Merge branch 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev

* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev: (54 commits)
  Revert "pata_sis: Implement MWDMA for the UDMA 133 capable chips"
  libata: Clarify ata_set_lba_range_entries function
  libata: Report zeroed read after TRIM and max discard size
  pata_hpt3x2n: fix overclocked MWDMA0 timing
  pata_it8213: MWDMA0 is unsupported
  [libata] MWDMA0 is unsupported on PIIX-like PATA controllers
  pata_via: clear UDMA transfer mode bit for PIO and MWDMA
  pata_sis: Power Management fix
  pata_rz1000: Power Management fix
  pata_radisys: fix UDMA handling
  pata_ns87415: Power Management fix
  pata_marvell: fix marvell_pre_reset() documentation
  pata_legacy: add pointers to QDI65x0 documentation
  pata_legacy: fix access to control register for QDI6580
  pata_legacy: fix QDI6580DP support
  pata_it8213: fix it8213_pre_reset() documentation
  pata_it8213: fix wrong MWDMA timings being programmed
  pata_it8213: fix PIO2 underclocking
  pata_it8213: fix wrong PIO timings being programmed
  pata_it8213: fix UDMA handling
  ...

39 files changed, 844 insertions, 272 deletions

diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index f2df6e2a224..676f08b004b 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -3,7 +3,7 @@
 #
 
 menuconfig ATA
-	tristate "Serial ATA (prod) and Parallel ATA (experimental) drivers"
+	tristate "Serial ATA and Parallel ATA drivers"
 	depends on HAS_IOMEM
 	depends on BLOCK
 	depends on !(M32R || M68K) || BROKEN
@@ -374,8 +374,8 @@ config PATA_HPT366
 	  If unsure, say N.
 
 config PATA_HPT37X
-	tristate "HPT 370/370A/371/372/374/302 PATA support (Experimental)"
-	depends on PCI && EXPERIMENTAL
+	tristate "HPT 370/370A/371/372/374/302 PATA support"
+	depends on PCI
 	help
 	  This option enables support for the majority of the later HPT
 	  PATA controllers via the new ATA layer.
@@ -383,8 +383,8 @@ config PATA_HPT37X
 	  If unsure, say N.
 
 config PATA_HPT3X2N
-	tristate "HPT 372N/302N PATA support (Experimental)"
-	depends on PCI && EXPERIMENTAL
+	tristate "HPT 372N/302N PATA support"
+	depends on PCI
 	help
 	  This option enables support for the N variant HPT PATA
 	  controllers via the new ATA layer
@@ -401,7 +401,7 @@ config PATA_HPT3X3
 	  If unsure, say N.
 
 config PATA_HPT3X3_DMA
-	bool "HPT 343/363 DMA support (Experimental)"
+	bool "HPT 343/363 DMA support"
 	depends on PATA_HPT3X3
 	help
 	  This option enables DMA support for the HPT343/363
@@ -510,8 +510,8 @@ config PATA_NETCELL
 	  If unsure, say N.
 
 config PATA_NINJA32
-	tristate "Ninja32/Delkin Cardbus ATA support (Experimental)"
-	depends on PCI && EXPERIMENTAL
+	tristate "Ninja32/Delkin Cardbus ATA support"
+	depends on PCI
 	help
 	  This option enables support for the Ninja32, Delkin and
 	  possibly other brands of Cardbus ATA adapter
@@ -573,6 +573,14 @@ config PATA_PCMCIA
 
 	  If unsure, say N.
 
+config PATA_PDC2027X
+	tristate "Promise PATA 2027x support"
+	depends on PCI
+	help
+	  This option enables support for Promise PATA pdc20268 to pdc20277 host adapters.
+
+	  If unsure, say N.
+
 config PATA_PDC_OLD
 	tristate "Older Promise PATA controller support"
 	depends on PCI
@@ -643,14 +651,6 @@ config PATA_SERVERWORKS
 
 	  If unsure, say N.
 
