| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Co-Authored-By: Hannes Rantzsch <[email protected]>
|
|
|
|
| |
Co-Authored-By: Hannes Rantzsch <[email protected]>
|
|
|
|
| |
Co-Authored-By: Hannes Rantzsch <[email protected]>
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| | |
Starting in OpenSSL 3.0, most of the functionality which we need to
implement the OpenSSL provider is deprecated. Rather than reimplement
the whole provider in order to allow it to continue to work in the
future, just remove it. Efforts would be better spent doing more
optimization work rather than chasing OpenSSL's API changes.
|
|/
|
|
| |
Co-Authored-By: Hannes Rantzsch <[email protected]>
|
| |
|
| |
|
| |
|
| |
|
|\ |
|
| | |
|
|/ |
|
| |
|
|
|
|
| |
Ref #2622
|
|
|
|
|
|
| |
Bakefile still only supports up to VC 2013 which we no longer support at all,
and the upstream looks pretty dead, so there doesn't seem to be much need to
continue supporting this.
|
| |
|
|
|
|
| |
This allows it to run concurrently with the compiler.
|
|
|
|
| |
Otherwise someone using eg ppc64le would not get the expected flags.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Verify we install exactly the includes we expect
Verify we install at least one library that looks like our name.
(This check could be tightened)
Change the install target to take all values via the build_config.json
instead of reflecting them through the makefile.
|
|
|
|
| |
See https://github.com/randombit/botan/pull/2526 for details
|
|
|
|
| |
Then the install target can use that instead of listdir
|
|
|
|
| |
Should be YYYYMMDD make some attempt to check this.
|
|
|
|
| |
GH #2496
|
| |
|
| |
|
|
|
|
| |
MinGW shell isn't happy about paths with backslashes
|
|
|
|
| |
Otherwise things will fail if python is not in PATH
|
|
|
|
|
|
|
|
|
| |
Now modules default to internal headers instead of defaulting to public; making
a new public API should be a visible and intentional choice.
Brings the public header count from over 300 to around 150.
Also removes the deprecated tls_blocking interface
|
|
|
|
|
|
|
|
|
| |
This extends the versioning from just an integer trip to also
support a suffix of the form -{alpha,beta,rc}N
Also fix a problem with reproducible releases caused by Python tarfile
switching its default format from GNU to PAX. Use PAX for the releases
that were (unintentionally) released as PAX and GNU for everything else.
|
|
|
| |
The 'compiler_include_dirs' variable contains include directories which needs to be setup in 'cmake.in' as 'COMPILER_INCLUDE_DIRS'
|
|
|
|
|
|
|
|
| |
If you take such amalgamation and insert it into a binary project
it won't behave as one might expect. Add a warning to provide a hint
to the user.
See GH #2373
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
This is a lot of complications just to work around a user
misaprehension about ld paths. And likely most macOS users are getting
it via Homebrew anyway.
Taken from #1505
|
|
|
|
| |
Fixes #2286. See also #2285.
|
|
|
|
|
|
| |
Allow multiple uses of --extra-cxxflags
See GH #2254 and #2256
|
|
|
|
|
|
|
|
|
|
| |
This is as if --single-file-amalgamation was always used, except
also now botan_all_internal.h is not created.
This effectively drops support for very old GCC/Clang in the
amalgamation (only). GCC 5+ and Clang 3.8+ support the target
attribute and work fine. MSVC not affected since it doesn't need
such attributes in the first place.
|
|
|
|
|
|
|
|
| |
We try to emit target annotations but generated incorrect input for
ARM and POWER because the -mfoo flags don't always match up with
target("foo") attribute names.
GH #2241
|
|
|
|
|
| |
Hide it in the configure.py --help output and add a big warning
block in the documentation.
|
|
|
|
| |
See #2226
|
|
|
|
|
|
| |
Also fix the zlib basename for Windows.
Resolves #2210
|
| |
|
| |
|
|
|
|
|
|
|
| |
Since most Windows make tools such as jom and nmake don't support it.
And a GCC build should be going for MinGW instead.
Closes #2160
|
|
|
|
| |
It warns or apparently with certain versions errors. (GH #2179)
|
|
|
|
|
|
|
|
| |
Useful when building an amalgamation which must build on multiple
targets.
Though of course it gives up any CPU optimizations so has pretty
major performance impact.
|
|\ |
|
| | |
|