From owner-freebsd-bugs@FreeBSD.ORG Tue Oct 8 00:40:00 2013 Return-Path: Delivered-To: freebsd-bugs@smarthost.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTP id 9B47DD29 for ; Tue, 8 Oct 2013 00:40:00 +0000 (UTC) (envelope-from gnats@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:1900:2254:206c::16:87]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mx1.freebsd.org (Postfix) with ESMTPS id 7A46021C1 for ; Tue, 8 Oct 2013 00:40:00 +0000 (UTC) Received: from freefall.freebsd.org (localhost [127.0.0.1]) by freefall.freebsd.org (8.14.7/8.14.7) with ESMTP id r980e0im007794 for ; Tue, 8 Oct 2013 00:40:00 GMT (envelope-from gnats@freefall.freebsd.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.14.7/8.14.7/Submit) id r980e0OE007793; Tue, 8 Oct 2013 00:40:00 GMT (envelope-from gnats) Resent-Date: Tue, 8 Oct 2013 00:40:00 GMT Resent-Message-Id: <201310080040.r980e0OE007793@freefall.freebsd.org> Resent-From: FreeBSD-gnats-submit@FreeBSD.org (GNATS Filer) Resent-To: freebsd-bugs@FreeBSD.org Resent-Reply-To: FreeBSD-gnats-submit@FreeBSD.org, Adam McDougall Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTP id CDFBAD23 for ; Tue, 8 Oct 2013 00:39:37 +0000 (UTC) (envelope-from nobody@FreeBSD.org) Received: from oldred.freebsd.org (oldred.freebsd.org [8.8.178.121]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.freebsd.org (Postfix) with ESMTPS id BBE5621BD for ; Tue, 8 Oct 2013 00:39:37 +0000 (UTC) Received: from oldred.freebsd.org ([127.0.1.6]) by oldred.freebsd.org (8.14.5/8.14.7) with ESMTP id r980dbxY075922 for ; Tue, 8 Oct 2013 00:39:37 GMT (envelope-from nobody@oldred.freebsd.org) Received: (from nobody@localhost) by oldred.freebsd.org (8.14.5/8.14.5/Submit) id r980dbtu075918; Tue, 8 Oct 2013 00:39:37 GMT (envelope-from nobody) Message-Id: <201310080039.r980dbtu075918@oldred.freebsd.org> Date: Tue, 8 Oct 2013 00:39:37 GMT From: Adam McDougall To: freebsd-gnats-submit@FreeBSD.org X-Send-Pr-Version: www-3.1 Subject: kern/182818: Slow writes to AHCI SATA on Soekris net6501 (Intel EG20T) in *BSD X-BeenThere: freebsd-bugs@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Bug reports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 08 Oct 2013 00:40:00 -0000 >Number: 182818 >Category: kern >Synopsis: Slow writes to AHCI SATA on Soekris net6501 (Intel EG20T) in *BSD >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Oct 08 00:40:00 UTC 2013 >Closed-Date: >Last-Modified: >Originator: Adam McDougall >Release: 10.0-ALPHA2 >Organization: >Environment: FreeBSD hoegaarden 10.0-ALPHA2 FreeBSD 10.0-ALPHA2 #0 r255793: Sun Sep 22 17:05:25 EDT 2013 root@build9:/usr/obj/proto/src10/src/sys/AMD64-10 amd64 >Description: I discovered these symptoms initially on 9.x, and this weekend upgraded to 10 just to check for improvement. The total AHCI SATA write speed on the net6501 (*BSD only?) has a strange 20MB/sec limitation but reads can go over 100MB/sec. If I write to one disk I get 20MB/sec, but if I write to both SATA disks (such as mirroring) I get 10MB/sec each for a total of 20. Writes are equally slow on a high end SSD. Both someone running OpenBSD and I running FreeBSD reported the same symptoms to the soekris-tech mailing list and received no useful replies towards getting that problem solved. I believe I duplicated the symptoms on NetBSD. I tested the write speed briefly with Linux and it did not appear to have the 20MB/sec limitation (I wrote enough data that I was satisfied it could not be cached in ram). I did confirm it was using MSI of some form with boot -v. The disks I tested with are fine in normal computers. I have three of these systems. They are all acting as firewalls where write speed is not critical to operation, although one is considered a spare that I can experiment with. They are not my property so I don't think I can mail one for personal inspection. >How-To-Repeat: Use something (dd?) to write to the raw devices or to a filesystem, use gstat -I 50000 to confirm a total limitation of 20MB/sec disk IO whether it is to one disk or two (10 each). >Fix: >Release-Note: >Audit-Trail: >Unformatted: