From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pavel Roskin Subject: Re: Sparse crash when mixing int and enum in ternary operator Date: Tue, 09 Mar 2010 20:09:22 -0500 Message-ID: <1268183362.2615.17.camel@mj> References: <1268097872.16227.10.camel@mj> <1268165517.23196.22.camel@mj> <201003092129.16118.kdudka@redhat.com> <201003100030.50165.kdudka@redhat.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from c60.cesmail.net ([216.154.195.49]:33272 "EHLO c60.cesmail.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751791Ab0CJBJY (ORCPT ); Tue, 9 Mar 2010 20:09:24 -0500 In-Reply-To: <201003100030.50165.kdudka@redhat.com> Sender: linux-sparse-owner@vger.kernel.org List-Id: linux-sparse@vger.kernel.org To: Kamil Dudka Cc: Josh Triplett , Christopher Li , linux-sparse@vger.kernel.org On Wed, 2010-03-10 at 00:30 +0100, Kamil Dudka wrote: > Hopefully I got it. The promissed fix, which includes new test-cases for ?:, > is attached. Looks good! "make check" works. Checking the whole kernel produces no core dumps. The warnings appear to be legitimate. I wonder is the quirk handling could be skipped for EXPR_SELECT, but I'm fine with it if it's harmless. When would expr or type be NULL? We could check type in warn_for_enum_conversions(). -- Regards, Pavel Roskin