vasild has quit [Remote host closed the connection]
vasild has joined #bitcoin-core-dev
PaperSword has quit [Quit: PaperSword]
<bitcoin-git>
[bitcoin] furszy opened pull request #29301: init: settings, do not load auto-generated warning msg (master...2024_settings_dont_load_warning_msg) https://github.com/bitcoin/bitcoin/pull/29301
<sdaftuar>
glozow: i spent some time trying to rework the testmempoolaccept() behavior, so that we would first cluster the transactions into connected components, and then iteratively invoke either AcceptMultipleTransactions or AcceptPackage on each cluster.
<sdaftuar>
I discovered that this is thornier than I thought! The code to do the clustering is simple, but after looking at tests that fail I realize that the behavior we're looking for is possibly not what I initially thought
AaronvanW has joined #bitcoin-core-dev
<sdaftuar>
it seems to me that the use case we're trying to meet with testmempoolaccept(tx1,..., txN) is to answer the question of whether the mempool would accept all of those transactions if we submitted them (one by one?)
<sdaftuar>
and by grouping them into separate clusters for processing, we answer a different question, of whether each connected component of (tx1, ..., txN) would be accepted
<sdaftuar>
and those are really answering different questions...
<sdaftuar>
interestingly, i think this question may get easier to answer in a cluster mempool world (it's very similar to what we have to do when evaluating an RBF, I think), so now I'm reluctant to suggest "breaking" the behavior we have now (even though i think it would make the rest of the v3 implementation simpler)
AaronvanW has quit [Ping timeout: 264 seconds]
Guest84 has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] marcofleon opened pull request #29302: wallet: clarify replaced_by_txid and replaces_txid in help output (master...2024/01/clarify-documentation) https://github.com/bitcoin/bitcoin/pull/29302
Guest84 has quit [Quit: Client closed]
bitdex has joined #bitcoin-core-dev
bitdex has quit [Quit: = ""]
metta has quit []
upekkha has joined #bitcoin-core-dev
pablomartin4btc has quit [Remote host closed the connection]
AaronvanW has joined #bitcoin-core-dev
AaronvanW has quit [Ping timeout: 264 seconds]
cmirror has quit [Remote host closed the connection]
cmirror has joined #bitcoin-core-dev
jarthur has quit [Quit: jarthur]
zato has joined #bitcoin-core-dev
AaronvanW has joined #bitcoin-core-dev
Calabrese has quit [Quit: Client closed]
vysn has joined #bitcoin-core-dev
AaronvanW has quit [Ping timeout: 276 seconds]
stmbln has joined #bitcoin-core-dev
AaronvanW has joined #bitcoin-core-dev
stmbln has left #bitcoin-core-dev [#bitcoin-core-dev]
Guyver2 has joined #bitcoin-core-dev
<glozow>
sdaftuar: ah makes sense. I imagine the best way to group the transactions should involve a linearization (e.g. submitting all of them would bust a package limit so put the higher feerate ones first)?
<glozow>
re: v3 complexity, I don't know how much we'd be able to chop off if we didn't have to deal with multiple connected components. A lot of it is contained within `PackageV3Checks`. We'd still need to be able to see an in-package ancestor's mempool ancestor, so we need to cache it (that's the ugliest part imo).
<bitcoin-git>
[bitcoin] dergoegge opened pull request #29304: fuzz: Exit and log stdout for parse_test_list errors (master...2024-01-fuzz-list-errs) https://github.com/bitcoin/bitcoin/pull/29304
<instagibbs>
Murch[m] this is the real mystery for me: https://github.com/bitcoin/bitcoin/pull/29264#discussion_r1463935963 adding the error string causes other coin selection routines to fail in rpc_psbt.py. something about how coin selection handles errors with and without error strings
<instagibbs>
would appreciate a coin selection sleuth to tell me what to do
qxs has quit [Ping timeout: 240 seconds]
qxs has joined #bitcoin-core-dev
puchka has quit [Ping timeout: 256 seconds]
szkl has quit [Quit: Connection closed for inactivity]
jonatack has joined #bitcoin-core-dev
SpellChecker has quit [Ping timeout: 240 seconds]
jon_atack has quit [Ping timeout: 255 seconds]
ghost43_ has joined #bitcoin-core-dev
ghost43 has quit [Quit: Leaving]
ghost43_ has quit [Remote host closed the connection]
ghost43 has joined #bitcoin-core-dev
puchka has joined #bitcoin-core-dev
PaperSword has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] dergoegge opened pull request #29305: fuzz: Use FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION for pow checks (master...2024-01-fuzz-unsafe-pow) https://github.com/bitcoin/bitcoin/pull/29305
SpellChecker has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] glozow opened pull request #29306: policy: enable sibling eviction for v3 transactions (master...2024-01-sibling-eviction) https://github.com/bitcoin/bitcoin/pull/29306
qxs has quit [Remote host closed the connection]
qxs has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] vasild opened pull request #29307: util: check for errors after close and read in AutoFile (master...AutoFile_error_check) https://github.com/bitcoin/bitcoin/pull/29307
zato has quit [Ping timeout: 256 seconds]
zato has joined #bitcoin-core-dev
jonatack has quit [Ping timeout: 256 seconds]
<sdaftuar>
glozow: PackageV3Checks was the thing that I found to be pretty complex :) Reviewing it now though with a better understanding of what the use case is for testmempoolaccept
<sdaftuar>
one question occurs to me as I read the code: if A is a v3 transaction, with 2 child v3 transactions (B, C), and A is in the mempool, and we call testmempoolaccept(B, C), it should fail right? I'm struggling to see what check would catch that.
