From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id EC352D19512 for ; Tue, 27 Jan 2026 08:19:17 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 3F4C610E4E7; Tue, 27 Jan 2026 08:19:17 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; secure) header.d=ziepe.ca header.i=@ziepe.ca header.b="DfSwyHQZ"; dkim-atps=neutral Received: from mail-qv1-f49.google.com (mail-qv1-f49.google.com [209.85.219.49]) by gabe.freedesktop.org (Postfix) with ESMTPS id E6BC010E496 for ; Mon, 26 Jan 2026 21:01:11 +0000 (UTC) Received: by mail-qv1-f49.google.com with SMTP id 6a1803df08f44-89461ccc46eso89763056d6.2 for ; Mon, 26 Jan 2026 13:01:11 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ziepe.ca; s=google; t=1769461271; x=1770066071; darn=lists.freedesktop.org; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:from:to:cc:subject:date:message-id:reply-to; bh=Tu3uAyHuSXXXZq6VSWtVG72sHh/RWKg7/MHSYDvcONg=; b=DfSwyHQZmNsEGg8EpRn1mki1izenW5u61iuEqI7EItSrQEJoDT7B9jwIEUl8hqT67E c2uNqDBKBVshbT9w9eUB5AB7ESoz96kMHiWlFjxxF8QZE8YkD0MrT3pV/pyEI35DpMwI 8qlLlKIhAc2XlON+51hHM7j/YfiSJwey7C3WJ+/sWl8Dee4W5ocgbuH+cQjmZiN+WsiW 355x854O7+2SkZN2EsVMMyMHyczGKYBMB8BtAh46KP0JoXjdAytYuhLBEGz0LTpnQqI2 2SjN3IJDKXYsL8CbY9vyVyo2WmcrcuROVnhnX8iLyvO06QlfH9jVIX9rZIXQXgx8nCyW bHew== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1769461271; x=1770066071; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:x-gm-gg:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=Tu3uAyHuSXXXZq6VSWtVG72sHh/RWKg7/MHSYDvcONg=; b=f07Q2nfrGs64r78jLL3V9pBsqjloWCH7Wo5MvQk+LPlGTGbW6E26bpkPbxOsMR9eTB QbEziYNJpo7kaDfK+Viw2MHapIyZ5IpfdQf+IrdBETkuHNDgvlws/Lzns2LzUnRZy2wc mc9m9oCp3D3BYgMxrWuazfvPGe7h+egsxBfY0+r3GmxqS1YH7xSHO2pepc/DeWEOoaFG 9YW1wgSpasuezIPFaz4f2iKfxwbekZWRc9/NwMNsnw3OE+CK2LbKte6IhqTpPsTmiNRE xbEUWkI85PQ2D/z0tFtmEZFu69R9NONbBdgOyNACyAzVLV7LrOg9GToCJpAVmlWD3bWA 2NnA== X-Forwarded-Encrypted: i=1; AJvYcCW5xph8HpJub53GYl3uAFTkF2Bm6ubfPe4zkzi4TSR5ZgLRGcUIDmasBIbvLqvu+PtDpkfp5GPn@lists.freedesktop.org X-Gm-Message-State: AOJu0Yx0hXbB4kLiq8JYdT03CHlbboKm83WVLSGH4ybvFWSgwnZz2zJv ld+xTB+C4heOCmFlJAEKltUqUL8TLCZs/p6EiympM2+eU6nZnxc5jdYM8CfLPhI461o= X-Gm-Gg: AZuq6aJdQdQPvmfP8GsObCwg+l8ZIePu+lGkUn4BUrDhpZ3Jw1YEjdAnV5EJ69WCMKU GY8UIn0nRPjCq+jdEQVNKOwf0XPGwfbVIX9EklPXcoXgvgxzGqiXVAZ1q2e96q+iplwy1JqkbS2 FbIXPnlJLtHqJKSVRpCCD/stNYa89DJIYPVLdVDj+djcOUlzDdQD5WiH5FfSPgaJdYW9xCyiwYB W96U8tkY3RmIzkuNHjk5u/hChIjOIFbn/wLXuTI+mWcJbvHkhL7rTVX7f1IlP7M8dzdavvkVf+X OwtFTawgJseAlXKU5JK9kCFzkaUmuY2qdpqiH4kkzH0UUcgO8j189y0KefuI35764H9XfN5fZxt 5lAtyiBGLlODMRQz0zaZYcbtW7/qg0zrPSu0DEtZeAIpNMUZTtL5c4dXhOMbTyCEU9R3pxe5TYs y/oWl+XePNyd4gAGTD5JDRedHD9LbU2a0Za5dzAP0fw4Au6rlVaySnS68pHA5TdomCKG0= X-Received: by 2002:a05:6214:1cc9:b0:894:7b34:dacd with SMTP id 6a1803df08f44-894b06f336dmr73487126d6.31.1769461270245; Mon, 26 Jan 2026 13:01:10 -0800 (PST) Received: from ziepe.ca (hlfxns017vw-142-162-112-119.dhcp-dynamic.fibreop.ns.bellaliant.net. [142.162.112.119]) by smtp.gmail.com with ESMTPSA id 6a1803df08f44-894918b1436sm104983856d6.35.2026.01.26.13.01.09 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 26 Jan 2026 13:01:09 -0800 (PST) Received: from jgg by wakko with local (Exim 4.97) (envelope-from ) id 1vkThs-00000008z30-31vX; Mon, 26 Jan 2026 17:01:08 -0400 Date: Mon, 26 Jan 2026 17:01:08 -0400 From: Jason Gunthorpe To: Pranjal Shrivastava Cc: Leon Romanovsky , Sumit Semwal , Christian =?utf-8?B?S8O2bmln?