From patchwork Wed Feb 3 15:29:21 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Daniel Vetter X-Patchwork-Id: 12064605 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-18.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 680D2C433E6 for ; Wed, 3 Feb 2021 15:33:36 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 21BD964E55 for ; Wed, 3 Feb 2021 15:33:36 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233847AbhBCPc0 (ORCPT ); Wed, 3 Feb 2021 10:32:26 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:40180 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234352AbhBCPa2 (ORCPT ); Wed, 3 Feb 2021 10:30:28 -0500 Received: from mail-wr1-x42e.google.com (mail-wr1-x42e.google.com [IPv6:2a00:1450:4864:20::42e]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 1A1BEC061352 for ; Wed, 3 Feb 2021 07:29:29 -0800 (PST) Received: by mail-wr1-x42e.google.com with SMTP id c4so22120192wru.9 for ; Wed, 03 Feb 2021 07:29:29 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ffwll.ch; s=google; h=from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=dX5VWhzdyCSXKOkzbFh/rppPL8SS0wtBsuXFORopB0E=; b=XopLrHKZbh8IvKLagLigdfsTrY8sxR+m6Xie6uCiR53EextjydiFKe5V6Zo+ctkjqU TM2/sYWjHi7tScNkKM4zbwFzfqrGn+XBtdHx+y27ZOUIXFJvowhu2Yw3RsI7uMPl/j+z tw0mjJGsjsICgamEykClEFgk+oHPA/vn3vJv4= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=dX5VWhzdyCSXKOkzbFh/rppPL8SS0wtBsuXFORopB0E=; b=EKAjQnSRu8hjBtfPTcsyf1L4lzqouwSnDvcWGg4Adwdwu5qtuTF+EgsgKs6f9Wr+xL +/dDZfvPRKLR4BF/QwVwkyJI6v7QMEoJr9yORzL1pxvYQvc28M8NUTOnqTSt4YZkny8k /ozbuEumd91JNliiMcCqyvecO65PO04/Tnh8nOhoqw1hBIe2j3Mtm40b3jt+7lqOXonZ 8qHJemynptLxpMHo+AxN5cQyYIAaNmvlXWoUpZi/SWieYUBTr+c0TbTBvuMnZB5prE71 d4piw6bmG5fRqiaoawP+s6GBhfakj5FZoOySzwc6zzKSpB+Xp47i0ZsHerhyKHT3N9UO mrzA== X-Gm-Message-State: AOAM5329/DyEkKnV6fVriahzvvUgukxkOrlA6g39tQjYNoHuWiRVfR/1 A9NDuJVKIRud/llll8wYqmtHXw== X-Google-Smtp-Source: ABdhPJzVLJmtYENS2YgeQZySm5SoeXYyzpaLIlSonXEHkElg5fxVGgKG2F6eexgXJaLvEh8wld5viA== X-Received: by 2002:adf:83a6:: with SMTP id 35mr4126060wre.274.1612366167884; Wed, 03 Feb 2021 07:29:27 -0800 (PST) Received: from phenom.ffwll.local ([2a02:168:57f4:0:efd0:b9e5:5ae6:c2fa]) by smtp.gmail.com with ESMTPSA id l1sm2779528wmq.17.2021.02.03.07.29.26 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 03 Feb 2021 07:29:27 -0800 (PST) From: Daniel Vetter To: DRI Development Cc: Daniel Vetter , Dave Airlie , Maarten Lankhorst , =?utf-8?q?Thomas_Hel?= =?utf-8?q?lstr=C3=B6m?= , =?utf-8?q?Christian_K?= =?utf-8?q?=C3=B6nig?= , Jerome Glisse , Felix Kuehling , Daniel Vetter , Sumit Semwal , linux-media@vger.kernel.org, linaro-mm-sig@lists.linaro.org Subject: [PATCH] dma-fence: Document recoverable page fault implications Date: Wed, 3 Feb 2021 16:29:21 +0100 Message-Id: <20210203152921.2429937-1-daniel.vetter@ffwll.ch> X-Mailer: git-send-email 2.30.0 MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-media@vger.kernel.org Recently there was a fairly long thread about recoreable hardware page faults, how they can deadlock, and what to do about that. While the discussion is still fresh I figured good time to try and document the conclusions a bit. This documentation section explains what's the potential problem, and the remedies we've discussed, roughly ordered from best to worst. v2: Linus -> Linux typoe (Dave) v3: - Make it clear drivers only need to implement one option (Christian) - Make it clearer that implicit sync is out the window with exclusive fences (Christian) - Add the fairly theoretical option of segementing the memory (either statically or through dynamic checks at runtime for which piece of memory is managed how) and explain why it's not a great idea (Felix) References: https://lore.kernel.org/dri-devel/20210107030127.20393-1-Felix.Kuehling@amd.com/ Cc: Dave Airlie Cc: Maarten Lankhorst Cc: Thomas Hellström Cc: "Christian König" Cc: Jerome Glisse Cc: Felix Kuehling Signed-off-by: Daniel Vetter Cc: Sumit Semwal Cc: linux-media@vger.kernel.org Cc: linaro-mm-sig@lists.linaro.org Reviewed-by: Christian König Acked-by: Thomas Hellström --- Documentation/driver-api/dma-buf.rst | 76 ++++++++++++++++++++++++++++ 1 file changed, 76 insertions(+) diff --git a/Documentation/driver-api/dma-buf.rst b/Documentation/driver-api/dma-buf.rst index a2133d69872c..7f37ec30d9fd 100644 --- a/Documentation/driver-api/dma-buf.rst +++ b/Documentation/driver-api/dma-buf.rst @@ -257,3 +257,79 @@ fences in the kernel. This means: userspace is allowed to use userspace fencing or long running compute workloads. This also means no implicit fencing for shared buffers in these cases. + +Recoverable Hardware Page Faults Implications +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +Modern hardware supports recoverable page faults, which has a lot of +implications for DMA fences. + +First, a pending page fault obviously holds up the work that's running on the +accelerator and a memory allocation is usually required to resolve the fault. +But memory allocations are not allowed to gate completion of DMA fences, which +means any workload using recoverable page faults cannot use DMA fences for +synchronization. Synchronization fences controlled by userspace must be used +instead. + +On GPUs this poses a problem, because current desktop compositor protocols on +Linux rely on DMA fences, which means without an entirely new userspace stack +built on top of userspace fences, they cannot benefit from recoverable page +faults. Specifically this means implicit synchronization will not be possible. +The exception is when page faults are only used as migration hints and never to +on-demand fill a memory request. For now this means recoverable page +faults on GPUs are limited to pure compute workloads. + +Furthermore GPUs usually have shared resources between the 3D rendering and +compute side, like compute units or command submission engines. If both a 3D +job with a DMA fence and a compute workload using recoverable page faults are +pending they could deadlock: + +- The 3D workload might need to wait for the compute job to finish and release + hardware resources first. + +- The compute workload might be stuck in a page fault, because the memory + allocation is waiting for the DMA fence of the 3D workload to complete. + +There are a few options to prevent this problem, one of which drivers need to +ensure: + +- Compute workloads can always be preempted, even when a page fault is pending + and not yet repaired. Not all hardware supports this. + +- DMA fence workloads and workloads which need page fault handling have + independent hardware resources to guarantee forward progress. This could be + achieved through e.g. through dedicated engines and minimal compute unit + reservations for DMA fence workloads. + +- The reservation approach could be further refined by only reserving the + hardware resources for DMA fence workloads when they are in-flight. This must + cover the time from when the DMA fence is visible to other threads up to + moment when fence is completed through dma_fence_signal(). + +- As a last resort, if the hardware provides no useful reservation mechanics, + all workloads must be flushed from the GPU when switching between jobs + requiring DMA fences or jobs requiring page fault handling: This means all DMA + fences must complete before a compute job with page fault handling can be + inserted into the scheduler queue. And vice versa, before a DMA fence can be + made visible anywhere in the system, all compute workloads must be preempted + to guarantee all pending GPU page faults are flushed. + +- Only a fairly theoretical option would be to untangle these dependencies when + allocating memory to repair hardware page faults, either through separate + memory blocks or runtime tracking of the full dependency graph of all DMA + fences. This results very wide impact on the kernel, since resolving the page + on the CPU side can itself involve a page fault. It is much more feasible and + robust to limit the impact of handling hardware page faults to the specific + driver. + +Note that workloads that run on independent hardware like copy engines or other +GPUs do not have any impact. This allows us to keep using DMA fences internally +in the kernel even for resolving hardware page faults, e.g. by using copy +engines to clear or copy memory needed to resolve the page fault. + +In some ways this page fault problem is a special case of the `Infinite DMA +Fences` discussions: Infinite fences from compute workloads are allowed to +depend on DMA fences, but not the other way around. And not even the page fault +problem is new, because some other CPU thread in userspace might +hit a page fault which holds up a userspace fence - supporting page faults on +GPUs doesn't anything fundamentally new.