Merge branch 'u-boot/master' into 'u-boot-arm/master'

17 files changed, 1119 insertions, 26 deletions

diff --git a/arch/arm/cpu/armv8/fsl-lsch3/Makefile b/arch/arm/cpu/armv8/fsl-lsch3/Makefile
index f920eebc56..9f7815bd52 100644
--- a/arch/arm/cpu/armv8/fsl-lsch3/Makefile
+++ b/arch/arm/cpu/armv8/fsl-lsch3/Makefile
@@ -6,6 +6,8 @@
 
 obj-y += cpu.o
 obj-y += lowlevel.o
+obj-y += soc.o
 obj-y += speed.o
+obj-$(CONFIG_SYS_HAS_SERDES) += fsl_lsch3_serdes.o ls2085a_serdes.o
 obj-$(CONFIG_MP) += mp.o
 obj-$(CONFIG_OF_LIBFDT) += fdt.o
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/README b/arch/arm/cpu/armv8/fsl-lsch3/README
index cc47466112..37f07fbb76 100644
--- a/arch/arm/cpu/armv8/fsl-lsch3/README
+++ b/arch/arm/cpu/armv8/fsl-lsch3/README
@@ -8,3 +8,141 @@ Freescale LayerScape with Chassis Generation 3
 
 This architecture supports Freescale ARMv8 SoCs with Chassis generation 3,
 for example LS2085A.
+
+Flash Layout
+============
+
+(1) A typical layout of various images (including Linux and other firmware images)
+   is shown below considering a 32MB NOR flash device present on most
+   pre-silicon platforms (simulator and emulator):
+
+	-------------------------
+	| 	FIT Image	|
+	| (linux + DTB + RFS)	|
+	------------------------- ----> 0x0120_0000
+	| 	Debug Server FW |
+	------------------------- ----> 0x00C0_0000
+	|	AIOP FW 	|
+	------------------------- ----> 0x0070_0000
+	|	MC FW 		|
+	------------------------- ----> 0x006C_0000
+	| 	MC DPL Blob 	|
+	------------------------- ----> 0x0020_0000
+	| 	BootLoader + Env|
+	------------------------- ----> 0x0000_1000
+	|	PBI 		|
+	------------------------- ----> 0x0000_0080
+	|	RCW 		|
+	------------------------- ----> 0x0000_0000
+
+	32-MB NOR flash layout for pre-silicon platforms (simulator and emulator)
+
+(2) A typical layout of various images (including Linux and other firmware images)
+    is shown below considering a 128MB NOR flash device present on QDS and RDB
+    boards:
+	----------------------------------------- ----> 0x5_8800_0000 ---
+	|	.. Unused .. (7M)		|			|
+	----------------------------------------- ----> 0x5_8790_0000	|
+	| FIT Image (linux + DTB + RFS)	(40M)	|			|
+	----------------------------------------- ----> 0x5_8510_0000	|
+	|	PHY firmware (2M)	 	|			|
+	----------------------------------------- ----> 0x5_84F0_0000	| 64K
+	|	Debug Server FW (2M)		|			| Alt
+	----------------------------------------- ----> 0x5_84D0_0000	| Bank
+	|	AIOP FW (4M)			|			|
+	----------------------------------------- ----> 0x5_8490_0000 (vbank4)
+	|	MC DPC Blob (1M) 		|			|
+	----------------------------------------- ----> 0x5_8480_0000	|
+	|	MC DPL Blob (1M)		|			|
+	----------------------------------------- ----> 0x5_8470_0000	|
+	| 	MC FW (4M)			|			|
+	----------------------------------------- ----> 0x5_8430_0000	|
+	|	BootLoader Environment (1M) 	|			|
+	----------------------------------------- ----> 0x5_8420_0000	|
+	|	BootLoader (1M)			|			|
+	----------------------------------------- ----> 0x5_8410_0000	|
+	|	RCW and PBI (1M) 		|			|
+	----------------------------------------- ----> 0x5_8400_0000 ---
+	|	.. Unused .. (7M)		|			|
+	----------------------------------------- ----> 0x5_8390_0000	|
+	| FIT Image (linux + DTB + RFS)	(40M)	|			|
+	----------------------------------------- ----> 0x5_8110_0000	|
+	|	PHY firmware (2M)	 	|			|
+	----------------------------------------- ----> 0x5_80F0_0000	| 64K
+	|	Debug Server FW (2M)		|			| Bank
+	----------------------------------------- ----> 0x5_80D0_0000	|
+	|	AIOP FW (4M)			|			|
+	----------------------------------------- ----> 0x5_8090_0000 (vbank0)
+	|	MC DPC Blob (1M) 		|			|
+	----------------------------------------- ----> 0x5_8080_0000	|
+	|	MC DPL Blob (1M)		|			|
+	----------------------------------------- ----> 0x5_8070_0000	|
+	| 	MC FW (4M)			|			|
+	----------------------------------------- ----> 0x5_8030_0000	|
+	|	BootLoader Environment (1M) 	|			|
+	----------------------------------------- ----> 0x5_8020_0000	|
+	|	BootLoader (1M)			|			|
+	----------------------------------------- ----> 0x5_8010_0000	|
+	|	RCW and PBI (1M) 		|			|
+	----------------------------------------- ----> 0x5_8000_0000 ---
