mstorsjo accepted this revision. mstorsjo added a comment. This revision is now accepted and ready to land.
In D154176#4466230 <https://reviews.llvm.org/D154176#4466230>, @HaohaiWen wrote: > In D154176#4466190 <https://reviews.llvm.org/D154176#4466190>, @mstorsjo > wrote: > >> In D154176#4466186 <https://reviews.llvm.org/D154176#4466186>, @mstorsjo >> wrote: >> >>> Ok, thanks for clarifying. However I still don’t understand the “why” >>> aspect here. You’re writing >>> >>>> so that lld specific flags can be append before inputs >>> >>> Are you planning on adding such flags in a later patch, position dependent >>> flags that need to be supplied before input files? >>> >>> Or does the change in order of command line arguments have an effect on the >>> behavior of the linker in this case? >> >> Ok, on second thought, I guess that parsing the vfsoverlay option before the >> inputs makes it have an effect where it didn’t before. Is that right? (Most >> flags in lld aren’t very order dependent since it just checks for any >> occurrence of a flag anywhere among the arguments.) > > I'm planing to add /dwodir to linker when user specify -gsplit-dwarf and > using lld and lto. > The order does not matter. Ok, thanks for clarifying that - since I didn't quite understand how this change would be needed for that option. > I just want it looks more consistent and elegent since most flags are before > inputs so that user can easily look for inputs when using -###. Ok, that sounds reasonable. Please clarify that intent in the commit message, and this seems ok to me. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D154176/new/ https://reviews.llvm.org/D154176 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
