Merge branch 'master' of git://git.denx.de/u-boot

Signed-off-by: Stefano Babic <sbabic@denx.de>

Conflicts:
	boards.cfg

8 files changed, 350 insertions, 138 deletions

diff --git a/arch/arm/cpu/armv7/exynos/clock.c b/arch/arm/cpu/armv7/exynos/clock.c
index 1fea4d6663..400d134d54 100644
--- a/arch/arm/cpu/armv7/exynos/clock.c
+++ b/arch/arm/cpu/armv7/exynos/clock.c
@@ -869,7 +869,7 @@ static void exynos4_set_mmc_clk(int dev_index, unsigned int div)
 {
 	struct exynos4_clock *clk =
 		(struct exynos4_clock *)samsung_get_base_clock();
-	unsigned int addr;
+	unsigned int addr, clear_bit, set_bit;
 
 	/*
 	 * CLK_DIV_FSYS1
@@ -877,44 +877,26 @@ static void exynos4_set_mmc_clk(int dev_index, unsigned int div)
 	 * CLK_DIV_FSYS2
 	 * MMC2_PRE_RATIO [15:8], MMC3_PRE_RATIO [31:24]
 	 * CLK_DIV_FSYS3
-	 * MMC4_PRE_RATIO [15:8]
+	 * MMC4_RATIO [3:0]
 	 */
 	if (dev_index < 2) {
 		addr = (unsigned int)&clk->div_fsys1;
-	}  else if (dev_index == 4) {
+		clear_bit = MASK_PRE_RATIO(dev_index);
+		set_bit = SET_PRE_RATIO(dev_index, div);
+	} else if (dev_index == 4) {
 		addr = (unsigned int)&clk->div_fsys3;
 		dev_index -= 4;
+		/* MMC4 is controlled with the MMC4_RATIO value */
+		clear_bit = MASK_RATIO(dev_index);
+		set_bit = SET_RATIO(dev_index, div);
 	} else {
 		addr = (unsigned int)&clk->div_fsys2;
 		dev_index -= 2;
+		clear_bit = MASK_PRE_RATIO(dev_index);
+		set_bit = SET_PRE_RATIO(dev_index, div);
 	}
 
-	clrsetbits_le32(addr, 0xff << ((dev_index << 4) + 8),
-			(div & 0xff) << ((dev_index << 4) + 8));
-}
-
-/* exynos4x12: set the mmc clock */
-static void exynos4x12_set_mmc_clk(int dev_index, unsigned int div)
-{
-	struct exynos4x12_clock *clk =
-		(struct exynos4x12_clock *)samsung_get_base_clock();
-	unsigned int addr;
-
-	/*
-	 * CLK_DIV_FSYS1
-	 * MMC0_PRE_RATIO [15:8], MMC1_PRE_RATIO [31:24]
-	 * CLK_DIV_FSYS2
-	 * MMC2_PRE_RATIO [15:8], MMC3_PRE_RATIO [31:24]
-	 */
-	if (dev_index < 2) {
-		addr = (unsigned int)&clk->div_fsys1;
-	} else {
-		addr = (unsigned int)&clk->div_fsys2;
-		dev_index -= 2;
-	}
-
-	clrsetbits_le32(addr, 0xff << ((dev_index << 4) + 8),
-			(div & 0xff) << ((dev_index << 4) + 8));
+	clrsetbits_le32(addr, clear_bit, set_bit);
 }
 
 /* exynos5: set the mmc clock */
@@ -1612,10 +1594,7 @@ void set_mmc_clk(int dev_index, unsigned int div)
 		else
 			exynos5_set_mmc_clk(dev_index, div);
 	} else {
-		if (proid_is_exynos4412())
-			exynos4x12_set_mmc_clk(dev_index, div);
-		else
-			exynos4_set_mmc_clk(dev_index, div);
+		exynos4_set_mmc_clk(dev_index, div);
 	}
 }
 
diff --git a/arch/arm/cpu/armv7/exynos/dmc_common.c b/arch/arm/cpu/armv7/exynos/dmc_common.c
index cca925e42c..9b6ee69f46 100644
--- a/arch/arm/cpu/armv7/exynos/dmc_common.c
+++ b/arch/arm/cpu/armv7/exynos/dmc_common.c
@@ -162,7 +162,7 @@ void mem_ctrl_init(int reset)
 
