From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-oi0-f65.google.com ([209.85.218.65]:49075 "EHLO mail-oi0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753325AbdJTR1X (ORCPT ); Fri, 20 Oct 2017 13:27:23 -0400 Received: by mail-oi0-f65.google.com with SMTP id m198so21119949oig.5 for ; Fri, 20 Oct 2017 10:27:23 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <20171020163221.GB26320@lst.de> References: <150846713528.24336.4459262264611579791.stgit@dwillia2-desk3.amr.corp.intel.com> <150846720244.24336.16885325309403883980.stgit@dwillia2-desk3.amr.corp.intel.com> <1508504726.5572.41.camel@kernel.org> <20171020163221.GB26320@lst.de> From: Dan Williams Date: Fri, 20 Oct 2017 10:27:22 -0700 Message-ID: Subject: Re: [PATCH v3 12/13] dax: handle truncate of dma-busy pages Content-Type: text/plain; charset="UTF-8" Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Christoph Hellwig Cc: Jeff Layton , Andrew Morton , Jan Kara , Matthew Wilcox , Dave Hansen , Dave Chinner , "linux-kernel@vger.kernel.org" , "J. Bruce Fields" , Linux MM , Jeff Moyer , Alexander Viro , linux-fsdevel , "Darrick J. Wong" , Ross Zwisler , linux-xfs@vger.kernel.org, "linux-nvdimm@lists.01.org" On Fri, Oct 20, 2017 at 9:32 AM, Christoph Hellwig wrote: > On Fri, Oct 20, 2017 at 08:42:00AM -0700, Dan Williams wrote: >> I agree, but it needs quite a bit more thought and restructuring of >> the truncate path. I also wonder how we reclaim those stranded >> filesystem blocks, but a first approximation is wait for the >> administrator to delete them or auto-delete them at the next mount. >> XFS seems well prepared to reflink-swap these DMA blocks around, but >> I'm not sure about EXT4. > > reflink still is an optional and experimental feature in XFS. That > being said we should not need to swap block pointers around on disk. > We just need to prevent the block allocator from reusing the blocks > for new allocations, and we have code for that, both for transactions > that haven't been committed to disk yet, and for deleted blocks > undergoing discard operations. > > But as mentioned in my second mail from this morning I'm not even > sure we need that. For short-term elevated page counts like normal > get_user_pages users I think we can just wait for the page count > to reach zero, while for abuses of get_user_pages for long term > pinning memory (not sure if anyone but rdma is doing that) we'll need > something like FL_LAYOUT leases to release the mapping. I'll take a look at hooking this up through a page-idle callback. Can I get some breadcrumbs to grep for from XFS folks on how to set/clear the busy state of extents?