+
+	128-MB NOR flash layout for QDS and RDB boards
+
+Environment Variables
+=====================
+mcboottimeout:	MC boot timeout in milliseconds. If this variable is not defined
+		the value CONFIG_SYS_LS_MC_BOOT_TIMEOUT_MS will be assumed.
+
+mcmemsize:	MC DRAM block size. If this variable is not defined, the value
+		CONFIG_SYS_LS_MC_DRAM_BLOCK_MIN_SIZE will be assumed.
+
+Booting from NAND
+-------------------
+Booting from NAND requires two images, RCW and u-boot-with-spl.bin.
+The difference between NAND boot RCW image and NOR boot image is the PBI
+command sequence. Below is one example for PBI commands for QDS which uses
+NAND device with 2KB/page, block size 128KB.
+
+1) CCSR 4-byte write to 0x00e00404, data=0x00000000
+2) CCSR 4-byte write to 0x00e00400, data=0x1800a000
+The above two commands set bootloc register to 0x00000000_1800a000 where
+the u-boot code will be running in OCRAM.
+
+3) Block Copy: SRC=0x0107, SRC_ADDR=0x00020000, DEST_ADDR=0x1800a000,
+BLOCK_SIZE=0x00014000
+This command copies u-boot image from NAND device into OCRAM. The values need
+to adjust accordingly.
+
+SRC		should match the cfg_rcw_src, the reset config pins. It depends
+		on the NAND device. See reference manual for cfg_rcw_src.
+SRC_ADDR	is the offset of u-boot-with-spl.bin image in NAND device. In
+		the example above, 128KB. For easy maintenance, we put it at
+		the beginning of next block from RCW.
+DEST_ADDR	is fixed at 0x1800a000, matching bootloc set above.
+BLOCK_SIZE	is the size to be copied by PBI.
+
+RCW image should be written to the beginning of NAND device. Example of using
+u-boot command
+
+nand write <rcw image in memory> 0 <size of rcw image>
+
+To form the NAND image, build u-boot with NAND config, for example,
+ls2085aqds_nand_defconfig. The image needed is u-boot-with-spl.bin.
+The u-boot image should be written to match SRC_ADDR, in above example 0x20000.
+
+nand write <u-boot image in memory> 200000 <size of u-boot image>
+
+With these two images in NAND device, the board can boot from NAND.
+
+Another example for RDB boards,
+
+1) CCSR 4-byte write to 0x00e00404, data=0x00000000
+2) CCSR 4-byte write to 0x00e00400, data=0x1800a000
+3) Block Copy: SRC=0x0119, SRC_ADDR=0x00080000, DEST_ADDR=0x1800a000,
+BLOCK_SIZE=0x00014000
+
+nand write <rcw image in memory> 0 <size of rcw image>
+nand write <u-boot image in memory> 80000 <size of u-boot image>
+
+Notice the difference from QDS is SRC, SRC_ADDR and the offset of u-boot image
+to match board NAND device with 4KB/page, block size 512KB.
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/cpu.c b/arch/arm/cpu/armv8/fsl-lsch3/cpu.c
index 49974878b9..67145778b9 100644
--- a/arch/arm/cpu/armv8/fsl-lsch3/cpu.c
+++ b/arch/arm/cpu/armv8/fsl-lsch3/cpu.c
@@ -10,7 +10,12 @@
 #include <asm/armv8/mmu.h>
 #include <asm/io.h>
 #include <asm/arch-fsl-lsch3/immap_lsch3.h>
+#include <fsl_debug_server.h>
 #include <fsl-mc/fsl_mc.h>
+#include <asm/arch/fsl_serdes.h>
+#ifdef CONFIG_FSL_ESDHC
+#include <fsl_esdhc.h>
+#endif
 #include "cpu.h"
 #include "mp.h"
 #include "speed.h"
@@ -24,8 +29,9 @@ DECLARE_GLOBAL_DATA_PTR;
  * levels of translation tables here to cover 40-bit address space.
  * We use 4KB granule size, with 40 bits physical address, T0SZ=24
  * Level 0 IA[39], table address @0
- * Level 1 IA[31:30], table address @01000, 0x2000
- * Level 2 IA[29:21], table address @0x3000
+ * Level 1 IA[31:30], table address @0x1000, 0x2000
+ * Level 2 IA[29:21], table address @0x3000, 0x4000
+ * Address above 0x5000 is free for other purpose.
  */
 
 #define SECTION_SHIFT_L0	39UL
@@ -60,12 +66,12 @@ static inline void early_mmu_setup(void)
 {
 	int el;
 	u64 i;
-	u64 section_l1t0, section_l1t1, section_l2;
+	u64 section_l1t0, section_l1t1, section_l2t0, section_l2t1;
 	u64 *level0_table = (u64 *)CONFIG_SYS_FSL_OCRAM_BASE;
 	u64 *level1_table_0 = (u64 *)(CONFIG_SYS_FSL_OCRAM_BASE + 0x1000);
 	u64 *level1_table_1 = (u64 *)(CONFIG_SYS_FSL_OCRAM_BASE + 0x2000);
-	u64 *level2_table = (u64 *)(CONFIG_SYS_FSL_OCRAM_BASE + 0x3000);
-
+	u64 *level2_table_0 = (u64 *)(CONFIG_SYS_FSL_OCRAM_BASE + 0x3000);
+	u64 *level2_table_1 = (u64 *)(CONFIG_SYS_FSL_OCRAM_BASE + 0x4000);
 
