[git patches] libata updates for 3.3

[git patches] libata updates for 3.3

[Jeff Garzik]

Summary (very little excitement at all this time):

0) Will play around with git signed tags with the next update.

1) PM improvements, including runtime suspend/resume work

2) AHCI DMA engine start/stop changes, intended to improve spec
compliance, thereby making a particular IP block work.  Tested for a
while in linux-next, but nonetheless worth watching for breakage in the
field.

3) other minor bits.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Please pull 318893e1429a9d50569a0379d1e20b0ecc45c555 from
git://github.com/jgarzik/libata-dev.git upstream-linus

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iQIVAwUBTwo0xiWzCDIBeCsvAQJrQhAAuhjtG/AG5My8cKj7RS7GIaEEPEsNIoWm
/TbRLUfgJTiAkfiuqF4WqfzL2sGNUDMlMeSbs3P1p4kwLLQN8cSfd+xZVfNm3Q6u
HATts2JBNT8GQvffxXG5Z2JO3VXqlMxpc6TAjx1R+AY09CeT65qJlOzI+Uc7SIny
AZ6q2ETzHPoAMrTMLBCmsWSkXOBuOzHOjoAYDMbFBsVYpn587NjkAYMLrSjuMKVz
ndcmSGS4geEA1zEcQXKjWFmb+w5ZuifXOpVvy+cCyWQN27KebyJyTYiTCaNlFLVC
c1jdpHcElq2fDcdOrREHRjVxeYeUNHLVaTL5O/H5gh9kFOn+EQDySkhhTjpg69tk
44Q1YBQbfeBXCgku2sL0osv3RLjsN782exg2iH7AaV8MTlJBuIrj2p+uTOOwdTEP
CKpIlv9aTcWikbNX69nBXRQ7cKrztfVcMPVKsjblaiIU7O3GYEdziYm8lxPqTksr
rmM3GQSPkwfdFzbVXcY0wIyL3P/WzypQ2CZ8i/F2VpywIdw58hUZ4TG2Csm47ML3
5Hf5kllJR0/lzrJXkwlpEtLR0Ir5rK0CR4YUykj7BL0TGynnNwn+AmHKEeI51Gcy
0t44ZNO9T1ezGDhDKUTzx3SgzhElbkDqzqWCJhYVIbF44F65SBpS3ieeuKAFabE0
zHlKy0fNvT4=
=z+FO
-----END PGP SIGNATURE-----

to receive the following updates:

 drivers/ata/Kconfig            |    2 +-
 drivers/ata/ahci.c             |   26 +++++-
 drivers/ata/ahci_platform.c    |   68 ++++++++++++++
 drivers/ata/libahci.c          |    5 +-
 drivers/ata/libata-core.c      |  188 ++++++++++++++++++++++++++--------------
 drivers/ata/libata-scsi.c      |    3 +-
 drivers/ata/libata-sff.c       |    4 +-
 drivers/ata/libata-transport.c |    5 +
 drivers/ata/libata.h           |    1 +
 drivers/ata/pata_arasan_cf.c   |   12 +---
 drivers/ata/pata_at91.c        |   15 +---
 drivers/ata/pata_bf54x.c       |   20 +----
 drivers/ata/pata_cs5536.c      |   99 +++++++++------------
 drivers/ata/pata_imx.c         |   12 +---
 drivers/ata/pata_ixp4xx_cf.c   |   13 +---
 drivers/ata/pata_mpc52xx.c     |   21 +----
 drivers/ata/pata_of_platform.c |   27 ++----
 drivers/ata/pata_palmld.c      |   13 +---
 drivers/ata/pata_platform.c    |   12 +---
 drivers/ata/pata_pxa.c         |   13 +---
 drivers/ata/pata_rb532_cf.c    |   21 +----
 drivers/ata/sata_dwc_460ex.c   |   13 +---
 drivers/ata/sata_fsl.c         |   14 +---
 drivers/scsi/scsi_error.c      |    5 +-
 drivers/scsi/scsi_pm.c         |   27 ++++++-
 drivers/scsi/sd.c              |    5 +
 include/linux/ahci_platform.h  |    2 +
 include/scsi/scsi_host.h       |    3 +
 28 files changed, 326 insertions(+), 323 deletions(-)

Alessandro Rubini (1):
      ahci: support the STA2X11 I/O Hub

Axel Lin (1):
      SATA/PATA: convert drivers/ata/* to use module_platform_driver()

Bartlomiej Zolnierkiewicz (1):
      pata_cs5536: forward port changes from cs5536

Brian Norris (1):
      ahci: platform support for suspend/resume

Lin Ming (9):
      ata: make ata port as parent device of scsi host
      [SCSI] add flag to skip the runtime PM calls on the host
      [SCSI] check runtime PM status in system PM
      [SCSI] sd: check runtime PM status in sd_shutdown
      ata: add ata port system PM callbacks
      ata: add ata port runtime PM callbacks
      [SCSI] runtime resume parent for child's system-resume
      ata: update ata port's runtime status during system resume
      ata: add ata port hibernate callbacks

Rob Herring (1):
      pata_of_platform: remove direct dependency on OF_IRQ

Sergei Shtylyov (3):
      libata-sff: use ATAPI_{COD|IO}
      libata-core: kill duplicate statement in ata_do_set_mode()
      pata_bf54x: fix BMIDE status register emulation

Tejun Heo (1):
      ahci: start engine only during soft/hard resets

diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index cf047c4..6bdedd7 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -820,7 +820,7 @@ config PATA_PLATFORM
 
 config PATA_OF_PLATFORM
 	tristate "OpenFirmware platform device PATA support"
-	depends on PATA_PLATFORM && OF && OF_IRQ
+	depends on PATA_PLATFORM && OF
 	help
 	  This option enables support for generic directly connected ATA
 	  devices commonly found on embedded systems with OpenFirmware
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index cf26222..d07bf03 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -52,7 +52,8 @@
 #define DRV_VERSION	"3.0"
 
 enum {
-	AHCI_PCI_BAR		= 5,
+	AHCI_PCI_BAR_STA2X11	= 0,
+	AHCI_PCI_BAR_STANDARD	= 5,
 };
 
 enum board_ids {
@@ -375,6 +376,9 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	{ PCI_VDEVICE(SI, 0x1185), board_ahci },		/* SiS 968 */
 	{ PCI_VDEVICE(SI, 0x0186), board_ahci },		/* SiS 968 */
 
+	/* ST Microelectronics */
+	{ PCI_VDEVICE(STMICRO, 0xCC06), board_ahci },		/* ST ConneXt */
+
 	/* Marvell */
 	{ PCI_VDEVICE(MARVELL, 0x6145), board_ahci_mv },	/* 6145 */
 	{ PCI_VDEVICE(MARVELL, 0x6121), board_ahci_mv },	/* 6121 */
@@ -622,6 +626,13 @@ static int ahci_configure_dma_masks(struct pci_dev *pdev, int using_dac)
 {
 	int rc;
 
+	/*
+	 * If the device fixup already set the dma_mask to some non-standard
+	 * value, don't extend it here. This happens on STA2X11, for example.
+	 */
+	if (pdev->dma_mask && pdev->dma_mask < DMA_BIT_MASK(32))
+		return 0;
+
 	if (using_dac &&
 	    !pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) {
 		rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
@@ -1026,6 +1037,7 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct ahci_host_priv *hpriv;
 	struct ata_host *host;
 	int n_ports, i, rc;
+	int ahci_pci_bar = AHCI_PCI_BAR_STANDARD;
 
 	VPRINTK("ENTER\n");
 
@@ -1057,6 +1069,10 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_info(&pdev->dev,
 			 "PDC42819 can only drive SATA devices with this driver\n");
 
+	/* The Connext uses non-standard BAR */
+	if (pdev->vendor == PCI_VENDOR_ID_STMICRO && pdev->device == 0xCC06)
+		ahci_pci_bar = AHCI_PCI_BAR_STA2X11;
+
 	/* acquire resources */
 	rc = pcim_enable_device(pdev);
 	if (rc)
@@ -1065,7 +1081,7 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* AHCI controllers often implement SFF compatible interface.
 	 * Grab all PCI BARs just in case.
 	 */
-	rc = pcim_iomap_regions_request_all(pdev, 1 << AHCI_PCI_BAR, DRV_NAME);
+	rc = pcim_iomap_regions_request_all(pdev, 1 << ahci_pci_bar, DRV_NAME);
 	if (rc == -EBUSY)
 		pcim_pin_device(pdev);
 	if (rc)
@@ -1108,7 +1124,7 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if ((hpriv->flags & AHCI_HFLAG_NO_MSI) || pci_enable_msi(pdev))
 		pci_intx(pdev, 1);
 
-	hpriv->mmio = pcim_iomap_table(pdev)[AHCI_PCI_BAR];
+	hpriv->mmio = pcim_iomap_table(pdev)[ahci_pci_bar];
 
 	/* save initial config */
 	ahci_pci_save_initial_config(pdev, hpriv);
@@ -1172,8 +1188,8 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	for (i = 0; i < host->n_ports; i++) {
 		struct ata_port *ap = host->ports[i];
 
-		ata_port_pbar_desc(ap, AHCI_PCI_BAR, -1, "abar");
-		ata_port_pbar_desc(ap, AHCI_PCI_BAR,
+		ata_port_pbar_desc(ap, ahci_pci_bar, -1, "abar");
+		ata_port_pbar_desc(ap, ahci_pci_bar,
 				   0x100 + ap->port_no * 0x80, "port");
 
 		/* set enclosure management message type */
diff --git a/drivers/ata/ahci_platform.c b/drivers/ata/ahci_platform.c
index 43b8758..48be4e1 100644
--- a/drivers/ata/ahci_platform.c
+++ b/drivers/ata/ahci_platform.c
@@ -202,6 +202,71 @@ static int __devexit ahci_remove(struct platform_device *pdev)
 	return 0;
 }
 