-config PATA_PDC2027X
-	tristate "Promise PATA 2027x support"
-	depends on PCI
-	help
-	  This option enables support for Promise PATA pdc20268 to pdc20277 host adapters.
-
-	  If unsure, say N.
-
 config PATA_SIL680
 	tristate "CMD / Silicon Image 680 PATA support"
 	depends on PCI
@@ -667,6 +667,15 @@ config PATA_SIS
 
 	  If unsure, say N.
 
+config PATA_TOSHIBA
+	tristate "Toshiba Piccolo support (Experimental)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  Support for the Toshiba Piccolo controllers. Currently only the
+	  primary channel is supported by this driver.
+
+	  If unsure, say N.
+
 config PATA_VIA
 	tristate "VIA PATA support"
 	depends on PCI
diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile
index 01e126f343b..d909435e9d8 100644
--- a/drivers/ata/Makefile
+++ b/drivers/ata/Makefile
@@ -63,6 +63,7 @@ obj-$(CONFIG_PATA_RZ1000)	+= pata_rz1000.o
 obj-$(CONFIG_PATA_SC1200)	+= pata_sc1200.o
 obj-$(CONFIG_PATA_SERVERWORKS)	+= pata_serverworks.o
 obj-$(CONFIG_PATA_SIL680)	+= pata_sil680.o
+obj-$(CONFIG_PATA_TOSHIBA)	+= pata_piccolo.o
 obj-$(CONFIG_PATA_VIA)		+= pata_via.o
 obj-$(CONFIG_PATA_WINBOND)	+= pata_sl82c105.o
 obj-$(CONFIG_PATA_WINBOND_VLB)	+= pata_winbond.o
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index a3241a1a710..b8bea100a16 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -113,6 +113,7 @@ enum {
 	board_ahci_mcp65	= 6,
 	board_ahci_nopmp	= 7,
 	board_ahci_yesncq	= 8,
+	board_ahci_nosntf	= 9,
 
 	/* global controller registers */
 	HOST_CAP		= 0x00, /* host capabilities */
@@ -235,6 +236,7 @@ enum {
 	AHCI_HFLAG_NO_SUSPEND		= (1 << 10), /* don't suspend */
 	AHCI_HFLAG_SRST_TOUT_IS_OFFLINE	= (1 << 11), /* treat SRST timeout as
 							link offline */
+	AHCI_HFLAG_NO_SNTF		= (1 << 12), /* no sntf */
 
 	/* ap->flags bits */
 
@@ -508,7 +510,7 @@ static const struct ata_port_info ahci_port_info[] = {
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_ops,
 	},
-	/* board_ahci_yesncq */
+	[board_ahci_yesncq] =
 	{
 		AHCI_HFLAGS	(AHCI_HFLAG_YES_NCQ),
 		.flags		= AHCI_FLAG_COMMON,
@@ -516,6 +518,14 @@ static const struct ata_port_info ahci_port_info[] = {
 		.udma_mask	= ATA_UDMA6,
 		.port_ops	= &ahci_ops,
 	},
+	[board_ahci_nosntf] =
+	{
+		AHCI_HFLAGS	(AHCI_HFLAG_NO_SNTF),
+		.flags		= AHCI_FLAG_COMMON,
+		.pio_mask	= ATA_PIO4,
+		.udma_mask	= ATA_UDMA6,
+		.port_ops	= &ahci_ops,
+	},
 };
 
 static const struct pci_device_id ahci_pci_tbl[] = {
@@ -531,7 +541,7 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, 0x2683), board_ahci }, /* ESB2 */
 	{ PCI_VDEVICE(INTEL, 0x27c6), board_ahci }, /* ICH7-M DH */
 	{ PCI_VDEVICE(INTEL, 0x2821), board_ahci }, /* ICH8 */
-	{ PCI_VDEVICE(INTEL, 0x2822), board_ahci }, /* ICH8 */
+	{ PCI_VDEVICE(INTEL, 0x2822), board_ahci_nosntf }, /* ICH8 */
 	{ PCI_VDEVICE(INTEL, 0x2824), board_ahci }, /* ICH8 */
 	{ PCI_VDEVICE(INTEL, 0x2829), board_ahci }, /* ICH8M */
 	{ PCI_VDEVICE(INTEL, 0x282a), board_ahci }, /* ICH8M */
@@ -849,6 +859,12 @@ static void ahci_save_initial_config(struct pci_dev *pdev,
 		cap &= ~HOST_CAP_PMP;
 	}
 
