ldionne added inline comments. Herald added a project: All.
================ Comment at: cfe/trunk/include/clang/Driver/ToolChain.h:312 + virtual std::string getCompilerRTPath() const; + ---------------- By introducing a second way of getting the path to compiler-rt, this created inconsistencies where some platforms that implement `getCompilerRT(...)` now won't return the right value for `getCompilerRTPath()`. Is this the intent: `getCompilerRT() == append(getCompilerRTPath(), basename(getCompilerRT(...))`? If so, I think it would make sense to refactor this a bit so that we have: ``` getCompilerRTDir(); // What's currently called getCompilerRTPath(). This would be overriden by toolchains. getCompilerRTBasename(args...); // Exists currently, but we would override this in the toolchains instead of having a default implementation. getCompilerRT(args...); // This would basically do append(getCompilerRTDir(), getCompilerRTBasename(args...)) ``` WDYT? Repository: rL LLVM CHANGES SINCE LAST ACTION https://reviews.llvm.org/D35820/new/ https://reviews.llvm.org/D35820 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
