From owner-svn-src-projects@freebsd.org Tue Oct 16 17:07:03 2018 Return-Path: Delivered-To: svn-src-projects@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id E6AE710DADFF for ; Tue, 16 Oct 2018 17:07:02 +0000 (UTC) (envelope-from leandro.lupori@gmail.com) Received: from mail-yb1-xb35.google.com (mail-yb1-xb35.google.com [IPv6:2607:f8b0:4864:20::b35]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 782BE75FC1; Tue, 16 Oct 2018 17:07:02 +0000 (UTC) (envelope-from leandro.lupori@gmail.com) Received: by mail-yb1-xb35.google.com with SMTP id o63-v6so9202967yba.2; Tue, 16 Oct 2018 10:07:02 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=h2KQrxGLhJkppGut9QjcAs8FB1grLsEYnNPDUBcjEd4=; b=DbYgxuWUWliBrfuTzj5DOh3EqEVplRRHCvr7vg+nFmm7DHea70DmhkUpaRzsnOWeY5 e6n5/OzOU0pAUMd10h2Ly17QeS0OolfIv6HkyayJ35oXwVeDiGsER3cMPO5Efk9jTIY1 nBul0fI3iGix9pbvzX2ig8VD5YgEygy/xOu+oVf7PdQRw4L/4yqZTO7DNR5sAYqaF1Nt rlmh1j8sxyzp3e9AM2eaCSHToBLTz5z0GKo+7EbXU79RPg/wzyRhKjYYOrkiwjz6HWmX C+C1z3fb4itMv00bgEN6J7uMMRz2lxl6/XtXT9CiD5iUI8gPK1bLQkc1fDR4FrDMKB63 0SuQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=h2KQrxGLhJkppGut9QjcAs8FB1grLsEYnNPDUBcjEd4=; b=RLYTrHsFqXhEc7hOPe4KVB5aG8Ipz/z7o3+F95NtBYchN0FoTDKks2uxAI8f5ryA4E 0VFzeYKtGJBU48PNPlToOZJ9OrMk+MOSS4mQJvx5fpo9a67DwFY+Om+SbXSBl7++ZiN3 URcazw4mZqXFxcoqf6EUj8rj5Zj9j1y1igexCLxm0YiM7s91j2bsrkjJScyGx5ytXPLi 4L2Y/S48hL5TvfI90qBKBu2B/fUl9XBG6Qg0yvYLR3bEtPCpLpTlM1MYhy2HIPuGZqKp TqpugBHJedAH/7LvEaXMHxbPytHIYic1Y85amfUj5SHh05GKEe/hvAXfAFPrktKNHAyv TWuA== X-Gm-Message-State: ABuFfojvqsIeK97GQKS/GSYB1Ny8/iwTY/t1DiCYQKBmw2viZFSp+Y3F cI3K31acoQsOxQGvDZEqF8WhAu4phNbVDA2qD+UwCP7O X-Google-Smtp-Source: ACcGV606ERknphWQRO8TCPxa95ar6pOb7R3fcT+lmAa4pmeMkl/6dTN94ANOgll58YY/y912+a2SiWDjI9pnjWgngZU= X-Received: by 2002:a25:3748:: with SMTP id e69-v6mr5915389yba.245.1539709621329; Tue, 16 Oct 2018 10:07:01 -0700 (PDT) MIME-Version: 1.0 References: <201810161636.w9GGa3Nl001743@repo.freebsd.org> <8a90e167-a4ac-c32f-ae9c-d71f0fcb5471@FreeBSD.org> In-Reply-To: <8a90e167-a4ac-c32f-ae9c-d71f0fcb5471@FreeBSD.org> From: Leandro Date: Tue, 16 Oct 2018 14:06:49 -0300 Message-ID: Subject: Re: svn commit: r339382 - projects/power8_bringup_hacks/sys/dev/bge To: jhb@freebsd.org Cc: luporl@freebsd.org, src-committers@freebsd.org, svn-src-projects@freebsd.org Content-Type: text/plain; charset="UTF-8" X-BeenThere: svn-src-projects@freebsd.org X-Mailman-Version: 2.1.27 Precedence: list List-Id: "SVN commit messages for the src " projects" tree" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 16 Oct 2018 17:07:03 -0000 The issue seems to be related to wrong configuration of DMA regions in opal_pci.c, which affects all PowerNV platforms. This is where I'm going to look next, that, if solved, we should be able to remove this hack. This is just to make the network interface stable, until the real issue is resolved. On Tue, Oct 16, 2018 at 1:55 PM John Baldwin wrote: > > On 10/16/18 9:36 AM, Leandro Lupori wrote: > > Author: luporl > > Date: Tue Oct 16 16:36:02 2018 > > New Revision: 339382 > > URL: https://svnweb.freebsd.org/changeset/base/339382 > > > > Log: > > bge(4): limit DMA address to 2GB > > > > This is a temporary hack, to avoid POWER8 Broadcom NICs from becoming > > unusable after transferring a reasonable amount of data. > > The issue is the same as that of AHCI, where DMA addresses above 2GB > > do not work and causes PCI errors. > > Is the issue bge specific or specific to the PCI chipset in these boxes? > If the latter, the better approach would be to set the limit in a tag the > chipset provides that all child devices would then inherit. > > -- > John Baldwin > >