[PATCH V2 0/8] memory: emif: miscellaneous bug fixes for EMIF driver

[PATCH V2 0/8] memory: emif: miscellaneous bug fixes for EMIF driver

[Lokesh Vutla]

This series resolves a few minor issues for EMIF driver.

Tested all patches on OMAP4430-sdp.
Patch : "memory: emif: setup LP settings on freq update"
is tested on a local tree, since freq update cannot be
tested on mainline.

Ambresh K (1):
  memory: emif: setup LP settings on freq update

Grygorii Strashko (1):
  memory: emif: errata i743: Prohibit usage of Power-Down mode

Lokesh Vutla (2):
  memory: emif: Fix the lpmode timeout calculation
  memory: emif: Load the correct custom config values from dt

Nishanth Menon (3):
  memory: emif: handle overflow for timing for LP mode
  memory: emif: Handle devices which are not rated for >85C
  memory: emif: use restart if power_off not present when out of spec

Oleksandr Dmytryshyn (1):
  memory: emif: Fix the incorrect 'size' parameter in memcpy

 drivers/memory/emif.c                   |  129 +++++++++++++++++++++++++++----
 include/linux/platform_data/emif_plat.h |    1 +
 2 files changed, 114 insertions(+), 16 deletions(-)

-- 
1.7.9.5

[PATCH V2 1/8] memory: emif: Fix the lpmode timeout calculation

[Lokesh Vutla]

The driver tries to round up the specified timeout cycles to
the next power of 2 value. This should be done defore updating
timeout variable.
Correcting this here.

Reported-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Changes since V1: 
	Updated Changelog and $subject
	Added a comment describing timeout calculation.

 drivers/memory/emif.c |   10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index df08736..508763c 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -728,13 +728,17 @@ static u32 get_pwr_mgmt_ctrl(u32 freq, struct emif_data *emif, u32 ip_rev)
 	/* Timeout based on DDR frequency */
 	timeout = freq >= freq_threshold ? timeout_perf : timeout_pwr;
 
-	/* The value to be set in register is "log2(timeout) - 3" */
+	/*
+	 * The value to be set in register is "log2(timeout) - 3"
+	 * if timeout < 16 load 0 in register
+	 * if timeout is not a power of 2, round to next highest power of 2
+	 */
 	if (timeout < 16) {
 		timeout = 0;
 	} else {
-		timeout = __fls(timeout) - 3;
 		if (timeout & (timeout - 1))
-			timeout++;
+			timeout <<= 1;
+		timeout = __fls(timeout) - 3;
 	}
 
 	switch (lpmode) {
-- 
1.7.9.5

[PATCH V2 2/8] memory: emif: setup LP settings on freq update

[Lokesh Vutla]

From: Ambresh K <ambresh@ti.com>

Program the power management shadow register on freq update
Else the concept of threshold frequencies dont really matter
as the system always uses the performance mode timing for LP
which is programmed in at init time.

Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Ambresh K <ambresh@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Changes since V1: No changes

 drivers/memory/emif.c |    2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index 508763c..5f3b7ed 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -819,6 +819,8 @@ static void setup_registers(struct emif_data *emif, struct emif_regs *regs)
 
 	writel(regs->sdram_tim2_shdw, base + EMIF_SDRAM_TIMING_2_SHDW);
 	writel(regs->phy_ctrl_1_shdw, base + EMIF_DDR_PHY_CTRL_1_SHDW);
+	writel(regs->pwr_mgmt_ctrl_shdw,
+	       base + EMIF_POWER_MANAGEMENT_CTRL_SHDW);
 
 	/* Settings specific for EMIF4D5 */
 	if (emif->plat_data->ip_rev != EMIF_4D5)
-- 
1.7.9.5

[PATCH V2 3/8] memory: emif: handle overflow for timing for LP mode

[Lokesh Vutla]

From: Nishanth Menon <nm@ti.com>

In case the custom timings provide values which overflow
the maximum possible field value, warn and use maximum
permissible value.

Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Changes since V1: No changes

 drivers/memory/emif.c |   36 ++++++++++++++++++++++++++++--------
 1 file changed, 28 insertions(+), 8 deletions(-)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index 5f3b7ed..e513889 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -715,6 +715,8 @@ static u32 get_pwr_mgmt_ctrl(u32 freq, struct emif_data *emif, u32 ip_rev)
 	u32 timeout_perf	= EMIF_LP_MODE_TIMEOUT_PERFORMANCE;
 	u32 timeout_pwr		= EMIF_LP_MODE_TIMEOUT_POWER;
 	u32 freq_threshold	= EMIF_LP_MODE_FREQ_THRESHOLD;
+	u32 mask;
+	u8 shift;
 
 	struct emif_custom_configs *cust_cfgs = emif->plat_data->custom_configs;
 
@@ -743,27 +745,45 @@ static u32 get_pwr_mgmt_ctrl(u32 freq, struct emif_data *emif, u32 ip_rev)
 
 	switch (lpmode) {
 	case EMIF_LP_MODE_CLOCK_STOP:
-		pwr_mgmt_ctrl = (timeout << CS_TIM_SHIFT) |
-					SR_TIM_MASK | PD_TIM_MASK;
+		shift = CS_TIM_SHIFT;
+		mask = CS_TIM_MASK;
 		break;
 	case EMIF_LP_MODE_SELF_REFRESH:
 		/* Workaround for errata i735 */
 		if (timeout < 6)
 			timeout = 6;
 
-		pwr_mgmt_ctrl = (timeout << SR_TIM_SHIFT) |
-					CS_TIM_MASK | PD_TIM_MASK;
+		shift = SR_TIM_SHIFT;
+		mask = SR_TIM_MASK;
 		break;
 	case EMIF_LP_MODE_PWR_DN:
-		pwr_mgmt_ctrl = (timeout << PD_TIM_SHIFT) |
-					CS_TIM_MASK | SR_TIM_MASK;
+		shift = PD_TIM_SHIFT;
+		mask = PD_TIM_MASK;
 		break;
 	case EMIF_LP_MODE_DISABLE:
 	default:
-		pwr_mgmt_ctrl = CS_TIM_MASK |
-					PD_TIM_MASK | SR_TIM_MASK;
+		mask = 0;
+		shift = 0;
+		break;
+	}
+	/* Round to maximum in case of overflow, BUT warn! */
+	if (lpmode != EMIF_LP_MODE_DISABLE && timeout > mask >> shift) {
+		pr_err("TIMEOUT Overflow - lpmode=%d perf=%d pwr=%d freq=%d\n",
+		       lpmode,
+		       timeout_perf,
+		       timeout_pwr,
+		       freq_threshold);
+		WARN(1, "timeout=0x%02x greater than 0x%02x. Using max\n",
+		     timeout, mask >> shift);
+		timeout = mask >> shift;
 	}
 
+	/* Setup required timing */
+	pwr_mgmt_ctrl = (timeout << shift) & mask;
+	/* setup a default mask for rest of the modes */
+	pwr_mgmt_ctrl |= (SR_TIM_MASK | CS_TIM_MASK | PD_TIM_MASK) &
+			  ~mask;
+
 	/* No CS_TIM in EMIF_4D5 */
 	if (ip_rev == EMIF_4D5)
 		pwr_mgmt_ctrl &= ~CS_TIM_MASK;
-- 
1.7.9.5

[PATCH V2 4/8] memory: emif: Handle devices which are not rated for >85C

[Lokesh Vutla]

From: Nishanth Menon <nm@ti.com>

As per JESD209-2E specification for LPDDR2,
      http://www.jedec.org/standards-documents/results/jesd209-2E
Table 73, LPDDR2 memories come in two flavors - Standard and
Extended. The Standard types can operate from -25C to +85C
However, beyond that and upto +105C can only be supported by
Extended types.

Unfortunately, it seems there is no info in MR0(device info) or
MR[1,2](device feature) for run time detection of this capability
as far as seen on the spec. Hence, we provide a custom_config
flag to be populated by platforms which have these "extended"
type memories.

