107e030723552cf272dc8da01bb682032a457a3d build: make protobuf optional in depends (fanquake)
ff6122f32b21fa00e9308e098b33b9657debc1d7 doc: clarify protobuf build requirements (fanquake)
Pull request description:
As mentioned by dongcarl in https://github.com/bitcoin/bitcoin/pull/15584#issuecomment-521780972, make building `protobuf` optional in depends. With this change it will only be built if you pass `PROTOBUF=1`.
ACKs for top commit:
laanwj:
code review ACK 107e030723552cf272dc8da01bb682032a457a3d
Sjors:
tACK 107e030 on macOS 10.14. When I build depends with `PROTOBUF=1` then `./configure` has `bip70` enabled.
Tree-SHA512: 49bc247a6879aaf55b943a3d0b930544ddef1e69a481955a8bebe0b02c9ad0fe168b93025f34168334cef34bb567478eb98eacab62ba909f2f64fb21119c71b8
Make fMixing atomic as it has concurrent access
Add tsan suppression for zmq namespace
Suppress deadlock false positive in ConnectTip
Switch ubsan target to linux32
Add new test job for linux64_cxx17 target without any sanitizers
Increase rpc time out for block reward reallocation test
Fix heap use after free in CConnman::GetExtraOutboundCount()
Different builds for linux32 and linux64 tsan and ubsan
Increase timeout for llmq_signing functional test
* build: Fix `--enable-glibc-back-compat`
Compiling on Ubuntu 20.04 results in binaries that can't be run on older systems we still support (e.g. Ubuntu 16.04) and `contrib/devtools/symbol-check.py` complains about it.
Available versions for `log` for example:
$ objdump -T /lib/x86_64-linux-gnu/libm.so.6 | egrep -w 'log'
00000000000431b0 g DF .text 0000000000000039 GLIBC_2.29 log
0000000000012360 g DF .text 0000000000000066 (GLIBC_2.2.5) log
(i.e. the default one is 2.29)
This commit fixes the issue by picking the version we support instead of the default one.
Before:
$ objdump -T dashd | egrep -w 'log'
0000000000000000 DF *UND* 0000000000000000 GLIBC_2.29 log
After:
$ objdump -T dashd | egrep -w 'log'
0000000000000000 DF *UND* 0000000000000000 GLIBC_2.2.5 log
* ci: Add `make check-symbols` to the `build` step
* ci: Do not specify `--enable-glibc-back-compat` for non-release builds
* ci: Set correct LDFLAGS for the release build
* doc: Update docs to mention the need for `LDFLAGS=-static-libstdc++` when compiling for same host but different distro
* ci: Add `--with-sanitizers=undefined` back to `linux64` build
c4606b84329d760d7cee144bebe05807857edaae Add Travis check for single parameter constructors not marked "explicit" (practicalswift)
Pull request description:
Make single parameter constructors `explicit` (C++11).
Rationale from the developer notes:
> - By default, declare single-argument constructors `explicit`.
> - *Rationale*: This is a precaution to avoid unintended conversions that might
> arise when single-argument constructors are used as implicit conversion
> functions.
ACKs for top commit:
laanwj:
ACK c4606b84329d760d7cee144bebe05807857edaae
Tree-SHA512: 3e6fd51935fd93b2604b2188664692973d0897469f814cd745b5147d71b99ea5d73c1081cfde9f6393f51f56969e412fcda35d2d54e938a3235b8d40945f31fd
fad0ce59e9 tests: Fail if RPC has been added without tests (MarcoFalke)
Pull request description:
Need to be run with --coverage
ACKs for commit fad0ce:
ryanofsky:
utACK fad0ce59e9154f9b7e61907a71c740a942c60282. New comment in travis.yml is the only change since last review.
Tree-SHA512: b53632dfe9865ec06991bfcba2fd67238bebbb866b355f09624eaf233257b2bca902caac6c24abb358b2f4c1c43f28ca75e30982765911e1a117102df65276d9
* Merge #18750: build: optionally skip external warnings
ba8950ee0134a7958e3e9b041cd54d222feb09a1 build: optionally skip external warnings (Vasil Dimov)
Pull request description:
Add an option to `./configure` to suppress compilation warnings from
external headers. The option is off by default (no change in behavior,
show warnings from external headers).
This option is useful if e.g. Boost or Qt is installed outside of
`/usr/include` (warnings from headers in `/usr/include` are already
suppressed by default) and those warnings stand in the way of compiling
Bitcoin Core with `-Werror[=...]` or they just clutter the build output
too much and make our own warnings hard to spot.
