On Wed, 10 Jan 2024 07:44:47 GMT, Matthias Baesken <mbaes...@openjdk.org> wrote:

> That might be a bit tricky, because on my macOS test machine, with a self 
> compiled autoconf 2.72, I did not even get this strange behavior. But it was 
> a different macOS version . So it seems to be a problem of very specific 
> environment.

I see, that's unfortunate (and confusing). But such is the case with autoconf 
after all I suppose :)



> Can we set this somewhere in the OpenJDK m4 files, would be better for us 
> than on the shell before configure.
> Should we set this in general , or do we want autoconf to add sometimes flags 
> like these to CXX (or maybe CC) ?

I believe can be set in m4 files as well (though it may need quoting, not 
sure). It would need to go before AC_PROG_CC and AC_PROG_CXX, and I also made a 
mistake earlier, we would also need to set ac_prog_cxx_stdcxx as well, not only 
ac_prog_cc_stdc (ac_prog_cc_stdc is for C, ac_prog_cxx_stdcxx is for C++). 
Perhaps something like the following might work:


readonly ac_prog_cc_stdc=
readonly ac_prog_cxx_stdcxx=

#
# Setup the compilers (CC and CXX)
#
TOOLCHAIN_FIND_COMPILER([CC], [C], $TOOLCHAIN_CC_BINARY)
# Now that we have resolved CC ourself, let autoconf have its go at it
AC_PROG_CC([$CC])

TOOLCHAIN_FIND_COMPILER([CXX], [C++], $TOOLCHAIN_CXX_BINARY)
# Now that we have resolved CXX ourself, let autoconf have its go at it
AC_PROG_CXX([$CXX])


Setting ac_prog_cc_stdc and ac_prog_cxx_stdcxx to empty values should work too, 
since the test explicitly only does the check if the value is a literal no. The 
readonly is because in the compiled configure script both are set to no 
unconditionally. It doesn't have to explicitly be right before AC_PROG_CC or 
AC_PROG_CXX, but it has to appear in front of them (In practice, before 
TOOLCHAIN_DETECT_TOOLCHAIN_CORE). But I stress that this is only a band aid 
solution. I don't think we have ever relied on autoconf adding options to our 
CFLAGS and CXXFLAGS thus far, though to confirm you might have to ask Erik or 
Kim

-------------

PR Comment: https://git.openjdk.org/jdk/pull/17301#issuecomment-1884357078

Reply via email to