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 X-Spam-Level: X-Spam-Status: No, score=-7.1 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI, SIGNED_OFF_BY,SPF_PASS 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 9FC68C43381 for ; Wed, 27 Mar 2019 17:12:04 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 538DD206BA for ; Wed, 27 Mar 2019 17:12:04 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=yadro.com header.i=@yadro.com header.b="A7B+rSp2" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727357AbfC0RMD (ORCPT ); Wed, 27 Mar 2019 13:12:03 -0400 Received: from mta-01.yadro.com ([89.207.88.251]:38640 "EHLO mta-01.yadro.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727234AbfC0RMD (ORCPT ); Wed, 27 Mar 2019 13:12:03 -0400 Received: from localhost (unknown [127.0.0.1]) by mta-01.yadro.com (Postfix) with ESMTP id EDAE241A41; Wed, 27 Mar 2019 17:11:59 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=yadro.com; h= content-type:content-type:in-reply-to:mime-version:user-agent :date:date:message-id:from:from:references:subject:subject :received:received:received; s=mta-01; t=1553706718; x= 1555521119; bh=qvH2JYhpsSI2xZn952KeehFyQj7LM2QFQxSV0zp1zmw=; b=A 7B+rSp2DYb6F/xBsUsBv+o31gPIBN8RHzIzBXaaC2EG5Q4VPTCMlLVNzWQhbDP0S 4aXlmuB96MSVNIYhcs3NE6YUI68sWVWZs8VlaC5ltWeV6RJycuSoLc13QkshMP4/ NWcxby/NuGYoktfQJuD6e149TMF1tVDDDQw7iDOUIs= X-Virus-Scanned: amavisd-new at yadro.com Received: from mta-01.yadro.com ([127.0.0.1]) by localhost (mta-01.yadro.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8Mud4U-nZkJX; Wed, 27 Mar 2019 20:11:58 +0300 (MSK) Received: from T-EXCH-02.corp.yadro.com (t-exch-02.corp.yadro.com [172.17.10.102]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by mta-01.yadro.com (Postfix) with ESMTPS id CB94441A0C; Wed, 27 Mar 2019 20:11:57 +0300 (MSK) Received: from [172.17.15.60] (172.17.15.60) by T-EXCH-02.corp.yadro.com (172.17.10.102) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA384_P384) id 15.1.669.32; Wed, 27 Mar 2019 20:11:57 +0300 Subject: Re: [PATCH RFC v4 02/21] PCI: Fix race condition in pci_enable/disable_device() To: Bjorn Helgaas CC: , , , Srinath Mannam , Marta Rybczynska , References: <20190311133122.11417-1-s.miroshnichenko@yadro.com> <20190311133122.11417-3-s.miroshnichenko@yadro.com> <20190326190038.GL24180@google.com> From: Sergey Miroshnichenko Openpgp: preference=signencrypt Autocrypt: addr=s.miroshnichenko@yadro.com; prefer-encrypt=mutual; keydata= xsFNBFm31LoBEAC1wCndw8xXjGaJOinDBeVD1/8TFlVehvafur6V9xH3gsHhs0weDcMgw2Ki r5ZVhS8BlltU0snpsnQHxYB5BF0gzCLwwPUjFPZ7E0/++ylbNJoGe53cVbE870NK5WqoSEUg QtTQev2/Y5q0v7kfMh9g5p5jzeqfQSZzOrEP4d1cg5tPNKYji5cCfB/NQTHWV9w4EPj3UJQT ZPp4xqMAXu0JU1W9/XecNobKaHfEv9T+UWdx2eufiNqCgfAkRVCl8V0tKhQ4PZlZdp0dQH/N BreUg1+QJ4/t2SyEsiIPqYxFBW6qWAgOP5fzGNG31VHaQeJCA31keh84/8t632HZ4FDRrS3N 6V7Oc0ew7h5AwgOca4d3TTn8ATfASQ5vAxHC2ZK9CZhfa3RgK+8X5+vwkqc8O70iTmE9Goap uDMtgvIc0r0PHTiB3eZlyHExMD+FIOBOp2GvL7BmFHMgyOjNDdh2vBNqUwiv1RTQVWPhNX/J 