JDevlieghere added a comment.
@amccarth I think this might explain the issues you were having last week.
Without this patch, `find_python_libs_windows` is not setting the expected
version variables. As a result, my check was complaining about a mismatch
between the Python interpreter that LLVM found, and the libs that LLDB found,
while LLDB was actually using a different Python interpreter, found by
`find_python_libs_windows`.
================
Comment at: lldb/cmake/modules/LLDBConfig.cmake:203-204
+ if (NOT CMAKE_CROSSCOMPILING)
+ # Manually split PYTHON_VERSION_STRING because find_python_libs_windows
+ # doesn't set PYTHON_VERSION_MAJOR or PYTHON_VERSION_MINOR.
+ string(REPLACE "." ";" python_version_list ${PYTHON_VERSION_STRING})
----------------
labath wrote:
> JDevlieghere wrote:
> > labath wrote:
> > > If this is the only reason, then maybe we could fix
> > > find_python_libs_windows to set the relevant variables instead (?)
> > FindPythonLibs only sets `PYTHONLIBS_VERSION_STRING`, so we'd still have
> > that to parse the major and minor version of that manually. It was a coin
> > flip between keeping the parsing together and updating
> > `find_python_libs_windows`. I'm happy to do either.
> Yea, I see what you mean. I think I'd go with changing
> `find_python_libs_windows` just to minimize the difference between the
> windows and non-windows branches of the python-finding code (and possible
> side-effects of those differences), but it's not a big deal either way...
Sure, sounds good!
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D65230/new/
https://reviews.llvm.org/D65230
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
