[PATCH 1/2] ahci: imx: use macros to define registers and bits

[PATCH 1/2] ahci: imx: use macros to define registers and bits

[Shawn Guo]

Comparing to enums, macros are more conventional to be used for
registers and bits definition.  Let's switch to macros.

While at it, the names of the registers and bit-fields are updated to
have proper namespace prefix and match the hardware reference manual.

No functional change is involved.

Signed-off-by: Shawn Guo <shawn.guo@freescale.com>
---
 drivers/ata/ahci_imx.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/ata/ahci_imx.c b/drivers/ata/ahci_imx.c
index 497c7ab..39629b4 100644
--- a/drivers/ata/ahci_imx.c
+++ b/drivers/ata/ahci_imx.c
@@ -28,11 +28,9 @@
 #include <linux/libata.h>
 #include "ahci.h"
 
-enum {
-	PORT_PHY_CTL = 0x178,			/* Port0 PHY Control */
-	PORT_PHY_CTL_PDDQ_LOC = 0x100000,	/* PORT_PHY_CTL bits */
-	HOST_TIMER1MS = 0xe0,			/* Timer 1-ms */
-};
+#define IMX_SATA_TIMER1MS			0x00e0
+#define IMX_SATA_P0PHYCR			0x0178
+#define  P0PHYCR_TEST_PDDQ			(1 << 20)
 
 enum ahci_imx_type {
 	AHCI_IMX53,
@@ -156,8 +154,8 @@ static void ahci_imx_error_handler(struct ata_port *ap)
 	 * without full reset once the pddq mode is enabled making it
 	 * impossible to use as part of libata LPM.
 	 */
-	reg_val = readl(mmio + PORT_PHY_CTL);
-	writel(reg_val | PORT_PHY_CTL_PDDQ_LOC, mmio + PORT_PHY_CTL);
+	reg_val = readl(mmio + IMX_SATA_P0PHYCR);
+	writel(reg_val | P0PHYCR_TEST_PDDQ, mmio + IMX_SATA_P0PHYCR);
 	imx_sata_disable(hpriv);
 	imxpriv->no_device = true;
 }
@@ -248,7 +246,7 @@ static int imx_ahci_probe(struct platform_device *pdev)
 
 	/*
 	 * Configure the HWINIT bits of the HOST_CAP and HOST_PORTS_IMPL,
-	 * and IP vendor specific register HOST_TIMER1MS.
+	 * and IP vendor specific register IMX_SATA_TIMER1MS.
 	 * Configure CAP_SSS (support stagered spin up).
 	 * Implement the port0.
 	 * Get the ahb clock rate, and configure the TIMER1MS register.
@@ -265,7 +263,7 @@ static int imx_ahci_probe(struct platform_device *pdev)
 	}
 
 	reg_val = clk_get_rate(imxpriv->ahb_clk) / 1000;
-	writel(reg_val, hpriv->mmio + HOST_TIMER1MS);
+	writel(reg_val, hpriv->mmio + IMX_SATA_TIMER1MS);
 
 	ret = ahci_platform_init_host(pdev, hpriv, &ahci_imx_port_info, 0, 0);
 	if (ret)
-- 
1.8.3.2

[PATCH 2/2] ahci: imx: software workaround for phy reset issue in resume

[Shawn Guo]

When suspending imx6q systems which have rootfs on SATA, the following
error will likely be seen in resume.  The SATA link will fail to come
up, and it results in an unusable system across the suspend/resume
cycle.

$ echo mem > /sys/power/state
PM: Syncing filesystems ... done.
PM: Preparing system for mem sleep
Freezing user space processes ... (elapsed 0.002 seconds) done.
Freezing remaining freezable tasks ... (elapsed 0.002 seconds) done.
PM: Entering mem sleep
sd 0:0:0:0: [sda] Synchronizing SCSI cache
sd 0:0:0:0: [sda] Stopping disk
PM: suspend of devices complete after 61.914 msecs
PM: suspend devices took 0.070 seconds
PM: late suspend of devices complete after 4.906 msecs
PM: noirq suspend of devices complete after 4.521 msecs
Disabling non-boot CPUs ...
CPU1: shutdown
CPU2: shutdown
CPU3: shutdown
Enabling non-boot CPUs ...
CPU1: Booted secondary processor
CPU1 is up
CPU2: Booted secondary processor
CPU2 is up
CPU3: Booted secondary processor
CPU3 is up
PM: noirq resume of devices complete after 10.486 msecs
PM: early resume of devices complete after 4.679 msecs
sd 0:0:0:0: [sda] Starting disk
PM: resume of devices complete after 22.674 msecs
PM: resume devices took 0.030 seconds
PM: Finishing wakeup.
Restarting tasks ... done.
$ ata1: SATA link down (SStatus 1 SControl 300)
ata1: SATA link down (SStatus 1 SControl 300)
ata1: limiting SATA link speed to 1.5 Gbps
ata1: SATA link down (SStatus 1 SControl 310)
ata1.00: disabled
ata1: exception Emask 0x10 SAct 0x0 SErr 0x4040000 action 0xe frozen t4
ata1: irq_stat 0x00000040, connection status changed
ata1: SError: { CommWake DevExch }
ata1: hard resetting link
sd 0:0:0:0: rejecting I/O to offline device
sd 0:0:0:0: killing request
sd 0:0:0:0: rejecting I/O to offline device
Aborting journal on device sda2-8.
sd 0:0:0:0: rejecting I/O to offline device
EXT4-fs warning (device sda2): ext4_end_bio:317: I/O error writing to inode 132577 (offset 0 size 0 starting block 26235)
Buffer I/O error on device sda2, logical block 10169
...

It's caused by a silicon issue that SATA phy does not get reset by
controller when coming back from LPM.  The patch adds a software
workaround for this issue.  It enforces a software reset on SATA phy
in imx_sata_enable() function, so that we can ensure SATA link will
come up properly in both power-on and resume.

The software reset is implemented by writing phy reset register through
the phy control register bus interface.  Functions
imx_phy_reg_[addressing|write|read]() implement this bus interface, while
imx_sata_phy_reset() performs the actually reset operation.

Signed-off-by: Richard Zhu <r65037@freescale.com>
Signed-off-by: Shawn Guo <shawn.guo@freescale.com>
---
 drivers/ata/ahci_imx.c | 175 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 175 insertions(+)

diff --git a/drivers/ata/ahci_imx.c b/drivers/ata/ahci_imx.c
index 39629b4..7d4756d 100644
--- a/drivers/ata/ahci_imx.c
+++ b/drivers/ata/ahci_imx.c
@@ -31,6 +31,18 @@
 #define IMX_SATA_TIMER1MS			0x00e0
 #define IMX_SATA_P0PHYCR			0x0178
 #define  P0PHYCR_TEST_PDDQ			(1 << 20)
+#define  P0PHYCR_CR_READ			(1 << 19)
+#define  P0PHYCR_CR_WRITE			(1 << 18)
+#define  P0PHYCR_CR_CAP_DATA			(1 << 17)
+#define  P0PHYCR_CR_CAP_ADDR			(1 << 16)
+#define IMX_SATA_P0PHYSR			0x017c
+#define  P0PHYSR_CR_ACK				(1 << 18)
+#define  P0PHYSR_CR_DATA_OUT			0xffff
+
+#define IMX_PHY_LANE0_OUT_STAT			0x2003
+#define  LANE0_OUT_STAT_RX_PLL_STATE		(1 << 1)
+#define IMX_PHY_CLOCK_RESET			0x7f3f
+#define  CLOCK_RESET_RESET			(1 << 0)
 
 enum ahci_imx_type {
 	AHCI_IMX53,
@@ -52,9 +64,165 @@ MODULE_PARM_DESC(hotplug, "AHCI IMX hot-plug support (0=Don't support, 1=support
 
 static void ahci_imx_host_stop(struct ata_host *host);
 
