commit 72ca0d8664ebabbefaf00aba205570ebbc877beb (HEAD -> develop, origin/develop, origin/HEAD)
Author: Peter Dimov <pdimov@gmail.com>
Date: Sat Dec 28 18:47:11 2019 +0200
Add CMake install support
and on top of boost/multiprecision commit
commit 632f68400c28e85d62171c40b16d71699ac6dfd7 (HEAD -> develop, origin/develop, origin/HEAD)
Author: Christopher Kormanyos <ckormanyos@yahoo.com>
Date: Fri Dec 13 22:46:54 2019 +0100
Repair excessive right shift in rotr. Change benchmark units to kops/s.
we have the following problem in boost.math:
g++-9 -g --std=c++17 -ffast-math -fno-finite-math-only -march=native -Wfatal-errors -MMD -fvisibility=hidden -O3 -I../../ -I/usr/local/include -o example/daubechies_coefficients.x example/daubechies_coefficients.cpp -pthread -L/usr/local/lib
In file included from ../../boost/multiprecision/cpp_int.hpp:15,
from ../../boost/multiprecision/cpp_bin_float.hpp:9,
from example/daubechies_coefficients.cpp:13:
../../boost/core/empty_value.hpp:142:1: error: 'BOOST_INLINE_VARIABLE' does not name a type
142 | BOOST_INLINE_VARIABLE
| ^~~~~~~~~~~~~~~~~~~~~
compilation terminated due to -Wfatal-errors.
Boost.math commit (though it is, I think, largely irrelevant):
commit b3e1f4d8a11b3ca4329464a60ec726b5acea6ebe (HEAD -> develop, origin/develop, origin/HEAD)
Merge: ce48f975e e5f019c62
Author: Nick <NAThompson@users.noreply.github.com>
Date: Sat Dec 28 09:32:13 2019 -0500
Merge pull request #290 from boostorg/distribution_entropy
Entropy of distributions.
git bisect shows the offending commit:
git bisect bad
38f71361fbbba1e8bf23876e218e8622e6cbc406 is the first bad commit
commit 38f71361fbbba1e8bf23876e218e8622e6cbc406
Author: Andrey Semashev <Lastique@users.noreply.github.com>
Date: Sun Nov 24 05:00:59 2019 +0300
Add a convenience instance of empty_init_t (#65)
* Added a convenience instance of empty_init_t.
Closes https://github.com/boostorg/core/issues/63.
:040000 040000 b06283a1bec80f56b9a76e7a872aa5631863f487 719b9902bfadfa5ba19ce86c480373794b50aa9e M include
On top of
boost/core
commit:and on top of
boost/multiprecision
commitwe have the following problem in
boost.math
:Boost.math commit (though it is, I think, largely irrelevant):
git bisect
shows the offending commit: