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

arch/arm/cpu/arm720t/tegra-common/spl.c

@@ -15,7 +15,7 @@
 #include <asm/arch/tegra.h>
 #include <asm/arch-tegra/apb_misc.h>
 #include <asm/arch-tegra/board.h>
-#include <asm/arch/spl.h>
+#include <asm/spl.h>
 #include "cpu.h"
 
 void spl_board_init(void)

arch/arm/cpu/arm926ejs/davinci/da850_lowlevel.c

@@ -14,7 +14,7 @@
 #include <asm/arch/hardware.h>
 #include <asm/arch/davinci_misc.h>
 #include <asm/arch/ddr2_defs.h>
-#include <asm/arch/emif_defs.h>
+#include <asm/ti-common/davinci_nand.h>
 #include <asm/arch/pll_defs.h>
 
 void davinci_enable_uart0(void)

arch/arm/cpu/arm926ejs/davinci/dm365_lowlevel.c

@@ -11,6 +11,7 @@
 #include <nand.h>
 #include <ns16550.h>
 #include <post.h>
+#include <asm/ti-common/davinci_nand.h>
 #include <asm/arch/dm365_lowlevel.h>
 #include <asm/arch/hardware.h>
 

arch/arm/cpu/armv7/am33xx/ddr.c

@@ -95,6 +95,7 @@ void config_sdram_emif4d5(const struct emif_regs *regs, int nr)
 	       &emif_reg[nr]->emif_rd_wr_exec_thresh);
 
 	writel(regs->ref_ctrl, &emif_reg[nr]->emif_sdram_ref_ctrl);
+	writel(regs->ref_ctrl, &emif_reg[nr]->emif_sdram_ref_ctrl_shdw);
 	writel(regs->sdram_config, &emif_reg[nr]->emif_sdram_config);
 	writel(regs->sdram_config, &cstat->secure_emif_sdram_config);
 

arch/arm/cpu/armv7/at91/cpu.c

@@ -61,6 +61,8 @@ int print_cpuinfo(void)
 
 void enable_caches(void)
 {
+	icache_enable();
+	dcache_enable();
 }
 
 unsigned int get_chip_id(void)

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)

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);

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);

arch/arm/cpu/armv7/exynos/lowlevel_init.c

@@ -49,8 +49,6 @@ int do_lowlevel_init(void)
 
 	arch_cpu_init();
 
-	set_ps_hold_ctrl();
-
 	reset_status = get_reset_status();
 
 	switch (reset_status) {

arch/arm/cpu/armv7/keystone/Makefile

@@ -5,7 +5,6 @@
 # SPDX-License-Identifier:     GPL-2.0+
 #
 
-obj-y	+= aemif.o
 obj-y	+= init.o
 obj-y	+= psc.o
 obj-y	+= clock.o

arch/arm/cpu/armv7/mx6/Makefile

@@ -8,4 +8,5 @@
 #
 
 obj-y	:= soc.o clock.o
+obj-$(CONFIG_SPL_BUILD)	     += ddr.o
 obj-$(CONFIG_SECURE_BOOT)    += hab.o