+static int imx_phy_ack_polling(void __iomem *mmio, bool assert)
+{
+	int timeout = 100;
+	u32 srval;
+
+	do {
+		srval = readl(mmio + IMX_SATA_P0PHYSR);
+		if ((assert ? srval : ~srval) & P0PHYSR_CR_ACK)
+			break;
+		usleep_range(100, 200);
+	} while (--timeout);
+
+	return timeout ? 0 : -ETIMEDOUT;
+}
+
+static int imx_phy_reg_addressing(u16 addr, void __iomem *mmio)
+{
+	u32 crval = addr;
+	int ret;
+
+	/* 1. Supply the address on cr_data_in */
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 2. Assert the cr_cap_addr signal */
+	crval |= P0PHYCR_CR_CAP_ADDR;
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 3. Wait for the cr_ack signal to be asserted */
+	ret = imx_phy_ack_polling(mmio, true);
+	if (ret)
+		return ret;
+
+	/* 4. Deassert cr_cap_addr */
+	crval &= ~P0PHYCR_CR_CAP_ADDR;
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 5. Wait for cr_ack to be deasserted */
+	ret = imx_phy_ack_polling(mmio, false);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int imx_phy_reg_write(u16 val, void __iomem *mmio)
+{
+	u32 crval = val;
+	int ret;
+
+	/* 1. Supply the data on cr_data_in */
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 2. Assert the cr_cap_data signal */
+	crval |= P0PHYCR_CR_CAP_DATA;
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 3. Wait for the cr_ack signal to be asserted */
+	ret = imx_phy_ack_polling(mmio, true);
+	if (ret)
+		return ret;
+
+	/* 4. Deassert cr_cap_data */
+	crval &= ~P0PHYCR_CR_CAP_DATA;
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 5. Wait for cr_ack to be deasserted */
+	ret = imx_phy_ack_polling(mmio, false);
+	if (ret)
+		return ret;
+
+	/* 6. Assert the cr_write signal */
+	crval |= P0PHYCR_CR_WRITE;
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 7. Wait for cr_ack to be asserted */
+	ret = imx_phy_ack_polling(mmio, true);
+	if (ret)
+		return ret;
+
+	/* 8. Deassert cr_write */
+	crval &= ~P0PHYCR_CR_WRITE;
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 9. Wait for cr_ack to be deasserted */
+	ret = imx_phy_ack_polling(mmio, false);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int imx_phy_reg_read(u16 *val, void __iomem *mmio)
+{
+	u32 crval = 0;
+	int ret;
+
+	/* 1. Assert the cr_read signal */
+	crval |= P0PHYCR_CR_READ;
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 2. Wait for the cr_ack signal to be asserted */
+	ret = imx_phy_ack_polling(mmio, true);
+	if (ret)
+		return ret;
+
+	/* 3. Capture the data from cr_data_out[] */
+	*val = readl(mmio + IMX_SATA_P0PHYSR) & P0PHYSR_CR_DATA_OUT;
+
+	/* 4. Deassert cr_read */
+	crval &= ~P0PHYCR_CR_READ;
+	writel(crval, mmio + IMX_SATA_P0PHYCR);
+
+	/* 5. Wait for cr_ack to be deasserted */
+	ret = imx_phy_ack_polling(mmio, false);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int imx_sata_phy_reset(struct ahci_host_priv *hpriv)
+{
+	void __iomem *mmio = hpriv->mmio;
+	int timeout = 10;
+	u16 val;
+	int ret;
+
+	/* Reset SATA PHY by setting RESET bit of PHY register CLOCK_RESET */
+	ret = imx_phy_reg_addressing(IMX_PHY_CLOCK_RESET, mmio);
+	if (ret)
+		return ret;
+	/*
+	 * For phy reset operation, we skip the timeout checking, because phy
+	 * will be unable to acknowledge in this case.
+	 */
+	imx_phy_reg_write(CLOCK_RESET_RESET, mmio);
+
+	usleep_range(100, 200);
+
+	/* Wait for PHY RX_PLL to be stable */
+	do {
+		ret = imx_phy_reg_addressing(IMX_PHY_LANE0_OUT_STAT, mmio);
+		if (ret)
+			return ret;
+		ret = imx_phy_reg_read(&val, mmio);
+		if (ret)
+			return ret;
+		if (val & LANE0_OUT_STAT_RX_PLL_STATE)
+			break;
+		usleep_range(100, 200);
+	} while (--timeout);
+
+	return timeout ? 0 : -ETIMEDOUT;
+}
+
 static int imx_sata_enable(struct ahci_host_priv *hpriv)
 {
 	struct imx_ahci_priv *imxpriv = hpriv->plat_data;
+	struct device *dev = &imxpriv->ahci_pdev->dev;
 	int ret;
 
 	if (imxpriv->no_device)
@@ -99,6 +267,12 @@ static int imx_sata_enable(struct ahci_host_priv *hpriv)
 		regmap_update_bits(imxpriv->gpr, IOMUXC_GPR13,
 				   IMX6Q_GPR13_SATA_MPLL_CLK_EN,
 				   IMX6Q_GPR13_SATA_MPLL_CLK_EN);
+
+		ret = imx_sata_phy_reset(hpriv);
+		if (ret) {
+			dev_err(dev, "failed to reset phy: %d\n", ret);
+			goto disable_regulator;
+		}
 	}
 
 	usleep_range(1000, 2000);
@@ -215,6 +389,7 @@ static int imx_ahci_probe(struct platform_device *pdev)
 	if (!imxpriv)
 		return -ENOMEM;
 
+	imxpriv->ahci_pdev = pdev;
 	imxpriv->no_device = false;
 	imxpriv->first_time = true;
 	imxpriv->type = (enum ahci_imx_type)of_id->data;
-- 
1.8.3.2

[PATCH 2/2] ahci: imx: software workaround for phy reset issue in resume

[Fabio Estevam]

