Notes |
|
(0028151)
|
Brad King
|
2012-01-03 16:32
|
|
We have always had to pass compilation flags to the compiler front-end when using it for linking. It matters at least for C++ because some compilers actually do real compilation at link time to do template instantiation (pre-linker). Since we've always done it for that reason we don't know where else it matters.
The behavior would have to be configured on a per-flag, per-toolchain basis. This is non-trivial. It really requires the long-requested feature of abstracting all optional behavior into properties that get mapped to flags as implementation details. |
|
|
(0029233)
|
David Cole
|
2012-04-19 08:42
|
|
Unset target version... This is not going to change unless somebody has a concrete proposal that "fixes it" but also preserves backward compatibility for the cases where the flags are required for linking... |
|
|
(0031257)
|
Amine Chadly
|
2012-10-18 04:35
|
|
What do you think of using a CMAKE_COMPILE_FLAGS_TO_LINKER or whatever cooler name you can come up with ? We could use it as a Boolean (true by default to keep today's behavior) that appends (or not) the compiler flags to the linker
Or we could use CMAKE_COMPILE_AND_LINK_FLAGS to specify flags that would be appended to both CMAKE_CXX_FLAGS and CMAKE_EXE_LINKERS_FLAGS.
Hope this helps. |
|
|
(0041952)
|
Kitware Robot
|
2016-06-10 14:28
|
|
Resolving issue as `moved`.
This issue tracker is no longer used. Further discussion of this issue may take place in the current CMake Issues page linked in the banner at the top of this page. |
|