<html><head><meta http-equiv="Content-Type" content="text/html charset=windows-1252"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><br><div><div>On Oct 29, 2013, at 19:14 , Richard Trieu <<a href="mailto:rtrieu@google.com">rtrieu@google.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><blockquote class="gmail_quote" style="font-family: Helvetica; font-size: medium; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: -webkit-auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; margin: 0px 0px 0px 0.8ex; border-left-width: 1px; border-left-color: rgb(204, 204, 204); border-left-style: solid; padding-left: 1ex; "><div style="word-wrap: break-word; "><div>With these points in mind, are there particular concerns about cases where the “CFG won’t check all the code a Sema based warning would”. If you addressed the last point, I think you’d pretty much get the coverage that you want. What do you think?</div><div><br></div></div></blockquote><div style="font-family: Helvetica; font-size: medium; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: -webkit-auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; ">I think that globals and global initializers are not represented in the CFG. That's my main concern about using only the CFG at the moment.</div></blockquote><br></div><div>I don't think it would be difficult to build a CFG from a single global initializer (or member initializer). No one's done it yet, but I don't think that means it can't be done.</div><div><br></div><div>On the more general issue I can see both sides: avoiding extra walks over the AST or CFG is good, knowing what's trivially dead code is good (except when it isn't), and not conflating concerns is good.</div><div><br></div><div>For these particular checks, though (and I haven't looked at the patch, just the checks), I think they are fundamentally syntactic checks, not flow-sensitive ones, and that we will actually get little benefit out of using this information to improve the CFG. Each case is warning about a likely-incorrect boolean expression, which implies that if the user fixed the expression we could easily get a different CFG. That doesn't help answer questions of general policy, but it might at least untangle this patch from the discussion.</div><div><br></div><div>Jordan</div></body></html>