4ZhTAZuAr5+6S/iRFpWspCqKvgonPxSzfWRS5dWJ2kavuvXkSB5eyPx9XRgrWxZwVdseuTpi CeTEW9/noDDl1edZdWHGWS9/4BC1nByitYYUcPXuzSkIsuae2tDw+lnsQfgAn+pXT6ESjEnZ LGnnWMQNLISf8yIaEh6bft+vXT67o1G2/U6VN1+suUPcDgYEVQARAQABzTJTZXJnZWkgTWly b3NobmljaGVua28gPHMubWlyb3NobmljaGVua29AeWFkcm8uY29tPsLBlAQTAQgAPhYhBB1u 0+6Lz/3BafPm9wx0PmjRU7O1BQJZt9S6AhsjBQkJZgGABQsJCAcCBhUICQoLAgQWAgMBAh4B AheAAAoJEAx0PmjRU7O1WfEP/jdWabDp11EdD9ZCK8LlwZ/SgXVfr9lZ5Kx3VVI68KAcfupH 3m+1lGTOktpRu7gQaj867KCbzRCWJjoVibrBgMMaFZQX2Bf2usxuBN9QxUnehg3R5Yr+c0KS 9v2oSduWaMJ/Fs3IVg5gh0bhH3lMHISqAQLtl3ncyB+1O+X+MgReRGznj5tkjQWC960t85SO hkNkhVMp0z2b1XfY51XxYRESdNkJswxv3UnpAvlgdh+ItzJU8fRmfUtOzRdGD6mukrkpkS1z lAGNLayBOiEWUk8E1gm3rK46l/sm6Gq9ExCh+bgkwQHRp/JhyHpsid9V/o5nLh+jbh/CLYIF onrG2RN6lePQpyh6TpiZfGbxz/4rny88HdCD31OdvTwbnNp5Fj48YXbUlo8WILg2OHWbSRQ9 w7OuTLcITPW084E/Uq/nL6+m316OZpY7iiVB+1e2reJRjnsqlK+TX7N1KsAamba3hGSqF8QC 61RAzXS99D1ohL98G0hJNYyuHaeWus4wJRt8JBEe6D4r0hrS/O97oa0juygwY+zP9mtpYRr4 t9Im1hpIkV+cC3aJrRiQNaXJN4S+8F8DQnXMUitf0590NNKwYRuQuTg5URoqjYBFZtXGgS7w vdyzevMt1bCBtZW6Rbdu6TcHoF3Aminx96wXlSizTGpo+xJ589xQ46U9KWXdzsFNBFm31LoB EADAsXCTRufklKBW9jdUMwjltZjXwu5muxcVRj8XICi77oa9DgsGhA5v7vosbpNXzZAL018h 1khPu6ca6X0shLm0Le2KQ6Q00VHEwrTjXQ0NN0aa+vRG3NKPb9t/SiXg6yNPKuQxTsYm0vP9 4fIH6nHDtJpBXq8LK5C6GTD6G2R3VTSPpJz6tFPrfLrV4jPARFRAZ483Wjs9iBRygFTtb6YJ r1YJnwmXcb8Z/ds3vPo5ULMcMlcXEA7NlkmN7r3LUkmE6Tjr1hZHGwEWRwSiw1CwkAQqLlMX xRul5+nPz0pPrB8hBxONjnlGX3f0Ky2xdKxrFxlzd8HtRzhWb4R0vqgWQRXXFeKc++uEyk6g KZ48zSjLq0Av4ZS8POCL1JisSV7Hbwe4Ik3qaeR61KEuVtBlySFijwvTs4p5b9PcG2fmNiyo aFBdFkbI/pTuORRBYCLbjXwyRWnCGBWZ8b0NSCs4sb9vNyObxoLYN4RdRnKKLpkXz3EXdPWZ WswxQQNopKs5pE3aAvYfTitIg0JmKSK57w3UJNS11s5xTRAmKDHj9PmLZcNLFhG7ceb9T41+ YLNCEu8/xvFEorp+AlJ6n0clfPsNsi8317ZJL0mgZ0XrD9efmuA+xvb/0T67D371qK6xDaZ2 xN71pfjhZl1OYNZ3FDJLpZSNZKNFluhRWOvTKQARAQABwsF8BBgBCAAmFiEEHW7T7ovP/cFp 8+b3DHQ+aNFTs7UFAlm31LoCGwwFCQlmAYAACgkQDHQ+aNFTs7XITg/9GHcaTLjsRP7Pacu0 PFs2ubddBvZPC19sIILUNDlQHsOVKTpuFTtEmA6F4o4gf/SY8AvnHyVVqe8YYsQkPwhwfwbH ihoDZyJxyr52mqanez3sQV6RQEqCZtKaJtMdZrtOZcjqrAxEG1arowCKnnoPF+ivtA4ZEtlm xt9x5S0UfytTIZR0KKsRfO7XZvqfzbg6/NVRnUibSzCz2yzC5kbsyjPoK+c+C142BlnCdgai 0It5xKX1BBoVT/YSeB5ACGijuRsuDH2mHzdOeEDlP/UOAB5gx9aBOdP8YMTAk2b4qfANX7Pc W8BnI99mWuOP04KVgdQf5vgwMRDlgdtsQJw7l5YBQxprq8edAH3xsKung03qsV2inbQDkMnl c+l79kx0ilh0oLwviRft5xVCOfCyVkvekUhN4qG+guGFJbxYffliFB02Kcf2e4CueCnGGZAw +OkhHbtDmgmyslv7cxf1qzsObQfYc9eR5f8uiX41bLPwTMy18YnYk2hxJSW0g+LkPqBVQcAO Nwdozk9DY6wY9cMQ8coYTctox5VsvYEz2rJCRiIc40NO76gdMVutEORjdSoeZK32srVNoBo9 L0EK2QCFFRDcslPDpZWE1uDZQPW+GC2Z/dmuEpaMzlrIgfZ8GLXxHbB+VdDQ7QE//lphXskF lHi50np+KDDPzZS51tw= Message-ID: Date: Wed, 27 Mar 2019 20:11:51 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.5.3 MIME-Version: 1.0 In-Reply-To: <20190326190038.GL24180@google.com> Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="MzcZ8NHdkHNPKzrPfEdUWBQ8QvIkrwKiD" X-Originating-IP: [172.17.15.60] X-ClientProxiedBy: T-EXCH-01.corp.yadro.com (172.17.10.101) To T-EXCH-02.corp.yadro.com (172.17.10.102) Sender: linux-pci-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org --MzcZ8NHdkHNPKzrPfEdUWBQ8QvIkrwKiD Content-Type: multipart/mixed; boundary="SZX8gVbqDGgIpm1j0BU84HpFvIMDCudKR"; protected-headers="v1" From: Sergey Miroshnichenko To: Bjorn Helgaas Cc: linux-pci@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, linux@yadro.com, Srinath Mannam , Marta Rybczynska , linux-kernel@vger.kernel.org Message-ID: Subject: Re: [PATCH RFC v4 02/21] PCI: Fix race condition in pci_enable/disable_device() References: <20190311133122.11417-1-s.miroshnichenko@yadro.com> <20190311133122.11417-3-s.miroshnichenko@yadro.com> <20190326190038.GL24180@google.com> In-Reply-To: <20190326190038.GL24180@google.com> --SZX8gVbqDGgIpm1j0BU84HpFvIMDCudKR Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: quoted-printable On 3/26/19 10:00 PM, Bjorn Helgaas wrote: > [+cc Srinath, Marta, LKML] >=20 > On Mon, Mar 11, 2019 at 04:31:03PM +0300, Sergey Miroshnichenko wrote: >> CPU0 CPU1 >> >> pci_enable_device_mem() pci_enable_device_mem() >> pci_enable_bridge() pci_enable_bridge() >> pci_is_enabled() >> return false; >> atomic_inc_return(enable_cnt) >> Start actual enabling the bridge >> ... pci_is_enabled() >> ... return true; >> ... Start memory requests <-- F= AIL >> ... >> Set the PCI_COMMAND_MEMORY bit <-- Must wait for this >> >> This patch protects the pci_enable/disable_device() and pci_enable_bri= dge() >> with mutexes. >=20 > This is a subtle issue that we've tried to fix before, but we've never > had a satisfactory solution, so I hope you've figured out the right > fix. >=20 > I'll include some links to previous discussion. This patch is very > similar to [2], which we didn't actually apply. We did apply the > patch from [3] as 40f11adc7cd9 ("PCI: Avoid race while enabling > upstream bridges"), but it caused the regressions reported in [4,5], > so we reverted it with 0f50a49e3008 ("Revert "PCI: Avoid race while > enabling upstream bridges""). >=20 Thanks for the links, I wasn't aware of these discussions and patches! On PowerNV this issue is partially hidden by db2173198b95 ("powerpc/power= nv/pci: Work around races in PCI bridge enabling"), and on x86 BIOS pre-initializes al= l the bridges, so it doesn't reproduce until hotplugging in a hotplugged bridge. This patch is indeed similar to 40f11adc7cd9 ("PCI: Avoid race while enab= ling upstream bridges"), but instead of a single static mutex it adds per-device mutexe= s and prevents the dev->enable_cnt from incrementing too early. So it's not needed anymo= re to carefully select a moment safe enough to enable the device. Serge > I think the underlying design problem is that we have a driver for > device B calling pci_enable_device(), and it is changing the state of > device A (an upstream bridge). The model generally is that a driver > should only touch the device it is bound to. >=20 > It's tricky to get the locking right when several children of device A > all need to