arch/arm/cpu/armv7/mx6/ddr.c

+/*
+ * Copyright (C) 2014 Gateworks Corporation
+ * Author: Tim Harvey <tharvey@gateworks.com>
+ *
+ * SPDX-License-Identifier:     GPL-2.0+
+ */
+
+#include <common.h>
+#include <linux/types.h>
+#include <asm/arch/mx6-ddr.h>
+#include <asm/arch/sys_proto.h>
+#include <asm/io.h>
+#include <asm/types.h>
+
+#if defined(CONFIG_MX6QDL) || defined(CONFIG_MX6Q) || defined(CONFIG_MX6D)
+/* Configure MX6DQ mmdc iomux */
+void mx6dq_dram_iocfg(unsigned width,
+		      const struct mx6dq_iomux_ddr_regs *ddr,
+		      const struct mx6dq_iomux_grp_regs *grp)
+{
+	volatile struct mx6dq_iomux_ddr_regs *mx6_ddr_iomux;
+	volatile struct mx6dq_iomux_grp_regs *mx6_grp_iomux;
+
+	mx6_ddr_iomux = (struct mx6dq_iomux_ddr_regs *)MX6DQ_IOM_DDR_BASE;
+	mx6_grp_iomux = (struct mx6dq_iomux_grp_regs *)MX6DQ_IOM_GRP_BASE;
+
+	/* DDR IO Type */
+	mx6_grp_iomux->grp_ddr_type = grp->grp_ddr_type;
+	mx6_grp_iomux->grp_ddrpke = grp->grp_ddrpke;
+
+	/* Clock */
+	mx6_ddr_iomux->dram_sdclk_0 = ddr->dram_sdclk_0;
+	mx6_ddr_iomux->dram_sdclk_1 = ddr->dram_sdclk_1;
+
+	/* Address */
+	mx6_ddr_iomux->dram_cas = ddr->dram_cas;
+	mx6_ddr_iomux->dram_ras = ddr->dram_ras;
+	mx6_grp_iomux->grp_addds = grp->grp_addds;
+
+	/* Control */
+	mx6_ddr_iomux->dram_reset = ddr->dram_reset;
+	mx6_ddr_iomux->dram_sdcke0 = ddr->dram_sdcke0;
+	mx6_ddr_iomux->dram_sdcke1 = ddr->dram_sdcke1;
+	mx6_ddr_iomux->dram_sdba2 = ddr->dram_sdba2;
+	mx6_ddr_iomux->dram_sdodt0 = ddr->dram_sdodt0;
+	mx6_ddr_iomux->dram_sdodt1 = ddr->dram_sdodt1;
+	mx6_grp_iomux->grp_ctlds = grp->grp_ctlds;
+
+	/* Data Strobes */
+	mx6_grp_iomux->grp_ddrmode_ctl = grp->grp_ddrmode_ctl;
+	mx6_ddr_iomux->dram_sdqs0 = ddr->dram_sdqs0;
+	mx6_ddr_iomux->dram_sdqs1 = ddr->dram_sdqs1;
+	if (width >= 32) {
+		mx6_ddr_iomux->dram_sdqs2 = ddr->dram_sdqs2;
+		mx6_ddr_iomux->dram_sdqs3 = ddr->dram_sdqs3;
+	}
+	if (width >= 64) {
+		mx6_ddr_iomux->dram_sdqs4 = ddr->dram_sdqs4;
+		mx6_ddr_iomux->dram_sdqs5 = ddr->dram_sdqs5;
+		mx6_ddr_iomux->dram_sdqs6 = ddr->dram_sdqs6;
+		mx6_ddr_iomux->dram_sdqs7 = ddr->dram_sdqs7;
+	}
+
+	/* Data */
+	mx6_grp_iomux->grp_ddrmode = grp->grp_ddrmode;
+	mx6_grp_iomux->grp_b0ds = grp->grp_b0ds;
+	mx6_grp_iomux->grp_b1ds = grp->grp_b1ds;
+	if (width >= 32) {
+		mx6_grp_iomux->grp_b2ds = grp->grp_b2ds;
+		mx6_grp_iomux->grp_b3ds = grp->grp_b3ds;
+	}
+	if (width >= 64) {
+		mx6_grp_iomux->grp_b4ds = grp->grp_b4ds;
+		mx6_grp_iomux->grp_b5ds = grp->grp_b5ds;
+		mx6_grp_iomux->grp_b6ds = grp->grp_b6ds;
+		mx6_grp_iomux->grp_b7ds = grp->grp_b7ds;
+	}
+	mx6_ddr_iomux->dram_dqm0 = ddr->dram_dqm0;
+	mx6_ddr_iomux->dram_dqm1 = ddr->dram_dqm1;
+	if (width >= 32) {
+		mx6_ddr_iomux->dram_dqm2 = ddr->dram_dqm2;
+		mx6_ddr_iomux->dram_dqm3 = ddr->dram_dqm3;
+	}
+	if (width >= 64) {
+		mx6_ddr_iomux->dram_dqm4 = ddr->dram_dqm4;
+		mx6_ddr_iomux->dram_dqm5 = ddr->dram_dqm5;
+		mx6_ddr_iomux->dram_dqm6 = ddr->dram_dqm6;
+		mx6_ddr_iomux->dram_dqm7 = ddr->dram_dqm7;
+	}
+}
+#endif
+
+#if defined(CONFIG_MX6QDL) || defined(CONFIG_MX6DL) || defined(CONFIG_MX6S)
+/* Configure MX6SDL mmdc iomux */
+void mx6sdl_dram_iocfg(unsigned width,
+		       const struct mx6sdl_iomux_ddr_regs *ddr,
+		       const struct mx6sdl_iomux_grp_regs *grp)
+{
+	volatile struct mx6sdl_iomux_ddr_regs *mx6_ddr_iomux;
+	volatile struct mx6sdl_iomux_grp_regs *mx6_grp_iomux;
+
+	mx6_ddr_iomux = (struct mx6sdl_iomux_ddr_regs *)MX6SDL_IOM_DDR_BASE;
+	mx6_grp_iomux = (struct mx6sdl_iomux_grp_regs *)MX6SDL_IOM_GRP_BASE;
+
+	/* DDR IO Type */
+	mx6_grp_iomux->grp_ddr_type = grp->grp_ddr_type;
+	mx6_grp_iomux->grp_ddrpke = grp->grp_ddrpke;
+
+	/* Clock */
+	mx6_ddr_iomux->dram_sdclk_0 = ddr->dram_sdclk_0;
+	mx6_ddr_iomux->dram_sdclk_1 = ddr->dram_sdclk_1;
+
+	/* Address */
+	mx6_ddr_iomux->dram_cas = ddr->dram_cas;
+	mx6_ddr_iomux->dram_ras = ddr->dram_ras;
+	mx6_grp_iomux->grp_addds = grp->grp_addds;
+
+	/* Control */
+	mx6_ddr_iomux->dram_reset = ddr->dram_reset;
+	mx6_ddr_iomux->dram_sdcke0 = ddr->dram_sdcke0;
+	mx6_ddr_iomux->dram_sdcke1 = ddr->dram_sdcke1;
+	mx6_ddr_iomux->dram_sdba2 = ddr->dram_sdba2;
+	mx6_ddr_iomux->dram_sdodt0 = ddr->dram_sdodt0;
+	mx6_ddr_iomux->dram_sdodt1 = ddr->dram_sdodt1;
+	mx6_grp_iomux->grp_ctlds = grp->grp_ctlds;
+
+	/* Data Strobes */
+	mx6_grp_iomux->grp_ddrmode_ctl = grp->grp_ddrmode_ctl;
+	mx6_ddr_iomux->dram_sdqs0 = ddr->dram_sdqs0;
+	mx6_ddr_iomux->dram_sdqs1 = ddr->dram_sdqs1;
+	if (width >= 32) {
+		mx6_ddr_iomux->dram_sdqs2 = ddr->dram_sdqs2;
+		mx6_ddr_iomux->dram_sdqs3 = ddr->dram_sdqs3;
+	}
+	if (width >= 64) {
+		mx6_ddr_iomux->dram_sdqs4 = ddr->dram_sdqs4;
+		mx6_ddr_iomux->dram_sdqs5 = ddr->dram_sdqs5;
+		mx6_ddr_iomux->dram_sdqs6 = ddr->dram_sdqs6;
+		mx6_ddr_iomux->dram_sdqs7 = ddr->dram_sdqs7;
+	}
+
+	/* Data */
+	mx6_grp_iomux->grp_ddrmode = grp->grp_ddrmode;
+	mx6_grp_iomux->grp_b0ds = grp->grp_b0ds;
+	mx6_grp_iomux->grp_b1ds = grp->grp_b1ds;
+	if (width >= 32) {
+		mx6_grp_iomux->grp_b2ds = grp->grp_b2ds;
+		mx6_grp_iomux->grp_b3ds = grp->grp_b3ds;
+	}
+	if (width >= 64) {
+		mx6_grp_iomux->grp_b4ds = grp->grp_b4ds;
+		mx6_grp_iomux->grp_b5ds = grp->grp_b5ds;
+		mx6_grp_iomux->grp_b6ds = grp->grp_b6ds;
+		mx6_grp_iomux->grp_b7ds = grp->grp_b7ds;
+	}
+	mx6_ddr_iomux->dram_dqm0 = ddr->dram_dqm0;
+	mx6_ddr_iomux->dram_dqm1 = ddr->dram_dqm1;
+	if (width >= 32) {
+		mx6_ddr_iomux->dram_dqm2 = ddr->dram_dqm2;
+		mx6_ddr_iomux->dram_dqm3 = ddr->dram_dqm3;
+	}
+	if (width >= 64) {
+		mx6_ddr_iomux->dram_dqm4 = ddr->dram_dqm4;
+		mx6_ddr_iomux->dram_dqm5 = ddr->dram_dqm5;
+		mx6_ddr_iomux->dram_dqm6 = ddr->dram_dqm6;
+		mx6_ddr_iomux->dram_dqm7 = ddr->dram_dqm7;
+	}
+}
+#endif
+
+/*
+ * Configure mx6 mmdc registers based on:
+ *  - board-specific memory configuration
+ *  - board-specific calibration data
+ *  - ddr3 chip details
+ *
+ * The various calculations here are derived from the Freescale
+ * i.Mx6DQSDL DDR3 Script Aid spreadsheet (DOC-94917) designed to generate MMDC
+ * configuration registers based on memory system and memory chip parameters.
+ *
+ * The defaults here are those which were specified in the spreadsheet.
+ * For details on each register, refer to the IMX6DQRM and/or IMX6SDLRM
+ * section titled MMDC initialization
+ */
+#define MR(val, ba, cmd, cs1) \
+	((val << 16) | (1 << 15) | (cmd << 4) | (cs1 << 3) | ba)
+void mx6_dram_cfg(const struct mx6_ddr_sysinfo *i,
+		  const struct mx6_mmdc_calibration *c,
+		  const struct mx6_ddr3_cfg *m)
+{
+	volatile struct mmdc_p_regs *mmdc0;
