From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [PATCH] Re: LLVM and PSEUDO_REG/PSEUDO_PHI Date: Sun, 28 Aug 2011 15:33:34 -0400 Message-ID: <4E5A980E.70308@garzik.org> References: <4E58731A.7010708@garzik.org> <4E58AE9E.1090601@garzik.org> <4E59478C.9000504@garzik.org> <4E5A129F.1090801@garzik.org> <20110828175255.GA10203@leaf> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail-qw0-f46.google.com ([209.85.216.46]:41076 "EHLO mail-qw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751420Ab1H1Tdg (ORCPT ); Sun, 28 Aug 2011 15:33:36 -0400 Received: by qwk3 with SMTP id 3so2916547qwk.19 for ; Sun, 28 Aug 2011 12:33:35 -0700 (PDT) In-Reply-To: Sender: linux-sparse-owner@vger.kernel.org List-Id: linux-sparse@vger.kernel.org To: Linus Torvalds Cc: Josh Triplett , Pekka Enberg , Sparse Mailing-list On 08/28/2011 02:23 PM, Linus Torvalds wrote: > On Sun, Aug 28, 2011 at 10:52 AM, Josh Triplett wrote: >>> >>> You should consider all types to be just "blocks of memory", and >>> sparse has already calculated all offsets etc for you. As far as LLVM >>> is concerned, the memory has no structure, it's just a blob. >> >> I suspect LLVM's optimization passes won't particularly care for that >> approach. > > That's fine. We've already done the CSE and alias analysis on the > thing. And as mentioned, trying to turn overlapping (or partial) > accesses into some "named accesses" is just *wrong*. They weren't > named in the C code either. They are accesses through pointer > arithmetic. Trying to make them be somehow named would just be crazy. LLVM just cares about layout. It does not care about naming (or lack thereof). Nobody/nothing is trying to make them named. Jeff