http://llvm.org/bugs/show_bug.cgi?id=20336
Hans Wennborg <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #9 from Hans Wennborg <[email protected]> --- (In reply to comment #8) > (In reply to comment #7) > > (In reply to comment #6) > > > Alternatively, could --show-includes use errs() instead of outs(), if it's > > > something that doesn't disable the rest of compilation? This is how > > > everything else I'm aware of in clang and other llvm tools work, which is > > > why the raw_ostream code in question has been able to last almost 4 years > > > in > > > the tree by now. > > > > Unfortunately not. MSVC prints --show-includes to stdout, and build tools > > rely on that, so clang-cl should do the same. > > Ok. Does MSVC support "-o -" or something like it? If not, perhaps you could > just issue an error if the user requests "-o -" and --show-includes at the > same time. I think it's really just /E (preprocess to stdout) that results in this situation. I've committed r213589 which makes us ignore /showIncludes (with the usual warning) when combined with /E or /EP. -- You are receiving this mail because: You are on the CC list for the bug.
_______________________________________________ LLVMbugs mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs
