: congrats on getting #19954 merged! 🎉
< bitcoin-git>
[bitcoin] practicalswift opened pull request #20122: Make Assert(…) usable in all contexts. Make implicit assumptions explicit. (master...Assert) https://github.com/bitcoin/bitcoin/pull/20122
< vasild>
wumpus: thanks, that was a true team effort!
< ja>
luke-jr: when would an ack be necessary for marking 'obsolete'? i thought it would be analogous to 'final' which you seem to wait for an explicit ack for here: https://github.com/bitcoin/bips/pull/926
< luke-jr>
Final doesn't need an ACK, but Draft->Proposed does
< ja>
luke-jr: is there something i can do to advance PR 926?
< luke-jr>
I can't think of any scenario where ACK has meaning for Obsolete
< luke-jr>
ja: ask morcos to ACK it?
< luke-jr>
[again]
< sipa>
this is an unusual way of messaging a channel, but i guess it works
< bitcoin-git>
[bitcoin] promag opened pull request #20125: rpc, wallet: Expose database format in getwalletinfo (master...2020-10-walletinfoformat) https://github.com/bitcoin/bitcoin/pull/20125
< luke-jr>
fanquake: hebasto isn't listed as an author..
< fanquake>
luke-jr? he opened the issue
< * luke-jr>
sometimes wishes we could relax BIP stuff, sigh
< luke-jr>
fanquake: only the author can approve Draft->Proposed status change
< fanquake>
luke-jr: no idea what you're talking about. I was replying to "fanquake: why is 13478 linked from https://github.com/bitcoin/bitcoin/issues/20104 ? it doesn't have recent discussion like claimed"
< achow101>
if I compile against the oldest version of sqlite I can find, and all of the tests pass, is it reasonable to assume that every version between that one and the latest are also compatible?
< achow101>
aka I don't want to test every version of sqlite
< sipa>
what is the oldest version you tried?
< achow101>
3.8.0
< achow101>
only because that was the oldest major release I could download
< sipa>
given how little functionality i assume you're using, i think that's very reasonable
< achow101>
turns out the actual oldest I could download (3.7.16) doesn't work anyways