 	level0_table[0] =
 		(u64)level1_table_0 | PMD_TYPE_TABLE;
@@ -79,21 +85,25 @@ static inline void early_mmu_setup(void)
 	 */
 	section_l1t0 = 0;
 	section_l1t1 = BLOCK_SIZE_L0;
-	section_l2 = 0;
+	section_l2t0 = 0;
+	section_l2t1 = CONFIG_SYS_FLASH_BASE;
 	for (i = 0; i < 512; i++) {
 		set_pgtable_section(level1_table_0, i, section_l1t0,
 				    MT_DEVICE_NGNRNE);
 		set_pgtable_section(level1_table_1, i, section_l1t1,
 				    MT_NORMAL);
-		set_pgtable_section(level2_table, i, section_l2,
+		set_pgtable_section(level2_table_0, i, section_l2t0,
+				    MT_DEVICE_NGNRNE);
+		set_pgtable_section(level2_table_1, i, section_l2t1,
 				    MT_DEVICE_NGNRNE);
 		section_l1t0 += BLOCK_SIZE_L1;
 		section_l1t1 += BLOCK_SIZE_L1;
-		section_l2 += BLOCK_SIZE_L2;
+		section_l2t0 += BLOCK_SIZE_L2;
+		section_l2t1 += BLOCK_SIZE_L2;
 	}
 
 	level1_table_0[0] =
-		(u64)level2_table | PMD_TYPE_TABLE;
+		(u64)level2_table_0 | PMD_TYPE_TABLE;
 	level1_table_0[1] =
 		0x40000000 | PMD_SECT_AF | PMD_TYPE_SECT |
 		PMD_ATTRINDX(MT_DEVICE_NGNRNE);
@@ -104,17 +114,34 @@ static inline void early_mmu_setup(void)
 		0xc0000000 | PMD_SECT_AF | PMD_TYPE_SECT |
 		PMD_ATTRINDX(MT_NORMAL);
 
-	/* Rewrite table to enable cache */
-	set_pgtable_section(level2_table,
+	/* Rewerite table to enable cache for OCRAM */
+	set_pgtable_section(level2_table_0,
 			    CONFIG_SYS_FSL_OCRAM_BASE >> SECTION_SHIFT_L2,
 			    CONFIG_SYS_FSL_OCRAM_BASE,
 			    MT_NORMAL);
-	for (i = CONFIG_SYS_IFC_BASE >> SECTION_SHIFT_L2;
-	     i < (CONFIG_SYS_IFC_BASE + CONFIG_SYS_IFC_SIZE)
-	     >> SECTION_SHIFT_L2; i++) {
-		section_l2 = i << SECTION_SHIFT_L2;
-		set_pgtable_section(level2_table, i,
-				    section_l2, MT_NORMAL);
+
+#if defined(CONFIG_SYS_NOR0_CSPR_EARLY) && defined(CONFIG_SYS_NOR_AMASK_EARLY)
+	/* Rewrite table to enable cache for two entries (4MB) */
+	section_l2t1 = CONFIG_SYS_IFC_BASE;
+	set_pgtable_section(level2_table_0,
+			    section_l2t1 >> SECTION_SHIFT_L2,
+			    section_l2t1,
+			    MT_NORMAL);
+	section_l2t1 += BLOCK_SIZE_L2;
+	set_pgtable_section(level2_table_0,
+			    section_l2t1 >> SECTION_SHIFT_L2,
+			    section_l2t1,
+			    MT_NORMAL);
+#endif
+
+	/* Create a mapping for 256MB IFC region to final flash location */
+	level1_table_0[CONFIG_SYS_FLASH_BASE >> SECTION_SHIFT_L1] =
+		(u64)level2_table_1 | PMD_TYPE_TABLE;
+	section_l2t1 = CONFIG_SYS_IFC_BASE;
+	for (i = 0; i < 0x10000000 >> SECTION_SHIFT_L2; i++) {
+		set_pgtable_section(level2_table_1, i,
+				    section_l2t1, MT_DEVICE_NGNRNE);
+		section_l2t1 += BLOCK_SIZE_L2;
 	}
 
 	el = current_el();
@@ -347,6 +374,7 @@ u32 fsl_qoriq_core_to_type(unsigned int core)
 #ifdef CONFIG_DISPLAY_CPUINFO
 int print_cpuinfo(void)
 {
+	struct ccsr_gur __iomem *gur = (void *)(CONFIG_SYS_FSL_GUTS_ADDR);
 	struct sys_info sysinfo;
 	char buf[32];
 	unsigned int i, core;
@@ -370,21 +398,40 @@ int print_cpuinfo(void)
 	printf("     DP-DDR:   %-4s MHz", strmhz(buf, sysinfo.freq_ddrbus2));
 	puts("\n");
 
+	/* Display the RCW, so that no one gets confused as to what RCW
+	 * we're actually using for this boot.
+	 */
+	puts("Reset Configuration Word (RCW):");
+	for (i = 0; i < ARRAY_SIZE(gur->rcwsr); i++) {
+		u32 rcw = in_le32(&gur->rcwsr[i]);
+
+		if ((i % 4) == 0)
+			printf("\n       %02x:", i * 4);
+		printf(" %08x", rcw);
+	}
+	puts("\n");
+
 	return 0;
 }
 #endif
 
+#ifdef CONFIG_FSL_ESDHC
+int cpu_mmc_init(bd_t *bis)
+{
+	return fsl_esdhc_mmc_init(bis);
+}
+#endif
+
 int cpu_eth_init(bd_t *bis)
 {
 	int error = 0;
 