+#ifdef CONFIG_PM
+static int ahci_suspend(struct device *dev)
+{
+	struct ahci_platform_data *pdata = dev_get_platdata(dev);
+	struct ata_host *host = dev_get_drvdata(dev);
+	struct ahci_host_priv *hpriv = host->private_data;
+	void __iomem *mmio = hpriv->mmio;
+	u32 ctl;
+	int rc;
+
+	if (hpriv->flags & AHCI_HFLAG_NO_SUSPEND) {
+		dev_err(dev, "firmware update required for suspend/resume\n");
+		return -EIO;
+	}
+
+	/*
+	 * AHCI spec rev1.1 section 8.3.3:
+	 * Software must disable interrupts prior to requesting a
+	 * transition of the HBA to D3 state.
+	 */
+	ctl = readl(mmio + HOST_CTL);
+	ctl &= ~HOST_IRQ_EN;
+	writel(ctl, mmio + HOST_CTL);
+	readl(mmio + HOST_CTL); /* flush */
+
+	rc = ata_host_suspend(host, PMSG_SUSPEND);
+	if (rc)
+		return rc;
+
+	if (pdata && pdata->suspend)
+		return pdata->suspend(dev);
+	return 0;
+}
+
+static int ahci_resume(struct device *dev)
+{
+	struct ahci_platform_data *pdata = dev_get_platdata(dev);
+	struct ata_host *host = dev_get_drvdata(dev);
+	int rc;
+
+	if (pdata && pdata->resume) {
+		rc = pdata->resume(dev);
+		if (rc)
+			return rc;
+	}
+
+	if (dev->power.power_state.event == PM_EVENT_SUSPEND) {
+		rc = ahci_reset_controller(host);
+		if (rc)
+			return rc;
+
+		ahci_init_controller(host);
+	}
+
+	ata_host_resume(host);
+
+	return 0;
+}
+
+static struct dev_pm_ops ahci_pm_ops = {
+	.suspend		= &ahci_suspend,
+	.resume			= &ahci_resume,
+};
+#endif
+
 static const struct of_device_id ahci_of_match[] = {
 	{ .compatible = "calxeda,hb-ahci", },
 	{},
@@ -214,6 +279,9 @@ static struct platform_driver ahci_driver = {
 		.name = "ahci",
 		.owner = THIS_MODULE,
 		.of_match_table = ahci_of_match,
+#ifdef CONFIG_PM
+		.pm = &ahci_pm_ops,
+#endif
 	},
 	.id_table	= ahci_devtype,
 };
diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index 3c92dbd..a72bfd0 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -746,9 +746,6 @@ static void ahci_start_port(struct ata_port *ap)
 	/* enable FIS reception */
 	ahci_start_fis_rx(ap);
 
-	/* enable DMA */
-	ahci_start_engine(ap);
-
 	/* turn on LEDs */
 	if (ap->flags & ATA_FLAG_EM) {
 		ata_for_each_link(link, ap, EDGE) {
@@ -2022,7 +2019,7 @@ static int ahci_port_suspend(struct ata_port *ap, pm_message_t mesg)
 		ahci_power_down(ap);
 	else {
 		ata_port_err(ap, "%s (%d)\n", emsg, rc);
-		ahci_start_port(ap);
+		ata_port_freeze(ap);
 	}
 
 	return rc;
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index c04ad68..11c9aea 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -66,6 +66,7 @@
 #include <asm/byteorder.h>
 #include <linux/cdrom.h>
 #include <linux/ratelimit.h>
+#include <linux/pm_runtime.h>
 
 #include "libata.h"
 #include "libata-transport.h"
@@ -3248,10 +3249,10 @@ int ata_do_set_mode(struct ata_link *link, struct ata_device **r_failed_dev)
 		ata_force_xfermask(dev);
 
 		pio_mask = ata_pack_xfermask(dev->pio_mask, 0, 0);
-		dma_mask = ata_pack_xfermask(0, dev->mwdma_mask, dev->udma_mask);
 
 		if (libata_dma_mask & mode_mask)
-			dma_mask = ata_pack_xfermask(0, dev->mwdma_mask, dev->udma_mask);
+			dma_mask = ata_pack_xfermask(0, dev->mwdma_mask,
+						     dev->udma_mask);
 		else
 			dma_mask = 0;
 
@@ -5234,73 +5235,55 @@ bool ata_link_offline(struct ata_link *link)
 }
 
 #ifdef CONFIG_PM
-static int ata_host_request_pm(struct ata_host *host, pm_message_t mesg,
+static int ata_port_request_pm(struct ata_port *ap, pm_message_t mesg,
 			       unsigned int action, unsigned int ehi_flags,
 			       int wait)
 {
+	struct ata_link *link;
 	unsigned long flags;
-	int i, rc;
-
-	for (i = 0; i < host->n_ports; i++) {
-		struct ata_port *ap = host->ports[i];
-		struct ata_link *link;
+	int rc;
 
-		/* Previous resume operation might still be in
-		 * progress.  Wait for PM_PENDING to clear.
-		 */
-		if (ap->pflags & ATA_PFLAG_PM_PENDING) {
-			ata_port_wait_eh(ap);
-			WARN_ON(ap->pflags & ATA_PFLAG_PM_PENDING);
-		}
+	/* Previous resume operation might still be in
+	 * progress.  Wait for PM_PENDING to clear.
+	 */
+	if (ap->pflags & ATA_PFLAG_PM_PENDING) {
+		ata_port_wait_eh(ap);
+		WARN_ON(ap->pflags & ATA_PFLAG_PM_PENDING);
+	}
 
-		/* request PM ops to EH */
-		spin_lock_irqsave(ap->lock, flags);
+	/* request PM ops to EH */
+	spin_lock_irqsave(ap->lock, flags);
 
-		ap->pm_mesg = mesg;
-		if (wait) {
-			rc = 0;
-			ap->pm_result = &rc;
-		}
+	ap->pm_mesg = mesg;
+	if (wait) {
+		rc = 0;
+		ap->pm_result = &rc;
+	}
 
-		ap->pflags |= ATA_PFLAG_PM_PENDING;
-		ata_for_each_link(link, ap, HOST_FIRST) {
-			link->eh_info.action |= action;
-			link->eh_info.flags |= ehi_flags;
-		}
+	ap->pflags |= ATA_PFLAG_PM_PENDING;
+	ata_for_each_link(link, ap, HOST_FIRST) {
+		link->eh_info.action |= action;
+		link->eh_info.flags |= ehi_flags;
+	}
 
-		ata_port_schedule_eh(ap);
+	ata_port_schedule_eh(ap);
 
-		spin_unlock_irqrestore(ap->lock, flags);
+	spin_unlock_irqrestore(ap->lock, flags);
 
-		/* wait and check result */
-		if (wait) {
-			ata_port_wait_eh(ap);
-			WARN_ON(ap->pflags & ATA_PFLAG_PM_PENDING);
-			if (rc)
-				return rc;
-		}
+	/* wait and check result */
+	if (wait) {
+		ata_port_wait_eh(ap);
+		WARN_ON(ap->pflags & ATA_PFLAG_PM_PENDING);
 	}
 
-	return 0;
+	return rc;
 }
 
-/**
- *	ata_host_suspend - suspend host
- *	@host: host to suspend
- *	@mesg: PM message
- *
- *	Suspend @host.  Actual operation is performed by EH.  This
- *	function requests EH to perform PM operations and waits for EH
- *	to finish.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep).
- *
- *	RETURNS:
- *	0 on success, -errno on failure.
- */
-int ata_host_suspend(struct ata_host *host, pm_message_t mesg)
+#define to_ata_port(d) container_of(d, struct ata_port, tdev)
+
+static int ata_port_suspend_common(struct device *dev, pm_message_t mesg)
 {
+	struct ata_port *ap = to_ata_port(dev);
 	unsigned int ehi_flags = ATA_EHI_QUIET;
 	int rc;
 
@@ -5315,31 +5298,108 @@ int ata_host_suspend(struct ata_host *host, pm_message_t mesg)
 	if (mesg.event == PM_EVENT_SUSPEND)
 		ehi_flags |= ATA_EHI_NO_AUTOPSY | ATA_EHI_NO_RECOVERY;
 
-	rc = ata_host_request_pm(host, mesg, 0, ehi_flags, 1);
-	if (rc == 0)
-		host->dev->power.power_state = mesg;
+	rc = ata_port_request_pm(ap, mesg, 0, ehi_flags, 1);
 	return rc;
 }
 
+static int ata_port_suspend(struct device *dev)
+{
+	if (pm_runtime_suspended(dev))
+		return 0;
+
+	return ata_port_suspend_common(dev, PMSG_SUSPEND);
+}
+
+static int ata_port_do_freeze(struct device *dev)
+{
+	if (pm_runtime_suspended(dev))
+		pm_runtime_resume(dev);
+
+	return ata_port_suspend_common(dev, PMSG_FREEZE);
+}
+
+static int ata_port_poweroff(struct device *dev)
+{
+	if (pm_runtime_suspended(dev))
+		return 0;
+
+	return ata_port_suspend_common(dev, PMSG_HIBERNATE);
+}
+
+static int ata_port_resume_common(struct device *dev)
+{
+	struct ata_port *ap = to_ata_port(dev);
+	int rc;
+
+	rc = ata_port_request_pm(ap, PMSG_ON, ATA_EH_RESET,
+		ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, 1);
+	return rc;
+}
+
+static int ata_port_resume(struct device *dev)
+{
+	int rc;
+
+	rc = ata_port_resume_common(dev);
+	if (!rc) {
+		pm_runtime_disable(dev);
+		pm_runtime_set_active(dev);
+		pm_runtime_enable(dev);
+	}
+
+	return rc;
+}
+
+static int ata_port_runtime_idle(struct device *dev)
+{
+	return pm_runtime_suspend(dev);
+}
+
+static const struct dev_pm_ops ata_port_pm_ops = {
+	.suspend = ata_port_suspend,
+	.resume = ata_port_resume,
+	.freeze = ata_port_do_freeze,
+	.thaw = ata_port_resume,
+	.poweroff = ata_port_poweroff,
+	.restore = ata_port_resume,
+
+	.runtime_suspend = ata_port_suspend,
+	.runtime_resume = ata_port_resume_common,
+	.runtime_idle = ata_port_runtime_idle,
+};
+
+/**
+ *	ata_host_suspend - suspend host
+ *	@host: host to suspend
+ *	@mesg: PM message
+ *
+ *	Suspend @host.  Actual operation is performed by port suspend.
+ */
+int ata_host_suspend(struct ata_host *host, pm_message_t mesg)
+{
+	host->dev->power.power_state = mesg;
+	return 0;
+}
+
 /**
  *	ata_host_resume - resume host
  *	@host: host to resume
  *
- *	Resume @host.  Actual operation is performed by EH.  This
- *	function requests EH to perform PM operations and returns.
- *	Note that all resume operations are performed parallelly.
- *
- *	LOCKING:
- *	Kernel thread context (may sleep).
+ *	Resume @host.  Actual operation is performed by port resume.
  */
 void ata_host_resume(struct ata_host *host)
 {
-	ata_host_request_pm(host, PMSG_ON, ATA_EH_RESET,
-			    ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, 0);
 	host->dev->power.power_state = PMSG_ON;
 }
 #endif
 
+struct device_type ata_port_type = {
+	.name = "ata_port",
+#ifdef CONFIG_PM
+	.pm = &ata_port_pm_ops,
+#endif
+};
+
 /**
  *	ata_dev_init - Initialize an ata_device structure
  *	@dev: Device structure to initialize
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 2a5412e..508a60b 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -3381,6 +3381,7 @@ int ata_scsi_add_hosts(struct ata_host *host, struct scsi_host_template *sht)
 		if (!shost)
 			goto err_alloc;
 
+		shost->eh_noresume = 1;
 		*(struct ata_port **)&shost->hostdata[0] = ap;
 		ap->scsi_host = shost;
 
@@ -3398,7 +3399,7 @@ int ata_scsi_add_hosts(struct ata_host *host, struct scsi_host_template *sht)
 		 */
 		shost->max_host_blocked = 1;
 