`-isystem /usr/include` bricks GCC's `#include_next`, so we use
`-idirafter` instead. This way we don't have to treat `/usr/include`
specially.
ACKs for top commit:
practicalswift:
ACK ba8950ee0134a7958e3e9b041cd54d222feb09a1: diff looks correct!
hebasto:
ACK ba8950ee0134a7958e3e9b041cd54d222feb09a1, tested on Linux Mint 20 (x86_64).
luke-jr:
utACK ba8950ee0134a7958e3e9b041cd54d222feb09a1
Tree-SHA512: 9b54fae8590be6c79f2688a5aca09e0a9067f481dabecdd49bb278c08a62ac2b0cc704c894fbd53240e77ac84da0c7a237845df0a696cfbdb0359e1c8e2e10c9
* add --enable-suppress-external-warnings to matrix.sh
Co-authored-by: Wladimir J. van der Laan <laanwj@protonmail.com>
* lint: Skip shell linting if gawk is not installed
* lint: Skip Gitian descriptor scripts checking if jq is not installed
* ci: Install gawk and jq
`yq` requires `jq`
* Fix shellcheck warnings
39d526bde48d98af4fa27906e85db0399b6aa8b1 test: Bump linter versions (Duncan Dean)
Pull request description:
As per #19346, `mypy==0.700` was incompatible with Python 3.8.
I've bumped the versions of all the linters to their latest stable versions.
Checked with both Python 3.7 and 3.8 and everything still seems to work fine.
ACKs for top commit:
hebasto:
ACK 39d526bde48d98af4fa27906e85db0399b6aa8b1, I have reviewed the code and it looks OK, I agree it can be merged.
Tree-SHA512: f3ee7fda8095aa25aa68685e863076d52a6b82649770d24b0064d652763c0ceb8ebcbf9024fc74fca45c754e67b2a831dd070b3af23bc099140e6d27e89a5319
3d0a82cff8cbb809876e82dbe62d14d2adc07d94 devtools: Accomodate block-style copyright blocks (Ben Woosley)
0ef0e51fe4bb592e67255776b5a0ba04679fb8c4 lint: Bump flake8 to 3.7.8 (Ben Woosley)
838920704ad90a71cf288b700052503db8abb17e lint: Disable flake8 W504 warning (Ben Woosley)
b21680baf5391a602b295b9d7d0ef66553661cb9 test/contrib: Fix invalid escapes in regex strings (Ben Woosley)
Pull request description:
This is a second go at #15221, fixing new lints in:
W504 line break after binary operator
W605 invalid escape sequence
F841 local variable 'e' is assigned to but never used
This time around:
* One commit per rule, for easier review
* I went with the PEP-8 style of breaking before binary operators
* I looked into the raw regex newline issue, and found that raw strings with newlines embedded do work appropriately. E.g. run `re.match(r" \n ", " \n ")` to check this for yourself. `re.MULTILINE` exists to modify `^` and `$` in multiline scenarios, but all of these searches are per-line.
ACKs for top commit:
practicalswift:
ACK 3d0a82cff8cbb809876e82dbe62d14d2adc07d94 -- diff looks correct
Tree-SHA512: bea0c144cadd72e4adf2e9a4b4ee0535dd91a8e694206924cf8a389dc9253f364a717edfe9abda88108fbb67fda19b9e823f46822d7303c0aaa72e48909a6105
3f8776a1391c3978ed66144df15fd9bcb9edd35d Re-add dead code detection (flack)
Pull request description:
This re-adds unreachable code detection for Python based on `vulture`.
Effectively, this reverts f4beb4996d27f2cdaf4f0a63e7dc044bf17decce. The difference to the previous version is that this runs with the `--min-confidence 100` setting. From https://pypi.org/project/vulture/:
> Use `--min-confidence 100` to only report code that is guaranteed to be unused within the analyzed files.
So this should avoid the previous issues where static analysis had wrong positives due to the dynamic nature of Python code by only reporting things that are unambiguous (such as code after a `return` statement). As such, there is not suppressions list.