 #ifdef CONFIG_FSL_MC_ENET
-	error = mc_init(bis);
+	error = fsl_mc_ldpaa_init(bis);
 #endif
 	return error;
 }
 
-
 int arch_early_init_r(void)
 {
 	int rv;
@@ -393,5 +440,43 @@ int arch_early_init_r(void)
 	if (rv)
 		printf("Did not wake secondary cores\n");
 
+#ifdef CONFIG_SYS_HAS_SERDES
+	fsl_serdes_init();
+#endif
+	return 0;
+}
+
+int timer_init(void)
+{
+	u32 __iomem *cntcr = (u32 *)CONFIG_SYS_FSL_TIMER_ADDR;
+	u32 __iomem *cltbenr = (u32 *)CONFIG_SYS_FSL_PMU_CLTBENR;
+#ifdef COUNTER_FREQUENCY_REAL
+	unsigned long cntfrq = COUNTER_FREQUENCY_REAL;
+
+	/* Update with accurate clock frequency */
+	asm volatile("msr cntfrq_el0, %0" : : "r" (cntfrq) : "memory");
+#endif
+
+	/* Enable timebase for all clusters.
+	 * It is safe to do so even some clusters are not enabled.
+	 */
+	out_le32(cltbenr, 0xf);
+
+	/* Enable clock for timer
+	 * This is a global setting.
+	 */
+	out_le32(cntcr, 0x1);
+
 	return 0;
 }
+
+void reset_cpu(ulong addr)
+{
+	u32 __iomem *rstcr = (u32 *)CONFIG_SYS_FSL_RST_ADDR;
+	u32 val;
+
+	/* Raise RESET_REQ_B */
+	val = in_le32(rstcr);
+	val |= 0x02;
+	out_le32(rstcr, val);
+}
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/fdt.c b/arch/arm/cpu/armv8/fsl-lsch3/fdt.c
index 7eb9b6aa4b..d37002333c 100644
--- a/arch/arm/cpu/armv8/fsl-lsch3/fdt.c
+++ b/arch/arm/cpu/armv8/fsl-lsch3/fdt.c
@@ -7,6 +7,9 @@
 #include <common.h>
 #include <libfdt.h>
 #include <fdt_support.h>
+#ifdef CONFIG_FSL_ESDHC
+#include <fsl_esdhc.h>
+#endif
 #include "mp.h"
 
 #ifdef CONFIG_MP
@@ -62,7 +65,11 @@ void ft_cpu_setup(void *blob, bd_t *bd)
 #endif
 