= , Alex Deucher , David Airlie , Simona Vetter , Gerd Hoffmann , Dmitry Osipenko , Gurchetan Singh , Chia-I Wu , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , Lucas De Marchi , Thomas =?utf-8?Q?Hellstr=C3=B6m?= , Rodrigo Vivi , Kevin Tian , Joerg Roedel , Will Deacon , Robin Murphy , Felix Kuehling , Alex Williamson , Ankit Agrawal , Vivek Kasireddy , linux-media@vger.kernel.org, dri-devel@lists.freedesktop.org, linaro-mm-sig@lists.linaro.org, linux-kernel@vger.kernel.org, amd-gfx@lists.freedesktop.org, virtualization@lists.linux.dev, intel-xe@lists.freedesktop.org, linux-rdma@vger.kernel.org, iommu@lists.linux.dev, kvm@vger.kernel.org Subject: Re: [PATCH v5 6/8] dma-buf: Add dma_buf_attach_revocable() Message-ID: <20260126210108.GD1641016@ziepe.ca> References: <20260124-dmabuf-revoke-v5-0-f98fca917e96@nvidia.com> <20260124-dmabuf-revoke-v5-6-f98fca917e96@nvidia.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-Mailman-Approved-At: Tue, 27 Jan 2026 08:19:15 +0000 X-BeenThere: amd-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Discussion list for AMD gfx List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: amd-gfx-bounces@lists.freedesktop.org Sender: "amd-gfx" On Mon, Jan 26, 2026 at 08:38:44PM +0000, Pranjal Shrivastava wrote: > I noticed that Patch 5 removes the invalidate_mappings stub from > umem_dmabuf.c, effectively making the callback NULL for an RDMA > importer. Consequently, dma_buf_attach_revocable() (introduced here) > will return false for these importers. Yes, that is the intention. > Since the cover letter mentions that VFIO will use > dma_buf_attach_revocable() to prevent unbounded waits, this appears to > effectively block paths like the VFIO-export -> RDMA-import path.. It remains usable with the ODP path and people are using that right now. > Given that RDMA is a significant consumer of dma-bufs, are there plans > to implement proper revocation support in the IB/RDMA core (umem_dmabuf)? This depends on each HW, they need a way to implement the revoke semantic. I can't guess what is possible, but I would hope that most HW could at least do a revoke on a real MR. Eg a MR rereg operation to a kernel owned empty PD is an effective "revoke", and MR rereg is at least defined by standards so HW should implement it. > It would be good to know if there's a plan for bringing such importers > into compliance with the new revocation semantics so they can interop > with VFIO OR are we completely ruling out users like RDMA / IB importing > any DMABUFs exported by VFIO? It will be driver dependent, there is no one shot update here. Jason