 	/* If there are any other memory variant, add their init call below */
 	if (param->mem_type == DDR_MODE_DDR3) {
-		ret = ddr3_mem_ctrl_init(mem, param->mem_iv_size, reset);
+		ret = ddr3_mem_ctrl_init(mem, reset);
 		if (ret) {
 			/* will hang if failed to init memory control */
 			while (1)
diff --git a/arch/arm/cpu/armv7/exynos/dmc_init_ddr3.c b/arch/arm/cpu/armv7/exynos/dmc_init_ddr3.c
index 487e6f423f..b86dd2d650 100644
--- a/arch/arm/cpu/armv7/exynos/dmc_init_ddr3.c
+++ b/arch/arm/cpu/armv7/exynos/dmc_init_ddr3.c
@@ -6,6 +6,7 @@
  * SPDX-License-Identifier:	GPL-2.0+
  */
 
+#include <common.h>
 #include <config.h>
 #include <asm/io.h>
 #include <asm/arch/clock.h>
@@ -16,7 +17,11 @@
 #include "exynos5_setup.h"
 #include "clock_init.h"
 
-#define TIMEOUT	10000
+#define TIMEOUT_US		10000
+#define NUM_BYTE_LANES		4
+#define DEFAULT_DQS		8
+#define DEFAULT_DQS_X4		(DEFAULT_DQS << 24) || (DEFAULT_DQS << 16) \
+				|| (DEFAULT_DQS << 8) || (DEFAULT_DQS << 0)
 
 #ifdef CONFIG_EXYNOS5250
 static void reset_phy_ctrl(void)
@@ -28,8 +33,7 @@ static void reset_phy_ctrl(void)
 	writel(DDR3PHY_CTRL_PHY_RESET, &clk->lpddr3phy_ctrl);
 }
 
-int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
-		       int reset)
+int ddr3_mem_ctrl_init(struct mem_timings *mem, int reset)
 {
 	unsigned int val;
 	struct exynos5_phy_control *phy0_ctrl, *phy1_ctrl;
@@ -177,7 +181,7 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 		writel(val, &phy1_ctrl->phy_con1);
 
 		writel(CTRL_RDLVL_GATE_ENABLE, &dmc->rdlvl_config);
-		i = TIMEOUT;
+		i = TIMEOUT_US;
 		while ((readl(&dmc->phystatus) &
 			(RDLVL_COMPLETE_CHO | RDLVL_COMPLETE_CH1)) !=
 			(RDLVL_COMPLETE_CHO | RDLVL_COMPLETE_CH1) && i > 0) {
@@ -221,8 +225,220 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 #endif
 
