0c55d8b581 depends: qt: Patch to remove dep on libX11 (Carl Dong)
222e6cc520 gitignore: Actually pay attention to depends patches (Carl Dong)
65f8da08df symbol-check: Disallow libX11-*.so.* shared libraries (Carl Dong)
924569914e depends: libXext isn't needed by anyone (Carl Dong)
689d3b4a03 build-aux: Remove check for x11-xcb (Carl Dong)
aa53cb7a2f depends: libX11: Make package headers-only (Carl Dong)
9a01ab04e1 depends: qt: Explicitly stop using Xlib/libX11 (Carl Dong)
1ec30b8fbe depends: xproto is only directly needed by libXau (Carl Dong)
Pull request description:
Related to: #16150
We noticed that we could build QT without using XLib/libX11 as a library. XLib/libX11's headers are still used, and a minimal `configure.ac` has been added to eliminate overly-enthusiastic configure-time dependencies that aren't actually required to obtain the headers.
This also means that we eliminate XLib/libX11 as required shared libraries at runtime, which is desirable.
See commit messages for more details.
---
Reviewers: I am least sure about the minimal `configure.ac`, as I'm not too familiar with the autoconf syntax. Any improvements w/re robustness would be welcome.
ACKs for top commit:
theuni:
ACK 0c55d8b581
fanquake:
ACK 0c55d8b581
Tree-SHA512: 41f653a0f91bc0e0faac49713c0c6dfd8cb605f9c4e34eb75a790dd808ebf3e5c160f1dd40bc8fbc911ee718ea319313b526d63733c98ff62d8dffecb58caa01
4f050b91c7 move-onlyish: move CCoinsViewErrorCatcher out of init.cpp (James O'Beirne)
Pull request description:
This is part of the [assumeutxo project](https://github.com/bitcoin/bitcoin/projects/11):
Parent PR: #15606
Issue: #15605
Specification: https://github.com/jamesob/assumeutxo-docs/tree/2019-04-proposal/proposal
---
This change moves `CCoinsViewErrorCatcher` out of `init` and into `coins` so that it can later be included in [a `CoinsView` instance](91284964ef (diff-349fbb003d5ae550a2e8fa658e475880R504)) under `CChainState`.
Instead of hardcoding read failure behavior that has knowledge of qt, it accepts error callbacks via `AddReadErrCallback()`.
ACKs for top commit:
dongcarl:
re-ACK 4f050b91c7
ryanofsky:
utACK 4f050b91c7. Only change since last review is fixing const.
Tree-SHA512: eaba21606d15d2b8d0e3db7cec57779ce181af953db1ef4af80a0bc1dfb57923d0befde9d61b7be55c32224744f7fb6bd47d4e4c72f3ccfe6eaf0f4ae3765c17
fa56b21c74 doc: Update bips 35, 37 and 111 status (MarcoFalke)
Pull request description:
Follow-up to
* #16152: Disable bloom filtering by default
ACKs for top commit:
laanwj:
ACK fa56b21c74, thanks for keeping this file up to date
fanquake:
ACK fa56b21c74
Tree-SHA512: 50c17067abbba096e8604b8abccbd0474ecc2dca973935751720c79a023a2d517bb025bb6c36d4fb0d29b7338322af7ae1c0d5a9aaf2712000d9d336c898c204
fa4010e112 travis: Print memory and number of cpus (MarcoFalke)
Pull request description:
For some reason it shows a different value than the one they advertise. This might be related to the flood of sanitizer warnings we see.
https://docs.travis-ci.com/user/reference/overview/#virtualisation-environment-vs-operating-system
Top commit has no ACKs.
Tree-SHA512: 285bdf6e7fe29990b980acf64ca658f4319d17c770f45cfd4339244e6b7ec11b7a39b9c4a54e71415c32fef08ee5da75ab7171861905494d633631779480c146
The "version" record that these functions read and write are not
used anywhere in the code except for one place. There is no reason
to expose these functions publicly. Furthermore, this avoids potential
confusion as developers may mistake these functions for actually
reading and writing the wallet version when they do not.
nFileVersion is not the actual file version and is not used except
in one place. So it is removed from CWalletScanState and changed so
that it is just read at the place it is needed. Furthermore, the
"version" record now only indicates the version of the highest
versioned client that has opened a wallet file so the variable
name is changed accordingly
When building the ZMQ static library, add AM_CPPFLAGS to the library
CPPFLAGS. Otherwise, we may miss important flags that are specified
elsewhere. For instance, if --enable-debug is passed and
-DDEBUG_LOCKORDER set, then that would not apply to the ZMQ library
before (causing potential for hard-to-find bugs).
and into coins.cpp. This move is necessary so that we can later include a
CCoinsViewErrorCatcher instance under CChainState.