 #ifdef CONFIG_SYS_NS16550
-	do_fixup_by_compat_u32(blob, "ns16550",
+	do_fixup_by_compat_u32(blob, "fsl,ns16550",
 			       "clock-frequency", CONFIG_SYS_NS16550_CLK, 1);
 #endif
+
+#if defined(CONFIG_FSL_ESDHC)
+	fdt_fixup_esdhc(blob, bd);
+#endif
 }
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/fsl_lsch3_serdes.c b/arch/arm/cpu/armv8/fsl-lsch3/fsl_lsch3_serdes.c
new file mode 100644
index 0000000000..02ca126ab8
--- /dev/null
+++ b/arch/arm/cpu/armv8/fsl-lsch3/fsl_lsch3_serdes.c
@@ -0,0 +1,115 @@
+/*
+ * Copyright 2015 Freescale Semiconductor, Inc.
+ *
+ * SPDX-License-Identifier:	GPL-2.0+
+ */
+
+#include <common.h>
+#include <asm/io.h>
+#include <asm/errno.h>
+#include <asm/arch/fsl_serdes.h>
+#include <asm/arch-fsl-lsch3/immap_lsch3.h>
+#include <fsl-mc/ldpaa_wriop.h>
+
+#ifdef CONFIG_SYS_FSL_SRDS_1
+static u8 serdes1_prtcl_map[SERDES_PRCTL_COUNT];
+#endif
+#ifdef CONFIG_SYS_FSL_SRDS_2
+static u8 serdes2_prtcl_map[SERDES_PRCTL_COUNT];
+#endif
+
+int is_serdes_configured(enum srds_prtcl device)
+{
+	int ret = 0;
+
+#ifdef CONFIG_SYS_FSL_SRDS_1
+	ret |= serdes1_prtcl_map[device];
+#endif
+#ifdef CONFIG_SYS_FSL_SRDS_2
+	ret |= serdes2_prtcl_map[device];
+#endif
+
+	return !!ret;
+}
+
+int serdes_get_first_lane(u32 sd, enum srds_prtcl device)
+{
+	struct ccsr_gur __iomem *gur = (void *)(CONFIG_SYS_FSL_GUTS_ADDR);
+	u32 cfg = in_le32(&gur->rcwsr[28]);
+	int i;
+
+	switch (sd) {
+#ifdef CONFIG_SYS_FSL_SRDS_1
+	case FSL_SRDS_1:
+		cfg &= FSL_CHASSIS3_RCWSR28_SRDS1_PRTCL_MASK;
+		cfg >>= FSL_CHASSIS3_RCWSR28_SRDS1_PRTCL_SHIFT;
+		break;
+#endif
+#ifdef CONFIG_SYS_FSL_SRDS_2
+	case FSL_SRDS_2:
+		cfg &= FSL_CHASSIS3_RCWSR28_SRDS2_PRTCL_MASK;
+		cfg >>= FSL_CHASSIS3_RCWSR28_SRDS2_PRTCL_SHIFT;
+		break;
+#endif
+	default:
+		printf("invalid SerDes%d\n", sd);
+		break;
+	}
+	/* Is serdes enabled at all? */
+	if (cfg == 0)
+		return -ENODEV;
+
+	for (i = 0; i < SRDS_MAX_LANES; i++) {
+		if (serdes_get_prtcl(sd, cfg, i) == device)
+			return i;
+	}
+
+	return -ENODEV;
+}
+
+void serdes_init(u32 sd, u32 sd_addr, u32 sd_prctl_mask, u32 sd_prctl_shift,
+		u8 serdes_prtcl_map[SERDES_PRCTL_COUNT])
+{
+	struct ccsr_gur __iomem *gur = (void *)(CONFIG_SYS_FSL_GUTS_ADDR);
+	u32 cfg;
+	int lane;
+
+	memset(serdes_prtcl_map, 0, sizeof(serdes_prtcl_map));
+
+	cfg = in_le32(&gur->rcwsr[28]) & sd_prctl_mask;
+	cfg >>= sd_prctl_shift;
+	printf("Using SERDES%d Protocol: %d (0x%x)\n", sd + 1, cfg, cfg);
+
+	if (!is_serdes_prtcl_valid(sd, cfg))
+		printf("SERDES%d[PRTCL] = 0x%x is not valid\n", sd + 1, cfg);
+
+	for (lane = 0; lane < SRDS_MAX_LANES; lane++) {
+		enum srds_prtcl lane_prtcl = serdes_get_prtcl(sd, cfg, lane);
+		if (unlikely(lane_prtcl >= SERDES_PRCTL_COUNT))
+			debug("Unknown SerDes lane protocol %d\n", lane_prtcl);
+		else {
+			serdes_prtcl_map[lane_prtcl] = 1;
+#ifdef CONFIG_FSL_MC_ENET
+			wriop_init_dpmac(sd, lane + 1, (int)lane_prtcl);
+#endif
+		}
+	}
+}
+
+void fsl_serdes_init(void)
+{
+#ifdef CONFIG_SYS_FSL_SRDS_1
+	serdes_init(FSL_SRDS_1,
+		    CONFIG_SYS_FSL_LSCH3_SERDES_ADDR,
+		    FSL_CHASSIS3_RCWSR28_SRDS1_PRTCL_MASK,
+		    FSL_CHASSIS3_RCWSR28_SRDS1_PRTCL_SHIFT,
+		    serdes1_prtcl_map);
+#endif
+#ifdef CONFIG_SYS_FSL_SRDS_2
+	serdes_init(FSL_SRDS_2,
+		    CONFIG_SYS_FSL_LSCH3_SERDES_ADDR + FSL_SRDS_2 * 0x10000,
+		    FSL_CHASSIS3_RCWSR28_SRDS2_PRTCL_MASK,
+		    FSL_CHASSIS3_RCWSR28_SRDS2_PRTCL_SHIFT,
+		    serdes2_prtcl_map);
+#endif
+}
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/lowlevel.S b/arch/arm/cpu/armv8/fsl-lsch3/lowlevel.S
index 886576ef99..018c61742e 100644
--- a/arch/arm/cpu/armv8/fsl-lsch3/lowlevel.S
+++ b/arch/arm/cpu/armv8/fsl-lsch3/lowlevel.S
@@ -15,6 +15,15 @@
 ENTRY(lowlevel_init)
 	mov	x29, lr			/* Save LR */
 
+	/* Add fully-coherent masters to DVM domain */
+	ldr	x1, =CCI_MN_BASE
+	ldr	x2, [x1, #CCI_MN_RNF_NODEID_LIST]
+	str	x2, [x1, #CCI_MN_DVM_DOMAIN_CTL_SET]
+1:	ldr	x3, [x1, #CCI_MN_DVM_DOMAIN_CTL_SET]
+	mvn	x0, x3
+	tst	x0, x3		/* Wait for domain addition to complete */
+	b.ne	1b
+
 	/* Set the SMMU page size in the sACR register */
 	ldr	x1, =SMMU_BASE
 	ldr	w0, [x1, #0x10]
@@ -224,6 +233,9 @@ ENTRY(secondary_boot_func)
 	/* physical address of this cpus spin table element */
 	add	x11, x1, x0
 
+	ldr	x0, =__real_cntfrq
+	ldr	x0, [x0]
+	msr	cntfrq_el0, x0	/* set with real frequency */
 	str	x9, [x11, #16]	/* LPID */
 	mov	x4, #1
 	str	x4, [x11, #8]	/* STATUS */
@@ -275,6 +287,9 @@ ENDPROC(secondary_switch_to_el1)
 
