From patchwork Thu Apr 13 19:40:42 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Alex Williamson X-Patchwork-Id: 13210598 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5D01CC77B61 for ; Thu, 13 Apr 2023 19:41:47 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230033AbjDMTlq (ORCPT ); Thu, 13 Apr 2023 15:41:46 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:38456 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229632AbjDMTll (ORCPT ); Thu, 13 Apr 2023 15:41:41 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id CECA465B6 for ; Thu, 13 Apr 2023 12:40:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1681414856; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=7hS3zAlwUKHt4i9Sj3OqwJTrufrugLSA6n35uNDz5V0=; b=G9poOaCvRj0qGa/qBr0XhJpasEC+JuuhC/riKtGmJBEC5hozY5up5JUsEbEmfOjzeCVFca A5Vsy16CqHlcUvPoUqkX8c9lDBBqAdA96dEjLM/Lx6EyHbHwMVYK+8hhPnSMGc7aPCIoUt b825KVOWO10XN4vxxy+bdOPxrXluPSQ= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-284-FlsL602-M2Kn7T8cxqnHEw-1; Thu, 13 Apr 2023 15:40:52 -0400 X-MC-Unique: FlsL602-M2Kn7T8cxqnHEw-1 Received: from smtp.corp.redhat.com (int-mx09.intmail.prod.int.rdu2.redhat.com [10.11.54.9]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 7BE758314E9; Thu, 13 Apr 2023 19:40:52 +0000 (UTC) Received: from omen.home.shazbot.org (unknown [10.22.34.53]) by smtp.corp.redhat.com (Postfix) with ESMTP id 0976E492C14; Thu, 13 Apr 2023 19:40:51 +0000 (UTC) From: Alex Williamson To: linux-pci@vger.kernel.org Cc: abhsahu@nvidia.com, targupta@nvidia.com, zhguo@redhat.com, alex.williamson@redhat.com, sdalvi@google.com Subject: [PATCH] PCI: Extend D3hot delay for NVIDIA HDA controllers Date: Thu, 13 Apr 2023 13:40:42 -0600 Message-Id: <20230413194042.605768-1-alex.williamson@redhat.com> In-Reply-To: <168004421186.935858.12296629041962399467.stgit@omen> References: <168004421186.935858.12296629041962399467.stgit@omen> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.1 on 10.11.54.9 Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org Assignment of NVIDIA Ampere-based GPUs have seen a regression since the below referenced commit, where the reduced D3hot transition delay appears to introduce a small window where a D3hot->D0 transition followed by a bus reset can wedge the device. The entire device is subsequently unavailable, returning -1 on config space read and is unrecoverable without a host reset. This has been observed with RTX A2000 and A5000 GPU and audio functions assigned to a Windows VM, where shutdown of the VM places the devices in D3hot prior to vfio-pci performing a bus reset when userspace releases the devices. The issue has roughly a 2-3% chance of occurring per shutdown. Restoring the HDA controller d3hot_delay to the effective value before the below commit has been shown to resolve the issue. NVIDIA confirms this change should be safe for all of their HDA controllers. Cc: Abhishek Sahu Cc: Tarun Gupta Fixes: 3e347969a577 ("PCI/PM: Reduce D3hot delay with usleep_range()") Reported-by: Zhiyi Guo Reviewed-by: Tarun Gupta Signed-off-by: Alex Williamson --- Unfortunately Tarun's reply with confirmation doesn't show up on lore, possibly due to html email, or else I'd provide that as a Link:. drivers/pci/quirks.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 44cab813bf95..f4e2a88729fd 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1939,6 +1939,19 @@ static void quirk_radeon_pm(struct pci_dev *dev) } DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x6741, quirk_radeon_pm); +/* + * NVIDIA Ampere-based HDA controllers can wedge the whole device if a bus + * reset is performed too soon after transition to D0, extend d3hot_delay + * to previous effective default for all NVIDIA HDA controllers. + */ +static void quirk_nvidia_hda_pm(struct pci_dev *dev) +{ + quirk_d3hot_delay(dev, 20); +} +DECLARE_PCI_FIXUP_CLASS_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_ANY_ID, + PCI_CLASS_MULTIMEDIA_HD_AUDIO, 8, + quirk_nvidia_hda_pm); + /* * Ryzen5/7 XHCI controllers fail upon resume from runtime suspend or s2idle. * https://bugzilla.kernel.org/show_bug.cgi?id=205587