From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCH #upstream-fixes] libata: implement and use HORKAGE_NOSETXFER Date: Thu, 09 Jul 2009 08:07:44 +0900 Message-ID: <4A5526C0.2030302@kernel.org> References: <1229894315.6931.17.camel@Thutmosis> <494EDDBE.3010306@shaw.ca> <494F65C1.80602@kernel.org> <1229947998.6045.10.camel@Thutmosis> <49505500.8080409@kernel.org> <1230126055.6430.17.camel@Thutmosis> <495886E7.7050204@kernel.org> <1230586376.11130.46.camel@Thutmosis> <1231282276.6376.13.camel@Thutmosis> <49640CE1.6070808@kernel.org> <20090107094043.580f5bf7@lxorguk.ukuu.org.uk> <496483A5.6020105@kernel.org> <20090107105339.289f47a4@lxorguk.ukuu.org.uk> <49648C41.6030002@kernel.org> <4A1A0C1B.7020801@kernel.org> <20090525091110.626da5f8@lxorguk.ukuu.org.uk> <4A545236.6080808@kernel.org> <20090708111413.7bb25e1e@lxorguk.ukuu.org.uk> <4A547FF6.7010609@kernel.org> <20090708132104.10e969a1@lxorguk.ukuu.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from hera.kernel.org ([140.211.167.34]:36573 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754280AbZGHXJF (ORCPT ); Wed, 8 Jul 2009 19:09:05 -0400 In-Reply-To: <20090708132104.10e969a1@lxorguk.ukuu.org.uk> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Alan Cox Cc: linux--ide-momail.e4ward.com-linux--ide-vger.kernel.org-CDD-6w99-4@reply.e4ward.com, Jeff Garzik , Robert Hancock , peter.klotz@aon.at, linux-ide@vger.kernel.org, m.nov4k@gmail.com, lars21ce@gmx.de Hello, Alan Cox wrote: >> 2. Skip SETXFER. Any direct SATA connection to the device would work >> fine. >> >> 2. Ignore SETXFER timeout. Probably would work for TF based >> controllers but it's largely untested for sata controllers and the >> behavior is likely undefined for more advanced controllers. > >> So, #2 seems like the logical choice here. If worst comes to worst > > You have two #2's Oops. > but I agree that providing its for this specific drive > and we document it carefully so people don't add it to stuff that is > wrong then your first #2 is probably safest. > > Possibly HORKAGE_NO_SETXFER_SATA and check word 93 ? How about something like the following? diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 045a486..2c6aeda 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -3392,17 +3392,27 @@ int ata_down_xfermask_limit(struct ata_device *dev, unsigned int sel) static int ata_dev_set_mode(struct ata_device *dev) { + struct ata_port *ap = dev->link->ap; struct ata_eh_context *ehc = &dev->link->eh_context; + const bool nosetxfer = dev->horkage & ATA_HORKAGE_NOSETXFER; const char *dev_err_whine = ""; int ign_dev_err = 0; - unsigned int err_mask; + unsigned int err_mask = 0; int rc; dev->flags &= ~ATA_DFLAG_PIO; if (dev->xfer_shift == ATA_SHIFT_PIO) dev->flags |= ATA_DFLAG_PIO; - err_mask = ata_dev_set_xfermode(dev); + if (nosetxfer && ap->flags & ATA_FLAG_SATA && ata_id_is_sata(dev->id)) + dev_err_whine = " (SET_XFERMODE skipped)"; + else { + if (nosetxfer) + ata_dev_printk(dev, KERN_WARNING, + "NOSETXFER but PATA detected - can't " + "skip SETXFER, might malfunction\n"); + err_mask = ata_dev_set_xfermode(dev); + } if (err_mask & ~AC_ERR_DEV) goto fail; @@ -4297,6 +4307,12 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { /* Devices which aren't very happy with higher link speeds */ { "WD My Book", NULL, ATA_HORKAGE_1_5_GBPS, }, + /* + * Devices which choke on SETXFER. Applies only if both the + * device and controller are SATA. + */ + { "PIONEER DVD-RW DVRTD08", "1.00", ATA_HORKAGE_NOSETXFER }, + /* End Marker */ { } }; diff --git a/include/linux/libata.h b/include/linux/libata.h index 3d501db..2b641af 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -385,6 +385,7 @@ enum { not multiple of 16 bytes */ ATA_HORKAGE_FIRMWARE_WARN = (1 << 12), /* firmware update warning */ ATA_HORKAGE_1_5_GBPS = (1 << 13), /* force 1.5 Gbps */ + ATA_HORKAGE_NOSETXFER = (1 << 14), /* DMA mask for user DMA control: User visible values; DO NOT renumber */ -- tejun