+	volatile struct mmdc_p_regs *mmdc1;
+	u32 reg;
+	u8 tcke, tcksrx, tcksre, txpdll, taofpd, taonpd, trrd;
+	u8 todtlon, taxpd, tanpd, tcwl, txp, tfaw, tcl;
+	u8 todt_idle_off = 0x4; /* from DDR3 Script Aid spreadsheet */
+	u16 trcd, trc, tras, twr, tmrd, trtp, trp, twtr, trfc, txs, txpr;
+	u16 CS0_END;
+	u16 tdllk = 0x1ff; /* DLL locking time: 512 cycles (JEDEC DDR3) */
+	int clkper; /* clock period in picoseconds */
+	int clock; /* clock freq in mHz */
+	int cs;
+
+	mmdc0 = (struct mmdc_p_regs *)MMDC_P0_BASE_ADDR;
+	mmdc1 = (struct mmdc_p_regs *)MMDC_P1_BASE_ADDR;
+
+	/* MX6D/MX6Q: 1066 MHz memory clock, clkper = 1.894ns = 1894ps */
+	if (is_cpu_type(MXC_CPU_MX6Q) || is_cpu_type(MXC_CPU_MX6D)) {
+		clock = 528;
+		tcwl = 4;
+	}
+	/* MX6S/MX6DL: 800 MHz memory clock, clkper = 2.5ns = 2500ps */
+	else {
+		clock = 400;
+		tcwl = 3;
+	}
+	clkper = (1000*1000)/clock; /* ps */
+	todtlon = tcwl;
+	taxpd = tcwl;
+	tanpd = tcwl;
+	tcwl = tcwl;
+
+	switch (m->density) {
+	case 1: /* 1Gb per chip */
+		trfc = DIV_ROUND_UP(110000, clkper) - 1;
+		txs = DIV_ROUND_UP(120000, clkper) - 1;
+		break;
+	case 2: /* 2Gb per chip */
+		trfc = DIV_ROUND_UP(160000, clkper) - 1;
+		txs = DIV_ROUND_UP(170000, clkper) - 1;
+		break;
+	case 4: /* 4Gb per chip */
+		trfc = DIV_ROUND_UP(260000, clkper) - 1;
+		txs = DIV_ROUND_UP(270000, clkper) - 1;
+		break;
+	case 8: /* 8Gb per chip */
+		trfc = DIV_ROUND_UP(350000, clkper) - 1;
+		txs = DIV_ROUND_UP(360000, clkper) - 1;
+		break;
+	default:
+		/* invalid density */
+		printf("invalid chip density\n");
+		hang();
+		break;
+	}
+	txpr = txs;
+
+	switch (m->mem_speed) {
+	case 800:
+		txp = DIV_ROUND_UP(MAX(3*clkper, 7500), clkper) - 1;
+		tcke = DIV_ROUND_UP(MAX(3*clkper, 7500), clkper) - 1;
+		if (m->pagesz == 1) {
+			tfaw = DIV_ROUND_UP(40000, clkper) - 1;
+			trrd = DIV_ROUND_UP(MAX(4*clkper, 10000), clkper) - 1;
+		} else {
+			tfaw = DIV_ROUND_UP(50000, clkper) - 1;
+			trrd = DIV_ROUND_UP(MAX(4*clkper, 10000), clkper) - 1;
+		}
+		break;
+	case 1066:
+		txp = DIV_ROUND_UP(MAX(3*clkper, 7500), clkper) - 1;
+		tcke = DIV_ROUND_UP(MAX(3*clkper, 5625), clkper) - 1;
+		if (m->pagesz == 1) {
+			tfaw = DIV_ROUND_UP(37500, clkper) - 1;
+			trrd = DIV_ROUND_UP(MAX(4*clkper, 7500), clkper) - 1;
+		} else {
+			tfaw = DIV_ROUND_UP(50000, clkper) - 1;
+			trrd = DIV_ROUND_UP(MAX(4*clkper, 10000), clkper) - 1;
+		}
+		break;
+	case 1333:
+		txp = DIV_ROUND_UP(MAX(3*clkper, 6000), clkper) - 1;
+		tcke = DIV_ROUND_UP(MAX(3*clkper, 5625), clkper) - 1;
+		if (m->pagesz == 1) {
+			tfaw = DIV_ROUND_UP(30000, clkper) - 1;
+			trrd = DIV_ROUND_UP(MAX(4*clkper, 6000), clkper) - 1;
+		} else {
+			tfaw = DIV_ROUND_UP(45000, clkper) - 1;
+			trrd = DIV_ROUND_UP(MAX(4*clkper, 7500), clkper) - 1;
+		}
+		break;
+	case 1600:
+		txp = DIV_ROUND_UP(MAX(3*clkper, 6000), clkper) - 1;
+		tcke = DIV_ROUND_UP(MAX(3*clkper, 5000), clkper) - 1;
+		if (m->pagesz == 1) {
+			tfaw = DIV_ROUND_UP(30000, clkper) - 1;
+			trrd = DIV_ROUND_UP(MAX(4*clkper, 6000), clkper) - 1;
+		} else {
+			tfaw = DIV_ROUND_UP(40000, clkper) - 1;
+			trrd = DIV_ROUND_UP(MAX(4*clkper, 7500), clkper) - 1;
+		}
+		break;
+	default:
+		printf("invalid memory speed\n");
+		hang();
+		break;
+	}
