From mboxrd@z Thu Jan 1 00:00:00 1970 From: Linus Torvalds Subject: Re: [PATCHSET] fouled-bitwise handling Date: Sun, 1 Oct 2006 10:33:44 -0700 (PDT) Message-ID: References: <20061001172253.GX29920@ftp.linux.org.uk> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-path: Received: from smtp.osdl.org ([65.172.181.4]:23261 "EHLO smtp.osdl.org") by vger.kernel.org with ESMTP id S932118AbWJAReL (ORCPT ); Sun, 1 Oct 2006 13:34:11 -0400 In-Reply-To: <20061001172253.GX29920@ftp.linux.org.uk> Sender: linux-sparse-owner@vger.kernel.org List-Id: linux-sparse@vger.kernel.org To: Al Viro Cc: linux-sparse@vger.kernel.org On Sun, 1 Oct 2006, Al Viro wrote: > > Hrm... The gut feeling: we mangle expression trees too much to be able > to do that decently. Having looked at it a bit now, I think I agree. We warn about the whole "~" thing too early, so we'd have to remove that warning anyway (easy enough to do), but then we'd have to track the ops that don't like using a "~" because of its C semantics. And at that point, your "fouled" bit is actually probably simpler. Let me think about this a bit more, but I'm starting to think that you're right. As usual. Linus