-		rc = scsi_add_host(ap->scsi_host, ap->host->dev);
+		rc = scsi_add_host(ap->scsi_host, &ap->tdev);
 		if (rc)
 			goto err_add;
 	}
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 4cadfa2..9691dd0 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -929,11 +929,11 @@ static void atapi_pio_bytes(struct ata_queued_cmd *qc)
 	bytes = (bc_hi << 8) | bc_lo;
 
 	/* shall be cleared to zero, indicating xfer of data */
-	if (unlikely(ireason & (1 << 0)))
+	if (unlikely(ireason & ATAPI_COD))
 		goto atapi_check;
 
 	/* make sure transfer direction matches expected */
-	i_write = ((ireason & (1 << 1)) == 0) ? 1 : 0;
+	i_write = ((ireason & ATAPI_IO) == 0) ? 1 : 0;
 	if (unlikely(do_write != i_write))
 		goto atapi_check;
 
diff --git a/drivers/ata/libata-transport.c b/drivers/ata/libata-transport.c
index ce9dc62..9a7f0ea 100644
--- a/drivers/ata/libata-transport.c
+++ b/drivers/ata/libata-transport.c
@@ -32,6 +32,7 @@
 #include <linux/libata.h>
 #include <linux/hdreg.h>
 #include <linux/uaccess.h>
+#include <linux/pm_runtime.h>
 
 #include "libata.h"
 #include "libata-transport.h"
@@ -279,6 +280,7 @@ int ata_tport_add(struct device *parent,
 	struct device *dev = &ap->tdev;
 
 	device_initialize(dev);
+	dev->type = &ata_port_type;
 
 	dev->parent = get_device(parent);
 	dev->release = ata_tport_release;
@@ -289,6 +291,9 @@ int ata_tport_add(struct device *parent,
 		goto tport_err;
 	}
 
+	pm_runtime_set_active(dev);
+	pm_runtime_enable(dev);
+
 	transport_add_device(dev);
 	transport_configure_device(dev);
 
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 773de97..814486d 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -58,6 +58,7 @@ extern int atapi_passthru16;
 extern int libata_fua;
 extern int libata_noacpi;
 extern int libata_allow_tpm;
+extern struct device_type ata_port_type;
 extern struct ata_link *ata_dev_phys_link(struct ata_device *dev);
 extern void ata_force_cbl(struct ata_port *ap);
 extern u64 ata_tf_to_lba(const struct ata_taskfile *tf);
diff --git a/drivers/ata/pata_arasan_cf.c b/drivers/ata/pata_arasan_cf.c
index e8574bb..048589f 100644
--- a/drivers/ata/pata_arasan_cf.c
+++ b/drivers/ata/pata_arasan_cf.c
@@ -963,17 +963,7 @@ static struct platform_driver arasan_cf_driver = {
 	},
 };
 
-static int __init arasan_cf_init(void)
-{
-	return platform_driver_register(&arasan_cf_driver);
-}
-module_init(arasan_cf_init);
-
-static void __exit arasan_cf_exit(void)
-{
-	platform_driver_unregister(&arasan_cf_driver);
-}
-module_exit(arasan_cf_exit);
+module_platform_driver(arasan_cf_driver);
 
 MODULE_AUTHOR("Viresh Kumar <viresh.kumar@st.com>");
 MODULE_DESCRIPTION("Arasan ATA Compact Flash driver");
diff --git a/drivers/ata/pata_at91.c b/drivers/ata/pata_at91.c
index a76f24a..d56dd45 100644
--- a/drivers/ata/pata_at91.c
+++ b/drivers/ata/pata_at91.c
@@ -454,20 +454,7 @@ static struct platform_driver pata_at91_driver = {
 	},
 };
 
-static int __init pata_at91_init(void)
-{
-	return platform_driver_register(&pata_at91_driver);
-}
-
-static void __exit pata_at91_exit(void)
-{
-	platform_driver_unregister(&pata_at91_driver);
-}
-
-
-module_init(pata_at91_init);
-module_exit(pata_at91_exit);
-
+module_platform_driver(pata_at91_driver);
 
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Driver for CF in True IDE mode on AT91SAM9260 SoC");
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index bd987bb..d6a4677 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -418,14 +418,6 @@ static void bfin_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 					(tcyc_tdvs<<8 | tdvs));
 				ATAPI_SET_ULTRA_TIM_2(base, (tmli<<8 | tss));
 				ATAPI_SET_ULTRA_TIM_3(base, (trp<<8 | tzah));
-
-				/* Enable host ATAPI Untra DMA interrupts */
-				ATAPI_SET_INT_MASK(base,
-					ATAPI_GET_INT_MASK(base)
-					| UDMAIN_DONE_MASK
-					| UDMAOUT_DONE_MASK
-					| UDMAIN_TERM_MASK
-					| UDMAOUT_TERM_MASK);
 			}
 		}
 	}
@@ -470,10 +462,6 @@ static void bfin_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 			ATAPI_SET_MULTI_TIM_0(base, (tm<<8 | td));
 			ATAPI_SET_MULTI_TIM_1(base, (tkr<<8 | tkw));
 			ATAPI_SET_MULTI_TIM_2(base, (teoc<<8 | th));
-
-			/* Enable host ATAPI Multi DMA interrupts */
-			ATAPI_SET_INT_MASK(base, ATAPI_GET_INT_MASK(base)
-				| MULTI_DONE_MASK | MULTI_TERM_MASK);
 			SSYNC();
 		}
 	}
@@ -1153,15 +1141,11 @@ static unsigned char bfin_bmdma_status(struct ata_port *ap)
 {
 	unsigned char host_stat = 0;
 	void __iomem *base = (void __iomem *)ap->ioaddr.ctl_addr;
-	unsigned short int_status = ATAPI_GET_INT_STATUS(base);
 
-	if (ATAPI_GET_STATUS(base) & (MULTI_XFER_ON|ULTRA_XFER_ON))
+	if (ATAPI_GET_STATUS(base) & (MULTI_XFER_ON | ULTRA_XFER_ON))
 		host_stat |= ATA_DMA_ACTIVE;
-	if (int_status & (MULTI_DONE_INT|UDMAIN_DONE_INT|UDMAOUT_DONE_INT|
-		ATAPI_DEV_INT))
+	if (ATAPI_GET_INT_STATUS(base) & ATAPI_DEV_INT)
 		host_stat |= ATA_DMA_INTR;
-	if (int_status & (MULTI_TERM_INT|UDMAIN_TERM_INT|UDMAOUT_TERM_INT))
-		host_stat |= ATA_DMA_ERR|ATA_DMA_INTR;
 
 	dev_dbg(ap->dev, "ATAPI: host_stat=0x%x\n", host_stat);
 
diff --git a/drivers/ata/pata_cs5536.c b/drivers/ata/pata_cs5536.c
index 628c8fa..7a402c7 100644
--- a/drivers/ata/pata_cs5536.c
+++ b/drivers/ata/pata_cs5536.c
@@ -1,6 +1,7 @@
 /*
  * pata_cs5536.c	- CS5536 PATA for new ATA layer
  *			  (C) 2007 Martin K. Petersen <mkp@mkp.net>
+ *			  (C) 2011 Bartlomiej Zolnierkiewicz
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -55,24 +56,16 @@ MODULE_PARM_DESC(msr, "Force using MSR to configure IDE function (Default: 0)");
 #define DRV_VERSION	"0.0.8"
 
 enum {
-	CFG			= 0,
-	DTC			= 1,
-	CAST			= 2,
-	ETC			= 3,
-
-	MSR_IDE_BASE		= 0x51300000,
-	MSR_IDE_CFG		= (MSR_IDE_BASE + 0x10),
-	MSR_IDE_DTC		= (MSR_IDE_BASE + 0x12),
-	MSR_IDE_CAST		= (MSR_IDE_BASE + 0x13),
-	MSR_IDE_ETC		= (MSR_IDE_BASE + 0x14),
-
+	MSR_IDE_CFG		= 0x51300010,
 	PCI_IDE_CFG		= 0x40,
-	PCI_IDE_DTC		= 0x48,
-	PCI_IDE_CAST		= 0x4c,
-	PCI_IDE_ETC		= 0x50,
 
-	IDE_CFG_CHANEN		= 0x2,
-	IDE_CFG_CABLE		= 0x10000,
+	CFG			= 0,
+	DTC			= 2,
+	CAST			= 3,
+	ETC			= 4,
+
+	IDE_CFG_CHANEN		= (1 << 1),
+	IDE_CFG_CABLE		= (1 << 17) | (1 << 16),
 
 	IDE_D0_SHIFT		= 24,
 	IDE_D1_SHIFT		= 16,
@@ -84,45 +77,50 @@ enum {
 	IDE_CAST_CMD_MASK	= 0xff,
 	IDE_CAST_CMD_SHIFT	= 24,
 
-	IDE_ETC_NODMA		= 0x03,
-};
-
-static const u32 msr_reg[4] = {
-	MSR_IDE_CFG, MSR_IDE_DTC, MSR_IDE_CAST, MSR_IDE_ETC,
-};
-
-static const u8 pci_reg[4] = {
-	PCI_IDE_CFG, PCI_IDE_DTC, PCI_IDE_CAST, PCI_IDE_ETC,
+	IDE_ETC_UDMA_MASK	= 0xc0,
 };
 
-static inline int cs5536_read(struct pci_dev *pdev, int reg, u32 *val)
+static int cs5536_read(struct pci_dev *pdev, int reg, u32 *val)
 {
 	if (unlikely(use_msr)) {
 		u32 dummy __maybe_unused;
 
-		rdmsr(msr_reg[reg], *val, dummy);
+		rdmsr(MSR_IDE_CFG + reg, *val, dummy);
 		return 0;
 	}
 
-	return pci_read_config_dword(pdev, pci_reg[reg], val);
+	return pci_read_config_dword(pdev, PCI_IDE_CFG + reg * 4, val);
 }
 
-static inline int cs5536_write(struct pci_dev *pdev, int reg, int val)
+static int cs5536_write(struct pci_dev *pdev, int reg, int val)
 {
 	if (unlikely(use_msr)) {
-		wrmsr(msr_reg[reg], val, 0);
+		wrmsr(MSR_IDE_CFG + reg, val, 0);
 		return 0;
 	}
 
-	return pci_write_config_dword(pdev, pci_reg[reg], val);
+	return pci_write_config_dword(pdev, PCI_IDE_CFG + reg * 4, val);
+}
+
+static void cs5536_program_dtc(struct ata_device *adev, u8 tim)
+{
+	struct pci_dev *pdev = to_pci_dev(adev->link->ap->host->dev);
+	int dshift = adev->devno ? IDE_D1_SHIFT : IDE_D0_SHIFT;
+	u32 dtc;
+
+	cs5536_read(pdev, DTC, &dtc);
+	dtc &= ~(IDE_DRV_MASK << dshift);
+	dtc |= tim << dshift;
+	cs5536_write(pdev, DTC, dtc);
 }
 