 #ifdef CONFIG_EXYNOS5420
-int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
-		       int reset)
+/**
+ * RAM address to use in the test.
+ *
+ * We'll use 4 words at this address and 4 at this address + 0x80 (Ares
+ * interleaves channels every 128 bytes).  This will allow us to evaluate all of
+ * the chips in a 1 chip per channel (2GB) system and half the chips in a 2
+ * chip per channel (4GB) system.  We can't test the 2nd chip since we need to
+ * do tests before the 2nd chip is enabled.  Looking at the 2nd chip isn't
+ * critical because the 1st and 2nd chip have very similar timings (they'd
+ * better have similar timings, since there's only a single adjustment that is
+ * shared by both chips).
+ */
+const unsigned int test_addr = CONFIG_SYS_SDRAM_BASE;
+
+/* Test pattern with which RAM will be tested */
+static const unsigned int test_pattern[] = {
+	0x5a5a5a5a,
+	0xa5a5a5a5,
+	0xf0f0f0f0,
+	0x0f0f0f0f,
+};
+
+/**
+ * This function is a test vector for sw read leveling,
+ * it compares the read data with the written data.
+ *
+ * @param ch			DMC channel number
+ * @param byte_lane		which DQS byte offset,
+ *				possible values are 0,1,2,3
+ * @return			TRUE if memory was good, FALSE if not.
+ */
+static bool dmc_valid_window_test_vector(int ch, int byte_lane)
+{
+	unsigned int read_data;
+	unsigned int mask;
+	int i;
+
+	mask = 0xFF << (8 * byte_lane);
+
+	for (i = 0; i < ARRAY_SIZE(test_pattern); i++) {
+		read_data = readl(test_addr + i * 4 + ch * 0x80);
+		if ((read_data & mask) != (test_pattern[i] & mask))
+			return false;
+	}
+
+	return true;
+}
+
+/**
+ * This function returns current read offset value.
+ *
+ * @param phy_ctrl	pointer to the current phy controller
+ */
+static unsigned int dmc_get_read_offset_value(struct exynos5420_phy_control
+					       *phy_ctrl)
+{
+	return readl(&phy_ctrl->phy_con4);
+}
+
+/**
+ * This function performs resync, so that slave DLL is updated.
+ *
+ * @param phy_ctrl	pointer to the current phy controller
+ */
+static void ddr_phy_set_do_resync(struct exynos5420_phy_control *phy_ctrl)
+{
+	setbits_le32(&phy_ctrl->phy_con10, PHY_CON10_CTRL_OFFSETR3);
+	clrbits_le32(&phy_ctrl->phy_con10, PHY_CON10_CTRL_OFFSETR3);
+}
+
+/**
+ * This function sets read offset value register with 'offset'.
+ *
+ * ...we also call call ddr_phy_set_do_resync().
+ *
+ * @param phy_ctrl	pointer to the current phy controller
+ * @param offset	offset to read DQS
+ */
+static void dmc_set_read_offset_value(struct exynos5420_phy_control *phy_ctrl,
+				      unsigned int offset)
+{
+	writel(offset, &phy_ctrl->phy_con4);
+	ddr_phy_set_do_resync(phy_ctrl);
+}
+
+/**
+ * Convert a 2s complement byte to a byte with a sign bit.
+ *
+ * NOTE: you shouldn't use normal math on the number returned by this function.
+ *   As an example, -10 = 0xf6.  After this function -10 = 0x8a.  If you wanted
+ *   to do math and get the average of 10 and -10 (should be 0):
+ *     0x8a + 0xa = 0x94 (-108)
+ *     0x94 / 2   = 0xca (-54)
+ *   ...and 0xca = sign bit plus 0x4a, or -74
+ *
+ * Also note that you lose the ability to represent -128 since there are two
+ * representations of 0.
+ *
+ * @param b	The byte to convert in two's complement.
+ * @return	The 7-bit value + sign bit.
+ */
+
+unsigned char make_signed_byte(signed char b)
+{
+	if (b < 0)
+		return 0x80 | -b;
+	else
+		return b;
+}
+
+/**
+ * Test various shifts starting at 'start' and going to 'end'.
+ *
+ * For each byte lane, we'll walk through shift starting at 'start' and going
+ * to 'end' (inclusive).  When we are finally able to read the test pattern
+ * we'll store the value in the results array.
+ *
+ * @param phy_ctrl		pointer to the current phy controller
+ * @param ch			channel number
+ * @param start			the start shift.  -127 to 127
+ * @param end			the end shift.  -127 to 127
+ * @param results		we'll store results for each byte lane.
+ */
+
+void test_shifts(struct exynos5420_phy_control *phy_ctrl, int ch,
+		 int start, int end, int results[NUM_BYTE_LANES])
+{
+	int incr = (start < end) ? 1 : -1;
+	int byte_lane;
+
+	for (byte_lane = 0; byte_lane < NUM_BYTE_LANES; byte_lane++) {
+		int shift;
+
+		dmc_set_read_offset_value(phy_ctrl, DEFAULT_DQS_X4);
+		results[byte_lane] = DEFAULT_DQS;
+
+		for (shift = start; shift != (end + incr); shift += incr) {
+			unsigned int byte_offsetr;
+			unsigned int offsetr;
+
+			byte_offsetr = make_signed_byte(shift);
+
+			offsetr = dmc_get_read_offset_value(phy_ctrl);
+			offsetr &= ~(0xFF << (8 * byte_lane));
+			offsetr |= (byte_offsetr << (8 * byte_lane));
+			dmc_set_read_offset_value(phy_ctrl, offsetr);
+
+			if (dmc_valid_window_test_vector(ch, byte_lane)) {
+				results[byte_lane] = shift;
+				break;
+			}
+		}
+	}
+}
+
+/**
+ * This function performs SW read leveling to compensate DQ-DQS skew at
+ * receiver it first finds the optimal read offset value on each DQS
+ * then applies the value to PHY.
+ *
+ * Read offset value has its min margin and max margin. If read offset
+ * value exceeds its min or max margin, read data will have corruption.
+ * To avoid this we are doing sw read leveling.
+ *
+ * SW read leveling is:
+ * 1> Finding offset value's left_limit and right_limit
+ * 2> and calculate its center value
+ * 3> finally programs that center value to PHY
+ * 4> then PHY gets its optimal offset value.
+ *
+ * @param phy_ctrl		pointer to the current phy controller
+ * @param ch			channel number
+ * @param coarse_lock_val	The coarse lock value read from PHY_CON13.
+ *				(0 - 0x7f)
+ */
+static void software_find_read_offset(struct exynos5420_phy_control *phy_ctrl,
+				      int ch, unsigned int coarse_lock_val)
+{
+	unsigned int offsetr_cent;
+	int byte_lane;
+	int left_limit;
+	int right_limit;
+	int left[NUM_BYTE_LANES];
+	int right[NUM_BYTE_LANES];
+	int i;
+
+	/* Fill the memory with test patterns */
+	for (i = 0; i < ARRAY_SIZE(test_pattern); i++)
+		writel(test_pattern[i], test_addr + i * 4 + ch * 0x80);
+
+	/* Figure out the limits we'll test with; keep -127 < limit < 127 */
+	left_limit = DEFAULT_DQS - coarse_lock_val;
+	right_limit = DEFAULT_DQS + coarse_lock_val;
+	if (right_limit > 127)
+		right_limit = 127;
+
+	/* Fill in the location where reads were OK from left and right */
+	test_shifts(phy_ctrl, ch, left_limit, right_limit, left);
+	test_shifts(phy_ctrl, ch, right_limit, left_limit, right);
+
+	/* Make a final value by taking the center between the left and right */
+	offsetr_cent = 0;
+	for (byte_lane = 0; byte_lane < NUM_BYTE_LANES; byte_lane++) {
+		int temp_center;
+		unsigned int vmwc;
+
+		temp_center = (left[byte_lane] + right[byte_lane]) / 2;
+		vmwc = make_signed_byte(temp_center);
+		offsetr_cent |= vmwc << (8 * byte_lane);
+	}
+	dmc_set_read_offset_value(phy_ctrl, offsetr_cent);
+}
+
+int ddr3_mem_ctrl_init(struct mem_timings *mem, int reset)
 {
 	struct exynos5420_clock *clk =
 		(struct exynos5420_clock *)samsung_get_base_clock();
@@ -231,7 +447,9 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 	struct exynos5420_phy_control *phy0_ctrl, *phy1_ctrl;
 	struct exynos5420_dmc *drex0, *drex1;
 	struct exynos5420_tzasc *tzasc0, *tzasc1;
+	struct exynos5_power *pmu;
 	uint32_t val, n_lock_r, n_lock_w_phy0, n_lock_w_phy1;
+	uint32_t lock0_info, lock1_info;
 	int chip;
 	int i;
 
