This mostly looks good to me. One comment, though: + !RetValExp->isValueDependent() &&
The returns_nonnull check doesn't bother to check if it's value-dependent first. Is that necessary? Should that be folded in somewhere else? Jordan On Jan 23, 2014, at 2:24 , Artyom Skrobov <[email protected]> wrote: > Jordan, good catch that this can be combined; attaching for a review. > > > > -----Original Message----- > From: Jordan Rose [mailto:[email protected]] > Sent: 22 January 2014 17:21 > To: Ted Kremenek > Cc: llvm cfe; Richard Smith; Artyom Skrobov > Subject: Re: r199790 - Add basic checking for returning null from > functions/methods marked 'returns_nonnull'. > > Should this be combined with the check that a throwing operator new never > returns non-null? (r199452) > > > On Jan 21, 2014, at 22:10 , Ted Kremenek <[email protected]> wrote: > >> Author: kremenek >> Date: Wed Jan 22 00:10:28 2014 >> New Revision: 199790 >> >> URL: http://llvm.org/viewvc/llvm-project?rev=199790&view=rev >> Log: >> Add basic checking for returning null from functions/methods marked > 'returns_nonnull'. >> >> This involved making CheckReturnStackAddr into a static function, which >> is now called by a top-level return value checking routine called >> CheckReturnValExpr. >> >> Modified: >> cfe/trunk/include/clang/Basic/DiagnosticSemaKinds.td >> cfe/trunk/include/clang/Sema/Sema.h >> cfe/trunk/lib/Sema/SemaChecking.cpp >> cfe/trunk/lib/Sema/SemaStmt.cpp >> cfe/trunk/test/Sema/nonnull.c >> cfe/trunk/test/SemaObjC/nonnull.m
combined-nonnull.patch
Description: Binary data
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