+	txpdll = DIV_ROUND_UP(MAX(10*clkper, 24000), clkper) - 1;
+	tcl = DIV_ROUND_UP(m->trcd, clkper/10) - 3;
+	tcksre = DIV_ROUND_UP(MAX(5*clkper, 10000), clkper);
+	tcksrx = tcksre;
+	taonpd = DIV_ROUND_UP(2000, clkper) - 1;
+	taofpd = taonpd;
+	trp = DIV_ROUND_UP(m->trcd, clkper/10) - 1;
+	trcd = trp;
+	trc = DIV_ROUND_UP(m->trcmin, clkper/10) - 1;
+	tras = DIV_ROUND_UP(m->trasmin, clkper/10) - 1;
+	twr = DIV_ROUND_UP(15000, clkper) - 1;
+	tmrd = DIV_ROUND_UP(MAX(12*clkper, 15000), clkper) - 1;
+	twtr = ROUND(MAX(4*clkper, 7500)/clkper, 1) - 1;
+	trtp = twtr;
+	CS0_END = ((4*i->cs_density) <= 120) ? (4*i->cs_density)+7 : 127;
+	debug("density:%d Gb (%d Gb per chip)\n", i->cs_density, m->density);
+	debug("clock: %dMHz (%d ps)\n", clock, clkper);
+	debug("memspd:%d\n", m->mem_speed);
+	debug("tcke=%d\n", tcke);
+	debug("tcksrx=%d\n", tcksrx);
+	debug("tcksre=%d\n", tcksre);
+	debug("taofpd=%d\n", taofpd);
+	debug("taonpd=%d\n", taonpd);
+	debug("todtlon=%d\n", todtlon);
+	debug("tanpd=%d\n", tanpd);
+	debug("taxpd=%d\n", taxpd);
+	debug("trfc=%d\n", trfc);
+	debug("txs=%d\n", txs);
+	debug("txp=%d\n", txp);
+	debug("txpdll=%d\n", txpdll);
+	debug("tfaw=%d\n", tfaw);
+	debug("tcl=%d\n", tcl);
+	debug("trcd=%d\n", trcd);
+	debug("trp=%d\n", trp);
+	debug("trc=%d\n", trc);
+	debug("tras=%d\n", tras);
+	debug("twr=%d\n", twr);
+	debug("tmrd=%d\n", tmrd);
+	debug("tcwl=%d\n", tcwl);
+	debug("tdllk=%d\n", tdllk);
+	debug("trtp=%d\n", trtp);
+	debug("twtr=%d\n", twtr);
+	debug("trrd=%d\n", trrd);
+	debug("txpr=%d\n", txpr);
+	debug("CS0_END=%d\n", CS0_END);
+	debug("ncs=%d\n", i->ncs);
+	debug("Rtt_wr=%d\n", i->rtt_wr);
+	debug("Rtt_nom=%d\n", i->rtt_nom);
+	debug("SRT=%d\n", m->SRT);
+	debug("tcl=%d\n", tcl);
+	debug("twr=%d\n", twr);
+
+	/*
+	 * board-specific configuration:
+	 *  These values are determined empirically and vary per board layout
+	 *  see:
+	 *   appnote, ddr3 spreadsheet
+	 */
+	mmdc0->mpwldectrl0 = c->p0_mpwldectrl0;
+	mmdc0->mpwldectrl1 = c->p0_mpwldectrl1;
+	mmdc0->mpdgctrl0 = c->p0_mpdgctrl0;
+	mmdc0->mpdgctrl1 = c->p0_mpdgctrl1;
+	mmdc0->mprddlctl = c->p0_mprddlctl;
+	mmdc0->mpwrdlctl = c->p0_mpwrdlctl;
+	if (i->dsize > 1) {
+		mmdc1->mpwldectrl0 = c->p1_mpwldectrl0;
+		mmdc1->mpwldectrl1 = c->p1_mpwldectrl1;
+		mmdc1->mpdgctrl0 = c->p1_mpdgctrl0;
+		mmdc1->mpdgctrl1 = c->p1_mpdgctrl1;
+		mmdc1->mprddlctl = c->p1_mprddlctl;
+		mmdc1->mpwrdlctl = c->p1_mpwrdlctl;
+	}
+
+	/* Read data DQ Byte0-3 delay */
+	mmdc0->mprddqby0dl = (u32)0x33333333;
+	mmdc0->mprddqby1dl = (u32)0x33333333;
+	if (i->dsize > 0) {
+		mmdc0->mprddqby2dl = (u32)0x33333333;
+		mmdc0->mprddqby3dl = (u32)0x33333333;
+	}
+	if (i->dsize > 1) {
+		mmdc1->mprddqby0dl = (u32)0x33333333;
+		mmdc1->mprddqby1dl = (u32)0x33333333;
+		mmdc1->mprddqby2dl = (u32)0x33333333;
+		mmdc1->mprddqby3dl = (u32)0x33333333;
+	}
+
+	/* MMDC Termination: rtt_nom:2 RZQ/2(120ohm), rtt_nom:1 RZQ/4(60ohm) */
+	reg = (i->rtt_nom == 2) ? 0x00011117 : 0x00022227;
+	mmdc0->mpodtctrl = reg;
+	if (i->dsize > 1)
+		mmdc1->mpodtctrl = reg;
+
+	/* complete calibration */
+	reg = (1 << 11); /* Force measurement on delay-lines */
+	mmdc0->mpmur0 = reg;
+	if (i->dsize > 1)
+		mmdc1->mpmur0 = reg;
+
+	/* Step 1: configuration request */