 /**
  *	cs5536_cable_detect	-	detect cable type
  *	@ap: Port to detect on
  *
- *	Perform cable detection for ATA66 capable cable. Return a libata
- *	cable type.
+ *	Perform cable detection for ATA66 capable cable.
+ *
+ *	Returns a cable type.
  */
 
 static int cs5536_cable_detect(struct ata_port *ap)
@@ -132,7 +130,7 @@ static int cs5536_cable_detect(struct ata_port *ap)
 
 	cs5536_read(pdev, CFG, &cfg);
 
-	if (cfg & (IDE_CFG_CABLE << ap->port_no))
+	if (cfg & IDE_CFG_CABLE)
 		return ATA_CBL_PATA80;
 	else
 		return ATA_CBL_PATA40;
@@ -162,19 +160,15 @@ static void cs5536_set_piomode(struct ata_port *ap, struct ata_device *adev)
 	struct ata_device *pair = ata_dev_pair(adev);
 	int mode = adev->pio_mode - XFER_PIO_0;
 	int cmdmode = mode;
-	int dshift = adev->devno ? IDE_D1_SHIFT : IDE_D0_SHIFT;
 	int cshift = adev->devno ? IDE_CAST_D1_SHIFT : IDE_CAST_D0_SHIFT;
-	u32 dtc, cast, etc;
+	u32 cast;
 
 	if (pair)
 		cmdmode = min(mode, pair->pio_mode - XFER_PIO_0);
 
-	cs5536_read(pdev, DTC, &dtc);
-	cs5536_read(pdev, CAST, &cast);
-	cs5536_read(pdev, ETC, &etc);
+	cs5536_program_dtc(adev, drv_timings[mode]);
 
-	dtc &= ~(IDE_DRV_MASK << dshift);
-	dtc |= drv_timings[mode] << dshift;
+	cs5536_read(pdev, CAST, &cast);
 
 	cast &= ~(IDE_CAST_DRV_MASK << cshift);
 	cast |= addr_timings[mode] << cshift;
@@ -182,12 +176,7 @@ static void cs5536_set_piomode(struct ata_port *ap, struct ata_device *adev)
 	cast &= ~(IDE_CAST_CMD_MASK << IDE_CAST_CMD_SHIFT);
 	cast |= cmd_timings[cmdmode] << IDE_CAST_CMD_SHIFT;
 
-	etc &= ~(IDE_DRV_MASK << dshift);
-	etc |= IDE_ETC_NODMA << dshift;
-
-	cs5536_write(pdev, DTC, dtc);
 	cs5536_write(pdev, CAST, cast);
-	cs5536_write(pdev, ETC, etc);
 }
 
 /**
@@ -208,25 +197,21 @@ static void cs5536_set_dmamode(struct ata_port *ap, struct ata_device *adev)
 	};
 
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
-	u32 dtc, etc;
+	u32 etc;
 	int mode = adev->dma_mode;
 	int dshift = adev->devno ? IDE_D1_SHIFT : IDE_D0_SHIFT;
 
-	if (mode >= XFER_UDMA_0) {
-		cs5536_read(pdev, ETC, &etc);
+	cs5536_read(pdev, ETC, &etc);
 
+	if (mode >= XFER_UDMA_0) {
 		etc &= ~(IDE_DRV_MASK << dshift);
 		etc |= udma_timings[mode - XFER_UDMA_0] << dshift;
-
-		cs5536_write(pdev, ETC, etc);
 	} else { /* MWDMA */
-		cs5536_read(pdev, DTC, &dtc);
-
-		dtc &= ~(IDE_DRV_MASK << dshift);
-		dtc |= mwdma_timings[mode - XFER_MW_DMA_0] << dshift;
-
-		cs5536_write(pdev, DTC, dtc);
+		etc &= ~(IDE_ETC_UDMA_MASK << dshift);
+		cs5536_program_dtc(adev, mwdma_timings[mode - XFER_MW_DMA_0]);
 	}
+
+	cs5536_write(pdev, ETC, etc);
 }
 
 static struct scsi_host_template cs5536_sht = {
diff --git a/drivers/ata/pata_imx.c b/drivers/ata/pata_imx.c
index ca9d9ca..c5af97f 100644
--- a/drivers/ata/pata_imx.c
+++ b/drivers/ata/pata_imx.c
@@ -235,17 +235,7 @@ static struct platform_driver pata_imx_driver = {
 	},
 };
 
-static int __init pata_imx_init(void)
-{
-	return platform_driver_register(&pata_imx_driver);
-}
-
-static void __exit pata_imx_exit(void)
-{
-	platform_driver_unregister(&pata_imx_driver);
-}
-module_init(pata_imx_init);
-module_exit(pata_imx_exit);
+module_platform_driver(pata_imx_driver);
 
 MODULE_AUTHOR("Arnaud Patard <arnaud.patard@rtp-net.org>");
 MODULE_DESCRIPTION("low-level driver for iMX PATA");
diff --git a/drivers/ata/pata_ixp4xx_cf.c b/drivers/ata/pata_ixp4xx_cf.c
index 15b6431..badb178 100644
--- a/drivers/ata/pata_ixp4xx_cf.c
+++ b/drivers/ata/pata_ixp4xx_cf.c
@@ -205,21 +205,10 @@ static struct platform_driver ixp4xx_pata_platform_driver = {
 	.remove		= __devexit_p(ixp4xx_pata_remove),
 };
 
-static int __init ixp4xx_pata_init(void)
-{
-	return platform_driver_register(&ixp4xx_pata_platform_driver);
-}
-
-static void __exit ixp4xx_pata_exit(void)
-{
-	platform_driver_unregister(&ixp4xx_pata_platform_driver);
-}
+module_platform_driver(ixp4xx_pata_platform_driver);
 
 MODULE_AUTHOR("Alessandro Zummo <a.zummo@towertech.it>");
 MODULE_DESCRIPTION("low-level driver for ixp4xx Compact Flash PATA");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_VERSION);
 MODULE_ALIAS("platform:" DRV_NAME);
-
-module_init(ixp4xx_pata_init);
-module_exit(ixp4xx_pata_exit);
diff --git a/drivers/ata/pata_mpc52xx.c b/drivers/ata/pata_mpc52xx.c
index 3e17463..00748ae 100644
--- a/drivers/ata/pata_mpc52xx.c
+++ b/drivers/ata/pata_mpc52xx.c
@@ -897,26 +897,7 @@ static struct platform_driver mpc52xx_ata_of_platform_driver = {
 	},
 };
 
-
-/* ======================================================================== */
-/* Module                                                                   */
-/* ======================================================================== */
-
-static int __init
-mpc52xx_ata_init(void)
-{
-	printk(KERN_INFO "ata: MPC52xx IDE/ATA libata driver\n");
-	return platform_driver_register(&mpc52xx_ata_of_platform_driver);
-}
-
-static void __exit
-mpc52xx_ata_exit(void)
-{
-	platform_driver_unregister(&mpc52xx_ata_of_platform_driver);
-}
-
-module_init(mpc52xx_ata_init);
-module_exit(mpc52xx_ata_exit);
+module_platform_driver(mpc52xx_ata_of_platform_driver);
 
 MODULE_AUTHOR("Sylvain Munaut <tnt@246tNt.com>");
 MODULE_DESCRIPTION("Freescale MPC52xx IDE/ATA libata driver");
diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c
index 2a472c5..1654dc2 100644
--- a/drivers/ata/pata_of_platform.c
+++ b/drivers/ata/pata_of_platform.c
@@ -12,8 +12,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/of_address.h>
-#include <linux/of_irq.h>
-#include <linux/of_platform.h>
+#include <linux/platform_device.h>
 #include <linux/ata_platform.h>
 
 static int __devinit pata_of_platform_probe(struct platform_device *ofdev)
@@ -22,7 +21,7 @@ static int __devinit pata_of_platform_probe(struct platform_device *ofdev)
 	struct device_node *dn = ofdev->dev.of_node;
 	struct resource io_res;
 	struct resource ctl_res;
-	struct resource irq_res;
+	struct resource *irq_res;
 	unsigned int reg_shift = 0;
 	int pio_mode = 0;
 	int pio_mask;
@@ -51,11 +50,9 @@ static int __devinit pata_of_platform_probe(struct platform_device *ofdev)
 		}
 	}
 
-	ret = of_irq_to_resource(dn, 0, &irq_res);
-	if (!ret)
-		irq_res.start = irq_res.end = 0;
-	else
-		irq_res.flags = 0;
+	irq_res = platform_get_resource(ofdev, IORESOURCE_IRQ, 0);
+	if (irq_res)
+		irq_res->flags = 0;
 
 	prop = of_get_property(dn, "reg-shift", NULL);
 	if (prop)
@@ -75,7 +72,7 @@ static int __devinit pata_of_platform_probe(struct platform_device *ofdev)
 	pio_mask = 1 << pio_mode;
 	pio_mask |= (1 << pio_mode) - 1;
 
-	return __pata_platform_probe(&ofdev->dev, &io_res, &ctl_res, &irq_res,
+	return __pata_platform_probe(&ofdev->dev, &io_res, &ctl_res, irq_res,
 				     reg_shift, pio_mask);
 }
 
@@ -101,17 +98,7 @@ static struct platform_driver pata_of_platform_driver = {
 	.remove		= __devexit_p(pata_of_platform_remove),
 };
 
-static int __init pata_of_platform_init(void)
-{
-	return platform_driver_register(&pata_of_platform_driver);
-}
-module_init(pata_of_platform_init);
-
-static void __exit pata_of_platform_exit(void)
-{
-	platform_driver_unregister(&pata_of_platform_driver);
-}
-module_exit(pata_of_platform_exit);
+module_platform_driver(pata_of_platform_driver);
 
 MODULE_DESCRIPTION("OF-platform PATA driver");
 MODULE_AUTHOR("Anton Vorontsov <avorontsov@ru.mvista.com>");
diff --git a/drivers/ata/pata_palmld.c b/drivers/ata/pata_palmld.c
index b86d7e2..5ff31b6 100644
--- a/drivers/ata/pata_palmld.c
+++ b/drivers/ata/pata_palmld.c
@@ -132,20 +132,9 @@ static struct platform_driver palmld_pata_platform_driver = {
 	.remove		= __devexit_p(palmld_pata_remove),
 };
 
-static int __init palmld_pata_init(void)
-{
-	return platform_driver_register(&palmld_pata_platform_driver);
-}
-
-static void __exit palmld_pata_exit(void)
-{
-	platform_driver_unregister(&palmld_pata_platform_driver);
-}
+module_platform_driver(palmld_pata_platform_driver);
 
 MODULE_AUTHOR("Marek Vasut <marek.vasut@gmail.com>");
 MODULE_DESCRIPTION("PalmLD PATA driver");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:" DRV_NAME);
-
-module_init(palmld_pata_init);
-module_exit(palmld_pata_exit);
diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c
index 2067308..f1848ae 100644
--- a/drivers/ata/pata_platform.c
+++ b/drivers/ata/pata_platform.c
@@ -256,17 +256,7 @@ static struct platform_driver pata_platform_driver = {
 	},
 };
 