operate on A. >=20 > That's all to say I'll have to think carefully about this particular > patch, so I'll go on to the others and come back to this one. >=20 > Bjorn >=20 > [1] https://lore.kernel.org/linux-pci/1494256190-28993-1-git-send-email= -srinath.mannam@broadcom.com/T/#u > [RFC PATCH] pci: Concurrency issue in NVMe Init through PCIe switch= >=20 > [2] https://lore.kernel.org/linux-pci/1496135297-19680-1-git-send-email= -srinath.mannam@broadcom.com/T/#u > [RFC PATCH v2] pci: Concurrency issue in NVMe Init through PCIe swi= tch >=20 > [3] https://lore.kernel.org/linux-pci/1501858648-22228-1-git-send-email= -srinath.mannam@broadcom.com/T/#u > [RFC PATCH v3] pci: Concurrency issue during pci enable bridge >=20 > [4] https://lore.kernel.org/linux-pci/150547971091.977464.1629404586617= 9907260.stgit@buzz/T/#u > [PATCH bisected regression in 4.14] PCI: fix race while enabling up= stream bridges concurrently >=20 > [5] https://lore.kernel.org/linux-wireless/04c9b578-693c-1dc6-9f0f-9045= 80231b21@kernel.dk/T/#u > iwlwifi firmware load broken in current -git >=20 > [6] https://lore.kernel.org/linux-pci/744877924.5841545.1521630049567.J= avaMail.zimbra@kalray.eu/T/#u > [RFC PATCH] nvme: avoid race-conditions when enabling devices >=20 >> Signed-off-by: Sergey Miroshnichenko >> --- >> drivers/pci/pci.c | 26 ++++++++++++++++++++++---- >> drivers/pci/probe.c | 1 + >> include/linux/pci.h | 1 + >> 3 files changed, 24 insertions(+), 4 deletions(-) >> >> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c >> index f006068be209..895201d4c9e6 100644 >> --- a/drivers/pci/pci.c >> +++ b/drivers/pci/pci.c >> @@ -1615,6 +1615,8 @@ static void pci_enable_bridge(struct pci_dev *de= v) >> struct pci_dev *bridge; >> int retval; >> =20 >> + mutex_lock(&dev->enable_mutex); >> + >> bridge =3D pci_upstream_bridge(dev); >> if (bridge) >> pci_enable_bridge(bridge); >> @@ -1622,6 +1624,7 @@ static void pci_enable_bridge(struct pci_dev *de= v) >> if (pci_is_enabled(dev)) { >> if (!dev->is_busmaster) >> pci_set_master(dev); >> + mutex_unlock(&dev->enable_mutex); >> return; >> } >> =20 >> @@ -1630,11 +1633,14 @@ static void pci_enable_bridge(struct pci_dev *= dev) >> pci_err(dev, "Error enabling bridge (%d), continuing\n", >> retval); >> pci_set_master(dev); >> + mutex_unlock(&dev->enable_mutex); >> } >> =20 >> static int pci_enable_device_flags(struct pci_dev *dev, unsigned long= flags) >> { >> struct pci_dev *bridge; >> + /* Enable-locking of bridges is performed within the pci_enable_brid= ge() */ >> + bool need_lock =3D !dev->subordinate; >> int err; >> int i, bars =3D 0; >> =20 >> @@ -1650,8 +1656,13 @@ static int pci_enable_device_flags(struct pci_d= ev *dev, unsigned long flags) >> dev->current_state =3D (pmcsr & PCI_PM_CTRL_STATE_MASK); >> } >> =20 >> - if (atomic_inc_return(&dev->enable_cnt) > 1) >> + if (need_lock) >> + mutex_lock(&dev->enable_mutex); >> + if (pci_is_enabled(dev)) { >> + if (need_lock) >> + mutex_unlock(&dev->enable_mutex); >> return 