From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-f72.google.com (mail-wm0-f72.google.com [74.125.82.72]) by kanga.kvack.org (Postfix) with ESMTP id 0FDD26B02E3 for ; Tue, 15 Nov 2016 17:57:16 -0500 (EST) Received: by mail-wm0-f72.google.com with SMTP id m203so10295742wma.2 for ; Tue, 15 Nov 2016 14:57:16 -0800 (PST) Received: from mail-wm0-x244.google.com (mail-wm0-x244.google.com. [2a00:1450:400c:c09::244]) by mx.google.com with ESMTPS id f125si4791408wmf.44.2016.11.15.14.57.14 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 15 Nov 2016 14:57:14 -0800 (PST) Received: by mail-wm0-x244.google.com with SMTP id m203so4990531wma.3 for ; Tue, 15 Nov 2016 14:57:14 -0800 (PST) Date: Wed, 16 Nov 2016 01:57:11 +0300 From: "Kirill A. Shutemov" Subject: Re: [PATCH 17/21] mm: Change return values of finish_mkwrite_fault() Message-ID: <20161115225711.GQ23021@node> References: <1478233517-3571-1-git-send-email-jack@suse.cz> <1478233517-3571-18-git-send-email-jack@suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1478233517-3571-18-git-send-email-jack@suse.cz> Sender: owner-linux-mm@kvack.org List-ID: To: Jan Kara Cc: linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, linux-nvdimm@lists.01.org, Andrew Morton , Ross Zwisler , "Kirill A. Shutemov" On Fri, Nov 04, 2016 at 05:25:13AM +0100, Jan Kara wrote: > Currently finish_mkwrite_fault() returns 0 when PTE got changed before > we acquired PTE lock and VM_FAULT_WRITE when we succeeded in modifying > the PTE. This is somewhat confusing since 0 generally means success, it > is also inconsistent with finish_fault() which returns 0 on success. > Change finish_mkwrite_fault() to return 0 on success and VM_FAULT_NOPAGE > when PTE changed. Practically, there should be no behavioral difference > since we bail out from the fault the same way regardless whether we > return 0, VM_FAULT_NOPAGE, or VM_FAULT_WRITE. Also note that > VM_FAULT_WRITE has no effect for shared mappings since the only two > places that check it - KSM and GUP - care about private mappings only. > Generally the meaning of VM_FAULT_WRITE for shared mappings is not well > defined and we should probably clean that up. > > Signed-off-by: Jan Kara Sounds right. Acked-by: Kirill A. Shutemov -- Kirill A. Shutemov -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org