@@ -244,6 +462,7 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 	tzasc0 = (struct exynos5420_tzasc *)samsung_get_base_dmc_tzasc();
 	tzasc1 = (struct exynos5420_tzasc *)(samsung_get_base_dmc_tzasc()
 							+ DMC_OFFSET);
+	pmu = (struct exynos5_power *)EXYNOS5420_POWER_BASE;
 
 	/* Enable PAUSE for DREX */
 	setbits_le32(&clk->pause, ENABLE_BIT);
@@ -394,7 +613,41 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 		 */
 		dmc_config_mrs(mem, &drex0->directcmd);
 		dmc_config_mrs(mem, &drex1->directcmd);
-	} else {
+	}
+
+	/*
+	 * Get PHY_CON13 from both phys.  Gate CLKM around reading since
+	 * PHY_CON13 is glitchy when CLKM is running.  We're paranoid and
+	 * wait until we get a "fine lock", though a coarse lock is probably
+	 * OK (we only use the coarse numbers below).  We try to gate the
+	 * clock for as short a time as possible in case SDRAM is somehow
+	 * sensitive.  sdelay(10) in the loop is arbitrary to make sure
+	 * there is some time for PHY_CON13 to get updated.  In practice
+	 * no delay appears to be needed.
+	 */
+	val = readl(&clk->gate_bus_cdrex);
+	while (true) {
+		writel(val & ~0x1, &clk->gate_bus_cdrex);
+		lock0_info = readl(&phy0_ctrl->phy_con13);
+		writel(val, &clk->gate_bus_cdrex);
+
+		if ((lock0_info & CTRL_FINE_LOCKED) == CTRL_FINE_LOCKED)
+			break;
+
+		sdelay(10);
+	}
+	while (true) {
+		writel(val & ~0x2, &clk->gate_bus_cdrex);
+		lock1_info = readl(&phy1_ctrl->phy_con13);
+		writel(val, &clk->gate_bus_cdrex);
+
+		if ((lock1_info & CTRL_FINE_LOCKED) == CTRL_FINE_LOCKED)
+			break;
+
+		sdelay(10);
+	}
+
+	if (!reset) {
 		/*
 		 * During Suspend-Resume & S/W-Reset, as soon as PMU releases
 		 * pad retention, CKE goes high. This causes memory contents
@@ -445,15 +698,13 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 		val |= (RDLVL_PASS_ADJ_VAL << RDLVL_PASS_ADJ_OFFSET);
 		writel(val, &phy1_ctrl->phy_con1);
 