0; /* already enabled */ >> + } >> =20 >> bridge =3D pci_upstream_bridge(dev); >> if (bridge) >> @@ -1666,8 +1677,10 @@ static int pci_enable_device_flags(struct pci_d= ev *dev, unsigned long flags) >> bars |=3D (1 << i); >> =20 >> err =3D do_pci_enable_device(dev, bars); >> - if (err < 0) >> - atomic_dec(&dev->enable_cnt); >> + if (err >=3D 0) >> + atomic_inc(&dev->enable_cnt); >> + if (need_lock) >> + mutex_unlock(&dev->enable_mutex); >> return err; >> } >> =20 >> @@ -1910,15 +1923,20 @@ void pci_disable_device(struct pci_dev *dev) >> if (dr) >> dr->enabled =3D 0; >> =20 >> + mutex_lock(&dev->enable_mutex); >> dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <=3D 0, >> "disabling already-disabled device"); >> =20 >> - if (atomic_dec_return(&dev->enable_cnt) !=3D 0) >> + if (atomic_dec_return(&dev->enable_cnt) !=3D 0) { >> + mutex_unlock(&dev->enable_mutex); >> return; >> + } >> =20 >> do_pci_disable_device(dev); >> =20 >> dev->is_busmaster =3D 0; >> + >> + mutex_unlock(&dev->enable_mutex); >> } >> EXPORT_SYMBOL(pci_disable_device); >> =20 >> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c >> index 2ec0df04e0dc..977a127ce791 100644 >> --- a/drivers/pci/probe.c >> +++ b/drivers/pci/probe.c >> @@ -2267,6 +2267,7 @@ struct pci_dev *pci_alloc_dev(struct pci_bus *bu= s) >> INIT_LIST_HEAD(&dev->bus_list); >> dev->dev.type =3D &pci_dev_type; >> dev->bus =3D pci_bus_get(bus); >> + mutex_init(&dev->enable_mutex); >> =20 >> return dev; >> } >> diff --git a/include/linux/pci.h b/include/linux/pci.h >> index 77448215ef5b..cb2760a31fe2 100644 >> --- a/include/linux/pci.h >> +++ b/include/linux/pci.h >> @@ -419,6 +419,7 @@ struct pci_dev { >> unsigned int no_vf_scan:1; /* Don't scan for VFs after IOV enableme= nt */ >> pci_dev_flags_t dev_flags; >> atomic_t enable_cnt; /* pci_enable_device has been called */ >> + struct mutex enable_mutex; >> =20 >> u32 saved_config_space[16]; /* Config space saved at suspend time *= / >> struct hlist_head saved_cap_space; >> --=20 >> 2.20.1 >> --SZX8gVbqDGgIpm1j0BU84HpFvIMDCudKR-- --MzcZ8NHdkHNPKzrPfEdUWBQ8QvIkrwKiD Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEHW7T7ovP/cFp8+b3DHQ+aNFTs7UFAlybrtcACgkQDHQ+aNFT s7Uzfw/+Pvmd871A1hqb3OmxCrLtfBb5rY2ePq4LkqoG0bddw4X4is1M/dj0EhxX NLvX6VPOYMSPz9U3GWW7fjrwSa0LKX8hOsAH1zSeYU7YEclY/JzsbuVRqu51bjOZ I5rXmXumyXmrlPoBKQGGcGjUkhN2QPVT4fbJMlzQfLAU9bmJse4zlIKTaIux4bf/ iNfSZ7SWBo48dKBR+4nldHgy3pHg8FEYCvfmlsjoWcEsuTVZMU6ZU/5waJ+Atw63 S7UFlduSmHKQ36DLnrhbt7P13paX2REIvs7XAyUqYTMrffv6gl3+sIiTPPbcexBS UGgok2lfIJ5EG+GPn5AtrmWniZsCSv7hmcHL/qGekzwP8P4JFVhmgIlh0HcLATZr z+oevX1xZuGHN8OWA8Av4qmVw5rnEPt3ON+vvEsDtpET4U1Y6zKFC5ePNNwQV0/l 38zc+aRXSBKq5tWE652jVhb3GQshQ2aXAPFfPReKYfiWeFuSOkJwdDzNjwBquLKn ijHdIw0we7hoFr9Z665RLLyAf0haaPiu3dGRqNHFiT5gjLjfefk7+Y4lhoeCM5cS BU3VC0GaYl2LU+RXBJK8Cvk/QK3JyleD6f3PCiKRV8jFPfNuU9+ZeJY/SMJQOMDq H3JdM7o/q+16GYcN0//BbH1Sldkw06HmDnEBMSYrbPSBfGH9ivE= =hfd7 -----END PGP SIGNATURE----- --MzcZ8NHdkHNPKzrPfEdUWBQ8QvIkrwKiD--