Attempt to codify the possible error statuses associated with script
validation. script/types.h has been created with the expectation that it will
be part of the public lib interface. The other flag enums will be moved here in
a future commit.
Logging has also been removed in order to drop the dependency on core.h. It can
be re-added to bitcoind as-needed. This makes script verification finally free
of application state and boost!
This is a long chain of errors, and there are likely other changes that could
be made to cope in other places along that chain.
If depends don't build successfully, don't bother trying again for the sake of
better logging. That's likely to hurt more than help. In this case, qt build
failed, and on the second attempt, it appeared to be successful. However, due
to a bad object from an internal gcc error on the first build, the resulting
lib was unusable. This caused bitcoin-qt to not be built, and tests and
packaging which expected bitcoin-qt to be there failed.
The root cause:
Mingw is especially crashy when using -jX, likely compounded by low-memory
environments. I've seen multiple problems with this combo in Gitian as well.
In this case:
i686-w64-mingw32-g++: internal compiler error: Killed (program cc1plus)
...
make[3]: *** [.obj/release/qdrawhelper.o] Error 4
The workaround:
Bump Travis down to using -j2 by default. Additionaly, enable --with-gui for
the windows builds. This will cause configure to fail if qt is not working
while also testing the config flag.
Other failures which may be worth revisiting separately:
- If a depends package fails, maybe remove the workdir so that it doesn't taint
subsequent runs
- See if there's anything repeatable about the ICE when building qt
tl;dr: This solves boost visibility problems for default/release build configs
on non-Linux platforms.
When Bitcoin builds against boost's header-only classes, it ends up with
objects containing symbols that the upstream boost libs also have. Since
Bitcoin builds by default with hidden symbol visibility, it can end up trying
to link against a copy of the same symbols with default visibility.
This is not a problem on Linux because 3rd party static libs are un-exported
by default (--exclude-libs,ALL), but that is not available for MinGW and OSX.
Those platforms (and maybe others?) end up confused about which version to use.
The OSX linker spews hundreds of: "ld: warning: direct access in <foo> to
global weak symbol guard variable for <bar> means the weak symbol cannot be
overridden at runtime. This was likely caused by different translation units
being compiled with different visibility settings."
MinGW's linker complains similarly.
Since the default symbol visibility for Bitcoin is hidden and releases are
built that way as well, build Boost with hidden visibility. Linux builds Boost
this way also, but only for the sake of continuity.
This means that the linker confusion logic is reversed, so the problem will
will now be encountered if Bitcoin is built with --disable-reduce-exports, but
that's better than the current situation.
This allows for a reversal of the current behavior.
This:
CScript foo;
CScriptID bar(foo.GetID());
Becomes:
CScript foo;
CScriptID bar(foo);
This way, CScript is no longer dependent on CScriptID or Hash();
There's only one case where a vector containing a fundamental type is
serialized all-at-once, unsigned char. Anything else would lead to
strange results.
Use a dummy argument to overload in that case.
Similar to the INCLUDES changes in 6b099402b4, split out LIBS into individual
entries for more fine-grained control.
Also add MINIUPNPC_LIBS which was missing before, and hook it up to
executables.
There's a brief race here, the process might've already exited and cleaned up
after itself. If that's the case, reading from the pidfile will harmlessly
fail. Keep those quiet.
Newer mingw supports the features necessary to enable this api, whereas older
versions didn't. However once enabled (automatically by configure), it triggers
an unrelated build bug.
Since it was not enabled previously anyway, and we don't depend on the
functionality, just disable it across the board.
Qt5 is bottled, so configure won't find it without some help. Use
brew to find out its prefix.
Also, qt5 added the host_bins variable to pkg-config, use it.
The entire debug log would be huge, and could cause issues for automated tools
like travis. Printing 200 lines is an initial guess at a reasonable number,
more may be required.
Windows needed a few fixups to get the tests running:
1. bitcoin-tx needs a file extension in Windows. Take this opportunity to
add an env file, which pulls variables out of our build config. This can
be extended as needed, for now it's very simple.
2. After #1, split the args out of the exec key in the test data.
3. Correct the line-endings from windows stdout
The approach from 65f3fa8d1 worked for signing on 10.9.4, but not newer
versions. 10.9.5 (and up) want each framework to stand alone.
Now in addition to copying the plist's from Qt for each framework, we put them
in per-version dirs and only symlink to the latest, rather than using symlinks
for any contents.
One might assume that CAutoFile would be ref-counted so that a copied object
would delay closing the underlying file until all copies have gone out of
scope. Since that's not the case with CAutoFile, explicitly disable copying.
For the all-off build, enable the wallet and debug. This ensures that debug
options will catch wallet problems as well.
In order to make sure the no-wallet path is still tested, disable the wallet
in the other x86_64 build.
Some sources are renamed after download, since the filenames don't play nice
with (for example) gitian. This fixes the rename.
Needed for OSX build as it renames a file.
libminiupnpc changed their required static define to the much more sane
"MINIUPNP_STATICLIB". Sadly, they don't respect the old "STATICLIB" for
back-compat. Define them both since the old one didn't seem to be conflicting
anywhere.
Also go ahead and split out the cppflags so that they can be applied only where
they're needed. This will help us to build dll's from our libs without having
their import/export declspecs poisoned.
Starting with 10.9, Framework versions must be signed individually, rather
than as a single bundle version, in order to be properly codesigned. This
change ensures that the proper plist files and symlinks are present prior to
packaging.
When using clang and ccache, builds spew lots of:
Clang: warning: argument unused during compilation
Upstream bug: https://bugzilla.samba.org/show_bug.cgi?id=8118
This is harmless, bug annoying. If ccache is being used and the
-Qunused-arguments flag is supported (clang), use it.
This was committed previously as 4975ae172 and reverted, because the flags were
applied even if the checks didn't pass. This is the same commit, fixed up to
actually disable the functionality when necessary.
Enabled automatically if boost >= 1.49.
See: https://svn.boost.org/trac/boost/ticket/2309
Also, check for a default visibility attribute, so that we can mark future
api functions correctly.
Enabled automatically if boost >= 1.49.
See: https://svn.boost.org/trac/boost/ticket/2309
Also, check for a default visibility attribute, so that we can mark future
api functions correctly.
When the libpath doesn't line up with the value from config.sub, we don't find
the correct path to boost's libs. This adds a hack to try another path before
giving up.
Should close#3219.
While we're at it, reduce the use of LIBS as well. This makes dependencies
explicit.
Fixes building with (the not-yet-merged) libsecp256k1 as well.
Github-Pull: #4689
Rebased-By: Wladimir J. van der laan <laanwj@gmail.com>
Rebased-From: 909b347 c0e5dda
When QT is linked statically, macdeploy can't infer its paths. While plugins
and frameworks don't need to be packaged, translations still do (for now).
If clock_gettime is implemented outside of libc (librt in this case), configure
would fail when testing boost. Since clock_gettime is not present on all OSs,
boost only uses it when it can. Check for it in librt and add it to LIBS if
found, but don't fail if it's not (since boost won't be expecting it in this
case).
Also, reverse the link order as necessary for static libs.
Note that it's possible that there are other similar cases for boost, which may
be handled the same way.
Cory Fields