 	/* 64 bit alignment for elements accessed as data */
 	.align 4
+	.global __real_cntfrq
+__real_cntfrq:
+	.quad COUNTER_FREQUENCY
 	.globl __secondary_boot_code_size
 	.type __secondary_boot_code_size, %object
 	/* Secondary Boot Code ends here */
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/ls2085a_serdes.c b/arch/arm/cpu/armv8/fsl-lsch3/ls2085a_serdes.c
new file mode 100644
index 0000000000..098745bd74
--- /dev/null
+++ b/arch/arm/cpu/armv8/fsl-lsch3/ls2085a_serdes.c
@@ -0,0 +1,117 @@
+/*
+ * Copyright 2015 Freescale Semiconductor, Inc.
+ *
+ * SPDX-License-Identifier:	GPL-2.0+
+ */
+
+#include <common.h>
+#include <asm/arch/fsl_serdes.h>
+#include <asm/arch-fsl-lsch3/immap_lsch3.h>
+
+struct serdes_config {
+	u8 protocol;
+	u8 lanes[SRDS_MAX_LANES];
+};
+
+static struct serdes_config serdes1_cfg_tbl[] = {
+	/* SerDes 1 */
+	{0x03, {PCIE1, PCIE1, PCIE1, PCIE1, PCIE2, PCIE2, PCIE2, PCIE2 } },
+	{0x05, {PCIE2, PCIE2, PCIE2, PCIE2, SGMII4, SGMII3, SGMII2, SGMII1 } },
+	{0x07, {SGMII8, SGMII7, SGMII6, SGMII5, SGMII4, SGMII3, SGMII2,
+		SGMII1 } },
+	{0x09, {SGMII8, SGMII7, SGMII6, SGMII5, SGMII4, SGMII3, SGMII2,
+		SGMII1 } },
+	{0x0A, {SGMII8, SGMII7, SGMII6, SGMII5, SGMII4, SGMII3, SGMII2,
+		SGMII1 } },
+	{0x0C, {SGMII8, SGMII7, SGMII6, SGMII5, SGMII4, SGMII3, SGMII2,
+		SGMII1 } },
+	{0x0E, {SGMII8, SGMII7, SGMII6, SGMII5, SGMII4, SGMII3, SGMII2,
+		SGMII1 } },
+	{0x26, {SGMII8, SGMII7, SGMII6, SGMII5, SGMII4, SGMII3, XFI2, XFI1 } },
+	{0x28, {SGMII8, SGMII7, SGMII6, SGMII5, XFI4, XFI3, XFI2, XFI1 } },
+	{0x2A, {XFI8, XFI7, XFI6, XFI5, XFI4, XFI3, XFI2, XFI1 } },
+	{0x2B, {SGMII8, SGMII7, SGMII6, SGMII5, XAUI1, XAUI1, XAUI1, XAUI1  } },
+	{0x32, {XAUI2, XAUI2, XAUI2, XAUI2, XAUI1, XAUI1, XAUI1, XAUI1  } },
+	{0x33, {PCIE2, PCIE2, PCIE2, PCIE2, QSGMII_D, QSGMII_C, QSGMII_B,
+		QSGMII_A} },
+	{0x35, {QSGMII_D, QSGMII_C, QSGMII_B, PCIE2, XFI4, XFI3, XFI2, XFI1 } },
+		{}
+};
+static struct serdes_config serdes2_cfg_tbl[] = {
+	/* SerDes 2 */
+	{0x07, {SGMII9, SGMII10, SGMII11, SGMII12, SGMII13, SGMII14, SGMII15,
+		SGMII16 } },
+	{0x09, {SGMII9, SGMII10, SGMII11, SGMII12, SGMII13, SGMII14, SGMII15,
+		SGMII16 } },
+	{0x0A, {SGMII9, SGMII10, SGMII11, SGMII12, SGMII13, SGMII14, SGMII15,
+		SGMII16 } },
+	{0x0C, {SGMII9, SGMII10, SGMII11, SGMII12, SGMII13, SGMII14, SGMII15,
+		SGMII16 } },
+	{0x0E, {SGMII9, SGMII10, SGMII11, SGMII12, SGMII13, SGMII14, SGMII15,
+		SGMII16 } },
+	{0x3D, {PCIE3, PCIE3, PCIE3, PCIE3, PCIE3, PCIE3, PCIE3, PCIE3 } },
+	{0x3E, {PCIE3, PCIE3, PCIE3, PCIE3, PCIE3, PCIE3, PCIE3, PCIE3 } },
+	{0x3F, {PCIE3, PCIE3, PCIE3, PCIE3, PCIE4, PCIE4, PCIE4, PCIE4 } },
+	{0x40, {PCIE3, PCIE3, PCIE3, PCIE3, PCIE4, PCIE4, PCIE4, PCIE4 } },
+	{0x41, {PCIE3, PCIE3, PCIE3, PCIE3, PCIE4, PCIE4, SATA1, SATA2 } },
+	{0x42, {PCIE3, PCIE3, PCIE3, PCIE3, PCIE4, PCIE4, SATA1, SATA2 } },
+	{0x43, {PCIE3, PCIE3, PCIE3, PCIE3, NONE, NONE, SATA1, SATA2 } },
+	{0x44, {PCIE3, PCIE3, PCIE3, PCIE3, NONE, NONE, SATA1, SATA2 } },
+	{0x45, {PCIE3, SGMII10, SGMII11, SGMII12, PCIE4, SGMII14, SGMII15,
+		SGMII16 } },
+	{0x47, {SGMII9, SGMII10, SGMII11, SGMII12, PCIE4, PCIE4, PCIE4,
+		PCIE4 } },
+	{0x49, {SGMII9, SGMII10, SGMII11, SGMII12, PCIE4, PCIE4, SATA1,
+		SATA2 } },
+	{0x4A, {SGMII9, SGMII10, SGMII11, SGMII12, PCIE4, PCIE4, SATA1,
+		SATA2 } },
+	{}
+};
+
+static struct serdes_config *serdes_cfg_tbl[] = {
+	serdes1_cfg_tbl,
+	serdes2_cfg_tbl,
+};
+
+enum srds_prtcl serdes_get_prtcl(int serdes, int cfg, int lane)
+{
+	struct serdes_config *ptr;
+
+	if (serdes >= ARRAY_SIZE(serdes_cfg_tbl))
+		return 0;
+
+	ptr = serdes_cfg_tbl[serdes];
+	while (ptr->protocol) {
+		if (ptr->protocol == cfg)
+			return ptr->lanes[lane];
+		ptr++;
+	}
+
+	return 0;
+}
+
+int is_serdes_prtcl_valid(int serdes, u32 prtcl)
+{
+	int i;
+	struct serdes_config *ptr;
+
+	if (serdes >= ARRAY_SIZE(serdes_cfg_tbl))
+		return 0;
+
+	ptr = serdes_cfg_tbl[serdes];
+	while (ptr->protocol) {
+		if (ptr->protocol == prtcl)
+			break;
+		ptr++;
+	}
+
+	if (!ptr->protocol)
+		return 0;
+
+	for (i = 0; i < SRDS_MAX_LANES; i++) {
+		if (ptr->lanes[i] != NONE)
+			return 1;
+	}
+
+	return 0;
+}
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/mp.c b/arch/arm/cpu/armv8/fsl-lsch3/mp.c
index ce9c0c1bdb..da7853a5af 100644
--- a/arch/arm/cpu/armv8/fsl-lsch3/mp.c
+++ b/arch/arm/cpu/armv8/fsl-lsch3/mp.c
@@ -31,6 +31,13 @@ int fsl_lsch3_wake_seconday_cores(void)
 	int i, timeout = 10;
 	u64 *table = get_spin_tbl_addr();
 
