[PATCH] D59923: [Driver] Simplify -g level computation and its interaction with -gsplit-dwarf

[Paul Robinson via Phabricator via cfe-commits]

probinson added a comment.

Commentary questions.



================
Comment at: lib/Driver/ToolChains/Clang.cpp:3166
+      // If you say "-gsplit-dwarf -gline-tables-only", -gsplit-dwarf loses.
+      // But -gsplit-dwarf is not a g_Group option, hence we have to check the
+      // order explicitly. If -gsplit-dwarf wins, we fix DebugInfoKind later.
----------------
Regarding `But -gsplit-dwarf is not a g_Group option, hence we have to check the order explicitly.` It looks like when you added OPT_gsplit_dwarf to the if-statement above, we no longer have to check it explicitly?


================
Comment at: lib/Driver/ToolChains/Clang.cpp:3167
+      // But -gsplit-dwarf is not a g_Group option, hence we have to check the
+      // order explicitly. If -gsplit-dwarf wins, we fix DebugInfoKind later.
+      // This gets a bit more complicated if you've disabled inline info in
----------------
Regarding `If -gsplit-dwarf wins, we fix DebugInfoKind later.`
Is that still true?  It looks like -gsplit-dwarf is handled here now, because you removed the compensating code from lines 3268-3269.


================
Comment at: lib/Driver/ToolChains/Clang.cpp:3253
   // -gsplit-dwarf should turn on -g and enable the backend dwarf
   // splitting and extraction.
   if (T.isOSBinFormatELF()) {
----------------
This block of code no longer turns on -g?


Repository:
  rC Clang

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D59923/new/

https://reviews.llvm.org/D59923