Co-authored-by: MarcoFalke <falke.marco@gmail.com>
50cede3f5a [mempool] Allow one extra single-ancestor transaction per package (Matt Corallo)
Pull request description:
This implements the proposed policy change from [1], which allows
certain classes of contract protocols involving revocation
punishments to use CPFP. Note that some such use-cases may still
want some form of one-deep package relay, though even this alone
may greatly simplify some lightning fee negotiation.
[1] https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2018-November/016518.html
ACKs for top commit:
ajtowns:
ACK 50cede3f5a -- looked over code again, compared with previous commit, compiles, etc.
sdaftuar:
ACK 50cede3f5a
ryanofsky:
utACK 50cede3f5a. Changes since last review: adding EXTRA_DESCENDANT_TX_SIZE_LIMIT constant, changing max ancestor size from 1,000,000 to nLimitAncestorSize constant (101,000), fixing test comment and getting rid of unused test node.
Tree-SHA512: b052c2a0f384855572b4579310131897b612201214b5abbb225167224e4f550049e300b471dbf320928652571e92ca2d650050b7cf39ac92b3bc1d2bcd386c1c
a52818cc56 net: Make poll in InterruptibleRecv only filter for POLLIN events. poll should block until there is data to be read or the timeout expires. (tecnovert)
Pull request description:
poll should block until there is data to be read or the timeout expires.
Filtering for the POLLOUT event causes poll to return immediately which leads to high CPU usage when trying to connect to non-responding peers through tor.
When USE_POLL is not defined select is used with the writefds parameter set to nullptr.
Removing POLLOUT causes the behavior of poll to match that of select.
Fixes: #16004.
ACKs for top commit:
laanwj:
code review ACK a52818cc56
jonasschnelli:
utACK a52818cc56
Tree-SHA512: 69934cc14e3327c7ff7f6c5942af8761e865220b2540d74ea1e176adad326307a73860417dddfd32d601b5c0e9e2ada1848bd7e3d27b0b7a9b42f11129af8eb1
5c3c24cf9e test: remove redundant setup in addrman_tests (zenosage)
Pull request description:
#10765 make this default behavior. No reason to keep these line.
Top commit has no ACKs.
Tree-SHA512: 545eea9c2d0741a75708f288f2c8752534ecaa6d54a9d014ef9afa295b0d075007704b64809eec090023703f47753e8ec755d22c9ccecf57b75f6898f6b708dd
bead32e31e Add release notes for DEFAULT_BLOOM change (Matt Corallo)
f27309f55c Move DEFAULT_PEERBLOOMFILTERS from validation.h to net_processing.h (Matt Corallo)
5efcb77283 Disable bloom filtering by default. (Matt Corallo)
Pull request description:
BIP 37 bloom filters have been well-known to be a significant DoS
target for some time. However, in order to provide continuity for
SPV clients relying on it, the NODE_BLOOM service flag was added,
and left as a default, to ensure sufficient nodes exist with such a
flag.
NODE_BLOOM is, at this point, well-established and, as long as
there exist 0.18 nodes with default config (which I'd anticipate
will be true for many years), will be available from some peers. By
that time, the continued slowdown of BIP 37-based filtering will
likely have rendered it useless (though this is already largely the
case). Further, BIP 37 was deliberately never updated to support
witness-based filtering as newer wallets are expected to migrate to
some yet-to-be-network-exposed filters.
ACKs for top commit:
jnewbery:
ACK bead32e31e
kallewoof:
ACK bead32e31e
Tree-SHA512: ecd901898e8efe1a7c82b471af0acc2373c2282ac633eb58d9aae7c35deda1999d0f79fb0485e6cecbda7246aeda00206cd82c7fa36866e2ac64705ba93f9390
024ecd7e01 QA: Fix race condition in wallet_encryption test (Jonas Schnelli)
Pull request description:
There is some imprecision probably in the internal HTTPRPCTimer class (haven't exactly figured out where).
But we can't expect that waiting excatly 2 seconds right after calling `walletpassphrase(2)` will result in a locked wallet due to the nature how we internally handle threads/timers.
The wallet_encryption test fails regularely in CIs.
Here is a logged session:
```shell
[0;34m node0 2019-07-18T18:51:22.569739Z [] ThreadRPCServer method=walletpassphrase user=__cookie__ [0m
[0;34m node0 2019-07-18T18:51:22.628656Z [] queue run of timer lockwallet() in 2 seconds (using HTTP) [0m
[0;34m node0 2019-07-18T18:51:22.629002Z [] Received a POST request for / from 127.0.0.1:46898 [0m
[0;34m node0 2019-07-18T18:51:22.629081Z [] ThreadRPCServer method=dumpprivkey user=__cookie__ [0m
[0;34m node0 2019-07-18T18:51:24.445620Z [] Flushing wallet.dat [0m
[0;34m node0 2019-07-18T18:51:24.451421Z [] Flushed wallet.dat 6ms [0m
[0;34m node0 2019-07-18T18:51:24.631703Z [] Received a POST request for / from 127.0.0.1:46898 [0m
[0;34m node0 2019-07-18T18:51:24.631737Z [] ThreadRPCServer method=dumpprivkey user=__cookie__ [0m
[0;36m test 2019-07-18T18:51:24.632000Z TestFramework (ERROR): Assertion failed [0m
[0;36m Traceback (most recent call last):[0m
[0;36m File "/home/ubuntu/src/test/functional/test_framework/test_framework.py", line 193, in main[0m
[0;36m self.run_test()[0m
[0;36m File "/home/ubuntu/src/test/functional/wallet_encryption.py", line 53, in run_test[0m
[0;36m assert_raises_rpc_error(-13, "Please enter the wallet passphrase with walletpassphrase first", self.nodes[0].dumpprivkey, address)[0m
```
ACKs for top commit:
promag:
ACK 024ecd7e01, simple fix, one second shouldn't hurt.
MarcoFalke:
ACK 024ecd7e01
fanquake:
ACK 024ecd7e01
Tree-SHA512: 0cda1b8969b084bb765d2b35e90a8611c565ee458a7be1f2dde675f8ddbd9b9e421514547a7683f836e2c996e0538eb66b8c5b935b5a81e9319fb2be27624374
The logic before would only include watchonly addresses if it was
explicitly set in the rpc argument.
This changes the logic like so:
If the include_watchonly argument is missing, check the
WALLET_FLAG_DISABLE_PRIVATE_KEYS flag to determine if we're working
with a watchonly wallet. If so, default include_watchonly to true.
If the include_watchonly argument is explicit set to false, we still
disable them from the listing. Although this would always return
nothing, it might be still useful in situations where you want to
explicitly filter out watchonly addresses regardless of what wallet
you are dealing with.
Signed-off-by: William Casarin <jb55@jb55.com>
No reason for this class to exist if it doesn't have any code to run in the
destructor. e10e1e8db0 from
https://github.com/bitcoin/bitcoin/pull/16208 recently removed code destructor
code that would return an unused key if the transaction wasn't committed.
ae311bc036 Fix autostart filenames on Linux (Hennadii Stepanov)
Pull request description:
Currently, on master the `bitcoin-test.lnk` and `bitcoin-regtest.lnk` files do not work as autostart application `.desktop` files.
This PR fixes it.
Refs:
- #7045
- [Autostart Of Applications During Startup](https://standards.freedesktop.org/autostart-spec/autostart-spec-latest.html)
ACKs for top commit:
promag:
utACK ae311bc, weird why extension `.lnk` was used in #7045.
laanwj:
Code review ACK ae311bc036
Tree-SHA512: 210cc346600d52b0a262c81ed5f258365a3cea2e5522f4b5f4798fd3b54f45ed82aba68eefae59a6b6f1d8e4d00221476c23bdffc038f16f2f45c1acc837f522
a981e749e6 fix: tor: Call event_base_loopbreak from the event's callback (João Barbosa)
Pull request description:
Calling `event_base_loopbreak` before `event_base_dispatch` has no effect. Fix this by calling `event_base_loopbreak` from the event's callback. From the [documentation](http://www.wangafu.net/~nickm/libevent-2.0/doxygen/html/event_8h.html#a07a7599e478e4031fa8cf52e26d8aa1e):
> event_base_loop() will abort the loop after the next event is completed; event_base_loopbreak() is typically invoked from this event's callback. This behavior is analogous to the "break;" statement.