+	if ((cap & HOST_CAP_SNTF) && (hpriv->flags & AHCI_HFLAG_NO_SNTF)) {
+		dev_printk(KERN_INFO, &pdev->dev,
+			   "controller can't do SNTF, turning off CAP_SNTF\n");
+		cap &= ~HOST_CAP_SNTF;
+	}
+
 	if (pdev->vendor == PCI_VENDOR_ID_JMICRON && pdev->device == 0x2361 &&
 	    port_map != 1) {
 		dev_printk(KERN_INFO, &pdev->dev,
@@ -2988,6 +3004,14 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (pdev->vendor == PCI_VENDOR_ID_MARVELL && !marvell_enable)
 		return -ENODEV;
 
+	/* Promise's PDC42819 is a SAS/SATA controller that has an AHCI mode.
+	 * At the moment, we can only use the AHCI mode. Let the users know
+	 * that for SAS drives they're out of luck.
+	 */
+	if (pdev->vendor == PCI_VENDOR_ID_PROMISE)
+		dev_printk(KERN_INFO, &pdev->dev, "PDC42819 "
+			   "can only drive SATA devices with this driver\n");
+
 	/* acquire resources */
 	rc = pcim_enable_device(pdev);
 	if (rc)
diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index ecfd22b4f1c..12e26c3c68e 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -168,9 +168,12 @@ static struct pci_device_id ata_generic[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_VIA,    PCI_DEVICE_ID_VIA_82C561), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_OPTI,   PCI_DEVICE_ID_OPTI_82C558), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_CENATEK,PCI_DEVICE_ID_CENATEK_IDE), },
-	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO), },
+#if !defined(CONFIG_PATA_TOSHIBA) && !defined(CONFIG_PATA_TOSHIBA_MODULE)
 	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_1), },
 	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_2),  },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_3),  },
+	{ PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_5),  },
+#endif	
 	/* Must come last. If you add entries adjust this table appropriately */
 	{ PCI_ANY_ID,		PCI_ANY_ID,			   PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xFFFFFF00UL, 1},
 	{ 0, },
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index 9ac4e378992..0c6155f5117 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -869,10 +869,10 @@ static void do_pata_set_dmamode(struct ata_port *ap, struct ata_device *adev, in
 				(timings[pio][1] << 8);
 		}
 
-		if (ap->udma_mask) {
+		if (ap->udma_mask)
 			udma_enable &= ~(1 << devid);
-			pci_write_config_word(dev, master_port, master_data);
-		}
+
+		pci_write_config_word(dev, master_port, master_data);
 	}
 	/* Don't scribble on 0x48 if the controller does not support UDMA */
 	if (ap->udma_mask)
diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c
index b0882cddfd4..1245838ac13 100644
--- a/drivers/ata/libata-acpi.c
+++ b/drivers/ata/libata-acpi.c
@@ -807,12 +807,11 @@ static int ata_acpi_exec_tfs(struct ata_device *dev, int *nr_executed)
  * EH context.
  *
  * RETURNS:
- * 0 on success, -errno on failure.
+ * 0 on success, -ENOENT if _SDD doesn't exist, -errno on failure.
  */
 static int ata_acpi_push_id(struct ata_device *dev)
 {
 	struct ata_port *ap = dev->link->ap;
-	int err;
 	acpi_status status;
 	struct acpi_object_list input;
 	union acpi_object in_params[1];
@@ -835,12 +834,16 @@ static int ata_acpi_push_id(struct ata_device *dev)
 	status = acpi_evaluate_object(dev->acpi_handle, "_SDD", &input, NULL);
 	swap_buf_le16(dev->id, ATA_ID_WORDS);
 