My motivation was mainly #21081 which would have been caught by this (as can be seen by the CI run failing). This is still marked as draft because #21081 is needed to get the linter to pass. Also, there is a second problem that this found (see https://github.com/bitcoin/bitcoin/pull/19509/files#r571454691). From what I can tell, this is a spurious type comment that could just be removed (or if that line has no side effects it could also be deleted altogether?). I could add a commit here to fix it, but I wanted to see if there is interest in having this linter again in the first place
ACKs for top commit:
practicalswift:
ACK 3f8776a1391c3978ed66144df15fd9bcb9edd35d
Tree-SHA512: 52314ad4f627d969de1eb15375ca677ed86a2e816fe773756a1ce22421214ba407b5a09a4bf701a3aab1a10c7b336f548e4cef3327edf154acba55e987db21f6
f4beb4996d27f2cdaf4f0a63e7dc044bf17decce test: Remove python dead code linter (Wladimir J. van der Laan)
Pull request description:
Primarily I'd like to remove this because it is very imprecise, due to Python's dynamic nature, giving it a large list of false positives that need to be listed as exceptions. See for example #16906.
It's also a frequent source of complaints. I'm doubtful of the usefulness of checking for dead code in a linter in the first place.
Having some dead code in the test framework for a while is not a
disaster.
ACKs for top commit:
sdaftuar:
utACK f4beb4996d27f2cdaf4f0a63e7dc044bf17decce
practicalswift:
ACK f4beb4996d27f2cdaf4f0a63e7dc044bf17decce -- diff looks correct
jamesob:
ACK f4beb4996d
Tree-SHA512: 329b1555210311d5d15799fd2cb794b3208b0ac4d8a2ffaf4dece1bcc3e0e8b1fe952d5e7a394f94a98919cab579fb579eae7db2a796cc9a1a42ef495dd17507
fac35b21e2c2d798face7e289dcff4c1cce0e1a6 test: lint: Add DisabledOpcodeTemplates to whitelist (MarcoFalke)
Pull request description:
Fixes#16906
Top commit has no ACKs.
Tree-SHA512: 98b175bb062425fd3a8bd0d0258f4c0e0d5106980f1e037df7c2b2b2e5aa6031b11b582c026265d7b2de56049ccbadb0b7add9130d323f15886f681c6268ba0a
b748bf6f50dda6bb57eadf697edc320b2695e01a Fix spelling errors identified by codespell 1.15.0 (Ben Woosley)
Pull request description:
Note all changes are to comments / documentation.
After this commit, the only remaining output is:
```
$ test/lint/lint-spelling.sh
src/test/base32_tests.cpp:14: fo ==> of, for
src/test/base64_tests.cpp:14: fo ==> of, for
^ Warning: codespell identified likely spelling errors. Any false positives? Add them to the list of ignored words in test/lint/lint-spelling.ignore-words.txt
```
Note:
* I ignore several valid alternative spellings ~, but changed homogenous
to homogeneous as the latter is a more specific term according to the
Google dictionary definitions I found~
* homogenous is present in tinyformat, hence should be addressed upstream
* process' is correct only if there are plural processes
ACKs for commit b748bf:
practicalswift:
ACK b748bf6f50dda6bb57eadf697edc320b2695e01a
fanquake:
ACK b748bf6f50
Tree-SHA512: 9add7044643ce015e0a44d8b27a3f300d72c485ffff550fb6491a17f14528085289ec5caddfe02f291ea9b2cded38a0dd3079652a054e2d7fe2ff4f7b53db5d7
fa193dc8e6f3b96fa2dba2f1c1668f7720fed320 doc: Remove win32 from the release process (MarcoFalke)
faf666f8148eeb305a9c4f78459aff2c7268016b Remove Windows 32 bit build (MarcoFalke)
Pull request description:
The Windows 32 bit build has been removed from https://bitcoincore.org/en/download/, so unless there are complaints, we don't need to build it even
ACKs for commit fa193d:
fanquake:
utACK fa193dc8e6
Tree-SHA512: d6f2976a2e0c407698f720b00ac23ec4056626de4eff8621f4c5581120af0460afd1bdef72329cc0e7d92afca48d94ae5fce6777cb36bfabb60b8034ff08fd88
eeeee58bc9 travis: Combine --disable-bip70 into existing job (MarcoFalke)
Pull request description:
We already have too many jobs, so instead of creating a separate job for the `--disable-bip70` configue option, combine it into an existing job
Tree-SHA512: 9e2fae73d90cb55b588c545bc118a14eba064f17fffd9b302c3fdbb8715e2319db03eac92ae51b3c16481f28a004a1c964dab75ca80a213e87574da8f73e3207
c82190cdb6 tests: Add Python dead code linter (vulture) (practicalswift)
590a57fdec tests: Remove unused testing code (practicalswift)
Pull request description:
Add Python dead code linter (`vulture`) to Travis.
