From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1031157AbXDZKl0 (ORCPT ); Thu, 26 Apr 2007 06:41:26 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1031160AbXDZKl0 (ORCPT ); Thu, 26 Apr 2007 06:41:26 -0400 Received: from gprs189-60.eurotel.cz ([160.218.189.60]:54730 "EHLO amd.ucw.cz" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1031157AbXDZKlZ (ORCPT ); Thu, 26 Apr 2007 06:41:25 -0400 Date: Thu, 26 Apr 2007 12:40:52 +0200 From: Pavel Machek To: Johannes Berg Cc: Linus Torvalds , Nick Piggin , Mike Galbraith , linux-kernel@vger.kernel.org, Thomas Gleixner , Con Kolivas , suspend2-devel@lists.suspend2.net, Ingo Molnar , Andrew Morton , Arjan van de Ven , "Rafael J. Wysocki" Subject: Re: suspend2 merge (was Re: [Suspend2-devel] Re: CFS and suspend2: hang in atomic copy) Message-ID: <20070426104052.GA19072@elf.ucw.cz> References: <20070418211632.GA7610@elte.hu> <200704182357.28107.mail@earthworm.de> <20070418220228.GA14536@elte.hu> <1176947576.5906.21.camel@nigel.suspend2.net> <20070419070437.GA25211@elte.hu> <20070424202336.GC16503@elf.ucw.cz> <20070424212408.GD16457@elf.ucw.cz> <1177582633.6814.29.camel@johannes.berg> <20070426103051.GP17387@elf.ucw.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070426103051.GP17387@elf.ucw.cz> X-Warning: Reading this can be dangerous to your mental health. User-Agent: Mutt/1.5.11+cvs20060126 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi! > > The interface isn't even 64/32-bit compatible... > > Which parts? > > ioctl(AVAIL_SWAP, > ...hmm, is this the one you are complaining about? It returns > loff_t through a pointer. Maybe there's another interface > that can return available swap, and we should use that, > instead? loff_t is 64bit on i386, so I do not see immediate problem here, but maybe we should just explicitely pass u64? > ioctl(GET_SWAP_PAGE, > returns sector_t through a pointer. NOt sure if that's good > idea, either. Ok, that's very bad idea, because sector_t can be 32-bit or 64-bit, depending on CONFIG_LBD. We need to use u64 here. > ioctl(SET_SWAP_FILE, > does old_decode_dev(arg). Is that ok? > > ioctl(SET_SWAP_AREA, > shares struct resume_swap_area between user and kernel. I > guess that's bad..? struct resume_swap_area { loff_t offset; u_int32_t dev; } __attribute__((packed)); ...I guess we should change loff_t -> u64 and problem is solved? Old_decode_dev takes u16. That sucks for majors/minors > 256, but fortunately those are not common. Does this seem to help? Pavel diff --git a/kernel/power/power.h b/kernel/power/power.h index eb461b8..dc13af5 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -114,7 +114,7 @@ extern int snapshot_image_loaded(struct * SNAPSHOT_SET_SWAP_AREA ioctl */ struct resume_swap_area { - loff_t offset; + u_int64_t offset; u_int32_t dev; } __attribute__((packed)); diff --git a/kernel/power/user.c b/kernel/power/user.c index 558e18e..d0730c1 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -215,8 +215,7 @@ static int snapshot_ioctl(struct inode * { int error = 0; struct snapshot_data *data; - loff_t avail; - sector_t offset; + u64 avail, offset; if (_IOC_TYPE(cmd) != SNAPSHOT_IOC_MAGIC) return -ENOTTY; @@ -286,7 +285,7 @@ static int snapshot_ioctl(struct inode * case SNAPSHOT_AVAIL_SWAP: avail = count_swap_pages(data->swap, 1); avail <<= PAGE_SHIFT; - error = put_user(avail, (loff_t __user *)arg); + error = put_user(avail, (u64 __user *)arg); break; case SNAPSHOT_GET_SWAP_PAGE: @@ -304,7 +303,7 @@ static int snapshot_ioctl(struct inode * offset = alloc_swapdev_block(data->swap, data->bitmap); if (offset) { offset <<= PAGE_SHIFT; - error = put_user(offset, (sector_t __user *)arg); + error = put_user(offset, (u64 __user *)arg); } else { error = -ENOSPC; } -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html