From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christopher Li Subject: [PATCH 0/2] V2 llvm constant value pseudo size fix Date: Mon, 13 Nov 2017 23:27:54 +0800 Message-ID: Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Return-path: Received: from mail-yw0-f179.google.com ([209.85.161.179]:43205 "EHLO mail-yw0-f179.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753003AbdKMP1z (ORCPT ); Mon, 13 Nov 2017 10:27:55 -0500 Received: by mail-yw0-f179.google.com with SMTP id y75so13727070ywg.0 for ; Mon, 13 Nov 2017 07:27:55 -0800 (PST) Sender: linux-sparse-owner@vger.kernel.org List-Id: linux-sparse@vger.kernel.org To: Luc Van Oostenryck , Dibyendu Majumdar , Linux-Sparse Per Luc's feed back. This is the more complete version of the pseudo-size-value branch. I add the second patch to allow sparse-llvm to use the new size store in constant value pseudo. The whole series is very simple, only two patches. It directly address the problem of sparse-llvm wasn't able to get the integer size from constant value pseudo. I am consider this as better alternative compare to OP_PUSH. - OP_PUSH break byte code compatibility at source level, this does not. - OP_PUSH has more instruction (more memory). - OP_PUSH only cover constant value used in function call. this cover all usage of constant value. It pass the "make check" test ( I have to fix a problem with the bool type which is 1 bit integer). The V1 pass the kernel checker test with no change to output. Dibyendu, can you give this series some test since you report it? The git url: https://git.kernel.org/pub/scm/devel/sparse/chrisl/sparse.git/log/?h=pseudo-sized-value-v2 Chris