Rationale for allowing dead code only after explicit opt-in (via `--ignore-names`):
* Less is more :-)
* Unused code is by definition "untested"
* Unused code can be an indication of bugs/logical errors. By making the contributor aware of newly introduced unused code it gives him/her an opportunity to investigate if the unused code they introduce is malignant or benign :-)
* Unused code is hard to spot for humans and is thus often missed during manual review
* [YAGNI](https://en.wikipedia.org/wiki/You_aren%27t_gonna_need_it)
Based on #14312 to make linter job pass.
Tree-SHA512: 4c581df7c34986e226e4ade479e0d3c549daf38f4a4dc4564b25564d63e773a1830ba55d1289c771b1fa325483e8855b82b56e61859fe8e4b7dfa54034b093b6
414326952c use export LC_ALL=C.UTF-8 (Julian Fleischer)
728c82d029 make script exit if a command fails (Julian Fleischer)
506890b24d move remaining travis build steps into individual files (Julian Fleischer)
272306ea57 number .travis/ script according to build lifecycle and add README to explain (Julian Fleischer)
519e2739cf move lint stage up to resemble travis build ui (Julian Fleischer)
86d34f0e65 abort script in END_FOLD on non-zero exit code (Julian Fleischer)
4f2f88c7b0 move script sections info individual files and comply with shellcheck (Julian Fleischer)
Pull request description:
This PR is extracted from https://github.com/bitcoin/bitcoin/pull/13816 to make that one easier to review. It follows on https://github.com/bitcoin/bitcoin/pull/13849 and https://github.com/bitcoin/bitcoin/pull/13851
In here the shell script parts from `travis.yml` are extracted into `.travis/before_install.sh`, `.travis/install.sh`, `.travis/before_script.sh`, `.travis/script.sh`, and `.travis/lint.sh`.
This has the benefit that `test/lint/lint-shell.sh` will also shellcheck these parts. Also it makes the individual script parts more readable.
Tree-SHA512: c497e1687ceb1c1d795de177d3fc35af908bc8e3f781a871afabdecf031e581d4db229290627249e35ef7c09952bc34884e4734ea91d40f57b4a9efb85bba2e3
fad8afa3883675880dfa7fa6664fe658f10b11b5 travis: Run unit tests --with-sanitizers=undefined (MarcoFalke)
0d00fd5901102d9ca2b99d6f17a3bd96c946e3b7 depends: allow CC/CXX to be overridden during configure (Cory Fields)
Pull request description:
See issue #12691
Tree-SHA512: 0772e8e037411cb4664270945a0fc9c0c1ed24eeaf5308460baee707868f7b700d8e740a6babb1d02bc818a8a4b71d9e33601ab2cc850ff315755fcf3d79d29e
# Conflicts:
# .travis.yml
5e1777777790e855a9f3c8604208bc9bd6c8c99f qa: Create unicode tempdir in test_runner (MarcoFalke)
Pull request description:
Now that wallet filenames are properly quoted when used for rpc (#13823), we can add some unicode symbols to the test_runner path. Thus, the "extern" wallet that uses a full path has a unicode symbol in its name.
Should add unicode coverage to
* `listwallets`
* `wallet.getwalletinfo`
* `(un)loadwallet`
Tree-SHA512: 1633fde56f8748df0cfef9c31a878c105dfaac85d1041b292261f44c4d40e96942aacbf7d6e839e8bbf979dc131d81c24ceb521e927fc8a5a71ba093f36b891b
3828a79711 scripted-diff: prefer MAC_OSX over __APPLE__ (fanquake)
fa6e841e89 gui: remove macOS ProgressBar workaround (fanquake)
68c272527f gui: remove SubstituteFonts (fanquake)
6c6dbd8af5 doc: mention that macOS 10.10 is now required (fanquake)
84b0cfa8b6 release: bump minimum required macOS to 10.10 (fanquake)
26b15df99d depends: set OSX_MIN_VERSION to 10.10 (fanquake)
Pull request description:
Closes#13362
d99abfddb0c8f2111340a6127e77cc686e0043d8
This workaround should no longer be required, as it should have only been in use when compiled with the 10.7 SDK, which we haven't been building with for a while now.