-	err = ACPI_FAILURE(status) ? -EIO : 0;
-	if (err < 0)
+	if (status == AE_NOT_FOUND)
+		return -ENOENT;
+
+	if (ACPI_FAILURE(status)) {
 		ata_dev_printk(dev, KERN_WARNING,
 			       "ACPI _SDD failed (AE 0x%x)\n", status);
+		return -EIO;
+	}
 
-	return err;
+	return 0;
 }
 
 /**
@@ -971,7 +974,7 @@ int ata_acpi_on_devcfg(struct ata_device *dev)
 	/* do _SDD if SATA */
 	if (acpi_sata) {
 		rc = ata_acpi_push_id(dev);
-		if (rc)
+		if (rc && rc != -ENOENT)
 			goto acpi_err;
 	}
 
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index dc72690ed5d..22ff51bdbc8 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -6616,6 +6616,13 @@ static int __init ata_init(void)
 {
 	ata_parse_force_param();
 
+	/*
+	 * FIXME: In UP case, there is only one workqueue thread and if you
+	 * have more than one PIO device, latency is bloody awful, with
+	 * occasional multi-second "hiccups" as one PIO device waits for
+	 * another.  It's an ugly wart that users DO occasionally complain
+	 * about; luckily most users have at most one PIO polled device.
+	 */
 	ata_wq = create_workqueue("ata");
 	if (!ata_wq)
 		goto free_force_tbl;
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index bba2ae5df1c..0ea97c942ce 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -110,6 +110,13 @@ static const unsigned long ata_eh_identify_timeouts[] = {
 	ULONG_MAX,
 };
 
+static const unsigned long ata_eh_flush_timeouts[] = {
+	15000,	/* be generous with flush */
+	15000,  /* ditto */
+	30000,	/* and even more generous */
+	ULONG_MAX,
+};
+
 static const unsigned long ata_eh_other_timeouts[] = {
 	 5000,	/* same rationale as identify timeout */
 	10000,	/* ditto */
@@ -147,6 +154,8 @@ ata_eh_cmd_timeout_table[ATA_EH_CMD_TIMEOUT_TABLE_SIZE] = {
 	  .timeouts = ata_eh_other_timeouts, },
 	{ .commands = CMDS(ATA_CMD_INIT_DEV_PARAMS),
 	  .timeouts = ata_eh_other_timeouts, },
+	{ .commands = CMDS(ATA_CMD_FLUSH, ATA_CMD_FLUSH_EXT),
+	  .timeouts = ata_eh_flush_timeouts },
 };
 #undef CMDS
 
@@ -3112,6 +3121,82 @@ static int atapi_eh_clear_ua(struct ata_device *dev)
 	return 0;
 }
 
