https://sourceware.org/bugzilla/show_bug.cgi?id=33801
--- Comment #12 from Sourceware Commits <cvs-commit at gcc dot gnu.org> --- The master branch has been updated by Alan Modra <[email protected]>: https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=666edc6b61343aacac7de82058f57b8e6597c0b3 commit 666edc6b61343aacac7de82058f57b8e6597c0b3 Author: Alan Modra <[email protected]> Date: Sat Jan 17 18:43:06 2026 +1030 PR 33801 strip doesn't remove .gnu.lto_.opts OK, so people want to be able to partly destroy gcc slim LTO object files. I don't see why not, despite HJ's policy-setting slim lto strip -R tests. I also don't see why it matters that objcopy/strip won't similarly break LLVM objects (pr33271). So out goes the slim lto strip -R tests, and the strip-debug test expecting slim lto files won't be modified in any way. strip-debug now removes a FILE symbol, which isn't unreasonable. PR 33801 PR 33271 PR 33246 binutils/ * objcopy.c (copy_archive, copy_file): Do not special case gcc LTO IR files. ld/ * testsuite/ld-plugin/lto-binutils.exp: Remove slim lto strip -R tests. (run_pr33246_test): Don't compare slim object files after strip-debug. * testsuite/ld-plugin/strip-1a-s-all.nd: Delete. -- You are receiving this mail because: You are on the CC list for the bug.