On Mon, Apr 14, 2014 at 11:41 PM, Shawn Guo <shawn.guo@freescale.com> wrote:
> When suspending imx6q systems which have rootfs on SATA, the following
> error will likely be seen in resume.  The SATA link will fail to come
> up, and it results in an unusable system across the suspend/resume
> cycle.
>
> $ echo mem > /sys/power/state
> PM: Syncing filesystems ... done.
> PM: Preparing system for mem sleep
> Freezing user space processes ... (elapsed 0.002 seconds) done.
> Freezing remaining freezable tasks ... (elapsed 0.002 seconds) done.
> PM: Entering mem sleep
> sd 0:0:0:0: [sda] Synchronizing SCSI cache
> sd 0:0:0:0: [sda] Stopping disk
> PM: suspend of devices complete after 61.914 msecs
> PM: suspend devices took 0.070 seconds
> PM: late suspend of devices complete after 4.906 msecs
> PM: noirq suspend of devices complete after 4.521 msecs
> Disabling non-boot CPUs ...
> CPU1: shutdown
> CPU2: shutdown
> CPU3: shutdown
> Enabling non-boot CPUs ...
> CPU1: Booted secondary processor
> CPU1 is up
> CPU2: Booted secondary processor
> CPU2 is up
> CPU3: Booted secondary processor
> CPU3 is up
> PM: noirq resume of devices complete after 10.486 msecs
> PM: early resume of devices complete after 4.679 msecs
> sd 0:0:0:0: [sda] Starting disk
> PM: resume of devices complete after 22.674 msecs
> PM: resume devices took 0.030 seconds
> PM: Finishing wakeup.
> Restarting tasks ... done.
> $ ata1: SATA link down (SStatus 1 SControl 300)
> ata1: SATA link down (SStatus 1 SControl 300)
> ata1: limiting SATA link speed to 1.5 Gbps
> ata1: SATA link down (SStatus 1 SControl 310)
> ata1.00: disabled
> ata1: exception Emask 0x10 SAct 0x0 SErr 0x4040000 action 0xe frozen t4
> ata1: irq_stat 0x00000040, connection status changed
> ata1: SError: { CommWake DevExch }
> ata1: hard resetting link
> sd 0:0:0:0: rejecting I/O to offline device
> sd 0:0:0:0: killing request
> sd 0:0:0:0: rejecting I/O to offline device
> Aborting journal on device sda2-8.
> sd 0:0:0:0: rejecting I/O to offline device
> EXT4-fs warning (device sda2): ext4_end_bio:317: I/O error writing to inode 132577 (offset 0 size 0 starting block 26235)
> Buffer I/O error on device sda2, logical block 10169
> ...
>
> It's caused by a silicon issue that SATA phy does not get reset by
> controller when coming back from LPM.  The patch adds a software
> workaround for this issue.  It enforces a software reset on SATA phy
> in imx_sata_enable() function, so that we can ensure SATA link will
> come up properly in both power-on and resume.
>
> The software reset is implemented by writing phy reset register through
> the phy control register bus interface.  Functions
> imx_phy_reg_[addressing|write|read]() implement this bus interface, while
> imx_sata_phy_reset() performs the actually reset operation.
>
> Signed-off-by: Richard Zhu <r65037@freescale.com>
> Signed-off-by: Shawn Guo <shawn.guo@freescale.com>


Tested-by: Fabio Estevam <fabio.estevam@freescale.com>

This fixes the warning I reported at:
http://www.spinics.net/lists/arm-kernel/msg318745.html

Thanks,

Fabio Estevam

[PATCH 2/2] ahci: imx: software workaround for phy reset issue in resume

[Tejun Heo]

On Tue, Apr 15, 2014 at 10:41:43AM +0800, Shawn Guo wrote:
> +static int imx_phy_ack_polling(void __iomem *mmio, bool assert)
> +{
> +	int timeout = 100;
> +	u32 srval;
> +
> +	do {
> +		srval = readl(mmio + IMX_SATA_P0PHYSR);
> +		if ((assert ? srval : ~srval) & P0PHYSR_CR_ACK)
> +			break;
> +		usleep_range(100, 200);

What's up with usleep_range() these days?  There's no point in polling
in sub-msec intervals.  Let's please stick to msleep().

