3025ca9e77 [net processing] Add RemovePeer() (John Newbery)
a20ab22786 [net processing] Make GetPeerRef const (John Newbery)
ed7e469cee [net_processing] Move peer_map to PeerManager (John Newbery)
a529fd3e3f [net processing] Move GetNodeStateStats into PeerManager (John Newbery)
Pull request description:
This moves `g_peer_map` from a global in net_processing.cpp's unnamed namespace to being a member `m_peer_map` of `PeerManager`.
ACKs for top commit:
theuni:
Re-ACK 3025ca9e77.
dongcarl:
Re-ACK 3025ca9
hebasto:
re-ACK 3025ca9e77, since my [previous](https://github.com/bitcoin/bitcoin/pull/19910#pullrequestreview-545574237) review only reverted the change that introduced NRVO in `PeerManager::GetPeerRef`, and comments are fixed in the proper commits.
Tree-SHA512: 6369eb3c688ac5b84f89f7674115f78ff02edbed76063ac2ebb1759894c9e973883e10821a35dab92bd3d738280acc095bd5368f552a060b83cd309330387d47
4e28753f60 feestimator: encapsulate estimation file logic (Antoine Poinsot)
e8ea6ad9c1 init: don't create a CBlockPolicyEstimator if we don't relay transactions (Antoine Poinsot)
86ff2cf202 Remove the remaining fee estimation globals (Antoine Poinsot)
03bfeee957 interface: remove unused estimateSmartFee method from node (Antoine Poinsot)
Pull request description:
If the `blocksonly` mode is turned on after running with transaction
relay enabled for a while, the fee estimation will serve outdated data
to both the internal wallet and to external applications that might be
feerate-sensitive and make use of `estimatesmartfee` (for example a
Lightning Network node).
This has already caused issues (for example https://github.com/bitcoin/bitcoin/issues/16840 (C-lightning), or https://github.com/lightningnetwork/lnd/issues/2562 (LND)) and it seems prudent to fail rather than to give inaccurate values.
This fixes#16840, and closes#16890 which tried to fix the symptoms (RPC) but not the cause as mentioned by sdaftuar :
> If this is a substantial problem, then I would think we should take action to protect our own wallet users as well (rather than hide the results of what our fee estimation would do!).
ACKs for top commit:
MarcoFalke:
re-ACK 4e28753f60👋
jnewbery:
utACK 4e28753f60
Tree-SHA512: c869cf03b86d8194002970bbc84662dae76874967949b9be0d9a4511a1eabcb1627c38aca3154da9dcece1a4c49ec02bd4f9fcca2ec310986e07904559e63ba8
This moves the CBlockPolicyEstimator to the NodeContext, which get rids
of two globals and allows us to conditionally create the
CBlockPolicyEstimator (and to remove a circular dep).
Signed-off-by: Antoine Poinsot <darosior@protonmail.com>
629a9299b2 Move WalletImpl from interfaces/wallet.cpp to wallet/interfaces.cpp (Russell Yanofsky)
2a26771d81 Move ChainImpl from interfaces/chain.cpp to node/interfaces.cpp (Russell Yanofsky)
12bd0fc9d7 Move NodeImpl from interfaces/node.cpp to node/interfaces.cpp (Russell Yanofsky)
Pull request description:
This PR is part of the [process separation project](https://github.com/bitcoin/bitcoin/projects/10).
---
Move `NodeImpl` from `interfaces/node.cpp` to `node/interfaces.cpp`
Move `ChainImpl` from `interfaces/chain.cpp` to `node/interfaces.cpp`
Move `WalletImpl` from `interfaces/wallet.cpp` to `wallet/interfaces.cpp`
No changes to any classes (can review with `git diff --color-moved=dimmed_zebra`)
Motivation for this change is to move node and wallet code to respective directories where it might fit in better than `src/interfaces/`, but also to remove all unnecessary code from `src/interfaces/` to unblock #19160 review, which has been hung up partially because of code organization. Building on top of this PR, #19160 should now be able to organize interface implementations more understandably in `src/node/` `src/wallet/` `src/ipc/` and `src/init/` directories instead of having so much functionality all in `src/interfaces/`
ACKs for top commit:
promag:
Code review ACK 629a9299b2.
MarcoFalke:
review ACK 629a9299b2🔺
Tree-SHA512: 87c2b8fd51519bbd4e5ad3539a79debcf88c3bf021eb28c63f3f555186538b62a0c4cc1a3f07cfb4ff13aea8b0b2fdde505d81f22a5e5fd12a6e375b55a92ab8
Changes API for AcceptToMemoryPool to no longer check maximum fee; we remove
this check from interfaces::CChain::testPeginClaimAcceptance. This is fine,
the check never should have been there (this method's job is to notice double
claim attempts and provide a more helpful message to the user; other mempool
failures should be caught later in the `claimrawpegin` logic when we call
`pwallet->commitTransaction` which actually subits to the mempool). It was
only checking fees because the API made it easier to do so than to not.
Added a method testPeginClaimAcceptance to interfaces::CChain. It is not clear
to me that this is meaningfully different from the existing method broadcastTransaction
(with the `relay` flag set to false so it doesn't actually broadcast), except
that this method returns the TxValidationState, which we display to the user
in the RPC error. Worth revisiting this.
Another nonobvious change was replacing a block of mempool-searching logic
in wallet/rpcwallet.cpp with the single line `pwallet->chain().findCoins(coins)`.
This I stole from the current state of upstream #17211, our favorite in-progress
PR from which the original logic came.
This resulted in double-initializing the asset directory (and from multiple
places), so I added a `ClearGlobalAssetDir` method and called it from the
qt test setup.
Added a new "signed but not finalized" vector to the functional tests, which took
some time to figure out how to generate. Evenetually I took the output of walletsignpsbt
from the nonconf->nonconf test and manually un-finalized that by pulling the
scriptwitness apart into a pubkey->signature mapping.
This fixes a segfault I introduced in 981df590811 (#17371) a couple hundred PRs
ago, where we might out-of-bounds access the vout arrray of a non_witness_utxo
transaction in a PSBT.
This logic should probably be refactored after the rebase -- we sift through
the non_witness_utxos to find a scriptpubkey, which we use to get a wallet
provider, so we can pass this to SignPSBTInput, which then repeats the same
logic (for its own reasons, and without the segfault bug)....and we aren't
even calling SignPSBTInput for signing reasons! It has something to do with
blinding.