-		n_lock_r = readl(&phy0_ctrl->phy_con13);
-		n_lock_w_phy0 = (n_lock_r & CTRL_LOCK_COARSE_MASK) >> 2;
+		n_lock_w_phy0 = (lock0_info & CTRL_LOCK_COARSE_MASK) >> 2;
 		n_lock_r = readl(&phy0_ctrl->phy_con12);
 		n_lock_r &= ~CTRL_DLL_ON;
 		n_lock_r |= n_lock_w_phy0;
 		writel(n_lock_r, &phy0_ctrl->phy_con12);
 
-		n_lock_r = readl(&phy1_ctrl->phy_con13);
-		n_lock_w_phy1 = (n_lock_r & CTRL_LOCK_COARSE_MASK) >> 2;
+		n_lock_w_phy1 = (lock1_info & CTRL_LOCK_COARSE_MASK) >> 2;
 		n_lock_r = readl(&phy1_ctrl->phy_con12);
 		n_lock_r &= ~CTRL_DLL_ON;
 		n_lock_r |= n_lock_w_phy1;
@@ -482,7 +733,7 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 		writel(val, &phy1_ctrl->phy_con1);
 
 		writel(CTRL_RDLVL_GATE_ENABLE, &drex0->rdlvl_config);
-		i = TIMEOUT;
+		i = TIMEOUT_US;
 		while (((readl(&drex0->phystatus) & RDLVL_COMPLETE_CHO) !=
 			RDLVL_COMPLETE_CHO) && (i > 0)) {
 			/*
@@ -497,7 +748,7 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 		writel(CTRL_RDLVL_GATE_DISABLE, &drex0->rdlvl_config);
 
 		writel(CTRL_RDLVL_GATE_ENABLE, &drex1->rdlvl_config);
-		i = TIMEOUT;
+		i = TIMEOUT_US;
 		while (((readl(&drex1->phystatus) & RDLVL_COMPLETE_CHO) !=
 			RDLVL_COMPLETE_CHO) && (i > 0)) {
 			/*
@@ -522,77 +773,6 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 			       &drex1->directcmd);
 		}
 
-		if (mem->read_leveling_enable) {
-			/* Set Read DQ Calibration */
-			val = (0x3 << DIRECT_CMD_BANK_SHIFT) | 0x4;
-			for (chip = 0; chip < mem->chips_to_configure; chip++) {
-				writel(val | (chip << DIRECT_CMD_CHIP_SHIFT),
-				       &drex0->directcmd);
-				writel(val | (chip << DIRECT_CMD_CHIP_SHIFT),
-				       &drex1->directcmd);
-			}
-
-			val = readl(&phy0_ctrl->phy_con1);
-			val |= READ_LEVELLING_DDR3;
-			writel(val, &phy0_ctrl->phy_con1);
-			val = readl(&phy1_ctrl->phy_con1);
-			val |= READ_LEVELLING_DDR3;
-			writel(val, &phy1_ctrl->phy_con1);
-
-			val = readl(&phy0_ctrl->phy_con2);
-			val |= (RDLVL_EN | RDLVL_INCR_ADJ);
-			writel(val, &phy0_ctrl->phy_con2);
-			val = readl(&phy1_ctrl->phy_con2);
-			val |= (RDLVL_EN | RDLVL_INCR_ADJ);
-			writel(val, &phy1_ctrl->phy_con2);
-
-			setbits_le32(&drex0->rdlvl_config,
-				     CTRL_RDLVL_DATA_ENABLE);
-			i = TIMEOUT;
-			while (((readl(&drex0->phystatus) & RDLVL_COMPLETE_CHO)
-				 != RDLVL_COMPLETE_CHO) && (i > 0)) {
-				/*
-				 * TODO(waihong): Comment on how long this take
-				 * to timeout
-				 */
-				sdelay(100);
-				i--;
-			}
-			if (!i)
-				return SETUP_ERR_RDLV_COMPLETE_TIMEOUT;
-
-			clrbits_le32(&drex0->rdlvl_config,
-				     CTRL_RDLVL_DATA_ENABLE);
-			setbits_le32(&drex1->rdlvl_config,
-				     CTRL_RDLVL_DATA_ENABLE);
-			i = TIMEOUT;
-			while (((readl(&drex1->phystatus) & RDLVL_COMPLETE_CHO)
-				 != RDLVL_COMPLETE_CHO) && (i > 0)) {
-				/*
-				 * TODO(waihong): Comment on how long this take
-				 * to timeout
-				 */
-				sdelay(100);
-				i--;
-			}
-			if (!i)
-				return SETUP_ERR_RDLV_COMPLETE_TIMEOUT;
-
-			clrbits_le32(&drex1->rdlvl_config,
-				     CTRL_RDLVL_DATA_ENABLE);
-
-			val = (0x3 << DIRECT_CMD_BANK_SHIFT);
-			for (chip = 0; chip < mem->chips_to_configure; chip++) {
-				writel(val | (chip << DIRECT_CMD_CHIP_SHIFT),
-				       &drex0->directcmd);
-				writel(val | (chip << DIRECT_CMD_CHIP_SHIFT),
-				       &drex1->directcmd);
-			}
-
-			update_reset_dll(&drex0->phycontrol0, DDR_MODE_DDR3);
-			update_reset_dll(&drex1->phycontrol0, DDR_MODE_DDR3);
-		}
-
 		/* Common Settings for Leveling */
 		val = PHY_CON12_RESET_VAL;
 		writel((val + n_lock_w_phy0), &phy0_ctrl->phy_con12);
@@ -602,6 +782,27 @@ int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
 		setbits_le32(&phy1_ctrl->phy_con2, DLL_DESKEW_EN);
 	}
 