> +static int imx_phy_reg_addressing(u16 addr, void __iomem *mmio)
> +{
> +	u32 crval = addr;
> +	int ret;
> +
> +	/* 1. Supply the address on cr_data_in */
> +	writel(crval, mmio + IMX_SATA_P0PHYCR);
> +
> +	/* 2. Assert the cr_cap_addr signal */
> +	crval |= P0PHYCR_CR_CAP_ADDR;
> +	writel(crval, mmio + IMX_SATA_P0PHYCR);
> +
> +	/* 3. Wait for the cr_ack signal to be asserted */
> +	ret = imx_phy_ack_polling(mmio, true);
> +	if (ret)
> +		return ret;
> +
> +	/* 4. Deassert cr_cap_addr */
> +	crval &= ~P0PHYCR_CR_CAP_ADDR;
> +	writel(crval, mmio + IMX_SATA_P0PHYCR);
> +
> +	/* 5. Wait for cr_ack to be deasserted */
> +	ret = imx_phy_ack_polling(mmio, false);
> +	if (ret)
> +		return ret;
> +
> +	return 0;
> +}

Wouldn't folding comment 3 and 5 into 2 and 4 respectively make it
easier on the eyes?  They're single operations anyway.

> +static int imx_phy_reg_write(u16 val, void __iomem *mmio)
> +{
...
> +}

Ditto.  Also, maybe it'd be better to create a wrapper to assert/clear
and wait for ack?

...
> +static int imx_sata_phy_reset(struct ahci_host_priv *hpriv)
> +{
> +	void __iomem *mmio = hpriv->mmio;
> +	int timeout = 10;
> +	u16 val;
> +	int ret;
> +
> +	/* Reset SATA PHY by setting RESET bit of PHY register CLOCK_RESET */
> +	ret = imx_phy_reg_addressing(IMX_PHY_CLOCK_RESET, mmio);
> +	if (ret)
> +		return ret;
> +	/*
> +	 * For phy reset operation, we skip the timeout checking, because phy
> +	 * will be unable to acknowledge in this case.
> +	 */
> +	imx_phy_reg_write(CLOCK_RESET_RESET, mmio);
> +
> +	usleep_range(100, 200);
> +
> +	/* Wait for PHY RX_PLL to be stable */
> +	do {
> +		ret = imx_phy_reg_addressing(IMX_PHY_LANE0_OUT_STAT, mmio);
> +		if (ret)
> +			return ret;
> +		ret = imx_phy_reg_read(&val, mmio);
> +		if (ret)
> +			return ret;
> +		if (val & LANE0_OUT_STAT_RX_PLL_STATE)
> +			break;
> +		usleep_range(100, 200);

Ditto with above.

Thanks.

-- 
tejun

[PATCH 2/2] ahci: imx: software workaround for phy reset issue in resume

[Shawn Guo]

On Tue, Apr 15, 2014 at 12:10:18PM -0400, Tejun Heo wrote:
> On Tue, Apr 15, 2014 at 10:41:43AM +0800, Shawn Guo wrote:
> > +static int imx_phy_ack_polling(void __iomem *mmio, bool assert)
> > +{
> > +	int timeout = 100;
> > +	u32 srval;
> > +
> > +	do {
> > +		srval = readl(mmio + IMX_SATA_P0PHYSR);
> > +		if ((assert ? srval : ~srval) & P0PHYSR_CR_ACK)
> > +			break;
> > +		usleep_range(100, 200);
> 
> What's up with usleep_range() these days?  There's no point in polling
> in sub-msec intervals.  Let's please stick to msleep().

The hardware manual suggests that an ACK will normally be given within
100us.  So using msleep() will have it sleep longer than necessary for
normal case.  Basically, we're following the suggestion from
Documentation/timers/timers-howto.txt to use usleep_range() over
msleep().

SLEEPING FOR ~USECS OR SMALL MSECS ( 10us - 20ms):
	* Use usleep_range

	- Why not msleep for (1ms - 20ms)?
		Explained originally here:
			http://lkml.org/lkml/2007/8/3/250
		msleep(1~20) may not do what the caller intends, and
		will often sleep longer (~20 ms actual sleep for any
		value given in the 1~20ms range). In many cases this
		is not the desired behavior.