+/**
+ *	ata_eh_maybe_retry_flush - Retry FLUSH if necessary
+ *	@dev: ATA device which may need FLUSH retry
+ *
+ *	If @dev failed FLUSH, it needs to be reported upper layer
+ *	immediately as it means that @dev failed to remap and already
+ *	lost at least a sector and further FLUSH retrials won't make
+ *	any difference to the lost sector.  However, if FLUSH failed
+ *	for other reasons, for example transmission error, FLUSH needs
+ *	to be retried.
+ *
+ *	This function determines whether FLUSH failure retry is
+ *	necessary and performs it if so.
+ *
+ *	RETURNS:
+ *	0 if EH can continue, -errno if EH needs to be repeated.
+ */
+static int ata_eh_maybe_retry_flush(struct ata_device *dev)
+{
+	struct ata_link *link = dev->link;
+	struct ata_port *ap = link->ap;
+	struct ata_queued_cmd *qc;
+	struct ata_taskfile tf;
+	unsigned int err_mask;
+	int rc = 0;
+
+	/* did flush fail for this device? */
+	if (!ata_tag_valid(link->active_tag))
+		return 0;
+
+	qc = __ata_qc_from_tag(ap, link->active_tag);
+	if (qc->dev != dev || (qc->tf.command != ATA_CMD_FLUSH_EXT &&
+			       qc->tf.command != ATA_CMD_FLUSH))
+		return 0;
+
+	/* if the device failed it, it should be reported to upper layers */
+	if (qc->err_mask & AC_ERR_DEV)
+		return 0;
+
+	/* flush failed for some other reason, give it another shot */
+	ata_tf_init(dev, &tf);
+
+	tf.command = qc->tf.command;
+	tf.flags |= ATA_TFLAG_DEVICE;
+	tf.protocol = ATA_PROT_NODATA;
+
+	ata_dev_printk(dev, KERN_WARNING, "retrying FLUSH 0x%x Emask 0x%x\n",
+		       tf.command, qc->err_mask);
+
+	err_mask = ata_exec_internal(dev, &tf, NULL, DMA_NONE, NULL, 0, 0);
+	if (!err_mask) {
+		/*
+		 * FLUSH is complete but there's no way to
+		 * successfully complete a failed command from EH.
+		 * Making sure retry is allowed at least once and
+		 * retrying it should do the trick - whatever was in
+		 * the cache is already on the platter and this won't
+		 * cause infinite loop.
+		 */
+		qc->scsicmd->allowed = max(qc->scsicmd->allowed, 1);
+	} else {
+		ata_dev_printk(dev, KERN_WARNING, "FLUSH failed Emask 0x%x\n",
+			       err_mask);
+		rc = -EIO;
+
+		/* if device failed it, report it to upper layers */
+		if (err_mask & AC_ERR_DEV) {
+			qc->err_mask |= AC_ERR_DEV;
+			qc->result_tf = tf;
+			if (!(ap->pflags & ATA_PFLAG_FROZEN))
+				rc = 0;
+		}
+	}
+	return rc;
+}
+
 static int ata_link_nr_enabled(struct ata_link *link)
 {
 	struct ata_device *dev;
@@ -3455,6 +3540,15 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 			}
 		}
 
+		/* retry flush if necessary */
+		ata_for_each_dev(dev, link, ALL) {
+			if (dev->class != ATA_DEV_ATA)
+				continue;
+			rc = ata_eh_maybe_retry_flush(dev);
+			if (rc)
+				goto dev_fail;
+		}
+
 		/* configure link power saving */
 		if (ehc->i.action & ATA_EH_LPM)
 			ata_for_each_dev(dev, link, ALL)
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index b4ee28dec52..62e6b9ea96a 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -47,6 +47,7 @@
 #include <linux/hdreg.h>
 #include <linux/uaccess.h>
 #include <linux/suspend.h>
+#include <asm/unaligned.h>
 
 #include "libata.h"
 
@@ -154,8 +155,7 @@ static ssize_t ata_scsi_lpm_put(struct device *dev,
 	 */
 	for (i = 1; i < ARRAY_SIZE(link_pm_policy); i++) {
 		const int len = strlen(link_pm_policy[i].name);
-		if (strncmp(link_pm_policy[i].name, buf, len) == 0 &&
-		   buf[len] == '\n') {
+		if (strncmp(link_pm_policy[i].name, buf, len) == 0) {
 			policy = link_pm_policy[i].value;
 			break;
 		}
@@ -1964,6 +1964,7 @@ static unsigned int ata_scsiop_inq_00(struct ata_scsi_args *args, u8 *rbuf)
 		0x80,	/* page 0x80, unit serial no page */
 		0x83,	/* page 0x83, device ident page */
 		0x89,	/* page 0x89, ata info page */
+		0xb0,	/* page 0xb0, block limits page */
 		0xb1,	/* page 0xb1, block device characteristics page */
 	};
 