For the "Standard" memories, we need to consider MR4 notifications
of temperature triggers >85C as equivalent to thermal shutdown
events (equivalent to Spec specified thermal shutdown events for
"extended" parts).

Reported-by: Richard Woodruff <r-woodruff2@ti.com>
Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Changes since V1: No changes

 drivers/memory/emif.c                   |   27 +++++++++++++++++++++++++++
 include/linux/platform_data/emif_plat.h |    1 +
 2 files changed, 28 insertions(+)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index e513889..827eb3b 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -918,6 +918,7 @@ static irqreturn_t handle_temp_alert(void __iomem *base, struct emif_data *emif)
 {
 	u32		old_temp_level;
 	irqreturn_t	ret = IRQ_HANDLED;
+	struct emif_custom_configs *custom_configs;
 
 	spin_lock_irqsave(&emif_lock, irq_state);
 	old_temp_level = emif->temperature_level;
@@ -930,6 +931,29 @@ static irqreturn_t handle_temp_alert(void __iomem *base, struct emif_data *emif)
 		goto out;
 	}
 
+	custom_configs = emif->plat_data->custom_configs;
+
+	/*
+	 * IF we detect higher than "nominal rating" from DDR sensor
+	 * on an unsupported DDR part, shutdown system
+	 */
+	if (custom_configs && !(custom_configs->mask &
+				EMIF_CUSTOM_CONFIG_EXTENDED_TEMP_PART)) {
+		if (emif->temperature_level >= SDRAM_TEMP_HIGH_DERATE_REFRESH) {
+			dev_err(emif->dev,
+				"%s:NOT Extended temperature capable memory."
+				"Converting MR4=0x%02x as shutdown event\n",
+				__func__, emif->temperature_level);
+			/*
+			 * Temperature far too high - do kernel_power_off()
+			 * from thread context
+			 */
+			emif->temperature_level = SDRAM_TEMP_VERY_HIGH_SHUTDOWN;
+			ret = IRQ_WAKE_THREAD;
+			goto out;
+		}
+	}
+
 	if (emif->temperature_level < old_temp_level ||
 		emif->temperature_level == SDRAM_TEMP_VERY_HIGH_SHUTDOWN) {
 		/*
@@ -1228,6 +1252,9 @@ static void __init_or_module of_get_custom_configs(struct device_node *np_emif,
 		cust_cfgs->temp_alert_poll_interval_ms = *poll_intvl;
 	}
 
+	if (of_find_property(np_emif, "extended-temp-part", &len))
+		cust_cfgs->mask |= EMIF_CUSTOM_CONFIG_EXTENDED_TEMP_PART;
+
 	if (!is_custom_config_valid(cust_cfgs, emif->dev)) {
 		devm_kfree(emif->dev, cust_cfgs);
 		return;
diff --git a/include/linux/platform_data/emif_plat.h b/include/linux/platform_data/emif_plat.h
index 03378ca..5c19a2a 100644
--- a/include/linux/platform_data/emif_plat.h
+++ b/include/linux/platform_data/emif_plat.h
@@ -40,6 +40,7 @@
 /* Custom config requests */
 #define EMIF_CUSTOM_CONFIG_LPMODE			0x00000001
 #define EMIF_CUSTOM_CONFIG_TEMP_ALERT_POLL_INTERVAL	0x00000002
+#define EMIF_CUSTOM_CONFIG_EXTENDED_TEMP_PART		0x00000004
 
 #ifndef __ASSEMBLY__
 /**
-- 
1.7.9.5

[PATCH V2 5/8] memory: emif: use restart if power_off not present when out of spec

[Lokesh Vutla]

From: Nishanth Menon <nm@ti.com>

Some machine or kernel variants might have missed implementation
of power off handlers. We DONOT want to let the system be in
"out of spec" state in this condition. So, WARN and attempt
a machine restart in the hopes of clearing the out-of-spec
temperature condition.

NOTE: This is not the safest option, but safer than leaving the
system in unstable conditions.

Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Changes since V1:
	Added a new extra line before if condition.

 drivers/memory/emif.c |   10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index 827eb3b..8db74e4 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -25,6 +25,7 @@
 #include <linux/module.h>
 #include <linux/list.h>
 #include <linux/spinlock.h>
+#include <linux/pm.h>
 #include <memory/jedec_ddr.h>
 #include "emif.h"
 #include "of_memory.h"
@@ -1015,7 +1016,14 @@ static irqreturn_t emif_threaded_isr(int irq, void *dev_id)
 
 	if (emif->temperature_level == SDRAM_TEMP_VERY_HIGH_SHUTDOWN) {
 		dev_emerg(emif->dev, "SDRAM temperature exceeds operating limit.. Needs shut down!!!\n");
-		kernel_power_off();
+
+		/* If we have Power OFF ability, use it, else try restarting */
+		if (pm_power_off) {
+			kernel_power_off();
+		} else {
+			WARN(1, "FIXME: NO pm_power_off!!! trying restart\n");
+			kernel_restart("SDRAM Over-temp Emergency restart");
+		}
 		return IRQ_HANDLED;
 	}
 
-- 
1.7.9.5

[PATCH V2 6/8] memory: emif: Fix the incorrect 'size' parameter in memcpy

[Lokesh Vutla]

From: Oleksandr Dmytryshyn <oleksandr.dmytryshyn@ti.com>

The issue was that only the first timings table was added to the
emif platform data at the emif driver registration. All other
timings tables was filled with zeros. Now all emif timings table
are added to the platform data.

Signed-off-by: Oleksandr Dmytryshyn <oleksandr.dmytryshyn@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Changes since V1: Updated $subject

 drivers/memory/emif.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index 8db74e4..d27111e 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -1468,7 +1468,7 @@ static struct emif_data *__init_or_module get_device_details(
 	if (pd->timings) {
 		temp = devm_kzalloc(dev, size, GFP_KERNEL);
 		if (temp) {
-			memcpy(temp, pd->timings, sizeof(*pd->timings));
+			memcpy(temp, pd->timings, size);
 			pd->timings = temp;
 		} else {
 			dev_warn(dev, "%s:%d: allocation error\n", __func__,
-- 
1.7.9.5

[PATCH V2 7/8] memory: emif: errata i743: Prohibit usage of Power-Down mode

[Lokesh Vutla]

From: Grygorii Strashko <grygorii.strashko@ti.com>

ERRATA DESCRIPTION :
The EMIF supports power-down state for low power. The EMIF
automatically puts the SDRAM into power-down after the memory is
not accessed for a defined number of cycles and the
EMIF_PWR_MGMT_CTRL[10:8] REG_LP_MODE bit field is set to 0x4.
As the EMIF supports automatic output impedance calibration, a ZQ
calibration long command is issued every time it exits active
power-down and precharge power-down modes. The EMIF waits and
blocks any other command during this calibration.
The EMIF does not allow selective disabling of ZQ calibration upon
exit of power-down mode. Due to very short periods of power-down
cycles, ZQ calibration overhead creates bandwidth issues and
increases overall system power consumption. On the other hand,
issuing ZQ calibration long commands when exiting self-refresh is
still required.

WORKAROUND :
Because there is no power consumption benefit of the power-down due
to the calibration and there is a performance risk, the guideline
is to not allow power-down state and, therefore, to not have set
the EMIF_PWR_MGMT_CTRL[10:8] REG_LP_MODE bit field to 0x4.

This is applicable only for EMIF4D IP used in OMAP4 Soc's.

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Vitaly Chernooky <vitaly.chernooky@ti.com>
Signed-off-by: Oleksandr Dmytryshyn <oleksandr.dmytryshyn@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Changes since V1: 
	Updated the change log and fixed a typo in a comment.

 drivers/memory/emif.c |   35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index d27111e..9b27c2b 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -257,6 +257,41 @@ static void set_lpmode(struct emif_data *emif, u8 lpmode)
 	u32 temp;
 	void __iomem *base = emif->base;
 