> 
> > +static int imx_phy_reg_addressing(u16 addr, void __iomem *mmio)
> > +{
> > +	u32 crval = addr;
> > +	int ret;
> > +
> > +	/* 1. Supply the address on cr_data_in */
> > +	writel(crval, mmio + IMX_SATA_P0PHYCR);
> > +
> > +	/* 2. Assert the cr_cap_addr signal */
> > +	crval |= P0PHYCR_CR_CAP_ADDR;
> > +	writel(crval, mmio + IMX_SATA_P0PHYCR);
> > +
> > +	/* 3. Wait for the cr_ack signal to be asserted */
> > +	ret = imx_phy_ack_polling(mmio, true);
> > +	if (ret)
> > +		return ret;
> > +
> > +	/* 4. Deassert cr_cap_addr */
> > +	crval &= ~P0PHYCR_CR_CAP_ADDR;
> > +	writel(crval, mmio + IMX_SATA_P0PHYCR);
> > +
> > +	/* 5. Wait for cr_ack to be deasserted */
> > +	ret = imx_phy_ack_polling(mmio, false);
> > +	if (ret)
> > +		return ret;
> > +
> > +	return 0;
> > +}
> 
> Wouldn't folding comment 3 and 5 into 2 and 4 respectively make it
> easier on the eyes?  They're single operations anyway.
> 
> > +static int imx_phy_reg_write(u16 val, void __iomem *mmio)
> > +{
> ...
> > +}
> 
> Ditto.  Also, maybe it'd be better to create a wrapper to assert/clear
> and wait for ack?

I was writing the code in the exact steps documented in the hardware
manual.  But your comment makes a lot of sense to me, so I will create
such a wrapper to make it easier on the eyes.

Thanks, Tejun.

Shawn

> 
> ...
> > +static int imx_sata_phy_reset(struct ahci_host_priv *hpriv)
> > +{
> > +	void __iomem *mmio = hpriv->mmio;
> > +	int timeout = 10;
> > +	u16 val;
> > +	int ret;
> > +
> > +	/* Reset SATA PHY by setting RESET bit of PHY register CLOCK_RESET */
> > +	ret = imx_phy_reg_addressing(IMX_PHY_CLOCK_RESET, mmio);
> > +	if (ret)
> > +		return ret;
> > +	/*
> > +	 * For phy reset operation, we skip the timeout checking, because phy
> > +	 * will be unable to acknowledge in this case.
> > +	 */
> > +	imx_phy_reg_write(CLOCK_RESET_RESET, mmio);
> > +
> > +	usleep_range(100, 200);
> > +
> > +	/* Wait for PHY RX_PLL to be stable */
> > +	do {
> > +		ret = imx_phy_reg_addressing(IMX_PHY_LANE0_OUT_STAT, mmio);
> > +		if (ret)
> > +			return ret;
> > +		ret = imx_phy_reg_read(&val, mmio);
> > +		if (ret)
> > +			return ret;
> > +		if (val & LANE0_OUT_STAT_RX_PLL_STATE)
> > +			break;
> > +		usleep_range(100, 200);
> 
> Ditto with above.
> 
> Thanks.
> 
> -- 
> tejun
> 
> 

[PATCH 2/2] ahci: imx: software workaround for phy reset issue in resume

[Tejun Heo]

Hello,

On Wed, Apr 16, 2014 at 03:38:24PM +0800, Shawn Guo wrote:
> The hardware manual suggests that an ACK will normally be given within
> 100us.  So using msleep() will have it sleep longer than necessary for
> normal case.  Basically, we're following the suggestion from
> Documentation/timers/timers-howto.txt to use usleep_range() over
> msleep().
> 
> SLEEPING FOR ~USECS OR SMALL MSECS ( 10us - 20ms):
> 	* Use usleep_range
> 
> 	- Why not msleep for (1ms - 20ms)?
> 		Explained originally here:
> 			http://lkml.org/lkml/2007/8/3/250
> 		msleep(1~20) may not do what the caller intends, and
> 		will often sleep longer (~20 ms actual sleep for any
> 		value given in the 1~20ms range). In many cases this
> 		is not the desired behavior.

But we don't care whether the minimum wait is actually 100us here, do
we?  It's a reset path.  Let's please just be simple.  msleep(1) not
actually sleeping 1ms is an implementation detail which may change in
the future.  If this were in the command execution path, sure, but not
here.

Thanks.

-- 
tejun

[PATCH 2/2] ahci: imx: software workaround for phy reset issue in resume

[Shawn Guo]

On Wed, Apr 16, 2014 at 10:01:12AM -0400, Tejun Heo wrote:
> Hello,
> 
> On Wed, Apr 16, 2014 at 03:38:24PM +0800, Shawn Guo wrote:
> > The hardware manual suggests that an ACK will normally be given within
> > 100us.  So using msleep() will have it sleep longer than necessary for
> > normal case.  Basically, we're following the suggestion from
> > Documentation/timers/timers-howto.txt to use usleep_range() over
> > msleep().
> > 
> > SLEEPING FOR ~USECS OR SMALL MSECS ( 10us - 20ms):
> > 	* Use usleep_range
> > 
> > 	- Why not msleep for (1ms - 20ms)?
> > 		Explained originally here:
> > 			http://lkml.org/lkml/2007/8/3/250
> > 		msleep(1~20) may not do what the caller intends, and
> > 		will often sleep longer (~20 ms actual sleep for any
> > 		value given in the 1~20ms range). In many cases this
> > 		is not the desired behavior.
> 
> But we don't care whether the minimum wait is actually 100us here, do
> we?  It's a reset path.  Let's please just be simple.  msleep(1) not
> actually sleeping 1ms is an implementation detail which may change in
> the future.  If this were in the command execution path, sure, but not
> here.

