There were some frustrating problems with the implementation of
-Wwrite-strings. First and foremost, once the positive form of the flag
was passed, it could never be disabled by passing -Wno-write-strings.
Also, the diagnostic engine couldn't in turn use -Wwrite-strings to
control diagnostics (as GCC does) because it was essentially hijacked to
drive the language semantics.
Fix this by giving CC1 a clean '-fconst-strings' flag to enable
const-qualified strings in C and ObjC compilations. Corresponding
'-fno-const-strings' is also added. Then the driver is taught to
introduce '-fconst-strings' in the CC1 command when '-Wwrite-strings'
dominates.
This entire flag is basically GCC-bug-compatibility driven, so we also
match GCC's bug where '-w' doesn't actually disable -Wwrite-strings. I'm
open to changing this though as it seems insane.
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130051 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/Frontend/CompilerInvocation.cpp b/lib/Frontend/CompilerInvocation.cpp
index 3f3c1d8..0dd2093 100644
--- a/lib/Frontend/CompilerInvocation.cpp
+++ b/lib/Frontend/CompilerInvocation.cpp
@@ -577,7 +577,7 @@
if (Opts.WritableStrings)
Res.push_back("-fwritable-strings");
if (Opts.ConstStrings)
- Res.push_back("-Wwrite-strings");
+ Res.push_back("-fconst-strings");
if (!Opts.LaxVectorConversions)
Res.push_back("-fno-lax-vector-conversions");
if (Opts.AltiVec)
@@ -1475,7 +1475,8 @@
Opts.MSCVersion = Args.getLastArgIntValue(OPT_fmsc_version, 0, Diags);
Opts.Borland = Args.hasArg(OPT_fborland_extensions);
Opts.WritableStrings = Args.hasArg(OPT_fwritable_strings);
- Opts.ConstStrings = Args.hasArg(OPT_Wwrite_strings);
+ Opts.ConstStrings = Args.hasFlag(OPT_fconst_strings, OPT_fno_const_strings,
+ Opts.ConstStrings);
if (Args.hasArg(OPT_fno_lax_vector_conversions))
Opts.LaxVectorConversions = 0;
if (Args.hasArg(OPT_fno_threadsafe_statics))