+#ifdef COUNTER_FREQUENCY_REAL
+	/* update for secondary cores */
+	__real_cntfrq = COUNTER_FREQUENCY_REAL;
+	flush_dcache_range((unsigned long)&__real_cntfrq,
+			   (unsigned long)&__real_cntfrq + 8);
+#endif
+
 	cores = cpu_mask();
 	/* Clear spin table so that secondary processors
 	 * observe the correct value after waking up from wfe.
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/mp.h b/arch/arm/cpu/armv8/fsl-lsch3/mp.h
index 66144d6101..c985d6a6ba 100644
--- a/arch/arm/cpu/armv8/fsl-lsch3/mp.h
+++ b/arch/arm/cpu/armv8/fsl-lsch3/mp.h
@@ -26,6 +26,7 @@
 #define id_to_core(x)	((x & 3) | (x >> 6))
 #ifndef __ASSEMBLY__
 extern u64 __spin_table[];
+extern u64 __real_cntfrq;
 extern u64 *secondary_boot_code;
 extern size_t __secondary_boot_code_size;
 int fsl_lsch3_wake_seconday_cores(void);
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/soc.c b/arch/arm/cpu/armv8/fsl-lsch3/soc.c
new file mode 100644
index 0000000000..2538001bca
--- /dev/null
+++ b/arch/arm/cpu/armv8/fsl-lsch3/soc.c
@@ -0,0 +1,107 @@
+/*
+ * Copyright 2015 Freescale Semiconductor
+ *
+ * SPDX-License-Identifier:	GPL-2.0+
+ */
+
+#include <common.h>
+#include <fsl_ifc.h>
+#include <nand.h>
+#include <spl.h>
+#include <asm/arch-fsl-lsch3/soc.h>
+#include <asm/io.h>
+#include <asm/global_data.h>
+
+DECLARE_GLOBAL_DATA_PTR;
+
+static void erratum_a008751(void)
+{
+#ifdef CONFIG_SYS_FSL_ERRATUM_A008751
+	u32 __iomem *scfg = (u32 __iomem *)SCFG_BASE;
+
+	writel(0x27672b2a, scfg + SCFG_USB3PRM1CR / 4);
+#endif
+}
+
+static void erratum_rcw_src(void)
+{
+#if defined(CONFIG_SPL)
+	u32 __iomem *dcfg_ccsr = (u32 __iomem *)DCFG_BASE;
+	u32 __iomem *dcfg_dcsr = (u32 __iomem *)DCFG_DCSR_BASE;
+	u32 val;
+
+	val = in_le32(dcfg_ccsr + DCFG_PORSR1 / 4);
+	val &= ~DCFG_PORSR1_RCW_SRC;
+	val |= DCFG_PORSR1_RCW_SRC_NOR;
+	out_le32(dcfg_dcsr + DCFG_DCSR_PORCR1 / 4, val);
+#endif
+}
+
+#define I2C_DEBUG_REG 0x6
+#define I2C_GLITCH_EN 0x8
+/*
+ * This erratum requires setting glitch_en bit to enable
+ * digital glitch filter to improve clock stability.
+ */
+static void erratum_a009203(void)
+{
+	u8 __iomem *ptr;
+#ifdef CONFIG_SYS_I2C
+#ifdef I2C1_BASE_ADDR
+	ptr = (u8 __iomem *)(I2C1_BASE_ADDR + I2C_DEBUG_REG);
+
+	writeb(I2C_GLITCH_EN, ptr);
+#endif
+#ifdef I2C2_BASE_ADDR
+	ptr = (u8 __iomem *)(I2C2_BASE_ADDR + I2C_DEBUG_REG);
+
+	writeb(I2C_GLITCH_EN, ptr);
+#endif
+#ifdef I2C3_BASE_ADDR
+	ptr = (u8 __iomem *)(I2C3_BASE_ADDR + I2C_DEBUG_REG);
+
+	writeb(I2C_GLITCH_EN, ptr);
+#endif
+#ifdef I2C4_BASE_ADDR
+	ptr = (u8 __iomem *)(I2C4_BASE_ADDR + I2C_DEBUG_REG);
+
+	writeb(I2C_GLITCH_EN, ptr);
+#endif
+#endif
+}
+
+void fsl_lsch3_early_init_f(void)
+{
+	erratum_a008751();
+	erratum_rcw_src();
+	init_early_memctl_regs();	/* tighten IFC timing */
+	erratum_a009203();
+}
+
+#ifdef CONFIG_SPL_BUILD
+void board_init_f(ulong dummy)
+{
+	/* Clear global data */
+	memset((void *)gd, 0, sizeof(gd_t));
+
+	arch_cpu_init();
+	board_early_init_f();
+	timer_init();
+	env_init();
+	gd->baudrate = getenv_ulong("baudrate", 10, CONFIG_BAUDRATE);
+
+	serial_init();
+	console_init_f();
+	dram_init();
+
+	/* Clear the BSS. */
+	memset(__bss_start, 0, __bss_end - __bss_start);
+
+	board_init_r(NULL, 0);
+}
+
+u32 spl_boot_device(void)
+{
+	return BOOT_DEVICE_NAND;
+}
+#endif
diff --git a/arch/arm/cpu/armv8/fsl-lsch3/speed.c b/arch/arm/cpu/armv8/fsl-lsch3/speed.c
index 72cd999c5f..cac4f925a4 100644
--- a/arch/arm/cpu/armv8/fsl-lsch3/speed.c
+++ b/arch/arm/cpu/armv8/fsl-lsch3/speed.c
@@ -26,7 +26,7 @@ void get_sys_info(struct sys_info *sys_info)
 {
 	struct ccsr_gur __iomem *gur = (void *)(CONFIG_SYS_FSL_GUTS_ADDR);
 #ifdef CONFIG_FSL_IFC
-	struct fsl_ifc *ifc_regs = (void *)CONFIG_SYS_IFC_ADDR;
+	struct fsl_ifc ifc_regs = {(void *)CONFIG_SYS_IFC_ADDR, (void *)NULL};
 	u32 ccr;
 #endif
 	struct ccsr_clk_cluster_group __iomem *clk_grp[2] = {
@@ -86,6 +86,8 @@ void get_sys_info(struct sys_info *sys_info)
 	sys_info->freq_systembus *= (in_le32(&gur->rcwsr[0]) >>
 			FSL_CHASSIS3_RCWSR0_SYS_PLL_RAT_SHIFT) &
 			FSL_CHASSIS3_RCWSR0_SYS_PLL_RAT_MASK;
+	/* Platform clock is half of platform PLL */
+	sys_info->freq_systembus /= 2;
 	sys_info->freq_ddrbus *= (in_le32(&gur->rcwsr[0]) >>
 			FSL_CHASSIS3_RCWSR0_MEM_PLL_RAT_SHIFT) &
 			FSL_CHASSIS3_RCWSR0_MEM_PLL_RAT_MASK;
@@ -102,10 +104,7 @@ void get_sys_info(struct sys_info *sys_info)
 			 offsetof(struct ccsr_clk_cluster_group,
 				  pllngsr[i%3].gsr));
 		ratio[i] = (in_le32(offset) >> 1) & 0x3f;
-		if (ratio[i] > 4)
-			freq_c_pll[i] = sysclk * ratio[i];
-		else
-			freq_c_pll[i] = sys_info->freq_systembus * ratio[i];
+		freq_c_pll[i] = sysclk * ratio[i];
 	}
 
