[PATCH] D131007: [NFCI] Refactor how KeywordStatus is calculated

[Erich Keane via Phabricator via cfe-commits]

erichkeane marked 3 inline comments as done.
erichkeane added inline comments.


================
Comment at: clang/lib/Basic/IdentifierTable.cpp:169
+  case KEYALTIVEC:
+    return LangOpts.AltiVec ? KS_Enabled : KS_Unknown;
+  case KEYBORLAND:
----------------
aaron.ballman wrote:
> Should this one be "extension" instead?
It was not before, and the intent is for this to be NFC:

`  if (LangOpts.AltiVec && (Flags & KEYALTIVEC)) return KS_Enabled;`

I'm not sure the logic that was applied though.  The only difference seems to be whether we emit a 'extension used' diagnostic.


================
Comment at: clang/lib/Basic/IdentifierTable.cpp:173
+  case KEYOPENCLC:
+    return LangOpts.OpenCL && !LangOpts.OpenCLCPlusPlus ? KS_Enabled
+                                                        : KS_Unknown;
----------------
aaron.ballman wrote:
> Is this an extension as well? I'll stop asking individually and just ask generally, when should extension be used?
In this case, when it was before :D  I have no idea what our opinion on that is at all though.




CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D131007/new/

https://reviews.llvm.org/D131007