From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 11 Dec 2002 00:40:50 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 11 Dec 2002 00:40:50 -0500 Received: from ore.jhcloos.com ([64.240.156.239]:33796 "EHLO ore.jhcloos.com") by vger.kernel.org with ESMTP id ; Wed, 11 Dec 2002 00:40:49 -0500 To: linux-kernel@vger.kernel.org Subject: incompatable pointer type warnings on some archs From: "James H. Cloos Jr." Date: 11 Dec 2002 00:48:26 -0500 Message-ID: User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.1 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I see that find_next_zero_bit() has a first arg of void* in about half of the archs and unsigned long* in the rest (incl asm-i386/bitops.h). Looking at incompatable pointer type warnings in a recent compile, I found one where the caller was passing a u64*, thus the error. Should all of the archs use a void* for this, or is there some reason not to? test_and_set_bit() and test_and_clear_bit() also have the same issue. -JimC