<instagibbs>
each individually would work.... hard to say heh
<instagibbs>
testmempoolaccept stuff is a bit of a philosophical rabbit hole imo
<sdaftuar>
right now i'm feeling pretty far into the anti-testmempoolaccept camp :)
<instagibbs>
:)
<instagibbs>
I had an alternative idea that I think I linked to many pages of comments back, glozow in case you recall?
<instagibbs>
as much as I enjoy 100 pages of LN spec discussion...
<instagibbs>
does that test pass or fail currently :)
<sdaftuar>
if you run the test, it will say all tests pass, but the test is broken -- i assert that the 2 v3 children get accepted.
Guyver2 has left #bitcoin-core-dev [Closing Window]
<sdaftuar>
overall it seems to me like testmempoolaccept is trying to do something that is currently way too hard, so i don't think i mind that this is broken, but probably worth documenting its limitations?
<instagibbs>
there's a lot of caveats to what it does vs say test package evaluation, would be nice to be documented either way
abubakarsadiq has joined #bitcoin-core-dev
AaronvanW has quit [Remote host closed the connection]
<bitcoin-git>
[bitcoin] ismaelsadeeq opened pull request #29308: doc: clarify that `BroadcastTransaction` comment (master...01-2024-clarify-BroadcastTransaction-comment) https://github.com/bitcoin/bitcoin/pull/29308
ghost43 has quit [Remote host closed the connection]
ghost43 has joined #bitcoin-core-dev
<bitcoin-git>
[bitcoin] dergoegge closed pull request #29305: fuzz: Use FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION for pow checks (master...2024-01-fuzz-unsafe-pow) https://github.com/bitcoin/bitcoin/pull/29305
AaronvanW has joined #bitcoin-core-dev
AaronvanW has quit [Ping timeout: 246 seconds]
puchka has quit [Ping timeout: 276 seconds]
qxs has quit [Ping timeout: 240 seconds]
vasild has quit [Ping timeout: 240 seconds]
vasild has joined #bitcoin-core-dev
qxs has joined #bitcoin-core-dev
preimage has joined #bitcoin-core-dev
vasild has quit [Remote host closed the connection]
qxs has quit [Remote host closed the connection]
vasild has joined #bitcoin-core-dev
qxs has joined #bitcoin-core-dev
vasild has quit [Remote host closed the connection]
vasild has joined #bitcoin-core-dev
vasild has quit [Client Quit]
vasild has joined #bitcoin-core-dev
<glozow>
sdaftuar: ooh good catch!
jonatack has joined #bitcoin-core-dev
bugs_ has joined #bitcoin-core-dev
<glozow>
Maybe requiring testmempoolaccept be 1 connected component (but still simulating a 1 by 1 submission) solve this? Though that might be more trouble than it's worth as well. testmempoolaccept has some other documented shortcomings like sometimes overestimating ancestor/descendant limits and not doing cpfp carve out. So maybe it's fine.
<bitcoin-git>
bitcoin/master ff54314 marco: wallet: clarify replaced_by_txid and replaces_txid in help output
<bitcoin-git>
bitcoin/master 207220c Ava Chow: Merge bitcoin/bitcoin#29302: wallet: clarify replaced_by_txid and replaces...
<bitcoin-git>
[bitcoin] achow101 merged pull request #29302: wallet: clarify replaced_by_txid and replaces_txid in help output (master...2024/01/clarify-documentation) https://github.com/bitcoin/bitcoin/pull/29302
AaronvanW has quit [Ping timeout: 256 seconds]
ghost43 has quit [Ping timeout: 240 seconds]
ghost43 has joined #bitcoin-core-dev
Talkless has joined #bitcoin-core-dev
gracilis has joined #bitcoin-core-dev
gracilis has quit [Remote host closed the connection]
qxs has quit [Remote host closed the connection]
qxs has joined #bitcoin-core-dev
abubakarsadiq has quit [Quit: Connection closed for inactivity]
puchka has joined #bitcoin-core-dev
<sdaftuar>
glozow: i'm reluctant to propose any changes right now... i think ultimately, it will make sense to push these topology checks / global limit evaluation into the mempool itself. but we can wait until we have a better design in place for the mempool that will make that easier.
jon_atack has joined #bitcoin-core-dev
jonatack has quit [Ping timeout: 256 seconds]
gracilis has joined #bitcoin-core-dev
gracilis has quit [Remote host closed the connection]
mudsip has joined #bitcoin-core-dev
mudsip has quit [Client Quit]
Talkless has quit [Ping timeout: 260 seconds]
justache has quit [Read error: Connection reset by peer]
justache has joined #bitcoin-core-dev
zato is now known as ao
justache has quit [Read error: Connection reset by peer]
ao is now known as samurai_
samurai_ is now known as ichi
ichi is now known as fufu
fufu is now known as lo
justache has joined #bitcoin-core-dev
jon_atack has quit [Ping timeout: 264 seconds]
lo is now known as tata
tata is now known as tat
brunoerg has quit [Remote host closed the connection]