-static int __init pata_platform_init(void)
-{
-	return platform_driver_register(&pata_platform_driver);
-}
-
-static void __exit pata_platform_exit(void)
-{
-	platform_driver_unregister(&pata_platform_driver);
-}
-module_init(pata_platform_init);
-module_exit(pata_platform_exit);
+module_platform_driver(pata_platform_driver);
 
 module_param(pio_mask, int, 0);
 
diff --git a/drivers/ata/pata_pxa.c b/drivers/ata/pata_pxa.c
index b4ede40..0bb0fb7 100644
--- a/drivers/ata/pata_pxa.c
+++ b/drivers/ata/pata_pxa.c
@@ -390,18 +390,7 @@ static struct platform_driver pxa_ata_driver = {
 	},
 };
 
-static int __init pxa_ata_init(void)
-{
-	return platform_driver_register(&pxa_ata_driver);
-}
-
-static void __exit pxa_ata_exit(void)
-{
-	platform_driver_unregister(&pxa_ata_driver);
-}
-
-module_init(pxa_ata_init);
-module_exit(pxa_ata_exit);
+module_platform_driver(pxa_ata_driver);
 
 MODULE_AUTHOR("Marek Vasut <marek.vasut@gmail.com>");
 MODULE_DESCRIPTION("DMA-capable driver for PATA on PXA CPU");
diff --git a/drivers/ata/pata_rb532_cf.c b/drivers/ata/pata_rb532_cf.c
index 1b9d10d..9417101 100644
--- a/drivers/ata/pata_rb532_cf.c
+++ b/drivers/ata/pata_rb532_cf.c
@@ -188,9 +188,6 @@ static __devexit int rb532_pata_driver_remove(struct platform_device *pdev)
 	return 0;
 }
 
-/* work with hotplug and coldplug */
-MODULE_ALIAS("platform:" DRV_NAME);
-
 static struct platform_driver rb532_pata_platform_driver = {
 	.probe		= rb532_pata_driver_probe,
 	.remove		= __devexit_p(rb532_pata_driver_remove),
@@ -200,27 +197,13 @@ static struct platform_driver rb532_pata_platform_driver = {
 	},
 };
 
-/* ------------------------------------------------------------------------ */
-
 #define DRV_INFO DRV_DESC " version " DRV_VERSION
 
-static int __init rb532_pata_module_init(void)
-{
-	printk(KERN_INFO DRV_INFO "\n");
-
-	return platform_driver_register(&rb532_pata_platform_driver);
-}
-
-static void __exit rb532_pata_module_exit(void)
-{
-	platform_driver_unregister(&rb532_pata_platform_driver);
-}
+module_platform_driver(rb532_pata_platform_driver);
 
 MODULE_AUTHOR("Gabor Juhos <juhosg at openwrt.org>");
 MODULE_AUTHOR("Florian Fainelli <florian@openwrt.org>");
 MODULE_DESCRIPTION(DRV_DESC);
 MODULE_VERSION(DRV_VERSION);
 MODULE_LICENSE("GPL");
-
-module_init(rb532_pata_module_init);
-module_exit(rb532_pata_module_exit);
+MODULE_ALIAS("platform:" DRV_NAME);
diff --git a/drivers/ata/sata_dwc_460ex.c b/drivers/ata/sata_dwc_460ex.c
index 5c42374..69f7cde 100644
--- a/drivers/ata/sata_dwc_460ex.c
+++ b/drivers/ata/sata_dwc_460ex.c
@@ -1777,18 +1777,7 @@ static struct platform_driver sata_dwc_driver = {
 	.remove = sata_dwc_remove,
 };
 
-static int __init sata_dwc_init(void)
-{
-	return platform_driver_register(&sata_dwc_driver);
-}
-
-static void __exit sata_dwc_exit(void)
-{
-	platform_driver_unregister(&sata_dwc_driver);
-}
-
-module_init(sata_dwc_init);
-module_exit(sata_dwc_exit);
+module_platform_driver(sata_dwc_driver);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Mark Miesfeld <mmiesfeld@amcc.com>");
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index 78ae7b6..5a2c95b 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -1452,21 +1452,9 @@ static struct platform_driver fsl_sata_driver = {
 #endif
 };
 
-static int __init sata_fsl_init(void)
-{
-	platform_driver_register(&fsl_sata_driver);
-	return 0;
-}
-
-static void __exit sata_fsl_exit(void)
-{
-	platform_driver_unregister(&fsl_sata_driver);
-}
+module_platform_driver(fsl_sata_driver);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Ashish Kalra, Freescale Semiconductor");
 MODULE_DESCRIPTION("Freescale 3.0Gbps SATA controller low level driver");
 MODULE_VERSION("1.10");
-
-module_init(sata_fsl_init);
-module_exit(sata_fsl_exit);
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index dc6131e..5f84a14 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -1812,7 +1812,7 @@ int scsi_error_handler(void *data)
 		 * what we need to do to get it up and online again (if we can).
 		 * If we fail, we end up taking the thing offline.
 		 */
-		if (scsi_autopm_get_host(shost) != 0) {
+		if (!shost->eh_noresume && scsi_autopm_get_host(shost) != 0) {
 			SCSI_LOG_ERROR_RECOVERY(1,
 				printk(KERN_ERR "Error handler scsi_eh_%d "
 						"unable to autoresume\n",
@@ -1833,7 +1833,8 @@ int scsi_error_handler(void *data)
 		 * which are still online.
 		 */
 		scsi_restart_operations(shost);
-		scsi_autopm_put_host(shost);
+		if (!shost->eh_noresume)
+			scsi_autopm_put_host(shost);
 		set_current_state(TASK_INTERRUPTIBLE);
 	}
 	__set_current_state(TASK_RUNNING);
diff --git a/drivers/scsi/scsi_pm.c b/drivers/scsi/scsi_pm.c
index d329f8b..bf8bf79 100644
--- a/drivers/scsi/scsi_pm.c
+++ b/drivers/scsi/scsi_pm.c
@@ -49,8 +49,22 @@ static int scsi_bus_suspend_common(struct device *dev, pm_message_t msg)
 {
 	int err = 0;
 
-	if (scsi_is_sdev_device(dev))
+	if (scsi_is_sdev_device(dev)) {
+		/*
+		 * sd is the only high-level SCSI driver to implement runtime
+		 * PM, and sd treats runtime suspend, system suspend, and
+		 * system hibernate identically (but not system freeze).
+		 */
+		if (pm_runtime_suspended(dev)) {
+			if (msg.event == PM_EVENT_SUSPEND ||
+			    msg.event == PM_EVENT_HIBERNATE)
+				return 0;	/* already suspended */
+
+			/* wake up device so that FREEZE will succeed */
+			pm_runtime_resume(dev);
+		}
 		err = scsi_dev_type_suspend(dev, msg);
+	}
 	return err;
 }
 
@@ -58,8 +72,17 @@ static int scsi_bus_resume_common(struct device *dev)
 {
 	int err = 0;
 
-	if (scsi_is_sdev_device(dev))
+	if (scsi_is_sdev_device(dev)) {
+		/*
+		 * Parent device may have runtime suspended as soon as
+		 * it is woken up during the system resume.
+		 *
+		 * Resume it on behalf of child.
+		 */
+		pm_runtime_get_sync(dev->parent);
 		err = scsi_dev_type_resume(dev);
+		pm_runtime_put_sync(dev->parent);
+	}
 
 	if (err == 0) {
 		pm_runtime_disable(dev);
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index fa3a591..7b3f807 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -50,6 +50,7 @@
 #include <linux/string_helpers.h>
 #include <linux/async.h>
 #include <linux/slab.h>
+#include <linux/pm_runtime.h>
 #include <asm/uaccess.h>
 #include <asm/unaligned.h>
 
@@ -2741,6 +2742,9 @@ static void sd_shutdown(struct device *dev)
 	if (!sdkp)
 		return;         /* this can happen */
 
+	if (pm_runtime_suspended(dev))
+		goto exit;
+
 	if (sdkp->WCE) {
 		sd_printk(KERN_NOTICE, sdkp, "Synchronizing SCSI cache\n");
 		sd_sync_cache(sdkp);
@@ -2751,6 +2755,7 @@ static void sd_shutdown(struct device *dev)
 		sd_start_stop_device(sdkp, 0);
 	}
 
+exit:
 	scsi_disk_put(sdkp);
 }
 
diff --git a/include/linux/ahci_platform.h b/include/linux/ahci_platform.h
index be3d9a7..73a2500 100644
--- a/include/linux/ahci_platform.h
+++ b/include/linux/ahci_platform.h
@@ -23,6 +23,8 @@ struct ata_port_info;
 struct ahci_platform_data {
 	int (*init)(struct device *dev, void __iomem *addr);
 	void (*exit)(struct device *dev);
+	int (*suspend)(struct device *dev);
+	int (*resume)(struct device *dev);
 	const struct ata_port_info *ata_port_info;
 	unsigned int force_port_map;
 	unsigned int mask_port_map;
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index 50266c9..5f7d5b3 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -669,6 +669,9 @@ struct Scsi_Host {
 	/* Asynchronous scan in progress */
 	unsigned async_scan:1;
 
+	/* Don't resume host in EH */
+	unsigned eh_noresume:1;
+
 	/*
 	 * Optional work queue to be utilized by the transport
 	 */

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Sun, Jan 8, 2012 at 4:32 PM, Jeff Garzik <jeff@garzik.org> wrote:
>
> Summary (very little excitement at all this time):
>
> 0) Will play around with git signed tags with the next update.
>
> 1) PM improvements, including runtime suspend/resume work

Hmm. I don't know if this comes from the PM improvements or even this
particular pull, but links that aren't connected are *really* slow.

Annoyingly so.

My Macbook Air that I finally can resume reliably again used to come
back almost immediately from resume. No longer. And the reason seems
to be this:

 [  243.306149] ata_piix 0000:00:1f.2: setting latency timer to 64
 [  243.306180] bcma: Found rev 6 PMU (capabilities 0x108C2606)
 [  246.579648] ata1.01: failed to resume link (SControl 0)
 [  246.735472] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
 [  246.735485] ata1.01: SATA link down (SStatus 0 SControl 0)
 [  246.743632] ata1.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES)
filtered out
 [  246.744353] ata1.00: configured for UDMA/100
 [  246.744537] sd 0:0:0:0: [sda] Starting disk
 [  247.769806] ata2.00: failed to resume link (SControl 0)
 [  248.796207] ata2.01: failed to resume link (SControl 0)
 [  248.807665] ata2.00: SATA link down (SStatus 4 SControl 0)
 [  248.807681] ata2.01: SATA link down (SStatus 0 SControl 0)
 [  248.808338] PM: resume of devices complete after 5511.027 msecs
 [  248.882074] PM: Finishing wakeup.

Notice the basically five-second timeout all basically for "failed to
resume link: for things that didn't have anything connected to them
anyway.

