| commit | fa654ccc7f0099ffcc63bbed45681b502a77c9f3 | |
| author | Mark Abraham <mark.j.abraham@gmail.com> | |
| Mon, 2 Nov 2015 11:59:39 +0000 (2 12:59 +0100) | ||
| committer | Gerrit Code Review <gerrit@gerrit.gromacs.org> | |
| Wed, 4 Nov 2015 09:13:58 +0000 (4 10:13 +0100) | ||
| tree | 5b53b504fa5cc176cb2f1ab71c964117fcfb3687 | treesnapshot (tar.gz zip) |
| parent | 7e264cc62945b431adf8ce6308210e05efdc2822 | commitdiff |
Fix icc 16 issues
Some minor compilation warnings.
align_bytes for LINCS was not being correctly initialized, and had
value zero. The timing of static initialization can be a problem in
C++, though we'd expect that std::max over two compile-time constants
should resolve at compilation time. This could be a problem of mixing
standard-library ABI on a build slave, but that's an issue a user
could have also. Particularly given that the implementation of
snew_aligned is currently hard-coded to 128 bytes, we lose nothing by
making that consistent and documenting it.
Change-Id: I7db035c355c7f8c66522c4c03853ef2e328515ab
Some minor compilation warnings.
align_bytes for LINCS was not being correctly initialized, and had
value zero. The timing of static initialization can be a problem in
C++, though we'd expect that std::max over two compile-time constants
should resolve at compilation time. This could be a problem of mixing
standard-library ABI on a build slave, but that's an issue a user
could have also. Particularly given that the implementation of
snew_aligned is currently hard-coded to 128 bytes, we lose nothing by
making that consistent and documenting it.
Change-Id: I7db035c355c7f8c66522c4c03853ef2e328515ab