From owner-freebsd-current@freebsd.org Wed Aug 17 06:51:13 2016 Return-Path: Delivered-To: freebsd-current@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 7679BBBDC7F for ; Wed, 17 Aug 2016 06:51:13 +0000 (UTC) (envelope-from junchoon@dec.sakura.ne.jp) Received: from dec.sakura.ne.jp (dec.sakura.ne.jp [210.188.226.8]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 461321785 for ; Wed, 17 Aug 2016 06:51:13 +0000 (UTC) (envelope-from junchoon@dec.sakura.ne.jp) Received: from fortune.joker.local (123-48-23-227.dz.commufa.jp [123.48.23.227]) (authenticated bits=0) by dec.sakura.ne.jp (8.15.2/8.15.2/[SAKURA-WEB]/20080708) with ESMTPA id u7H6p7ps036070; Wed, 17 Aug 2016 15:51:08 +0900 (JST) (envelope-from junchoon@dec.sakura.ne.jp) Date: Wed, 17 Aug 2016 15:51:07 +0900 From: Tomoaki AOKI To: freebsd-current@freebsd.org Cc: imp@bsdimp.com Subject: Re: Missingly recognized ADA_Q_NCQ_TRIM_BROKEN for Crucial M550 MU02 Message-Id: <20160817155107.e4c94e19bafcce862ea1dc80@dec.sakura.ne.jp> In-Reply-To: <20160817100028.4248dccb54fd4c34964d8e3a@dec.sakura.ne.jp> References: <20160816230113.f0cf8e9bfe740244ad16856c@dec.sakura.ne.jp> <20160817100028.4248dccb54fd4c34964d8e3a@dec.sakura.ne.jp> Organization: Junchoon corps X-Mailer: Sylpheed 3.5.1 (GTK+ 2.24.29; amd64-portbld-freebsd11.0) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-2022-JP Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 17 Aug 2016 06:51:13 -0000 Hi. As I already added comment to Bug 210686, the proposed patch worked fine for me. The essential point of the patch is NOT using "*" in the middle. This causes, for example, M550 SSDs 1TB model and others must be treated separately as the proposed patch does. CT256M550SSD1 -> Crucial CT???M550* CT1000M550SSD1 -> Crucial CT????M550* To re-integrate these, wildcard processing in kernel needs to be rewritten, maybe causing slightly larger kernel. Re-confirming original codes, below would cause all SSDs starting with "Crucial CT" except M500 MU07 (appeares before this) to be recognized as ADA_Q_NCQ_TRIM_BROKEN. { /* * Crucial M500 SSDs all other firmware * NCQ Trim doesn't work */ { T_DIRECT, SIP_MEDIA_FIXED, "*", "Crucial CT*M500*", "*" }, /*quirks*/ADA_Q_NCQ_TRIM_BROKEN }, On Wed, 17 Aug 2016 10:00:28 +0900 Tomoaki AOKI wrote: > Hi. > > On Tue, 16 Aug 2016 13:48:40 -0600 > Warner Losh wrote: > > > What does camcontrol devlist say? I'm guessing firmware MU02, but I > > want to make sure... > > % camcontrol devlist > at scbus0 target 0 lun 0 (ada0,pass0) > at scbus1 target 0 lun 0 > (ada1,pass1) at scbus4 target 0 lun > 0 (ses0,pass2) > > Above includes another drive and ses0. > Another drive is recongized properly. > > ada1 at ahcich1 bus 0 scbus1 target 0 lun 0 > ada1: ACS-2 ATA SATA 3.x device > ada1: Serial Number *************** > ada1: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 512bytes) > ada1: Command Queueing enabled > ada1: 238475MB (488397168 512 byte sectors) > ada1: quirks=0x3<4K,NCQ_TRIM_BROKEN> > Steering write from 0 kBps to 300000 kBps > > Samsung consumer SSD should have 4k and NCQ_TRIM_BROKEN quirk, and > recognized so. > > > > It shouldn't match this one. The MU07 exception works for my M500's, > > so I'm confused... > > Me too very confused. :-( > This shouldn't happen with this code path. > > > > Please file a bugzilla ticket. > > Found already filed PR for other model while attempting to file one. > > Bug 210686 - NCQ_TRIM_BROKEN quirk mismatch > https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=210686 > > Please work on it. (So do I.) > > I've not yet tested the proposed patch (found just now), but will test > later and add feedback to it. > > > > > Warner > > > > On Tue, Aug 16, 2016 at 8:01 AM, Tomoaki AOKI wrote: > > > Hi. > > > > > > I noticed that my Crucial M550 SSD (firmware MU02) is missingly > > > recognized as quirk ADA_Q_NCQ_TRIM_BROKEN. > > > If I understand the source (introduced first at r298002, before > > > stable/11 branched) correctly, only firmware MU01 should be recognized > > > so for the SSD model, so I have no idea why. :-( > > > > > > Below are the related portion of dmesg. stable/11 at r304189, amd64. > > > (Built with options CAM_IOSCHED_DYNAMIC.) > > > > > > ada0 at ahcich0 bus 0 scbus0 target 0 lun 0 > > > ada0: ACS-2 ATA SATA 3.x device > > > ada0: Serial Number ************ > > > ada0: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes) > > > ada0: Command Queueing enabled > > > ada0: 976762MB (2000409264 512 byte sectors) > > > ada0: quirks=0x2 > > > Steering write from 0 kBps to 300000 kBps > > > > > > One thing to note: The SSD was shipped with MU01 and updated to MU02 > > > using CD image obtained from Crucial Japan website. > > > > > > Need Bugzilla ticket although it's not a severe problem (goes safer > > > side, and easy to workaround via loader.conf)? > > > > > > -- > > > Tomoaki AOKI > > _______________________________________________ > > freebsd-current@freebsd.org mailing list > > https://lists.freebsd.org/mailman/listinfo/freebsd-current > > To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org" > > > > > -- > Tomoaki AOKI > _______________________________________________ > freebsd-current@freebsd.org mailing list > https://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org" > -- 青木 知明 [Tomoaki AOKI]