This can be tested by running the following with and without this change:
```sh
bitcoind -- -regtest -proxy=127.0.0.1:9050 -listen=1 -bind=127.0.0.1 -whitebind=127.0.0.1:0
```
Fixes#16376.
ACKs for top commit:
laanwj:
code review ACK a981e749e6
fanquake:
ACK a981e749e6
Tree-SHA512: 328fe71366404d5be8177d7081d5b4868cee73412df631a1865d24fb1c153427210762738109e06b737f037f4c68966812fba041831bb9e8129861f19ce61a63
poll should block until there is data to be read or the timeout expires.
Filtering for the POLLOUT event causes poll to return immediately which leads to high CPU usage when trying to connect to non-responding peers through tor.
Removing POLLOUT matches how select is used when USE_POLL isn't defined.
e142ee03e7 doc: describe how to pass wildcard names to test runner (Jon Atack)
6a7a70b8cf test: enable passing wildcards with path to test runner (Jon Atack)
Pull request description:
Currently, passing wildcard testname args to the test runner from outside the test/functional/ directory does not work, even though developers expect it to. See these recent IRC discussions for more background: http://www.erisian.com.au/bitcoin-core-dev/log-2019-07-10.html#l-262 (lines 262 to 323) and http://www.erisian.com.au/bitcoin-core-dev/log-2019-07-11.html#l-134.
1. [BUGFIX] Enable passing wildcards with paths. Examples:
- `test/functional/test_runner.py test/functional/wallet*`
- `functional/test_runner.py functional/wallet*`
- `test/functional/test_runner.py ./test/functional/tool* test/functional/mempool*`
- A current limitation this PR does not change: 9 test files with arguments in their filename are not picked up by wildcard search.
2. [Docs] Describe how to pass wildcard names (multiple and with paths) to the test runner in test/README.md.
ACKs for top commit:
jnewbery:
tested ACK e142ee03e7
jachiang:
Tested ACK e142ee03e7. Thanks a lot for this fix!
MarcoFalke:
ACK e142ee03e7, fine with me
Tree-SHA512: cb3d994880cdc9b8918546b573a25faa5b4c7339826ac7cfe20f076aac6e731a34271609c0cf5a7ee5e4a2d5ae205298319d24bf36ef5b5d569a1a0c57883e54
libXext was only needed (as a library) by QT when it was using
XLib/libX11 (as a library), now that we're building QT without
XLib/libX11, we can safely remove libXext.
We're no longer building QT with libX11/XLib, however, libX11/XLib
headers are still required for parts of QT. In this commit we add a
minimal configure.ac for libX11/XLib that is headers-only.
This change allows us to remove all of libX11/XLib's dependencies.
Previously, in 683b7d7a3f and
0e752637a2, we accidentally broke QT's
ability to pick up Xlib thru the config.gui.tests.xlib configuration
test, which also means that config.gui.libraries.xcb_xlib wasn't run.
This resulted in a QT build that was implicitly -no-xcb-lib and
-no-feature-xlib.
This is actually a desired behaviour, as it means less required shared
objects for our final bitcoin-qt binary. Specifically, it eliminated the
libX11-xcb.so.1 and libX11.so.6 requirements.
In this commit, we explicitly build without Xlib. We should continue to
track upstream ticket https://bugreports.qt.io/browse/QTBUG-61452 which
talks about adding a -no-xlib (non-hidden) flag instead of the
-no-feature-xlib (hidden) flag.
e10e1e8db0 Restrict lifetime of ReserveDestination to CWallet::CreateTransaction (Gregory Sanders)
d9ff862f2d CreateTransaction calls KeepDestination on ReserveDestination before success (Gregory Sanders)
Pull request description:
The typical usage pattern of `ReserveDestination` is to explicitly `KeepDestination`, or `ReturnDestination` when it's detected it will not be used.
