From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by mail19.linbit.com (LINBIT Mail Daemon) with ESMTP id AFF6542066B for ; Tue, 29 Nov 2022 17:48:02 +0100 (CET) From: Benjamin Coddington To: Christoph Hellwig Date: Tue, 29 Nov 2022 11:47:47 -0500 Message-ID: <794DBAB0-EDAF-4DA2-A837-C1F99916BC8E@redhat.com> In-Reply-To: <20221129140242.GA15747@lst.de> References: <20221129140242.GA15747@lst.de> MIME-Version: 1.0 Content-Type: text/plain Cc: Latchesar Ionkov , samba-technical@lists.samba.org, Dominique Martinet , Valentina Manea , linux-nvme@lists.infradead.org, Philipp Reisner , David Howells , Joseph Qi , Eric Dumazet , linux-nfs@vger.kernel.org, Marc Dionne , Shuah Khan , linux-afs@lists.infradead.org, Mike Christie , drbd-dev@lists.linbit.com, linux-cifs@vger.kernel.org, Sagi Grimberg , linux-scsi@vger.kernel.org, Mark Fasheh , cluster-devel@redhat.com, Christine Caulfield , Jakub Kicinski , Ilya Dryomov , Paolo Abeni , Anna Schumaker , Eric Van Hensbergen , "James E.J. Bottomley" , Josef Bacik , nbd@other.debian.org, linux-block@vger.kernel.org, David Teigland , Joel Becker , v9fs-developer@lists.sourceforge.net, Keith Busch , ceph-devel@vger.kernel.org, Xiubo Li , Trond Myklebust , Jens Axboe , Chris Leech , open-iscsi@googlegroups.com, "Martin K. Petersen" , netdev@vger.kernel.org, linux-usb@vger.kernel.org, Jeff Layton , linux-kernel@vger.kernel.org, Steve French , Chuck Lever , Lee Duncan , Greg Kroah-Hartman , Lars Ellenberg , "David S. Miller" , ocfs2-devel@oss.oracle.com Subject: Re: [Drbd-dev] [PATCH v1 2/3] Treewide: Stop corrupting socket's task_frag List-Id: "*Coordination* of development, patches, contributions -- *Questions* \(even to developers\) go to drbd-user, please." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On 29 Nov 2022, at 9:02, Christoph Hellwig wrote: > Hmm. Having to set a flag to not accidentally corrupt per-task > state seems a bit fragile. Wouldn't it make sense to find a way to opt > into the feature only for sockets created from the syscall layer? It's totally fragile, and that's why it's currently broken in production. The fragile ship sailed when networking decided to depend on users setting the socket's GFP_ flags correctly to avoid corruption. Meantime, this problem needs fixing in a way that makes everyone happy. This fix doesn't make it less fragile, but it may (hopefully) address the previous criticisms enough that something gets done to fix it. Ben