This is a bog-standard Intel controller, there's nothing odd there.

I'm pretty sure this used to be much faster, but I haven't bisected
any of it (and with all the problems I had with resume both due to
wireless and MCE, I really wouldn't want to even try).

Taking 5.5 seconds to come back from suspend-to-ram really is too
long. Not *all* of it is the SATA part, but a lot of it is.

For ATA suspend/resume, could we perhaps only resume the ports that
*used* to have something on them? And then, if somebody has plugged
something into the others, not consider that a resume thing at all,
but a hotplug thing that happens *after* the resume?

If it takes five seconds to notice new hardware after a resume, nobody
cares. But the disk we had before obviously needs to get resumed.. But
it does seem like it's the "no link" part that takes long.

Hmm? Or any other ideas?

         Linus

Re: [git patches] libata updates for 3.3

[Jeff Garzik]

On 01/14/2012 12:21 AM, Linus Torvalds wrote:
> On Sun, Jan 8, 2012 at 4:32 PM, Jeff Garzik<jeff@garzik.org>  wrote:
>>
>> Summary (very little excitement at all this time):
>>
>> 0) Will play around with git signed tags with the next update.
>>
>> 1) PM improvements, including runtime suspend/resume work
>
> Hmm. I don't know if this comes from the PM improvements or even this
> particular pull, but links that aren't connected are *really* slow.
>
> Annoyingly so.
>
> My Macbook Air that I finally can resume reliably again used to come
> back almost immediately from resume. No longer. And the reason seems
> to be this:
>
>   [  243.306149] ata_piix 0000:00:1f.2: setting latency timer to 64
>   [  243.306180] bcma: Found rev 6 PMU (capabilities 0x108C2606)
>   [  246.579648] ata1.01: failed to resume link (SControl 0)
>   [  246.735472] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
>   [  246.735485] ata1.01: SATA link down (SStatus 0 SControl 0)
>   [  246.743632] ata1.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES)
> filtered out
>   [  246.744353] ata1.00: configured for UDMA/100
>   [  246.744537] sd 0:0:0:0: [sda] Starting disk
>   [  247.769806] ata2.00: failed to resume link (SControl 0)
>   [  248.796207] ata2.01: failed to resume link (SControl 0)
>   [  248.807665] ata2.00: SATA link down (SStatus 4 SControl 0)
>   [  248.807681] ata2.01: SATA link down (SStatus 0 SControl 0)
>   [  248.808338] PM: resume of devices complete after 5511.027 msecs
>   [  248.882074] PM: Finishing wakeup.
>
> Notice the basically five-second timeout all basically for "failed to
> resume link: for things that didn't have anything connected to them
> anyway.
>
> This is a bog-standard Intel controller, there's nothing odd there.
>
> I'm pretty sure this used to be much faster, but I haven't bisected
> any of it (and with all the problems I had with resume both due to
> wireless and MCE, I really wouldn't want to even try).
>
> Taking 5.5 seconds to come back from suspend-to-ram really is too
> long. Not *all* of it is the SATA part, but a lot of it is.
>
> For ATA suspend/resume, could we perhaps only resume the ports that
> *used* to have something on them? And then, if somebody has plugged
> something into the others, not consider that a resume thing at all,
> but a hotplug thing that happens *after* the resume?
>
> If it takes five seconds to notice new hardware after a resume, nobody
> cares. But the disk we had before obviously needs to get resumed.. But
> it does seem like it's the "no link" part that takes long.

We definitely notice new hardware after a resume, but you're right -- it 
should not take that long to work through ports that are empty.

Will take a look tomorrow (kid->doctor+relatives today, uff) at the most 
recent PM push; my quick testing did not show any problems, but 
suspend/resume varies widely across hardware platforms.  I think I might 
even have a MacBook I can test.  Apple platforms test to be weird too...  ;)

	Jeff

Re: [git patches] libata updates for 3.3

[Lin Ming]

On Sun, 2012-01-15 at 09:41 -0500, Jeff Garzik wrote:
> On 01/14/2012 12:21 AM, Linus Torvalds wrote:
> > On Sun, Jan 8, 2012 at 4:32 PM, Jeff Garzik<jeff@garzik.org>  wrote:
> >>
> >> Summary (very little excitement at all this time):
> >>
> >> 0) Will play around with git signed tags with the next update.
> >>
> >> 1) PM improvements, including runtime suspend/resume work
> >
> > Hmm. I don't know if this comes from the PM improvements or even this
> > particular pull, but links that aren't connected are *really* slow.
> >
> > Annoyingly so.
> >
> > My Macbook Air that I finally can resume reliably again used to come
> > back almost immediately from resume. No longer. And the reason seems
> > to be this:
> >
> >   [  243.306149] ata_piix 0000:00:1f.2: setting latency timer to 64
> >   [  243.306180] bcma: Found rev 6 PMU (capabilities 0x108C2606)
> >   [  246.579648] ata1.01: failed to resume link (SControl 0)
> >   [  246.735472] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
> >   [  246.735485] ata1.01: SATA link down (SStatus 0 SControl 0)
> >   [  246.743632] ata1.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES)
> > filtered out
> >   [  246.744353] ata1.00: configured for UDMA/100
> >   [  246.744537] sd 0:0:0:0: [sda] Starting disk
> >   [  247.769806] ata2.00: failed to resume link (SControl 0)
> >   [  248.796207] ata2.01: failed to resume link (SControl 0)
> >   [  248.807665] ata2.00: SATA link down (SStatus 4 SControl 0)
> >   [  248.807681] ata2.01: SATA link down (SStatus 0 SControl 0)
> >   [  248.808338] PM: resume of devices complete after 5511.027 msecs
> >   [  248.882074] PM: Finishing wakeup.
> >
> > Notice the basically five-second timeout all basically for "failed to
> > resume link: for things that didn't have anything connected to them
> > anyway.
> >
> > This is a bog-standard Intel controller, there's nothing odd there.
> >
> > I'm pretty sure this used to be much faster, but I haven't bisected
> > any of it (and with all the problems I had with resume both due to
> > wireless and MCE, I really wouldn't want to even try).
> >
> > Taking 5.5 seconds to come back from suspend-to-ram really is too
> > long. Not *all* of it is the SATA part, but a lot of it is.
> >
> > For ATA suspend/resume, could we perhaps only resume the ports that
> > *used* to have something on them? And then, if somebody has plugged
> > something into the others, not consider that a resume thing at all,
> > but a hotplug thing that happens *after* the resume?
> >
> > If it takes five seconds to notice new hardware after a resume, nobody
> > cares. But the disk we had before obviously needs to get resumed.. But
> > it does seem like it's the "no link" part that takes long.
> 
> We definitely notice new hardware after a resume, but you're right -- it 
> should not take that long to work through ports that are empty.
> 
> Will take a look tomorrow (kid->doctor+relatives today, uff) at the most 
> recent PM push; my quick testing did not show any problems, but 
> suspend/resume varies widely across hardware platforms.  I think I might 
> even have a MacBook I can test.  Apple platforms test to be weird too...  ;)

I just did a quick test with latest git head(122804e) and didn't find
the problem either.

I'll test other machines.

Lin Ming

> 
> 	Jeff
> 
> 
> 
> 
> 

Re: [git patches] libata updates for 3.3

[Lin Ming]

On Mon, 2012-01-16 at 09:15 +0800, Lin Ming wrote:
> On Sun, 2012-01-15 at 09:41 -0500, Jeff Garzik wrote:
> > On 01/14/2012 12:21 AM, Linus Torvalds wrote:
> > > On Sun, Jan 8, 2012 at 4:32 PM, Jeff Garzik<jeff@garzik.org>  wrote:
> > >>
> > >> Summary (very little excitement at all this time):
> > >>
> > >> 0) Will play around with git signed tags with the next update.
> > >>
> > >> 1) PM improvements, including runtime suspend/resume work
> > >
> > > Hmm. I don't know if this comes from the PM improvements or even this
> > > particular pull, but links that aren't connected are *really* slow.
> > >
> > > Annoyingly so.
> > >
> > > My Macbook Air that I finally can resume reliably again used to come
> > > back almost immediately from resume. No longer. And the reason seems
> > > to be this:
> > >
> > >   [  243.306149] ata_piix 0000:00:1f.2: setting latency timer to 64
> > >   [  243.306180] bcma: Found rev 6 PMU (capabilities 0x108C2606)
> > >   [  246.579648] ata1.01: failed to resume link (SControl 0)
> > >   [  246.735472] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
> > >   [  246.735485] ata1.01: SATA link down (SStatus 0 SControl 0)
> > >   [  246.743632] ata1.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES)
> > > filtered out
> > >   [  246.744353] ata1.00: configured for UDMA/100
> > >   [  246.744537] sd 0:0:0:0: [sda] Starting disk
> > >   [  247.769806] ata2.00: failed to resume link (SControl 0)
> > >   [  248.796207] ata2.01: failed to resume link (SControl 0)
> > >   [  248.807665] ata2.00: SATA link down (SStatus 4 SControl 0)
> > >   [  248.807681] ata2.01: SATA link down (SStatus 0 SControl 0)
> > >   [  248.808338] PM: resume of devices complete after 5511.027 msecs
> > >   [  248.882074] PM: Finishing wakeup.
> > >
> > > Notice the basically five-second timeout all basically for "failed to
> > > resume link: for things that didn't have anything connected to them
> > > anyway.
> > >
> > > This is a bog-standard Intel controller, there's nothing odd there.
> > >
> > > I'm pretty sure this used to be much faster, but I haven't bisected
> > > any of it (and with all the problems I had with resume both due to
> > > wireless and MCE, I really wouldn't want to even try).
> > >
> > > Taking 5.5 seconds to come back from suspend-to-ram really is too
> > > long. Not *all* of it is the SATA part, but a lot of it is.
> > >
> > > For ATA suspend/resume, could we perhaps only resume the ports that
> > > *used* to have something on them? And then, if somebody has plugged
> > > something into the others, not consider that a resume thing at all,
> > > but a hotplug thing that happens *after* the resume?
> > >
> > > If it takes five seconds to notice new hardware after a resume, nobody
> > > cares. But the disk we had before obviously needs to get resumed.. But
> > > it does seem like it's the "no link" part that takes long.
> > 
> > We definitely notice new hardware after a resume, but you're right -- it 
> > should not take that long to work through ports that are empty.
> > 
> > Will take a look tomorrow (kid->doctor+relatives today, uff) at the most 
> > recent PM push; my quick testing did not show any problems, but 
> > suspend/resume varies widely across hardware platforms.  I think I might 
> > even have a MacBook I can test.  Apple platforms test to be weird too...  ;)
> 
> I just did a quick test with latest git head(122804e) and didn't find
> the problem either.
> 
> I'll test other machines.

Set SATA mode to IDE on my machine can reproduce this problem.
The cause is that ata port async suspend was not enabled yet.

Linus,

Could you please try below patch?