Functions imx_phy_reg_[addressing|write|read]() implement the phy
control register bus interface.  Though the interface is used only for
phy reset right now, it's not limited to phy reset but anything that
needs to access phy.

Even for reset path, we still care about the time.  I just quickly
tested it with replacing all those usleep_range() calls with msleep(1),
and found that the resume time increases dramatically, from 10.880 msecs
to 69.431 msecs.  The system resume speed is quite critical to some of
our customers.  If every device sleeps longer than necessary like that,
the system resume speed will be unacceptably slow.  So, please, allow
the use of usleep_range() here.

Shawn

=== usleep_range ===
$ echo mem > /sys/power/state
PM: Syncing filesystems ... done.
PM: Preparing system for mem sleep
Freezing user space processes ... (elapsed 0.001 seconds) done.
Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
PM: Entering mem sleep
sd 0:0:0:0: [sda] Synchronizing SCSI cache
sd 0:0:0:0: [sda] Stopping disk
PM: suspend of devices complete after 51.209 msecs
PM: suspend devices took 0.060 seconds
PM: late suspend of devices complete after 4.651 msecs
PM: noirq suspend of devices complete after 4.869 msecs
Disabling non-boot CPUs ...
CPU1: shutdown
CPU2: shutdown
CPU3: shutdown
Enabling non-boot CPUs ...
CPU1: Booted secondary processor
CPU1 is up
CPU2: Booted secondary processor
CPU2 is up
CPU3: Booted secondary processor
CPU3 is up
PM: noirq resume of devices complete after 4.010 msecs
PM: early resume of devices complete after 4.619 msecs
sd 0:0:0:0: [sda] Starting disk
PM: resume of devices complete after 10.880 msecs
PM: resume devices took 0.020 seconds
PM: Finishing wakeup.
Restarting tasks ... done.
$ ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
ata1.00: configured for UDMA/133

=== msleep(1) ===
$ echo mem > /sys/power/state
PM: Syncing filesystems ... done.
PM: Preparing system for mem sleep
Freezing user space processes ... (elapsed 0.002 seconds) done.
Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
PM: Entering mem sleep
sd 0:0:0:0: [sda] Synchronizing SCSI cache
sd 0:0:0:0: [sda] Stopping disk
PM: suspend of devices complete after 35.751 msecs
PM: suspend devices took 0.040 seconds
PM: late suspend of devices complete after 4.190 msecs
PM: noirq suspend of devices complete after 4.450 msecs
Disabling non-boot CPUs ...
CPU1: shutdown
CPU2: shutdown
CPU3: shutdown
Enabling non-boot CPUs ...
CPU1: Booted secondary processor
CPU1 is up
CPU2: Booted secondary processor
CPU2 is up
CPU3: Booted secondary processor
CPU3 is up
PM: noirq resume of devices complete after 2.328 msecs
PM: early resume of devices complete after 2.774 msecs
sd 0:0:0:0: [sda] Starting disk
PM: resume of devices complete after 69.431 msecs
PM: resume devices took 0.070 seconds
PM: Finishing wakeup.
Restarting tasks ... done.
$ ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
ata1.00: configured for UDMA/133

[PATCH 1/2] ahci: imx: use macros to define registers and bits

[Tejun Heo]

On Tue, Apr 15, 2014 at 10:41:42AM +0800, Shawn Guo wrote:
> Comparing to enums, macros are more conventional to be used for
> registers and bits definition.  Let's switch to macros.
> 
> While at it, the names of the registers and bit-fields are updated to
> have proper namespace prefix and match the hardware reference manual.
> 
> No functional change is involved.
> 
> Signed-off-by: Shawn Guo <shawn.guo@freescale.com>

Please don't switch to defines.  A lot of libata constants use enums
and they tend to have more advantages anyway.

Thanks.

-- 
tejun

[PATCH 1/2] ahci: imx: use macros to define registers and bits

[Shawn Guo]

On Tue, Apr 15, 2014 at 12:03:17PM -0400, Tejun Heo wrote:
> On Tue, Apr 15, 2014 at 10:41:42AM +0800, Shawn Guo wrote:
> > Comparing to enums, macros are more conventional to be used for
> > registers and bits definition.  Let's switch to macros.
> > 
> > While at it, the names of the registers and bit-fields are updated to
> > have proper namespace prefix and match the hardware reference manual.
> > 
> > No functional change is involved.
> > 
> > Signed-off-by: Shawn Guo <shawn.guo@freescale.com>
> 
> Please don't switch to defines.  A lot of libata constants use enums
> and they tend to have more advantages anyway.