5bc5ae30982a0f0f6a9804b05d99434af770c724
The bugreport linked with this code is for an unrelated? issue, however from what I can tell the correct QTBUG is this one https://bugreports.qt.io/browse/QTBUG-20880. Reading though the discussion there, it seems that the way progress bars are animated changed in macOS 10.10.
Qt was patched [here (5.5+)](https://codereview.qt-project.org/#/c/112379/):
> Disable progress bar animations on 10.10 Yosemite and higher - the native style does not animate them any more. Keep the indeterminate progress bar animation.
Given all of that, I don't think this is worth keeping around, as it would seem to only be useful in the case that a macOS user is compiling with a Qt < 5.5. That should be pretty unlikely, as we don't support downloaded Qt binaries, and brew currently provides [5.11.1](571b46213c/Formula/qt.rb).
Tree-SHA512: 4278cb30cc9bcb313e166129ecf032c808995f8b51a3123637c47860a0010ac88f86f82ec44792153b6b1e5cca595f25013b2eaeae80194647b9ce4f7eaf32c1
fa3c910bfeab00703c947c5200a64c21225b50ef test: Move linters to test/lint, add readme (MarcoFalke)
Pull request description:
This moves the checks and linters from `devtools` to a subfolder in `test`. (Motivated by my opinion that the dev tools are mostly for generating code and updating the repo whereas the linters are read-only checks.)
Also, adds a readme to clarify that checks and linters are only meant to prevent bugs and user facing issues, not merely stylistic preference or inconsistencies. (This is motivated by the diversity in developers and work flows as well as existing code styles. It would be too disruptive to change all existing code to a single style or too burdensome to force all developers to adhere to a single style. Also note that our style guide is changing, so locking in at the wrong style "too early" would only waste resources.)
Tree-SHA512: 9b10e89f2aeaf0c8a9ae248aa891d74e0abf0569f8e5dfd266446efa8bfaf19f0ea0980abf0b0b22f0d8416ee90d7435d21a9f9285b66df43f370b7979173406
3d69853090 travis: Change Mac goal to all deploy so that travis can build all executables for Mac. (Chun Kuan Lee)
Pull request description:
Currently, travis only build bitcoin-qt for Mac with make deploy, so I think that we can add a make step to build cli, daemon, test and bench executables.
Tree-SHA512: 346c207836ee7313a57e0cd03367ada7e95c182d0f58b06c4ee5b8f2865ba4ee077b030a5494b58a808c015afe0569fa3252d0a5a687c47b7094dc77599ce50e
Signed-off-by: pasta <pasta@dashboost.org>
# Conflicts:
# .travis.yml
* Remove unused jenkins stuff
* Install all dependencies in builder image
Instead of only target specific dependencies.
* Use docker builder image for builds
* Optimize apt installations
* Move building of dependencies into separate stage
The build-depends-xxx jobs will create artifacts (depends/$HOST) which are
then pulled in by the build jobs with the help of "needs"
* Remove use of caches from develop branch
* Use gitlab specific extends instead of YAML anchors
* Move before_script of build_template into base_template
* Add hack for parallel installation of i686 and arm cross compilation
* Install python3-setuptools in builder image
* Remove unnecessary change-dir
* Use variables to pass BUILD_TARGET instead of relying on the job name
* Move integration tests into separate stage
* Don't use --quiet for integration tests on Gitlab
fa43626611 test_runner: Remove travis specific code (MarcoFalke)
Pull request description:
The tests are no longer run on travis, but in a docker, developer machines or a windows vm.
The code was essentially dead for months now. Fix that by explicitly passing in `--ci` to the test runner on our docker and appveyor windows vm.
Tree-SHA512: 5d48693c03e8eb27536658ccf9ba738fe93a72abd4b72c80caac084b5b2cdffa77a1031a671eeefe70b71d63500f55917803d4be54d01849722afdccb700a9e6
58f9a0a Use --failfast when running functional tests on Travis (James O'Beirne)
bf720c1 Add --failfast option to functional test runner (James O'Beirne)
Pull request description:
Add the option (`--failfast`) to stop the functional test runner's execution when it encounters the first failure.
Also cleans up run_test's arguments list ([no more mutable default for `args`](http://docs.python-guide.org/en/latest/writing/gotchas/#mutable-default-arguments)) and call site.