@@ -2085,6 +2086,43 @@ static unsigned int ata_scsiop_inq_89(struct ata_scsi_args *args, u8 *rbuf)
 	return 0;
 }
 
+static unsigned int ata_scsiop_inq_b0(struct ata_scsi_args *args, u8 *rbuf)
+{
+	u32 min_io_sectors;
+
+	rbuf[1] = 0xb0;
+	rbuf[3] = 0x3c;		/* required VPD size with unmap support */
+
+	/*
+	 * Optimal transfer length granularity.
+	 *
+	 * This is always one physical block, but for disks with a smaller
+	 * logical than physical sector size we need to figure out what the
+	 * latter is.
+	 */
+	if (ata_id_has_large_logical_sectors(args->id))
+		min_io_sectors = ata_id_logical_per_physical_sectors(args->id);
+	else
+		min_io_sectors = 1;
+	put_unaligned_be16(min_io_sectors, &rbuf[6]);
+
+	/*
+	 * Optimal unmap granularity.
+	 *
+	 * The ATA spec doesn't even know about a granularity or alignment
+	 * for the TRIM command.  We can leave away most of the unmap related
+	 * VPD page entries, but we have specifify a granularity to signal
+	 * that we support some form of unmap - in thise case via WRITE SAME
+	 * with the unmap bit set.
+	 */
+	if (ata_id_has_trim(args->id)) {
+		put_unaligned_be32(65535 * 512 / 8, &rbuf[20]);
+		put_unaligned_be32(1, &rbuf[28]);
+	}
+
+	return 0;
+}
+
 static unsigned int ata_scsiop_inq_b1(struct ata_scsi_args *args, u8 *rbuf)
 {
 	int form_factor = ata_id_form_factor(args->id);
@@ -2374,6 +2412,13 @@ static unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf)
 		rbuf[13] = log_per_phys;
 		rbuf[14] = (lowest_aligned >> 8) & 0x3f;
 		rbuf[15] = lowest_aligned;
+
+		if (ata_id_has_trim(args->id)) {
+			rbuf[14] |= 0x80; /* TPE */
+
+			if (ata_id_has_zero_after_trim(args->id))
+				rbuf[14] |= 0x40; /* TPRZ */
+		}
 	}
 
 	return 0;
@@ -2896,6 +2941,58 @@ static unsigned int ata_scsi_pass_thru(struct ata_queued_cmd *qc)
 	return 1;
 }
 
+static unsigned int ata_scsi_write_same_xlat(struct ata_queued_cmd *qc)
+{
+	struct ata_taskfile *tf = &qc->tf;
+	struct scsi_cmnd *scmd = qc->scsicmd;
+	struct ata_device *dev = qc->dev;
+	const u8 *cdb = scmd->cmnd;
+	u64 block;
+	u32 n_block;
+	u32 size;
+	void *buf;
+
+	/* we may not issue DMA commands if no DMA mode is set */
+	if (unlikely(!dev->dma_mode))
+		goto invalid_fld;
+
+	if (unlikely(scmd->cmd_len < 16))
+		goto invalid_fld;
+	scsi_16_lba_len(cdb, &block, &n_block);
+
+	/* for now we only support WRITE SAME with the unmap bit set */
+	if (unlikely(!(cdb[1] & 0x8)))
+		goto invalid_fld;
+
+	/*
+	 * WRITE SAME always has a sector sized buffer as payload, this
+	 * should never be a multiple entry S/G list.
+	 */
+	if (!scsi_sg_count(scmd))
+		goto invalid_fld;
+
+	buf = page_address(sg_page(scsi_sglist(scmd)));
+	size = ata_set_lba_range_entries(buf, 512, block, n_block);
+
+	tf->protocol = ATA_PROT_DMA;
+	tf->hob_feature = 0;
+	tf->feature = ATA_DSM_TRIM;
+	tf->hob_nsect = (size / 512) >> 8;
+	tf->nsect = size / 512;
+	tf->command = ATA_CMD_DSM;
+	tf->flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE | ATA_TFLAG_LBA48 |
+		     ATA_TFLAG_WRITE;
+
+	ata_qc_set_pc_nbytes(qc);
+
+	return 0;
+
+ invalid_fld:
+	ata_scsi_set_sense(scmd, ILLEGAL_REQUEST, 0x24, 0x00);
+	/* "Invalid field in cdb" */
+	return 1;
+}
+
 /**
  *	ata_get_xlat_func - check if SCSI to ATA translation is possible
  *	@dev: ATA device
@@ -2920,6 +3017,9 @@ static inline ata_xlat_func_t ata_get_xlat_func(struct ata_device *dev, u8 cmd)
 	case WRITE_16:
 		return ata_scsi_rw_xlat;
 
