From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [RFC][PATCH 0/3] implement pseudo->ctype Date: Sat, 09 Jun 2012 08:06:46 -0400 Message-ID: <4FD33C56.1080207@garzik.org> References: <1338792878-25898-1-git-send-email-xi.wang@gmail.com> <2055A8E4-5846-4913-9E57-5BBDCB868284@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail-yx0-f174.google.com ([209.85.213.174]:60958 "EHLO mail-yx0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751515Ab2FIMGu (ORCPT ); Sat, 9 Jun 2012 08:06:50 -0400 Received: by yenm10 with SMTP id m10so1798010yen.19 for ; Sat, 09 Jun 2012 05:06:50 -0700 (PDT) In-Reply-To: <2055A8E4-5846-4913-9E57-5BBDCB868284@gmail.com> Sender: linux-sparse-owner@vger.kernel.org List-Id: linux-sparse@vger.kernel.org To: Xi Wang Cc: Pekka Enberg , linux-sparse@vger.kernel.org, Benjamin Herrenschmidt , Christopher Li , Linus Torvalds On 06/09/2012 07:52 AM, Xi Wang wrote: > Besides, the phi nodes in sparse and LLVM are different. LLVM's > phi 1) requires one entry for each predecessor, and 2) must stay > at the beginning of a block. I didn't find an easy way to translate > sparse's phi to LLVM's, so instead generated load/store for phi > nodes. That is reasonable. Jeff