Tree-SHA512: e854b1b1634bf613ae8ae88e715df1460982fa68db9d785aafeb5eccf5bf324c7f20dded2ca6840ebf18a28347ecac2138d6c7592507b34939b02609ef55e1b3
* Add .gitlab-ci.yml
* Use | instead of > for multiline commands
This honor new-lines and makes ; unnecessary
* Use ubuntu:bionic as base image
* Move cache initialization before apt-get installs
* Cache apt packages
* Move installation of wget and unzip up as we need it for the cache
* Prevent apt from deleting caches
* Collect test logs into artifact
* Make combine_logs.py always look for the template in the correct dir
* Move final cache stuff into after_script
* Reintroduce PYTHON_DEBUG=1, but only for .travis.yml
* Install jinja2 in Travis builder image
* Enable ChainLocks after quorums have been created
Creating 4 quorums causes a lot of blocks to be created and signed by
ChainLocks, which then causes timeouts later.
* Increase timeout in wallet-dump.py test
The first dumpwallet is quite slow sometimes, which then makes the
later called dumpwallet throw a wallet locked exception.
9576b01 Enable xvfb in travis to allow running test_bitcoin-qt (Russell Yanofsky)
9e6817e Add new test_bitcoin-qt static library dependencies (Russell Yanofsky)
2754ef1 Add simple qt wallet test sending a transaction (Russell Yanofsky)
b61b34c Add braces to if statements in Qt test_main (Russell Yanofsky)
cc9503c Make qt test compatible with TestChain100Setup framework (Russell Yanofsky)
91e3035 Make test_bitcoin.cpp compatible with Qt Test framework (Russell Yanofsky)
Tree-SHA512: da491181848b8c39138e997ae5ff2df0b16eef2d9cdd0a965229b1a28d4fa862d5f1ef314a1736e5050e88858f329124d15c689659fc6e50fefde769ba24e523
remove line, testing
bitcoin -> dash, testing
bitcoin -> dash, testing
resolve name conflict, testing
bitcoin -> dash
re-add test fixture line
code review, fix tests
Signed-off-by: Pasta <Pasta@dash.org>
move ExceptionInitializer into test_dash_main.cpp
remove witness from nulldummy.py
Signed-off-by: Pasta <Pasta@dash.org>
change error text to match expected
Signed-off-by: Pasta <Pasta@dash.org>
* Manually pull builder base image and let travis retry it on failure
* Split base package installation in Dockerfile.builder into multiple RUN lines
This allows better local cache usage on failure and retry.
* Use travis_retry for docker build
* Add libbacktrace to depends
This is currently only useful to extract symbols. It fails to gather
stacktraces when compiled with MinGW, so we can only use it to get symbol
information from a stack trace which we gathered outside of libbacktrace.
* Add -mbig-obj to CXXFLAGS for MinGW builds
* Implement stacktraces for C++ exceptions
This is a hack and should only be used for debugging. It works by wrapping
the C++ ABI __wrap___cxa_allocate_exception. The wrapper records a backtrace
and stores it in a global map. Later the stacktrace can be retrieved with
GetExceptionStacktraceStr.
This commit also adds handlers to pretty print uncaught exceptions and
signals.
* Use GetPrettyExceptionStr for all unhandled exceptions
* Use --enable-stacktraces in CI for linux32/linux64
* Register exception translators to pretty print exceptions in unit tests
* Catch and print python exceptions when stopping nodes
Otherwise the code at the bottom is never executed when nodes crash,
leading to no output of debug.log files on Travis.
* Remove now unneeded/unused TestCrash methods
* GCC-7 and glibc-2.27 compat code
* Statically link libstdc++ for GCC based builds
Makes sure binaries which are built on a newer build host still work
on older distros.
* Use python3 when installing MacOS native tools
* Move actual build logic out of Travis and upgrade to gcc-7
Travis will now simply call a few scripts which do the actual work.
These scripts will first create a "builder image" which contains the
necessary environment for the actual build. Then scripts are called
inside this builder image to do the build.
This should make us more independant from Travis and also allows us
to do local CI testing.
The build matrix is also moved out of .travis.yml and instead moved
into ci/matrix.sh. This script is sourced with only "BUILD_TARGET" being
set so that it internally can figure out which other environment
variables need to be set.
This commit also upgrades the used GCC version to 7. This is due to the
use of ubuntu:bionic as base image for the builder image.
* Add Jenkinsfiles for regular CI and nightly gitian builds
* Automatically download OSX SDK in gitian-build.sh
* Remove bogus "export MAKEJOBS=-j5"
* Forward cache/src dirs into builder container
Fixes caching issues on Travis.
* fix
* Fail build immediately when building depends took too long