+	mmdc0->mdscr = (u32)(1 << 15); /* config request */
+
+	/* Step 2: Timing configuration */
+	reg = (trfc << 24) | (txs << 16) | (txp << 13) | (txpdll << 9) |
+	      (tfaw << 4) | tcl;
+	mmdc0->mdcfg0 = reg;
+	reg = (trcd << 29) | (trp << 26) | (trc << 21) | (tras << 16) |
+	      (1 << 15) |		/* trpa */
+	      (twr << 9) | (tmrd << 5) | tcwl;
+	mmdc0->mdcfg1 = reg;
+	reg = (tdllk << 16) | (trtp << 6) | (twtr << 3) | trrd;
+	mmdc0->mdcfg2 = reg;
+	reg = (taofpd << 27) | (taonpd << 24) | (tanpd << 20) | (taxpd << 16) |
+	      (todtlon << 12) | (todt_idle_off << 4);
+	mmdc0->mdotc = reg;
+	mmdc0->mdasp = CS0_END; /* CS addressing */
+
+	/* Step 3: Configure DDR type */
+	reg = (i->cs1_mirror << 19) | (i->walat << 16) | (i->bi_on << 12) |
+	      (i->mif3_mode << 9) | (i->ralat << 6);
+	mmdc0->mdmisc = reg;
+
+	/* Step 4: Configure delay while leaving reset */
+	reg = (txpr << 16) | (i->sde_to_rst << 8) | (i->rst_to_cke << 0);
+	mmdc0->mdor = reg;
+
+	/* Step 5: Configure DDR physical parameters (density and burst len) */
+	reg = (m->rowaddr - 11) << 24 |		/* ROW */
+	      (m->coladdr - 9) << 20 |		/* COL */
+	      (1 << 19) |			/* Burst Length = 8 for DDR3 */
+	      (i->dsize << 16);			/* DDR data bus size */
+	mmdc0->mdctl = reg;
+
+	/* Step 6: Perform ZQ calibration */
+	reg = (u32)0xa1390001; /* one-time HW ZQ calib */
+	mmdc0->mpzqhwctrl = reg;
+	if (i->dsize > 1)
+		mmdc1->mpzqhwctrl = reg;
+
+	/* Step 7: Enable MMDC with desired chip select */
+	reg = mmdc0->mdctl |
+	      (1 << 31) |			/* SDE_0 for CS0 */
+	      ((i->ncs == 2) ? 1 : 0) << 30;	/* SDE_1 for CS1 */
+	mmdc0->mdctl = reg;
+
+	/* Step 8: Write Mode Registers to Init DDR3 devices */
+	for (cs = 0; cs < i->ncs; cs++) {
+		/* MR2 */
+		reg = (i->rtt_wr & 3) << 9 | (m->SRT & 1) << 7 |
+		      ((tcwl - 3) & 3) << 3;
+		mmdc0->mdscr = (u32)MR(reg, 2, 3, cs);
+		/* MR3 */
+		mmdc0->mdscr = (u32)MR(0, 3, 3, cs);
+		/* MR1 */
+		reg = ((i->rtt_nom & 1) ? 1 : 0) << 2 |
+		      ((i->rtt_nom & 2) ? 1 : 0) << 6;
+		mmdc0->mdscr = (u32)MR(reg, 1, 3, cs);
+		reg = ((tcl - 1) << 4) |	/* CAS */
+		      (1 << 8)   |		/* DLL Reset */
+		      ((twr - 3) << 9);		/* Write Recovery */
+		/* MR0 */
+		mmdc0->mdscr = (u32)MR(reg, 0, 3, cs);
+		/* ZQ calibration */
+		reg = (1 << 10);
+		mmdc0->mdscr = (u32)MR(reg, 0, 4, cs);
+	}
+
+	/* Step 10: Power down control and self-refresh */
+	reg = (tcke & 0x7) << 16 |
+	      5            << 12 |  /* PWDT_1: 256 cycles */
+	      5            <<  8 |  /* PWDT_0: 256 cycles */
+	      1            <<  6 |  /* BOTH_CS_PD */
+	      (tcksrx & 0x7) << 3 |
+	      (tcksre & 0x7);
+	mmdc0->mdpdc = reg;
+	mmdc0->mapsr = (u32)0x00011006; /* ADOPT power down enabled */
+
+	/* Step 11: Configure ZQ calibration: one-time and periodic 1ms */
+	mmdc0->mpzqhwctrl = (u32)0xa1390003;
+	if (i->dsize > 1)
+		mmdc1->mpzqhwctrl = (u32)0xa1390003;
+
+	/* Step 12: Configure and activate periodic refresh */
+	reg = (1 << 14) |	/* REF_SEL: Periodic refresh cycles of 32kHz */
+	      (7 << 11);	/* REFR: Refresh Rate - 8 refreshes */
+	mmdc0->mdref = reg;
+
+	/* Step 13: Deassert config request - init complete */
+	mmdc0->mdscr = (u32)0x00000000;
+
+	/* wait for auto-ZQ calibration to complete */
+	mdelay(1);
+}