+	/*
+	 * Do software read leveling
+	 *
+	 * Do this before we turn on auto refresh since the auto refresh can
+	 * be in conflict with the resync operation that's part of setting
+	 * read leveling.
+	 */
+	if (!reset) {
+		/* restore calibrated value after resume */
+		dmc_set_read_offset_value(phy0_ctrl, readl(&pmu->pmu_spare1));
+		dmc_set_read_offset_value(phy1_ctrl, readl(&pmu->pmu_spare2));
+	} else {
+		software_find_read_offset(phy0_ctrl, 0,
+					  CTRL_LOCK_COARSE(lock0_info));
+		software_find_read_offset(phy1_ctrl, 1,
+					  CTRL_LOCK_COARSE(lock1_info));
+		/* save calibrated value to restore after resume */
+		writel(dmc_get_read_offset_value(phy0_ctrl), &pmu->pmu_spare1);
+		writel(dmc_get_read_offset_value(phy1_ctrl), &pmu->pmu_spare2);
+	}
+
 	/* Send PALL command */
 	dmc_config_prech(mem, &drex0->directcmd);
 	dmc_config_prech(mem, &drex1->directcmd);
diff --git a/arch/arm/cpu/armv7/exynos/exynos5_setup.h b/arch/arm/cpu/armv7/exynos/exynos5_setup.h
index 53b0ace6e3..3242093855 100644
--- a/arch/arm/cpu/armv7/exynos/exynos5_setup.h
+++ b/arch/arm/cpu/armv7/exynos/exynos5_setup.h
@@ -282,8 +282,12 @@
 #define PHY_CON12_VAL		0x10107F50
 #define CTRL_START		(1 << 6)
 #define CTRL_DLL_ON		(1 << 5)
+#define CTRL_LOCK_COARSE_OFFSET	10
+#define CTRL_LOCK_COARSE_MASK	(0x7F << CTRL_LOCK_COARSE_OFFSET)
+#define CTRL_LOCK_COARSE(x)	(((x) & CTRL_LOCK_COARSE_MASK) >> \
+				 CTRL_LOCK_COARSE_OFFSET)
 #define CTRL_FORCE_MASK		(0x7F << 8)
-#define CTRL_LOCK_COARSE_MASK	(0x7F << 10)
+#define CTRL_FINE_LOCKED	0x7
 
 #define CTRL_OFFSETD_RESET_VAL	0x8
 #define CTRL_OFFSETD_VAL	0x7F
@@ -431,10 +435,10 @@
 
 /*
  * Definitions that differ with SoC's.
- * Below is the part defining macros for smdk5250.
- * Else part introduces macros for smdk5420.
+ * Below is the part defining macros for Exynos5250.
+ * Else part introduces macros for Exynos5420.
  */
-#ifndef CONFIG_SMDK5420
+#ifndef CONFIG_EXYNOS5420
 
