Received: by 2002:a05:6a10:a0d1:0:0:0:0 with SMTP id j17csp1758634pxa; Thu, 6 Aug 2020 15:30:13 -0700 (PDT) X-Google-Smtp-Source: ABdhPJyCYMRdYwJa+Fbd7ODU4SBjoB2/z++Bz8o2i7mBHjcXXXLXFZQIukGt6rY6D70bZTbn77aK X-Received: by 2002:a17:906:3a19:: with SMTP id z25mr307405eje.117.1596753013388; Thu, 06 Aug 2020 15:30:13 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1596753013; cv=none; d=google.com; s=arc-20160816; b=0FIjpqgu8/PbRt2ZnXB0F96nHWqFok7JGAX0mpbqhDVW99wmDEhB6kmcFmhyzpkpjs /WjLsdKYoxX+D7LzORn/UxWWkPeWrnj2xyrK/hHHaKTG0I9O4oJ4Bj12a0LSsAIb/Rfm 6/r7RN2S5cQC2uCk3mStHMJJGcmOtFT+ndco1p9pMu64kR3tOXmbzBxX8WRB2Ey/pvfT tAgrNUBNtU7EWHCobNFHgNXa+I8GPrN1K7+2jlHTOCNiUcG16xOM9NJG5yII9NSTM0Ib o44FNGHf8kFkdz364h0+WN2egCjxjouGJBSk5MKRUHsKFkN/LVy6CJ5kWqOX+tJ909Hn d1kg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:in-reply-to:content-disposition :mime-version:message-id:subject:cc:to:from:date:dkim-signature; bh=sqnkMHiDUZmI4Aymw6r1hUO6KOniy5R3cJzybGjE4MU=; b=nAx7WrvCAq8dgz5xoW1YJ5/2/FzWRoPYzpH2PhO245+ZCiSLd/ACtEGFi46oXlXoeR a0dlVhw6XM6CvYHGE26qvukFtVoMG74I0iiSHB/DzzOFp5IUZ2xUeuCQwqU0SB8BYs/P yoLrDIuj/TwgAzKVRhrb/HNU0SUokOYwDQ5UC2CxtFqNY7Bc1cArepeCrG2K9ZRrxA5N RqDmQTVXYpH4R0S5/Z198LYLFrWYNfEmE6bHpI2drP7ddBZ2iqgZ9ti4KF6ktiwrtXWu ahXEnZIJjmNugKxL24KfgSRkj2McBehqVaOQ1MzjDkETG7yKyW9q7t8Zd9MbmhTQ+3FS Q7pQ== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@kernel.org header.s=default header.b=qPv7UMmu; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [23.128.96.18]) by mx.google.com with ESMTP id a1si3845522edy.514.2020.08.06.15.29.50; Thu, 06 Aug 2020 15:30:13 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18; Authentication-Results: mx.google.com; dkim=pass header.i=@kernel.org header.s=default header.b=qPv7UMmu; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726204AbgHFW1P (ORCPT + 99 others); Thu, 6 Aug 2020 18:27:15 -0400 Received: from mail.kernel.org ([198.145.29.99]:49868 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725927AbgHFW1P (ORCPT ); Thu, 6 Aug 2020 18:27:15 -0400 Received: from localhost (130.sub-72-107-113.myvzw.com [72.107.113.130]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPSA id 74D17221E2; Thu, 6 Aug 2020 22:27:14 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1596752834; bh=WgdGpbjoOqKRB5xCJJB1QUUR3NAm5AiI/g2A57rj4KI=; h=Date:From:To:Cc:Subject:In-Reply-To:From; b=qPv7UMmumdkk5aGIxvQqVqqb18le/xYKnvT2NLDxW1vNEmC3Ph4Qht7skoNahofop f4yilJWzb9U+BSlCfhrRXW+mYxFr3x4aM7SMY6d+2qsXOAomi/LJl0WMvWd4RgfyQy b/Rd8PhebQXbHyz/CLBputTdShNapNUxGxWiVOIc= Date: Thu, 6 Aug 2020 17:27:13 -0500 From: Bjorn Helgaas To: Tuan Phan Cc: patches@amperecomputing.com, Bjorn Helgaas , "Rafael J. Wysocki" , Len Brown , linux-pci@vger.kernel.org, linux-acpi@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] PCI/ACPI: Add Ampere Altra SOC MCFG quirk Message-ID: <20200806222713.GA704188@bjorn-Precision-5520> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1596751055-12316-1-git-send-email-tuanphan@os.amperecomputing.com> Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Aug 06, 2020 at 02:57:34PM -0700, Tuan Phan wrote: > Ampere Altra SOC supports only 32-bit ECAM reading. Therefore, > add an MCFG quirk for the platform. This