arch/arm/cpu/armv7/mx6/hab.c

@@ -7,15 +7,69 @@
 #include <common.h>
 #include <asm/io.h>
 #include <asm/arch/hab.h>
+#include <asm/arch/sys_proto.h>
 
 /* -------- start of HAB API updates ------------*/
-#define hab_rvt_report_event ((hab_rvt_report_event_t *)HAB_RVT_REPORT_EVENT)
-#define hab_rvt_report_status ((hab_rvt_report_status_t *)HAB_RVT_REPORT_STATUS)
-#define hab_rvt_authenticate_image \
-	((hab_rvt_authenticate_image_t *)HAB_RVT_AUTHENTICATE_IMAGE)
-#define hab_rvt_entry ((hab_rvt_entry_t *)HAB_RVT_ENTRY)
-#define hab_rvt_exit ((hab_rvt_exit_t *)HAB_RVT_EXIT)
-#define hab_rvt_clock_init HAB_RVT_CLOCK_INIT
+
+#define hab_rvt_report_event_p					\
+(								\
+	((is_cpu_type(MXC_CPU_MX6Q) ||				\
+	  is_cpu_type(MXC_CPU_MX6D)) &&				\
+	  (soc_rev() >= CHIP_REV_1_5)) ?			\
+	((hab_rvt_report_event_t *)HAB_RVT_REPORT_EVENT_NEW) :	\
+	(is_cpu_type(MXC_CPU_MX6DL) &&				\
+	 (soc_rev() >= CHIP_REV_1_2)) ?				\
+	((hab_rvt_report_event_t *)HAB_RVT_REPORT_EVENT_NEW) :	\
+	((hab_rvt_report_event_t *)HAB_RVT_REPORT_EVENT)	\
+)
+
+#define hab_rvt_report_status_p					\
+(								\
+	((is_cpu_type(MXC_CPU_MX6Q) ||				\
+	  is_cpu_type(MXC_CPU_MX6D)) &&				\
+	  (soc_rev() >= CHIP_REV_1_5)) ?			\
+	((hab_rvt_report_status_t *)HAB_RVT_REPORT_STATUS_NEW) :\
+	(is_cpu_type(MXC_CPU_MX6DL) &&				\
+	 (soc_rev() >= CHIP_REV_1_2)) ?				\
+	((hab_rvt_report_status_t *)HAB_RVT_REPORT_STATUS_NEW) :\
+	((hab_rvt_report_status_t *)HAB_RVT_REPORT_STATUS)	\
+)
+
+#define hab_rvt_authenticate_image_p				\
+(								\
+	((is_cpu_type(MXC_CPU_MX6Q) ||				\
+	  is_cpu_type(MXC_CPU_MX6D)) &&				\
+	  (soc_rev() >= CHIP_REV_1_5)) ?			\
+	((hab_rvt_authenticate_image_t *)HAB_RVT_AUTHENTICATE_IMAGE_NEW) : \
+	(is_cpu_type(MXC_CPU_MX6DL) &&				\
+	 (soc_rev() >= CHIP_REV_1_2)) ?				\
+	((hab_rvt_authenticate_image_t *)HAB_RVT_AUTHENTICATE_IMAGE_NEW) : \
+	((hab_rvt_authenticate_image_t *)HAB_RVT_AUTHENTICATE_IMAGE)	\
+)
+
+#define hab_rvt_entry_p						\
+(								\
+	((is_cpu_type(MXC_CPU_MX6Q) ||				\
+	  is_cpu_type(MXC_CPU_MX6D)) &&				\
+	  (soc_rev() >= CHIP_REV_1_5)) ?			\
+	((hab_rvt_entry_t *)HAB_RVT_ENTRY_NEW) :		\
+	(is_cpu_type(MXC_CPU_MX6DL) &&				\
+	 (soc_rev() >= CHIP_REV_1_2)) ?				\
+	((hab_rvt_entry_t *)HAB_RVT_ENTRY_NEW) :		\
+	((hab_rvt_entry_t *)HAB_RVT_ENTRY)			\
+)
+
+#define hab_rvt_exit_p						\
+(								\
+	((is_cpu_type(MXC_CPU_MX6Q) ||				\
+	  is_cpu_type(MXC_CPU_MX6D)) &&				\
+	  (soc_rev() >= CHIP_REV_1_5)) ?			\
+	((hab_rvt_exit_t *)HAB_RVT_EXIT_NEW) :			\
+	(is_cpu_type(MXC_CPU_MX6DL) &&				\
+	 (soc_rev() >= CHIP_REV_1_2)) ?				\
+	((hab_rvt_exit_t *)HAB_RVT_EXIT_NEW) :			\
+	((hab_rvt_exit_t *)HAB_RVT_EXIT)			\
+)
 
 bool is_hab_enabled(void)
 {
@@ -52,6 +106,11 @@ int get_hab_status(void)
 	size_t bytes = sizeof(event_data); /* Event size in bytes */
 	enum hab_config config = 0;
 	enum hab_state state = 0;
+	hab_rvt_report_event_t *hab_rvt_report_event;
+	hab_rvt_report_status_t *hab_rvt_report_status;
+
+	hab_rvt_report_event = hab_rvt_report_event_p;
+	hab_rvt_report_status = hab_rvt_report_status_p;
 
 	if (is_hab_enabled())
 		puts("\nSecure boot enabled\n");

arch/arm/cpu/armv7/omap-common/mem-common.c

@@ -121,7 +121,8 @@ void gpmc_init(void)
 	writel(0x00000008, &gpmc_cfg->sysconfig);
 	writel(0x00000000, &gpmc_cfg->irqstatus);
 	writel(0x00000000, &gpmc_cfg->irqenable);
-	writel(0x00000000, &gpmc_cfg->timeout_control);
+	/* disable timeout, set a safe reset value */
+	writel(0x00001ff0, &gpmc_cfg->timeout_control);
 #ifdef CONFIG_NOR
 	writel(0x00000200, &gpmc_cfg->config);
 #else
@@ -133,5 +134,6 @@ void gpmc_init(void)
 	writel(0, &gpmc_cfg->cs[0].config7);
 	sdelay(1000);
 	/* enable chip-select specific configurations */
-	enable_gpmc_cs_config(gpmc_regs, &gpmc_cfg->cs[0], base, size);
+	if (base != 0)
+		enable_gpmc_cs_config(gpmc_regs, &gpmc_cfg->cs[0], base, size);
 }

arch/arm/cpu/armv7/omap3/board.c

@@ -147,7 +147,7 @@ void secure_unlock_mem(void)
  *		configure secure registers and exit secure world
  *              general use.
  *****************************************************************************/
-void secureworld_exit()
+void secureworld_exit(void)
 {
 	unsigned long i;
 
@@ -178,7 +178,7 @@ void secureworld_exit()
  * Description: If chip is GP/EMU(special) type, unlock the SRAM for
  *              general use.
  *****************************************************************************/
-void try_unlock_memory()
+void try_unlock_memory(void)
 {
 	int mode;
 	int in_sdram = is_running_in_sdram();

arch/arm/cpu/armv7/tegra20/display.c

@@ -328,7 +328,7 @@ static int tegra_display_decode_config(const void *blob,
 	rgb = fdt_subnode_offset(blob, node, "rgb");
 
 	config->panel_node = fdtdec_lookup_phandle(blob, rgb, "nvidia,panel");
-	if (!config->panel_node < 0) {
+	if (config->panel_node < 0) {
 		debug("%s: Cannot find panel information\n", __func__);
 		return -1;
 	}

