[PATCH] D59919: [Attributor] Deduce "returned" argument attribute

nicholas added a comment. > CHANGED: build-libcalls NumNoUnwind > 4526 -> 3382 ( -25.276%) Why did the number of nounwinds drop? Comment at: llvm/lib/Transforms/IPO/Attributor.cpp:115 +template +static void

[PATCH] D48100: Append new attributes to the end of an AttributeList.

nicholas added inline comments. Comment at: lib/Sema/SemaOverload.cpp:6194 static SmallVector getOrderedEnableIfAttrs(const FunctionDecl *Function) { SmallVector Result; This function shouldn't be necessary any more. All it's doing now is making an

[PATCH] D40850: suppress undefined-template warnings when the pattern is declared in a system header

nlewycky added a comment. Ping! https://reviews.llvm.org/D40850 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D40850: suppress undefined-template warnings when the pattern is declared in a system header

nlewycky created this revision. The way to fix an undefined-template warning is to add lines to the header file that defines the template pattern. We should suppress the warnings when the template pattern is in a system header because we don't expect users to edit those.

[PATCH] D31839: make -Winteger-overflow find overflows in function arguments

nlewycky added a comment. Ping! https://reviews.llvm.org/D31839 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D31839: make -Winteger-overflow find overflows in function arguments

nlewycky added a comment. Ping! https://reviews.llvm.org/D31839 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D31839: make -Winteger-overflow find overflows in function arguments

nlewycky updated this revision to Diff 97253. nlewycky added a comment. Use an RAII object to always evaluate the arguments, except if HandleFunctionCall does it. https://reviews.llvm.org/D31839 Files: lib/AST/ExprConstant.cpp test/Sema/integer-overflow.c Index:

[PATCH] D32675: in expression evaluator, treat non-literal types as discarded value expressions if EvalInfo says to continue evaluating them

nlewycky created this revision. Make the integer overflow evaluator continue into expressions with non-literal types, notably void. In passing it fixes a crash attempting to codegen: struct A { char x; }; struct B : virtual A {}; A = ((A&)*(B*)0); which we nearly have a test for except

[PATCH] D32412: analyze all kinds of expressions for integer overflow

nlewycky updated this revision to Diff 97179. nlewycky added a comment. [No changes, just full context this time.] https://reviews.llvm.org/D32412 Files: include/clang/Sema/Sema.h lib/AST/ExprConstant.cpp lib/Sema/SemaChecking.cpp

[PATCH] D32412: analyze all kinds of expressions for integer overflow

nlewycky added a comment. > If we're now catching integer overflow in more cases, please add some > relevant testcases. Both more and fewer. More because we no longer have a whitelist of three kinds of expressions that we recurse into. Fewer because we no longer call IgnoreParenCasts() on the

[PATCH] D32412: analyze all kinds of expressions for integer overflow

nlewycky updated this revision to Diff 97178. nlewycky edited the summary of this revision. nlewycky added a comment. Rebase. Now that ObjCBoxedExpr change is in, we can remove Sema::CheckForIntOverflow entirely. https://reviews.llvm.org/D32412 Files: include/clang/Sema/Sema.h

[PATCH] D32410: change the way the expr evaluator handles objcboxedexpr

nlewycky updated this revision to Diff 97173. nlewycky added a comment. If the boxing method can't be constexpr then we can never evaluate it for a constant value, which means that we should unconditionally return Error, and use noteFailure to decide whether to visit the subexpr.

[PATCH] D32455: detect integer overflow inside arms of conditional operator with non-constant expression

nlewycky created this revision. Herald added subscribers: rengolin, aemerson. Descend into both the true and false expressions of a ConditionalOperator when the condition can't be evaluated and we're in an evaluation-mode that says we should continue evaluating.

[PATCH] D32412: analyze all kinds of expressions for integer overflow

nlewycky created this revision. Remove clang::Sema::CheckForIntOverflow(E) by calling into E->EvaluateForOverflow instead. CheckForIntOverflow implemented a whitelist of top-level expressions to check, currently BinaryOperator and InitListExpr. Two changes are made to avoid regression with the

[PATCH] D32410: change the way objcboxedexpr is handled

nlewycky created this revision. Make ObjCBoxedExpr less of a special case, teach the expression evaluator to handle it in general, sometimes descending through to its subexpr. Remove the code that called CheckForIntOverflow from outside BuildObjCBoxedExpr, leaving its only caller

[PATCH] D31839: make -Winteger-overflow find overflows in function arguments

nlewycky added a comment. In https://reviews.llvm.org/D31839#724551, @ahatanak wrote: > OK, thanks for looking into it. Warnings for ObjCMessageExpr can probably be > implemented in a separate patch. > > It looks like clang still doesn't issue overflow warnings when the called > functions have

[PATCH] D31839: make -Winteger-overflow find overflows in function arguments

nlewycky added a comment. In https://reviews.llvm.org/D31839#722763, @ahatanak wrote: > Is it possible to fix ObjCMessageExpr too while you are in here? I looked into this, but it turns out to be different enough to belong in a separate patch. An ObjCMessageExpr has void type which means that

[PATCH] D31839: make -Winteger-overflow find overflows in function arguments

nlewycky created this revision. When checkingForOverflow(), look through call arguments (and the callee itself if calculated). Make the statement visitor go through ObjCBoxedExpr by default because it has a single subexpr node. Don't do that when looking for a pointer object, stop because the

[PATCH] D29915: emit constant expression for new expression array size if it is one

nlewycky created this revision. When the new expr's array size is an ICE, emit it as a constant expression. This bypasses integer sanitization checks which are redundant on the expression since it's been checked by Sema. Fixes a clang codegen assertion on "void test() { new int[0+1]{0}; }"