 /* APLL_CON1 */
 #define APLL_CON1_VAL	(0x00203800)
@@ -890,16 +894,11 @@ enum {
 /*
  * Memory variant specific initialization code for DDR3
  *
- * @param mem		Memory timings for this memory type.
- * @param mem_iv_size	Memory interleaving size is a configurable parameter
- *			which the DMC uses to decide how to split a memory
- *			chunk into smaller chunks to support concurrent
- *			accesses; may vary across boards.
+ * @param mem          Memory timings for this memory type.
  * @param reset         Reset DDR PHY during initialization.
  * @return 0 if ok, SETUP_ERR_... if there is a problem
  */
-int ddr3_mem_ctrl_init(struct mem_timings *mem, unsigned long mem_iv_size,
-			int reset);
+int ddr3_mem_ctrl_init(struct mem_timings *mem, int reset);
 
 /* Memory variant specific initialization code for LPDDR3 */
 void lpddr3_mem_ctrl_init(void);
diff --git a/arch/arm/cpu/armv7/exynos/lowlevel_init.c b/arch/arm/cpu/armv7/exynos/lowlevel_init.c
index 11fe5b8d04..83e1dcfc1e 100644
--- a/arch/arm/cpu/armv7/exynos/lowlevel_init.c
+++ b/arch/arm/cpu/armv7/exynos/lowlevel_init.c
@@ -39,6 +39,7 @@ enum {
 	DO_CLOCKS	= 1 << 1,
 	DO_MEM_RESET	= 1 << 2,
 	DO_UART		= 1 << 3,
+	DO_POWER	= 1 << 4,
 };
 
 int do_lowlevel_init(void)
@@ -60,9 +61,12 @@ int do_lowlevel_init(void)
 		break;
 	default:
 		/* This is a normal boot (not a wake from sleep) */
-		actions = DO_CLOCKS | DO_MEM_RESET;
+		actions = DO_CLOCKS | DO_MEM_RESET | DO_POWER;
 	}
 
+	if (actions & DO_POWER)
+		set_ps_hold_ctrl();
+
 	if (actions & DO_CLOCKS) {
 		system_clock_init();
 		mem_ctrl_init(actions & DO_MEM_RESET);
diff --git a/arch/arm/cpu/armv7/exynos/pinmux.c b/arch/arm/cpu/armv7/exynos/pinmux.c
index ee7c2e5a4b..86a0c75326 100644
--- a/arch/arm/cpu/armv7/exynos/pinmux.c
+++ b/arch/arm/cpu/armv7/exynos/pinmux.c
@@ -573,15 +573,26 @@ static void exynos4_i2c_config(int peripheral, int flags)
 static int exynos4_mmc_config(int peripheral, int flags)
 {
 	int i, start = 0, start_ext = 0;
+	unsigned int func, ext_func;
 
 	switch (peripheral) {
 	case PERIPH_ID_SDMMC0:
 		start = EXYNOS4_GPIO_K00;
 		start_ext = EXYNOS4_GPIO_K13;
+		func = S5P_GPIO_FUNC(0x2);
+		ext_func = S5P_GPIO_FUNC(0x3);
 		break;
 	case PERIPH_ID_SDMMC2:
 		start = EXYNOS4_GPIO_K20;
 		start_ext = EXYNOS4_GPIO_K33;
+		func = S5P_GPIO_FUNC(0x2);
+		ext_func = S5P_GPIO_FUNC(0x3);
+		break;
+	case PERIPH_ID_SDMMC4:
+		start = EXYNOS4_GPIO_K00;
+		start_ext = EXYNOS4_GPIO_K13;
+		func = S5P_GPIO_FUNC(0x3);
+		ext_func = S5P_GPIO_FUNC(0x4);
 		break;
 	default:
 		return -1;
@@ -589,13 +600,14 @@ static int exynos4_mmc_config(int peripheral, int flags)
 	for (i = start; i < (start + 7); i++) {
 		if (i == (start + 2))
 			continue;
-		gpio_cfg_pin(i,  S5P_GPIO_FUNC(0x2));
+		gpio_cfg_pin(i,  func);
 		gpio_set_pull(i, S5P_GPIO_PULL_NONE);
 		gpio_set_drv(i, S5P_GPIO_DRV_4X);
 	}
+	/* SDMMC2 do not use 8bit mode at exynos4 */
 	if (flags & PINMUX_FLAG_8BIT_MODE) {
 		for (i = start_ext; i < (start_ext + 4); i++) {
-			gpio_cfg_pin(i,  S5P_GPIO_FUNC(0x3));
+			gpio_cfg_pin(i,  ext_func);
 			gpio_set_pull(i, S5P_GPIO_PULL_NONE);
 			gpio_set_drv(i, S5P_GPIO_DRV_4X);
 		}
@@ -676,15 +688,26 @@ static void exynos4x12_i2c_config(int peripheral, int flags)
 static int exynos4x12_mmc_config(int peripheral, int flags)
 {
 	int i, start = 0, start_ext = 0;
+	unsigned int func, ext_func;
 