>From ba9cba4df9e9389c4f11202e335bee6383b2868c Mon Sep 17 00:00:00 2001
From: Lin Ming <ming.m.lin@intel.com>
Date: Mon, 16 Jan 2012 13:09:02 +0800
Subject: [PATCH] ata: enable ata port async suspend

This saves devices suspend/resume time.

Tested system suspend/resume with SATA IDE/AHCI mode 3 times.
Below is the time took for devices suspend/resume.

SATA mode    vanilla-kernel           patched-kernel
---------    ---------------------    ---------------------
IDE          suspend: 0.744           suspend: 0.432
             (0.716, 0.768, 0.748)    (0.440, 0.428, 0.428)

             resume: 5.084            resume: 2.209
             (5.100, 5.064, 5.088)    (2.168, 2.232, 2.228)

AHCI:        suspend: 0.725           suspend: 0.449
             (0.740, 0.708, 0.728)    (0.456, 0.448, 0.444)

             resume: 2.556            resume: 1.896
             (2.604, 2.492, 2.572)    (1.932, 1.872, 1.884)

Signed-off-by: Lin Ming <ming.m.lin@intel.com>
---
 drivers/ata/libata-transport.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/ata/libata-transport.c b/drivers/ata/libata-transport.c
index 9a7f0ea..74aaee3 100644
--- a/drivers/ata/libata-transport.c
+++ b/drivers/ata/libata-transport.c
@@ -291,6 +291,7 @@ int ata_tport_add(struct device *parent,
 		goto tport_err;
 	}
 
+	device_enable_async_suspend(dev);
 	pm_runtime_set_active(dev);
 	pm_runtime_enable(dev);
 
-- 
1.7.2.5

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Sun, Jan 15, 2012 at 9:23 PM, Lin Ming <ming.m.lin@intel.com> wrote:
>
> Set SATA mode to IDE on my machine can reproduce this problem.
> The cause is that ata port async suspend was not enabled yet.

Yup. Your patch helps.

But why don't we fix the silly legacy mode thing? Yeah, Apple's
firmware is crap and doesn't tend to initialize things, but still - we
should have at least an option to not use the legacy PIIX mode just
because the chipset wasn't programmed for AHCI by the bootup.

That would also help with people who have a BIOS that has the option,
but where the person hadn't noticed, and left it in legacy mode.

The IO ports are all there, the *only* reason we don't use AHCI seems
to be that the firmware didn't turn off the legacy bit, so it doesn't
show as an AHCI interface. That's kind of sad.

                Linus

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Mon, Jan 16, 2012 at 11:25 AM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> But why don't we fix the silly legacy mode thing? Yeah, Apple's
> firmware is crap and doesn't tend to initialize things, but still - we
> should have at least an option to not use the legacy PIIX mode just
> because the chipset wasn't programmed for AHCI by the bootup.

Here's the lspci output for that device, in case anybody cares:


  00:1f.2 IDE interface [0101]: Intel Corporation 6 Series/C200 Series
Chipset Family 4 port SATA IDE Controller [8086:1c01] (rev 05)
(prog-if 8f [Master SecP SecO PriP PriO])
        Subsystem: Intel Corporation Device [8086:7270]
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium
>TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Interrupt: pin B routed to IRQ 19
        Region 0: I/O ports at 2168 [size=8]
        Region 1: I/O ports at 217c [size=4]
        Region 2: I/O ports at 2160 [size=8]
        Region 3: I/O ports at 2178 [size=4]
        Region 4: I/O ports at 2060 [size=16]
        Region 5: I/O ports at ffe0 [size=16]
        Capabilities: <access denied>
        Kernel driver in use: ata_piix
  00: 86 80 01 1c 07 00 b0 02 05 8f 01 01 00 00 00 00
  10: 69 21 00 00 7d 21 00 00 61 21 00 00 79 21 00 00
  20: 61 20 00 00 e1 ff 00 00 00 00 00 00 86 80 70 72
  30: 00 00 00 00 70 00 00 00 00 00 00 00 0b 02 00 00

and I do think that resume used to be faster in v3.2 than it is even
with your patch. But maybe that's some rose-colored glasses.

I'll recompile an old kernel to check.

               Linus

Re: [git patches] libata updates for 3.3

[Alan Cox]

> The IO ports are all there, the *only* reason we don't use AHCI seems
> to be that the firmware didn't turn off the legacy bit, so it doesn't
> show as an AHCI interface. That's kind of sad.

There are two sets of problems with that

The first is that quite a few machines crap themselves if you do this
because nobody has ever tested things like the IRQ routing or the
firmware on suspend/resume when they find the hardwared controller has
gone for a walk.

Putting it back on suspend might help in some cases but then you get to
pick your way through the documentation minefield, deal with device
changes while in non AHCI mode, figure out how to set up registers the
BIOS didn't etc.

Would be a good project for someone with a lot of patience anyway as the
AHCI mode is also 64bit capable and *much* faster overall as it doesn't
keep stalling the CPU poking imaginary registers and turning them into
messages on the SATA bus.

Alan

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Mon, Jan 16, 2012 at 11:42 AM, Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
>
> There are two sets of problems with that
>
> The first is that quite a few machines crap themselves if you do this
> because nobody has ever tested things like the IRQ routing or the
> firmware on suspend/resume when they find the hardwared controller has
> gone for a walk.

I think this could trivially be solved at least on these kinds of
Apple machines by just making it a config option, and just not doing
it if it doesn't work. Some kernel command line to say "ahci=force" or
whatever.

That doesn't solve the "user never even realized" problem, but at
least it gives the user the possibility of solving the "crap firmware
doesn't even allow this".

> Putting it back on suspend might help in some cases but then you get to
> pick your way through the documentation minefield, deal with device
> changes while in non AHCI mode, figure out how to set up registers the
> BIOS didn't etc.

Yeah, no, that would just be horrible. I doubt it happens in practice,
though.  I bet the normal PCI/AHCI resume will just do the right
thing. But I haven't tried..

                        Linus

Re: [git patches] libata updates for 3.3

[Alan Cox]

On Mon, 16 Jan 2012 11:47:12 -0800
Linus Torvalds <torvalds@linux-foundation.org> wrote:

> On Mon, Jan 16, 2012 at 11:42 AM, Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
> >
> > There are two sets of problems with that
> >
> > The first is that quite a few machines crap themselves if you do this
> > because nobody has ever tested things like the IRQ routing or the
> > firmware on suspend/resume when they find the hardwared controller has
> > gone for a walk.
> 
> I think this could trivially be solved at least on these kinds of
> Apple machines by just making it a config option, and just not doing
> it if it doesn't work. Some kernel command line to say "ahci=force" or
> whatever.

There are patches for this from way back including a white list of
devices that were tested

Just ping Matthew about

http://www.codon.org.uk/~mjg59/tmp/ahci_quirk_cleanup.diff

> That doesn't solve the "user never even realized" problem, but at
> least it gives the user the possibility of solving the "crap firmware
> doesn't even allow this".

For the "memory found over 4GB" case it may even be worth printing a
warning.

> > Putting it back on suspend might help in some cases but then you get to
> > pick your way through the documentation minefield, deal with device
> > changes while in non AHCI mode, figure out how to set up registers the
> > BIOS didn't etc.
> 
> Yeah, no, that would just be horrible. I doubt it happens in practice,
> though.  I bet the normal PCI/AHCI resume will just do the right
> thing. But I haven't tried..

Some BIOSes go off and try and re-unlock the drive with the bios password
etc. They get most unhappy when the controller isn't in the mode they
expected. Matthew used a white list for some he tested which didn't have
problems. Most netbooks seem ok with it.

Alan

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Mon, Jan 16, 2012 at 11:29 AM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> and I do think that resume used to be faster in v3.2 than it is even
> with your patch. But maybe that's some rose-colored glasses.
>
> I'll recompile an old kernel to check.

Confirmed.

Plain v3.2 really *is* faster than current git, even current git with
your patch.

In v3.2 I get this:

 [   92.035600] ata_piix 0000:00:1f.2: PCI INT B -> GSI 19 (level,
low) -> IRQ 19
 [   92.035610] ata_piix 0000:00:1f.2: setting latency timer to 64
 [   92.036213] sd 0:0:0:0: [sda] Starting disk
 [   93.060471] ata2.00: failed to resume link (SControl 0)
 [   93.379963] ata1.01: failed to resume link (SControl 0)
 [   93.535802] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
 [   93.535815] ata1.01: SATA link down (SStatus 0 SControl 0)
 [   93.543968] ata1.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES)
filtered out
 [   93.544747] ata1.00: configured for UDMA/100
 [   93.916453] PM: resume of devices complete after 1883.974 msecs
 [   93.916677] PM: Finishing wakeup.
 [   94.086800] ata2.01: failed to resume link (SControl 0)
 [   94.098408] ata2.00: SATA link down (SStatus 4 SControl 0)
 [   94.098429] ata2.01: SATA link down (SStatus 0 SControl 0)

while current git (with your patch) gives me

 [  108.115373] ata_piix 0000:00:1f.2: setting latency timer to 64
 [  109.142010] ata2.00: failed to resume link (SControl 0)
 [  109.462004] ata1.01: failed to resume link (SControl 0)
 [  109.618065] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
 [  109.618078] ata1.01: SATA link down (SStatus 0 SControl 0)
 [  109.626242] ata1.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES)
filtered out
 [  109.627060] ata1.00: configured for UDMA/100
 [  109.627240] sd 0:0:0:0: [sda] Starting disk
 [  110.170015] ata2.01: failed to resume link (SControl 0)
 [  110.181480] ata2.00: SATA link down (SStatus 4 SControl 0)
 [  110.181496] ata2.01: SATA link down (SStatus 0 SControl 0)
 [  110.182124] PM: resume of devices complete after 2066.945 msecs
 [  110.224533] PM: Finishing wakeup.

so old kernels used to be a tiny bit faster despite not doing that
async thing (still slower than I'd like: I'd think that we should be
able to resume devices in less than a second, but I don't know where
all the time goes)

             Linus

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Mon, Jan 16, 2012 at 11:54 AM, Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
>
> Just ping Matthew about
>
> http://www.codon.org.uk/~mjg59/tmp/ahci_quirk_cleanup.diff

Hmm. This with just a setup option to enable it might well be good
enough. No need for whitelists/blacklists, since we couldn't maintain
those sanely anyway.

Matthew?

Googling shows that there's a few old bugreports for this too (big
surprise: macs):

  https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/96692

and various other ones..

                 Linus

                      Linus

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Mon, Jan 16, 2012 at 12:02 PM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>>
>> http://www.codon.org.uk/~mjg59/tmp/ahci_quirk_cleanup.diff
>
> Hmm. This with just a setup option to enable it might well be good
> enough. No need for whitelists/blacklists, since we couldn't maintain
> those sanely anyway.

That patch doesn't work for me. I get a panic in ahci_enable_ahci,
looks like 'mmio' is bogus there (attempting to dereference 0x01ffe6).

I notice that the PIIX device doesn't have any MMIO at all, it's all
PIO. Maybe there is some "need to enable MMIO thing" too. Or maybe I
used the wrong PCI ID when I added it to the quirk table.