+	/*
+	 * Workaround for errata i743 - LPDDR2 Power-Down State is Not
+	 * Efficient
+	 *
+	 * i743 DESCRIPTION:
+	 * The EMIF supports power-down state for low power. The EMIF
+	 * automatically puts the SDRAM into power-down after the memory is
+	 * not accessed for a defined number of cycles and the
+	 * EMIF_PWR_MGMT_CTRL[10:8] REG_LP_MODE bit field is set to 0x4.
+	 * As the EMIF supports automatic output impedance calibration, a ZQ
+	 * calibration long command is issued every time it exits active
+	 * power-down and precharge power-down modes. The EMIF waits and
+	 * blocks any other command during this calibration.
+	 * The EMIF does not allow selective disabling of ZQ calibration upon
+	 * exit of power-down mode. Due to very short periods of power-down
+	 * cycles, ZQ calibration overhead creates bandwidth issues and
+	 * increases overall system power consumption. On the other hand,
+	 * issuing ZQ calibration long commands when exiting self-refresh is
+	 * still required.
+	 *
+	 * WORKAROUND
+	 * Because there is no power consumption benefit of the power-down due
+	 * to the calibration and there is a performance risk, the guideline
+	 * is to not allow power-down state and, therefore, to not have set
+	 * the EMIF_PWR_MGMT_CTRL[10:8] REG_LP_MODE bit field to 0x4.
+	 */
+	if ((emif->plat_data->ip_rev == EMIF_4D) &&
+	    (EMIF_LP_MODE_PWR_DN == lpmode)) {
+		WARN_ONCE(1,
+			  "REG_LP_MODE = LP_MODE_PWR_DN(4) is prohibited by"
+			  "erratum i743 switch to LP_MODE_SELF_REFRESH(2)\n");
+		/* rollback LP_MODE to Self-refresh mode */
+		lpmode = EMIF_LP_MODE_SELF_REFRESH;
+	}
+
 	temp = readl(base + EMIF_POWER_MANAGEMENT_CONTROL);
 	temp &= ~LP_MODE_MASK;
 	temp |= (lpmode << LP_MODE_SHIFT);
-- 
1.7.9.5

[PATCH V2 8/8] memory: emif: Load the correct custom config values from dt

[Lokesh Vutla]

of_get_property returns value in Big Endian format.
Before using this value it should be converted to little endian
using be32_to_cpup().
Custom configs of emif are read from dt using of_get_property,
but these are not converted to litte endian format.
Correcting the same here.

Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Changes since V1: No changes

 drivers/memory/emif.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/memory/emif.c b/drivers/memory/emif.c
index 9b27c2b..475174a 100644
--- a/drivers/memory/emif.c
+++ b/drivers/memory/emif.c
@@ -1263,7 +1263,7 @@ static void __init_or_module of_get_custom_configs(struct device_node *np_emif,
 {
 	struct emif_custom_configs	*cust_cfgs = NULL;
 	int				len;
-	const int			*lpmode, *poll_intvl;
+	const __be32			*lpmode, *poll_intvl;
 
 	lpmode = of_get_property(np_emif, "low-power-mode", &len);
 	poll_intvl = of_get_property(np_emif, "temp-alert-poll-interval", &len);
@@ -1277,7 +1277,7 @@ static void __init_or_module of_get_custom_configs(struct device_node *np_emif,
 
 	if (lpmode) {
 		cust_cfgs->mask |= EMIF_CUSTOM_CONFIG_LPMODE;
-		cust_cfgs->lpmode = *lpmode;
+		cust_cfgs->lpmode = be32_to_cpup(lpmode);
 		of_property_read_u32(np_emif,
 				"low-power-mode-timeout-performance",
 				&cust_cfgs->lpmode_timeout_performance);
@@ -1292,7 +1292,8 @@ static void __init_or_module of_get_custom_configs(struct device_node *np_emif,
 	if (poll_intvl) {
 		cust_cfgs->mask |=
 				EMIF_CUSTOM_CONFIG_TEMP_ALERT_POLL_INTERVAL;
-		cust_cfgs->temp_alert_poll_interval_ms = *poll_intvl;
+		cust_cfgs->temp_alert_poll_interval_ms =
+						be32_to_cpup(poll_intvl);
 	}
 
 	if (of_find_property(np_emif, "extended-temp-part", &len))
-- 
1.7.9.5