+	case 0x93 /*WRITE_SAME_16*/:
+		return ata_scsi_write_same_xlat;
+
 	case SYNCHRONIZE_CACHE:
 		if (ata_try_flush_cache(dev))
 			return ata_scsi_flush_xlat;
@@ -3109,6 +3209,9 @@ void ata_scsi_simulate(struct ata_device *dev, struct scsi_cmnd *cmd,
 		case 0x89:
 			ata_scsi_rbuf_fill(&args, ata_scsiop_inq_89);
 			break;
+		case 0xb0:
+			ata_scsi_rbuf_fill(&args, ata_scsiop_inq_b0);
+			break;
 		case 0xb1:
 			ata_scsi_rbuf_fill(&args, ata_scsiop_inq_b1);
 			break;
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index bbbb1fab175..51eb1e29860 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -2384,7 +2384,7 @@ void ata_sff_post_internal_cmd(struct ata_queued_cmd *qc)
 	ap->hsm_task_state = HSM_ST_IDLE;
 
 	if (ap->ioaddr.bmdma_addr)
-		ata_bmdma_stop(qc);
+		ap->ops->bmdma_stop(qc);
 
 	spin_unlock_irqrestore(ap->lock, flags);
 }
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index 1432dc9d0ab..9434114b2ca 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -453,7 +453,9 @@ static void ali_init_chipset(struct pci_dev *pdev)
 			/* Clear CD-ROM DMA write bit */
 			tmp &= 0x7F;
 		/* Cable and UDMA */
-		pci_write_config_byte(pdev, 0x4B, tmp | 0x09);
+		if (pdev->revision >= 0xc2)
+			tmp |= 0x01;
+		pci_write_config_byte(pdev, 0x4B, tmp | 0x08);
 		/*
 		 * CD_ROM DMA on (0x53 bit 0). Enable this even if we want
 		 * to use PIO. 0x53 bit 1 (rev 20 only) - enable FIFO control
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
index f98dffedf4b..dadfc358ba1 100644
--- a/drivers/ata/pata_cmd64x.c
+++ b/drivers/ata/pata_cmd64x.c
@@ -31,7 +31,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_cmd64x"
-#define DRV_VERSION "0.2.5"
+#define DRV_VERSION "0.3.1"
 
 /*
  * CMD64x specific registers definition.
@@ -254,17 +254,109 @@ static void cmd648_bmdma_stop(struct ata_queued_cmd *qc)
 }
 
 /**
- *	cmd646r1_dma_stop	-	DMA stop callback
+ *	cmd64x_bmdma_stop	-	DMA stop callback
  *	@qc: Command in progress
  *
- *	Stub for now while investigating the r1 quirk in the old driver.
+ *	Track the completion of live DMA commands and clear the
+ *	host->private_data DMA tracking flag as we do.
  */
 
-static void cmd646r1_bmdma_stop(struct ata_queued_cmd *qc)
+static void cmd64x_bmdma_stop(struct ata_queued_cmd *qc)
 {
+	struct ata_port *ap = qc->ap;
 	ata_bmdma_stop(qc);
+	WARN_ON(ap->host->private_data != ap);
+	ap->host->private_data = NULL;
 }
 