 	switch (peripheral) {
 	case PERIPH_ID_SDMMC0:
 		start = EXYNOS4X12_GPIO_K00;
 		start_ext = EXYNOS4X12_GPIO_K13;
+		func = S5P_GPIO_FUNC(0x2);
+		ext_func = S5P_GPIO_FUNC(0x3);
 		break;
 	case PERIPH_ID_SDMMC2:
 		start = EXYNOS4X12_GPIO_K20;
 		start_ext = EXYNOS4X12_GPIO_K33;
+		func = S5P_GPIO_FUNC(0x2);
+		ext_func = S5P_GPIO_FUNC(0x3);
+		break;
+	case PERIPH_ID_SDMMC4:
+		start = EXYNOS4_GPIO_K00;
+		start_ext = EXYNOS4_GPIO_K13;
+		func = S5P_GPIO_FUNC(0x3);
+		ext_func = S5P_GPIO_FUNC(0x4);
 		break;
 	default:
 		return -1;
@@ -692,13 +715,13 @@ static int exynos4x12_mmc_config(int peripheral, int flags)
 	for (i = start; i < (start + 7); i++) {
 		if (i == (start + 2))
 			continue;
-		gpio_cfg_pin(i,  S5P_GPIO_FUNC(0x2));
+		gpio_cfg_pin(i,  func);
 		gpio_set_pull(i, S5P_GPIO_PULL_NONE);
 		gpio_set_drv(i, S5P_GPIO_DRV_4X);
 	}
 	if (flags & PINMUX_FLAG_8BIT_MODE) {
 		for (i = start_ext; i < (start_ext + 4); i++) {
-			gpio_cfg_pin(i,  S5P_GPIO_FUNC(0x3));
+			gpio_cfg_pin(i,  ext_func);
 			gpio_set_pull(i, S5P_GPIO_PULL_NONE);
 			gpio_set_drv(i, S5P_GPIO_DRV_4X);
 		}
@@ -759,10 +782,10 @@ static int exynos4_pinmux_config(int peripheral, int flags)
 		break;
 	case PERIPH_ID_SDMMC0:
 	case PERIPH_ID_SDMMC2:
+	case PERIPH_ID_SDMMC4:
 		return exynos4_mmc_config(peripheral, flags);
 	case PERIPH_ID_SDMMC1:
 	case PERIPH_ID_SDMMC3:
-	case PERIPH_ID_SDMMC4:
 		debug("SDMMC device %d not implemented\n", peripheral);
 		return -1;
 	default:
@@ -794,10 +817,10 @@ static int exynos4x12_pinmux_config(int peripheral, int flags)
 		break;
 	case PERIPH_ID_SDMMC0:
 	case PERIPH_ID_SDMMC2:
+	case PERIPH_ID_SDMMC4:
 		return exynos4x12_mmc_config(peripheral, flags);
 	case PERIPH_ID_SDMMC1:
 	case PERIPH_ID_SDMMC3:
-	case PERIPH_ID_SDMMC4:
 		debug("SDMMC device %d not implemented\n", peripheral);
 		return -1;
 	default:
diff --git a/arch/arm/cpu/armv7/exynos/power.c b/arch/arm/cpu/armv7/exynos/power.c
index 563abd750f..638ee0b30b 100644
--- a/arch/arm/cpu/armv7/exynos/power.c
+++ b/arch/arm/cpu/armv7/exynos/power.c
@@ -112,6 +112,12 @@ static void exynos5_set_ps_hold_ctrl(void)
 			EXYNOS_PS_HOLD_CONTROL_DATA_HIGH);
 }
 
+/*
+ * Set ps_hold data driving value high
+ * This enables the machine to stay powered on
+ * after the initial power-on condition goes away
+ * (e.g. power button).
+ */
 void set_ps_hold_ctrl(void)
 {
 	if (cpu_is_exynos5())
diff --git a/arch/arm/cpu/armv7/exynos/spl_boot.c b/arch/arm/cpu/armv7/exynos/spl_boot.c
index ade45fd5d3..79166303d1 100644
--- a/arch/arm/cpu/armv7/exynos/spl_boot.c
+++ b/arch/arm/cpu/armv7/exynos/spl_boot.c
@@ -4,8 +4,8 @@
  * SPDX-License-Identifier:	GPL-2.0+
  */
 
-#include<common.h>
-#include<config.h>
+#include <common.h>
+#include <config.h>
 
 #include <asm/arch/clock.h>
 #include <asm/arch/clk.h>