is interesting. So this host bridge supports sub 32-bit config *writes*, but not reads? I actually don't know whether that complies with the spec or not. If config registers are not allowed to have side effects on read, this *would* be compliant. PCIe r5.0, sec 7.4, doesn't list any register types with read side effects, so there shouldn't be any in the registers defined by the spec. But I would think device-specific registers could do whatever they wanted, e.g., reading an interrupt status register or something could clear it. And I think sec 7.2.2 about ECAM implicitly requires sub 32-bit accesses because it mentions the access size and byte enables. Is this a one-off situation where future hardware will allow sub 32-bit reads and writes? We don't want a stream of quirks for future devices. > Signed-off-by: Tuan Phan > --- > drivers/acpi/pci_mcfg.c | 20 ++++++++++++++++++++ > drivers/pci/ecam.c | 10 ++++++++++ > include/linux/pci-ecam.h | 1 + > 3 files changed, 31 insertions(+) > > diff --git a/drivers/acpi/pci_mcfg.c b/drivers/acpi/pci_mcfg.c > index 54b36b7ad47d..e526571e0ebd 100644 > --- a/drivers/acpi/pci_mcfg.c > +++ b/drivers/acpi/pci_mcfg.c > @@ -142,6 +142,26 @@ static struct mcfg_fixup mcfg_quirks[] = { > XGENE_V2_ECAM_MCFG(4, 0), > XGENE_V2_ECAM_MCFG(4, 1), > XGENE_V2_ECAM_MCFG(4, 2), > + > +#define ALTRA_ECAM_QUIRK(rev, seg) \ > + { "Ampere", "Altra ", rev, seg, MCFG_BUS_ANY, &pci_32b_read_ops } > + > + ALTRA_ECAM_QUIRK(1, 0), > + ALTRA_ECAM_QUIRK(1, 1), > + ALTRA_ECAM_QUIRK(1, 2), > + ALTRA_ECAM_QUIRK(1, 3), > + ALTRA_ECAM_QUIRK(1, 4), > + ALTRA_ECAM_QUIRK(1, 5), > + ALTRA_ECAM_QUIRK(1, 6), > + ALTRA_ECAM_QUIRK(1, 7), > + ALTRA_ECAM_QUIRK(1, 8), > + ALTRA_ECAM_QUIRK(1, 9), > + ALTRA_ECAM_QUIRK(1, 10), > + ALTRA_ECAM_QUIRK(1, 11), > + ALTRA_ECAM_QUIRK(1, 12), > + ALTRA_ECAM_QUIRK(1, 13), > + ALTRA_ECAM_QUIRK(1, 14), > + ALTRA_ECAM_QUIRK(1, 15), > }; > > static char mcfg_oem_id[ACPI_OEM_ID_SIZE]; > diff --git a/drivers/pci/ecam.c b/drivers/pci/ecam.c > index 8f065a42fc1a..b54d32a31669 100644 > --- a/drivers/pci/ecam.c > +++ b/drivers/pci/ecam.c > @@ -168,4 +168,14 @@ const struct pci_ecam_ops pci_32b_ops = { > .write = pci_generic_config_write32, > } > }; > + > +/* ECAM ops for 32-bit read only (non-compliant) */ > +const struct pci_ecam_ops pci_32b_read_ops = { > + .bus_shift = 20, > + .pci_ops = { > + .map_bus = pci_ecam_map_bus, > + .read = pci_generic_config_read32, > + .write = pci_generic_config_write, > + } > +}; > #endif > diff --git a/include/linux/pci-ecam.h b/include/linux/pci-ecam.h > index 1af5cb02ef7f..033ce74f02e8 100644 > --- a/include/linux/pci-ecam.h > +++ b/include/linux/pci-ecam.h > @@ -51,6 +51,7 @@ extern const struct pci_ecam_ops pci_generic_ecam_ops; > > #if defined(CONFIG_ACPI) && defined(CONFIG_PCI_QUIRKS) > extern const struct pci_ecam_ops pci_32b_ops; /* 32-bit accesses only */ > +extern const struct pci_ecam_ops pci_32b_read_ops; /* 32-bit read only */ > extern const struct pci_ecam_ops hisi_pcie_ops; /* HiSilicon */ > extern const struct pci_ecam_ops thunder_pem_ecam_ops; /* Cavium ThunderX 1.x & 2.x */ > extern const struct pci_ecam_ops pci_thunder_ecam_ops; /* Cavium ThunderX 1.x */ > -- > 2.18.4 >