+/**
+ *	cmd64x_qc_defer		-	Defer logic for chip limits
+ *	@qc: queued command
+ *
+ *	Decide whether we can issue the command. Called under the host lock.
+ */
+
+static int cmd64x_qc_defer(struct ata_queued_cmd *qc)
+{
+	struct ata_host *host = qc->ap->host;
+	struct ata_port *alt = host->ports[1 ^ qc->ap->port_no];
+	int rc;
+	int dma = 0;
+
+	/* Apply the ATA rules first */
+	rc = ata_std_qc_defer(qc);
+	if (rc)
+		return rc;
+
+	if (qc->tf.protocol == ATAPI_PROT_DMA ||
+			qc->tf.protocol == ATA_PROT_DMA)
+		dma = 1;
+
+	/* If the other port is not live then issue the command */
+	if (alt == NULL || !alt->qc_active) {
+		if (dma)
+			host->private_data = qc->ap;
+		return 0;
+	}
+	/* If there is a live DMA command then wait */
+	if (host->private_data != NULL)
+		return 	ATA_DEFER_PORT;
+	if (dma)
+		/* Cannot overlap our DMA command */
+		return ATA_DEFER_PORT;
+	return 0;
+}
+
+/**
+ *	cmd64x_interrupt - ATA host interrupt handler
+ *	@irq: irq line (unused)
+ *	@dev_instance: pointer to our ata_host information structure
+ *
+ *	Our interrupt handler for PCI IDE devices.  Calls
+ *	ata_sff_host_intr() for each port that is flagging an IRQ. We cannot
+ *	use the defaults as we need to avoid touching status/altstatus during
+ *	a DMA.
+ *
+ *	LOCKING:
+ *	Obtains host lock during operation.
+ *
+ *	RETURNS:
+ *	IRQ_NONE or IRQ_HANDLED.
+ */
+irqreturn_t cmd64x_interrupt(int irq, void *dev_instance)
+{
+	struct ata_host *host = dev_instance;
+	struct pci_dev *pdev = to_pci_dev(host->dev);
+	unsigned int i;
+	unsigned int handled = 0;
+	unsigned long flags;
+	static const u8 irq_reg[2] = { CFR, ARTTIM23 };
+	static const u8 irq_mask[2] = { 1 << 2, 1 << 4 };
+
+	/* TODO: make _irqsave conditional on x86 PCI IDE legacy mode */
+	spin_lock_irqsave(&host->lock, flags);
+
+	for (i = 0; i < host->n_ports; i++) {
+		struct ata_port *ap;
+		u8 reg;
+
+		pci_read_config_byte(pdev, irq_reg[i], &reg);
+		ap = host->ports[i];
+		if (ap && (reg & irq_mask[i]) &&
+		    !(ap->flags & ATA_FLAG_DISABLED)) {
+			struct ata_queued_cmd *qc;
+
+			qc = ata_qc_from_tag(ap, ap->link.active_tag);
+			if (qc && (!(qc->tf.flags & ATA_TFLAG_POLLING)) &&
+			    (qc->flags & ATA_QCFLAG_ACTIVE))
+				handled |= ata_sff_host_intr(ap, qc);
+		}
+	}
+
+	spin_unlock_irqrestore(&host->lock, flags);
+
+	return IRQ_RETVAL(handled);
+}
 static struct scsi_host_template cmd64x_sht = {
 	ATA_BMDMA_SHT(DRV_NAME),
 };
@@ -273,6 +365,8 @@ static const struct ata_port_operations cmd64x_base_ops = {
 	.inherits	= &ata_bmdma_port_ops,
 	.set_piomode	= cmd64x_set_piomode,
 	.set_dmamode	= cmd64x_set_dmamode,
+	.bmdma_stop	= cmd64x_bmdma_stop,
+	.qc_defer	= cmd64x_qc_defer,
 };