arch/arm/cpu/armv7/zynq/u-boot.lds

@@ -18,6 +18,7 @@ SECTIONS
 	.text :
 	{
 		*(.__image_copy_start)
+		*(.vectors)
 		CPUDIR/start.o (.text*)
 		*(.text*)
 	}

arch/arm/cpu/armv8/transition.S

@@ -43,7 +43,7 @@ ENTRY(armv8_switch_to_el1)
 	mrs	x0, cnthctl_el2
 	orr	x0, x0, #0x3		/* Enable EL1 access to timers */
 	msr	cnthctl_el2, x0
-	msr	cntvoff_el2, x0
+	msr	cntvoff_el2, xzr
 	mrs	x0, cntkctl_el1
 	orr	x0, x0, #0x3		/* Enable EL0 access to timers */
 	msr	cntkctl_el1, x0

arch/arm/dts/Makefile

@@ -6,7 +6,8 @@ dtb-$(CONFIG_EXYNOS4) += exynos4210-origen.dtb \
 dtb-$(CONFIG_EXYNOS5) += exynos5250-arndale.dtb \
 	exynos5250-snow.dtb \
 	exynos5250-smdk5250.dtb \
-	exynos5420-smdk5420.dtb
+	exynos5420-smdk5420.dtb \
+	exynos5420-peach-pit.dtb
 dtb-$(CONFIG_MX6) += imx6q-sabreauto.dtb
 dtb-$(CONFIG_TEGRA) += tegra20-harmony.dtb \
 	tegra20-medcom-wide.dtb \

arch/arm/dts/exynos5420-peach-pit.dts

+/*
+ * SAMSUNG/GOOGLE Peach-Pit board device tree source
+ *
+ * Copyright (c) 2013 Samsung Electronics Co., Ltd.
+ *		http://www.samsung.com
+ *
+ * SPDX-License-Identifier:	GPL-2.0+
+ */
+
+/dts-v1/;
+/include/ "exynos54xx.dtsi"
+
+/ {
+	model = "Samsung/Google Peach Pit board based on Exynos5420";
+
+	compatible = "google,pit-rev#", "google,pit",
+		"google,peach", "samsung,exynos5420", "samsung,exynos5";
+
+	config {
+		google,bad-wake-gpios = <&gpio 0x56 0>; /* gpx0-6 */
+		hwid = "PIT TEST A-A 7848";
+		lazy-init = <1>;
+	};
+
+	aliases {
+		serial0 = "/serial@12C30000";
+		console = "/serial@12C30000";
+		pmic = "/i2c@12ca0000";
+	};
+
+	dmc {
+		mem-manuf = "samsung";
+		mem-type = "ddr3";
+		clock-frequency = <800000000>;
+		arm-frequency = <1700000000>;
+	};
+
+	tmu@10060000 {
+		samsung,min-temp	= <25>;
+		samsung,max-temp	= <125>;
+		samsung,start-warning	= <95>;
+		samsung,start-tripping	= <105>;
+		samsung,hw-tripping	= <110>;
+		samsung,efuse-min-value	= <40>;
+		samsung,efuse-value	= <55>;
+		samsung,efuse-max-value	= <100>;
+		samsung,slope		= <274761730>;
+		samsung,dc-value	= <25>;
+	};
+
+	/* MAX77802 is on i2c bus 4 */
+	i2c@12ca0000 {
+		clock-frequency = <400000>;
+		power-regulator@9 {
+			compatible = "maxim,max77802-pmic";
+			reg = <0x9>;
+		};
+	};
+
+	i2c@12cd0000 { /* i2c7 */
+		clock-frequency = <100000>;
+	       soundcodec@20 {
+	              reg = <0x20>;
+	              compatible = "maxim,max98090-codec";
+	       };
+	};
+
+        sound@3830000 {
+                samsung,codec-type = "max98090";
+        };
+
+	i2c@12e10000 { /* i2c9 */
+		clock-frequency = <400000>;
+                tpm@20 {
+                        compatible = "infineon,slb9645-tpm";
+                        reg = <0x20>;
+		};
+	};
+
+	spi@12d30000 { /* spi1 */
+		spi-max-frequency = <50000000>;
+		firmware_storage_spi: flash@0 {
+			reg = <0>;
+
+			/*
+			 * A region for the kernel to store a panic event
+			 * which the firmware will add to the log.
+			*/
+			elog-panic-event-offset = <0x01e00000 0x100000>;
+
+			elog-shrink-size = <0x400>;
+			elog-full-threshold = <0xc00>;
+		};
+	};
+
+	spi@12d40000 { /* spi2 */
+		spi-max-frequency = <4000000>;
+		spi-deactivate-delay = <200>;
+		cros-ec@0 {
+			reg = <0>;
+			compatible = "google,cros-ec";
+			spi-half-duplex;
+			spi-max-timeout-ms = <1100>;
+			spi-frame-header = <0xec>;
+			ec-interrupt = <&gpio 93 1>; /* GPX1_5 */
+
+			/*
+			 * This describes the flash memory within the EC. Note
+			 * that the STM32L flash erases to 0, not 0xff.
+			 */
+			#address-cells = <1>;
+			#size-cells = <1>;
+			flash@8000000 {
+				reg = <0x08000000 0x20000>;
+				erase-value = <0>;
+			};
+		};
+	};
+
+	xhci@12000000 {
+		samsung,vbus-gpio = <&gpio 0x40 0>; /* H00 */
+	};
+
+	xhci@12400000 {
+		samsung,vbus-gpio = <&gpio 0x41 0>; /* H01 */
+	};
+};