                       Linus

Re: [git patches] libata updates for 3.3

[Jeff Garzik]

On 01/16/2012 03:21 PM, Linus Torvalds wrote:
> On Mon, Jan 16, 2012 at 12:02 PM, Linus Torvalds
> <torvalds@linux-foundation.org>  wrote:
>>>
>>> http://www.codon.org.uk/~mjg59/tmp/ahci_quirk_cleanup.diff
>>
>> Hmm. This with just a setup option to enable it might well be good
>> enough. No need for whitelists/blacklists, since we couldn't maintain
>> those sanely anyway.
>
> That patch doesn't work for me. I get a panic in ahci_enable_ahci,
> looks like 'mmio' is bogus there (attempting to dereference 0x01ffe6).
>
> I notice that the PIIX device doesn't have any MMIO at all, it's all
> PIO. Maybe there is some "need to enable MMIO thing" too. Or maybe I
> used the wrong PCI ID when I added it to the quirk table.

Part of the problem with force-enable is that the MMIO BAR may need a 
value, and not have it.  That, and an expectations mismatch between BIOS 
and kernel WRT mode (IDE/AHCI) has always been the reason why 
'ahci=force' never made it in.  It just never seemed to be reliable broadly.

I'm happy with anything we can get working reliably, even with the 
obvious proviso that 'ahci=force' would be a default-off, user-enabled 
option.  It's always been motherboard/BIOS issues that got in the way.

	Jeff

Re: [git patches] libata updates for 3.3

[Matthew Garrett]

On Mon, Jan 16, 2012 at 12:02:32PM -0800, Linus Torvalds wrote:
> On Mon, Jan 16, 2012 at 11:54 AM, Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
> >
> > Just ping Matthew about
> >
> > http://www.codon.org.uk/~mjg59/tmp/ahci_quirk_cleanup.diff
> 
> Hmm. This with just a setup option to enable it might well be good
> enough. No need for whitelists/blacklists, since we couldn't maintain
> those sanely anyway.
> 
> Matthew?

Last time I tested it it wasn't dealing well with setting up combination 
mode for devices with optical media, and I don't have much hardware 
lying around which still uses piix. But if I can find some time I'll dig 
something up and see if it still works. Or, if anyone else wants to do 
the same...

> Googling shows that there's a few old bugreports for this too (big
> surprise: macs):

Macs will come up with ahci if you boot via EFI rather than the BIOS 
compatibility module.

-- 
Matthew Garrett | mjg59@srcf.ucam.org

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Mon, Jan 16, 2012 at 1:26 PM, Matthew Garrett <mjg59@srcf.ucam.org> wrote:
>
> Macs will come up with ahci if you boot via EFI rather than the BIOS
> compatibility module.

Apparently not this one. Or there is something else subtly wrong. I
use rEFIt to boot Linux on this thing, which should mean that it
booted as an EFI image, even if the kernel itself is not natively
using EFI (or even built to care).

But maybe rEFIt does something else than I expected it to do, so who knows..

                  Linus

Re: [git patches] libata updates for 3.3

[Matthew Garrett]

On Mon, Jan 16, 2012 at 01:34:53PM -0800, Linus Torvalds wrote:
> On Mon, Jan 16, 2012 at 1:26 PM, Matthew Garrett <mjg59@srcf.ucam.org> wrote:
> >
> > Macs will come up with ahci if you boot via EFI rather than the BIOS
> > compatibility module.
> 
> Apparently not this one. Or there is something else subtly wrong. I
> use rEFIt to boot Linux on this thing, which should mean that it
> booted as an EFI image, even if the kernel itself is not natively
> using EFI (or even built to care).

refit is an EFI app, but it's usually used to trigger a BIOS-based boot.

-- 
Matthew Garrett | mjg59@srcf.ucam.org

Re: [git patches] libata updates for 3.3

[Alan Cox]

> Part of the problem with force-enable is that the MMIO BAR may need a 
> value, and not have it.  That, and an expectations mismatch between BIOS 

That's fine - we can happily assign one at boot time.

Alan

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Mon, Jan 16, 2012 at 3:54 PM, Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
>> Part of the problem with force-enable is that the MMIO BAR may need a
>> value, and not have it.  That, and an expectations mismatch between BIOS
>
> That's fine - we can happily assign one at boot time.

Actually, I think the reason that Matthews patch doesn't work for me
is that on my device, BAR #5 really is a _port_ BAR.

I didn't play with it much - busy merging and looking at various other
issues - but I'm starting to wonder whether maybe that 8086:1c01 chip
doesn't support AHCI at all. Or maybe it does something differently.

Other reports of this have BAR#5 either clear, or an MMIO BAR.   That

   Region 5: I/O ports at ffe0 [size=16]

looks really odd. Matthew's patch uses

  pci_assign_resource(pdev, 5);

but if it is a PIO region, that won't help anything..

That may be why it oopses for me - the AHCI driver is trying to use a
PIO address as an MMIO one with readl().

                   Linus

Re: [git patches] libata updates for 3.3

[Matthew Garrett]

On Mon, Jan 16, 2012 at 04:02:31PM -0800, Linus Torvalds wrote:
> On Mon, Jan 16, 2012 at 3:54 PM, Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
> >> Part of the problem with force-enable is that the MMIO BAR may need a
> >> value, and not have it.  That, and an expectations mismatch between BIOS
> >
> > That's fine - we can happily assign one at boot time.
> 
> Actually, I think the reason that Matthews patch doesn't work for me
> is that on my device, BAR #5 really is a _port_ BAR.
> 
> I didn't play with it much - busy merging and looking at various other
> issues - but I'm starting to wonder whether maybe that 8086:1c01 chip
> doesn't support AHCI at all. Or maybe it does something differently.
> 
> Other reports of this have BAR#5 either clear, or an MMIO BAR.   That
> 
>    Region 5: I/O ports at ffe0 [size=16]
> 
> looks really odd. Matthew's patch uses
> 
>   pci_assign_resource(pdev, 5);
> 
> but if it is a PIO region, that won't help anything..

More recent ICHs may need different setup here. The datasheet should 
have enough to figure it out, but I can take a look once I get back from 
LCA.

-- 
Matthew Garrett | mjg59@srcf.ucam.org

Re: [git patches] libata updates for 3.3

[Lin Ming]

On Mon, 2012-01-16 at 11:55 -0800, Linus Torvalds wrote:
> On Mon, Jan 16, 2012 at 11:29 AM, Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
> >
> > and I do think that resume used to be faster in v3.2 than it is even
> > with your patch. But maybe that's some rose-colored glasses.
> >
> > I'll recompile an old kernel to check.
> 
> Confirmed.
> 
> Plain v3.2 really *is* faster than current git, even current git with
> your patch.
> 
> In v3.2 I get this:
> 
>  [   92.035600] ata_piix 0000:00:1f.2: PCI INT B -> GSI 19 (level,
> low) -> IRQ 19
>  [   92.035610] ata_piix 0000:00:1f.2: setting latency timer to 64
>  [   92.036213] sd 0:0:0:0: [sda] Starting disk
>  [   93.060471] ata2.00: failed to resume link (SControl 0)
>  [   93.379963] ata1.01: failed to resume link (SControl 0)
>  [   93.535802] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
>  [   93.535815] ata1.01: SATA link down (SStatus 0 SControl 0)
>  [   93.543968] ata1.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES)
> filtered out
>  [   93.544747] ata1.00: configured for UDMA/100
>  [   93.916453] PM: resume of devices complete after 1883.974 msecs
>  [   93.916677] PM: Finishing wakeup.
>  [   94.086800] ata2.01: failed to resume link (SControl 0)
>  [   94.098408] ata2.00: SATA link down (SStatus 4 SControl 0)
>  [   94.098429] ata2.01: SATA link down (SStatus 0 SControl 0)
> 
> while current git (with your patch) gives me
> 
>  [  108.115373] ata_piix 0000:00:1f.2: setting latency timer to 64
>  [  109.142010] ata2.00: failed to resume link (SControl 0)
>  [  109.462004] ata1.01: failed to resume link (SControl 0)
>  [  109.618065] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
>  [  109.618078] ata1.01: SATA link down (SStatus 0 SControl 0)
>  [  109.626242] ata1.00: ACPI cmd ef/03:46:00:00:00:a0 (SET FEATURES)
> filtered out
>  [  109.627060] ata1.00: configured for UDMA/100
>  [  109.627240] sd 0:0:0:0: [sda] Starting disk
>  [  110.170015] ata2.01: failed to resume link (SControl 0)
>  [  110.181480] ata2.00: SATA link down (SStatus 4 SControl 0)
>  [  110.181496] ata2.01: SATA link down (SStatus 0 SControl 0)
>  [  110.182124] PM: resume of devices complete after 2066.945 msecs
>  [  110.224533] PM: Finishing wakeup.
> 
> so old kernels used to be a tiny bit faster despite not doing that
> async thing (still slower than I'd like: I'd think that we should be

No, old kernel did that async thing.

Old kernel embedded the ata port suspend/resume code in host
suspend/resume code and host has async suspend enabled.

I split the ata port suspend/resume code out.

ata: add ata port system PM callbacks
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commit;h=5ef4108

But ata port async suspend was not enabled in above patch.
So I think this is a regression.

Thanks,
Lin Ming

> able to resume devices in less than a second, but I don't know where
> all the time goes)
> 
>              Linus

Re: [git patches] libata updates for 3.3

[Linus Torvalds]

On Mon, Jan 16, 2012 at 9:16 PM, Lin Ming <ming.m.lin@intel.com> wrote:
>
> But ata port async suspend was not enabled in above patch.
> So I think this is a regression.

Ok. Jeff? Should I take that patch directly, or will you take it?

                  Linus

Re: [git patches] libata updates for 3.3

[Jeff Garzik]

On 01/17/2012 12:19 AM, Linus Torvalds wrote:
> On Mon, Jan 16, 2012 at 9:16 PM, Lin Ming<ming.m.lin@intel.com>  wrote:
>>
>> But ata port async suspend was not enabled in above patch.
>> So I think this is a regression.
>
> Ok. Jeff? Should I take that patch directly, or will you take it?

Coming today, with at least one other change...  still keeping an eye 
out for possible PM problems on older non-AHCI SATA hardware.

	Jeff

Re: [git patches] libata updates for 3.3

[Jeff Garzik]

On 01/16/2012 02:55 PM, Linus Torvalds wrote:
> so old kernels used to be a tiny bit faster despite not doing that
> async thing (still slower than I'd like: I'd think that we should be
> able to resume devices in less than a second, but I don't know where
> all the time goes)

In libata specifically, we remain peppered liberally with a bunch of 
msleep()'s all over the place, particularly when dealing with SATA phys. 
  10ms here, 20ms there, 200ms in sata_link_resume(), it all adds up 
when you're talking about fast boot.

For most hardware, we can scan ports in parallel, which gets you past 
the biggest annoyances, but not all of them.

	Jeff