From owner-freebsd-hackers@freebsd.org Fri May 19 16:00:13 2017 Return-Path: Delivered-To: freebsd-hackers@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 1175DD7313A for ; Fri, 19 May 2017 16:00:13 +0000 (UTC) (envelope-from rpokala@mac.com) Received: from mr11p00im-asmtp002.me.com (mr11p00im-asmtp002.me.com [17.110.69.253]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id EBE2B7D5; Fri, 19 May 2017 16:00:12 +0000 (UTC) (envelope-from rpokala@mac.com) Received: from process-dkim-sign-daemon.mr11p00im-asmtp002.me.com by mr11p00im-asmtp002.me.com (Oracle Communications Messaging Server 7.0.5.38.0 64bit (built Feb 26 2016)) id <0OQ700D00FMHZG00@mr11p00im-asmtp002.me.com>; Fri, 19 May 2017 15:00:06 +0000 (GMT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mac.com; s=4d515a; t=1495206006; bh=/ZBAL3AbgIJg785SsY4duuZs6WcxIdsyQMZMLPe9L0I=; h=Date:Subject:From:To:Message-id:MIME-version:Content-type; b=MDTDn7YnjrxsktVoV1bVjKaTSZPcsPgprTt+di5IvpUtkywoedsnTVjonOgabYNXs Xu/vWNjK6zeGovYXh5xtGCIaNUHAIXB0Qv8PmcnksDgQ2z3kNBk/DGIgOnawaDUD8l xuMgG5v3Co3/Oej5smGc5y1a7EJDLR1Z5p62Di/KpWcKUXpxmYOCojtpQxsy0j8aKc To3KF1OMAt7G1q+Cj6cHV0PE/Q+LDCWmYhtmeAQg9f55RQp98Zq1p27HiFAXanpE5i Tl7Mb0PQ//IEOD0JXgzVThdrM0+44/DyJmbd2nYzysRm33Z8a5BTPsbUYZ7+W0mHk2 3ESCyzPJo/pAA== Received: from icloud.com ([127.0.0.1]) by mr11p00im-asmtp002.me.com (Oracle Communications Messaging Server 7.0.5.38.0 64bit (built Feb 26 2016)) with ESMTPSA id <0OQ7002EUGC2O530@mr11p00im-asmtp002.me.com>; Fri, 19 May 2017 15:00:03 +0000 (GMT) X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:,, definitions=2017-05-19_09:,, signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 clxscore=1034 suspectscore=0 malwarescore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1701120000 definitions=main-1705190094 User-Agent: Microsoft-MacOutlook/f.21.0.170409 Date: Fri, 19 May 2017 08:00:01 -0700 Subject: Re: iMC SMBus controller driver From: Ravi Pokala To: Andriy Gapon , "freebsd-hackers@freebsd.org" Message-id: <2428158C-45B3-434A-A2FD-B24A655C7564@panasas.com> Thread-topic: iMC SMBus controller driver References: <532facb6-00bb-52d2-3e99-4e8c1bc16670@FreeBSD.org> In-reply-to: <532facb6-00bb-52d2-3e99-4e8c1bc16670@FreeBSD.org> MIME-version: 1.0 Content-type: text/plain; charset=UTF-8 Content-transfer-encoding: 7bit X-Mailman-Approved-At: Fri, 19 May 2017 17:32:34 +0000 X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 May 2017 16:00:13 -0000 -----Original Message----- > From: Andriy Gapon > Date: 2017-05-19, Friday at 02:21 > To: Ravi Pokala , "freebsd-hackers@freebsd.org" > Subject: Re: iMC SMBus controller driver > > On 19/05/2017 00:22, Ravi Pokala wrote: >> Hi folks, >> >> In recent years, Intel has put an SMBus controller on the CPU's integrated Memory Controller (iMC). Does anyone have a driver for that? Even if it's not something that you're willing to open up, just knowing that someone has done it would be nice. >> >> (I'm specifically interested in Haswell/Broadwell Xeon, but I'd imagine the programming model is the same for the iMC/SMBus controller on all the recent CPUs.) > > Could ismt be what you are looking for? Yes and no. ismt(4) is a non-ichsmb SMBus controller, but the driver only supports Atom S1200 (Briarwood) and Atom C2000 (Avoton / Rangeley) SoCs. But maybe it just needs to have newer devices added to the list...? As it happens, I have the datasheet for the Rangeley at my fingertips. I searched for some of the magic numbers from ismt(4), and found where the "SMT" SMBus controller (PCI 0:13:0) is documented. I also did some digging on Intel's site, and found Volume 2 of the Broadwell Xeon datasheet, where the iMC (PCI 0:19:0 and 0:22:0) is documented. Unfortunately, Rangeley's SMT SMBus controller and Broadwell's iMC's SMBus controller bear no resemblance to each other. So, no, ismt(4) is not the droid ^W driver I'm looking for. :-/ Thanks, Ravi (rpokala@) > -- > Andriy Gapon