[flang-commits] [clang] [flang] [Driver] Move CommonArgs to a location visible by the Frontend Drivers (PR #142800)

[Tarun Prabhu via flang-commits]

================
@@ -3167,3 +3167,29 @@ void tools::handleInterchangeLoopsArgs(const ArgList &Args,
                    options::OPT_fno_loop_interchange, EnableInterchange))
     CmdArgs.push_back("-floop-interchange");
 }
+
+std::optional<StringRef> tools::ParseMPreferVectorWidthOption(
+    clang::DiagnosticsEngine &Diags, const llvm::opt::ArgList &Args,
+    ArgStringList &CmdArgs, bool isCompilerDriver) {
+  // If this was invoked by the Compiler Driver, we pass through the option
+  // as-is. Otherwise, if this is the Frontend Driver, we want just the value.
+  StringRef Out = isCompilerDriver ? "-mprefer-vector-width=" : "";
+
+  Arg *A = Args.getLastArg(clang::driver::options::OPT_mprefer_vector_width_EQ);
+  if (!A)
+    return std::nullopt;
+
+  StringRef Value = A->getValue();
+  unsigned Width __attribute((uninitialized));
+
+  // Only "none" and Integer values are accepted by
+  // -mprefer-vector-width=<value>.
+  if (Value != "none" && Value.getAsInteger(10, Width)) {
+    Diags.Report(clang::diag::err_drv_invalid_value)
+        << A->getOption().getName() << Value;
+    return std::nullopt;
+  }
+
+  CmdArgs.push_back(Args.MakeArgString(Out + Value));
----------------
tarunprabhu wrote:

The alternative would be to not add to `CmdArgs` here, but to do it in the caller instead. I think that is ok too. It would be a bit of a departure from the way the rest of the code works, but it feels like a more "natural" way of handling this. The rest of the code could, gradually, be made more consistent with this approach.

https://github.com/llvm/llvm-project/pull/142800