RFR: 8295990: Improve make handling of strip flags [v2]

[Julian Waters]

On Sat, 29 Oct 2022 07:12:40 GMT, Julian Waters <jwaters at openjdk.org> wrote:

>> Strip currently has its flags set by guessing based on the OS, it would be more appropriate to instead set them based on properly checking the strip binary instead of speculating what the correct flags are based on the compiler and OS.
>
> Julian Waters has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains 13 additional commits since the last revision:
> 
>  - Merge branch 'openjdk:master' into patch-1
>  - Merge branch 'openjdk:master' into patch-1
>  - Add more documentation
>  - Replace with the actual new check
>  - Little documentation doesn't hurt
>  - Prepare for actual changes
>  - Update flags-other.m4
>  - This is getting annoying
>  - -V doesn't work with MacOS strip
>  - Get output from strip to search for useful information
>  - ... and 3 more: https://git.openjdk.org/jdk/compare/633a3ca9...43f615d4

Tested successfully on Linux with gcc and clang, also with the platform specific strip utils targeting MacOS. Windows with Visual C++ (but importantly only with Visual C++) correctly does not perform the check since the toolchain never sets strip. This change seems to have unintentionally split the strip version from the toolchain type as well, instead categorizing it into 3 independent groups: GNU-like, IBM, and Xcode, which may be helpful in the future should any changes to strip be desired

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

PR: https://git.openjdk.org/jdk/pull/10882