Implementers such as myself may fail to complete this pattern, and could result in key re-use: https://github.com/bitcoin/bitcoin/pull/15557#discussion_r271956393
Since ReserveDestination is currently only used directly in the `CreateTransaction`/`CommitTransaction` flow(or fee bumping where it's just used in `CreateTransaction`), I instead make the assumption that if a transaction is returned by `CreateTransaction` it's highly likely that it will be accepted by the caller, and the `ReserveDestination` kept. This simplifies the API as well. There are very few cases where this would not be the case which may result in keys being burned.
Those failure cases appear to be:
`CommitTransaction` failing to get the transaction into the mempool
Belt and suspenders check in `WalletModel::prepareTransaction`
Alternative to https://github.com/bitcoin/bitcoin/pull/15796
ACKs for top commit:
achow101:
ACK e10e1e8db0 Reviewed the diff
stevenroose:
utACK e10e1e8db0
meshcollider:
utACK e10e1e8db0
Tree-SHA512: 78d047a00f39ab41cfa297052cc1e9c224d5f47d3d2299face650d71827635de077ac33fb4ab9f7dc6fc5a27f4a68415a1bc9ca33a3cb09a78f4f15b2a48411b
fa89badf88 test: Require standard txs in regtest (MarcoFalke)
fa9b419160 test: Add test that mainnet requires standard txs (MarcoFalke)
fa613ca0a8 chainparams: Remove unused fMineBlocksOnDemand (MarcoFalke)
Pull request description:
I don't see a reason why regtest should allow non-standard txs, as it makes testing mainnet behaviour such as #15846 unnecessarily hard and unintuitive.
Of course, testnet policy remains unchanged to allow propagation of non-standard txs.
ACKs for top commit:
ajtowns:
ACK fa89badf88
Tree-SHA512: c4c675affb054868850bd2683aa07f4c741a448cbacb2ea8334191e105f426b0790fe6a468be61e9c5880d24154f7bf1c7075051697172dce92180c1bc3a1c90
682a1d0f20 refactoring: remove mapBlockIndex global (James O'Beirne)
55d525ab90 refactoring: make pindexBestInvalid internal to validation.cpp (James O'Beirne)
4ed55dfcd7 refactoring: add block_index_candidates arg to LoadBlockIndex (James O'Beirne)
613c46fe9e refactoring: move block metadata structures into BlockManager (James O'Beirne)
Pull request description:
This is part of the [assumeutxo project](https://github.com/bitcoin/bitcoin/projects/11):
Parent PR: #15606
Issue: #15605
Specification: https://github.com/jamesob/assumeutxo-docs/tree/2019-04-proposal/proposal
---
Under an assumeutxo model, we have multiple CChainState instances in use at once in order to support background validation. Currently, each CChainState instance has its own mapBlockIndex, a collection of linked block headers, in addition to a few other data structures that are related to maintenance of the block tree but not necessarily to any given chainstate.
In order to avoid duplicating this data across chainstates, this change moves chainstate-agnostic block metadata (and related behavior) into a class, `BlockManager`. Chainstates are parameterized with a reference to a blockmanager instance and in practice they share the same instance.
Most of this change is conceptually move-only, though the diff is somewhat muddled. The first commit can be reviewed slightly more easily with `--color-moved=dimmed_zebra`. Admittedly, that commit is pretty unwieldy; I tried to split it up after the fact with `git add --patch`, but that was difficult because of git's inability to split hunks past a certain point. Some of the moves also ended up being obscured when done over separate commits.
ACKs for top commit:
MarcoFalke:
ACK 682a1d0f20
ryanofsky:
utACK 682a1d0f20, only changes since last review were rebase and fixing conflict on a moved line
ariard:
utACK 682a1d0. Most of the changes are move-only, with main problem being to avoid creating circular dependencies between `BlockManager` and `CChainState`. Tested, comments are mostly nits, feel free to ignore them
Tree-SHA512: 738d8d06539ba53acf4bd2d48ae000473e645bbc4e63d798d55d247a4d5a4f781b73538ed590f6407be9ab402ea9d395570ea20bff0a4b9ce747bcc1600c5108
