The LLVM coding conventions do suggest using "static" for TU-local functions and only using anonymous namespaces around classes for this (readibility) reason, I believe. You might want to reduce the scope of the anonymous namespace so it just wraps the class(es) and not the functions, then.
On Mon, Dec 2, 2013 at 11:58 PM, Bas van den Berg <[email protected]> wrote: > I did not measure it. The main reason was to narrow the scope so > reading/analysing the code > is easier. > > > On Tue, Dec 3, 2013 at 8:55 AM, Alp Toker <[email protected]> wrote: >> >> >> On 03/12/2013 07:44, Bas van den Berg wrote: >>> >>> CheckImplicitConversion() can be static to allow more compiler >>> optimizations. >>> >>> --- >>> lib/Sema/SemaChecking.cpp | 2 +- >>> 1 file changed, 1 insertion(+), 1 deletion(-) >>> >>> diff --git a/lib/Sema/SemaChecking.cpp b/lib/Sema/SemaChecking.cpp >>> index 0b95c48..830936d 100644 >>> --- a/lib/Sema/SemaChecking.cpp >>> +++ b/lib/Sema/SemaChecking.cpp >>> @@ -5210,7 +5210,7 @@ void CheckImplicitArgumentConversions(Sema &S, >>> CallExpr *TheCall, >>> } >>> } >>> -void CheckImplicitConversion(Sema &S, Expr *E, QualType T, >>> +static void CheckImplicitConversion(Sema &S, Expr *E, QualType T, >>> SourceLocation CC, bool *ICContext = 0) { >>> if (E->isTypeDependent() || E->isValueDependent()) return; >>> >> >> >> These functions are in an anonymous namespace so should get the same >> optimisations as 'static'. >> >> Did you see a difference in generated code / performance or was this just >> a speculative change? >> >> Alp. >> >> >> -- >> http://www.nuanti.com >> the browser experts >> > > > _______________________________________________ > cfe-commits mailing list > [email protected] > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits > _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