I'm seeing a lot of ata drivers use defines though.  And I'm not sure I
understand the advantages of using enums over defines in ahci_imx
driver, where register offset and bit position are defined in the same
'enum' without a particular enum type.

enum {
	PORT_PHY_CTL = 0x178,			/* Port0 PHY Control */
	PORT_PHY_CTL_PDDQ_LOC = 0x100000,	/* PORT_PHY_CTL bits */
	HOST_TIMER1MS = 0xe0,			/* Timer 1-ms */
};

To me it's a misuse of enum.

Shawn

[PATCH 1/2] ahci: imx: use macros to define registers and bits

[Tejun Heo]

Hello,

On Wed, Apr 16, 2014 at 02:35:12PM +0800, Shawn Guo wrote:
> I'm seeing a lot of ata drivers use defines though.  And I'm not sure I
> understand the advantages of using enums over defines in ahci_imx
> driver, where register offset and bit position are defined in the same
> 'enum' without a particular enum type.
> 
> enum {
> 	PORT_PHY_CTL = 0x178,			/* Port0 PHY Control */
> 	PORT_PHY_CTL_PDDQ_LOC = 0x100000,	/* PORT_PHY_CTL bits */
> 	HOST_TIMER1MS = 0xe0,			/* Timer 1-ms */
> };
> 
> To me it's a misuse of enum.

The difference isn't that significant but enums have the advantages of
the definitions being evaluated once rather than each time it's used
which prevents things like missing parentheses subtly messing up the
containing expressions and the compiler, and thus the debugger, being
aware of the constants in use.

The technical advantages not being enormous, it's a good bike shedding
material and argue about it back and forth forever without being
productive at all, so, please just stick with enums.

Thanks.

-- 
tejun

[PATCH 1/2] ahci: imx: use macros to define registers and bits

[Uwe Kleine-König]

On Tue, Apr 15, 2014 at 10:41:42AM +0800, Shawn Guo wrote:
> diff --git a/drivers/ata/ahci_imx.c b/drivers/ata/ahci_imx.c
> index 497c7ab..39629b4 100644
> --- a/drivers/ata/ahci_imx.c
> +++ b/drivers/ata/ahci_imx.c
> @@ -28,11 +28,9 @@
>  #include <linux/libata.h>
>  #include "ahci.h"
>  
> -enum {
> -	PORT_PHY_CTL = 0x178,			/* Port0 PHY Control */
> -	PORT_PHY_CTL_PDDQ_LOC = 0x100000,	/* PORT_PHY_CTL bits */
> -	HOST_TIMER1MS = 0xe0,			/* Timer 1-ms */
> -};
> +#define IMX_SATA_TIMER1MS			0x00e0
> +#define IMX_SATA_P0PHYCR			0x0178
> +#define  P0PHYCR_TEST_PDDQ			(1 << 20)
Did you intentionally not namespace the bit mask?

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-K?nig            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

[PATCH 1/2] ahci: imx: use macros to define registers and bits

[Shawn Guo]

On Wed, Apr 16, 2014 at 09:59:57AM +0200, Uwe Kleine-K?nig wrote:
> On Tue, Apr 15, 2014 at 10:41:42AM +0800, Shawn Guo wrote:
> > diff --git a/drivers/ata/ahci_imx.c b/drivers/ata/ahci_imx.c
> > index 497c7ab..39629b4 100644
> > --- a/drivers/ata/ahci_imx.c
> > +++ b/drivers/ata/ahci_imx.c
> > @@ -28,11 +28,9 @@
> >  #include <linux/libata.h>
> >  #include "ahci.h"
> >  
> > -enum {
> > -	PORT_PHY_CTL = 0x178,			/* Port0 PHY Control */
> > -	PORT_PHY_CTL_PDDQ_LOC = 0x100000,	/* PORT_PHY_CTL bits */
> > -	HOST_TIMER1MS = 0xe0,			/* Timer 1-ms */
> > -};
> > +#define IMX_SATA_TIMER1MS			0x00e0
> > +#define IMX_SATA_P0PHYCR			0x0178
> > +#define  P0PHYCR_TEST_PDDQ			(1 << 20)
> Did you intentionally not namespace the bit mask?

Yes, with the intention to make the bit names a little shorter.

Shawn