 	for_each_cpu(i, cpu, cpu_numcores(), cpu_mask()) {
@@ -119,7 +118,7 @@ void get_sys_info(struct sys_info *sys_info)
 	}
 
 #if defined(CONFIG_FSL_IFC)
-	ccr = in_le32(&ifc_regs->ifc_ccr);
+	ccr = in_le32(&ifc_regs.gregs->ifc_ccr);
 	ccr = ((ccr & IFC_CCR_CLK_DIV_MASK) >> IFC_CCR_CLK_DIV_SHIFT) + 1;
 
 	sys_info->freq_localbus = sys_info->freq_systembus / ccr;
diff --git a/arch/arm/cpu/armv8/generic_timer.c b/arch/arm/cpu/armv8/generic_timer.c
index 223b95e210..8e60baebc5 100644
--- a/arch/arm/cpu/armv8/generic_timer.c
+++ b/arch/arm/cpu/armv8/generic_timer.c
@@ -25,7 +25,18 @@ unsigned long get_tbclk(void)
 unsigned long timer_read_counter(void)
 {
 	unsigned long cntpct;
+#ifdef CONFIG_SYS_FSL_ERRATUM_A008585
+	/* This erratum number needs to be confirmed to match ARM document */
+	unsigned long temp;
+#endif
 	isb();
 	asm volatile("mrs %0, cntpct_el0" : "=r" (cntpct));
+#ifdef CONFIG_SYS_FSL_ERRATUM_A008585
+	asm volatile("mrs %0, cntpct_el0" : "=r" (temp));
+	while (temp != cntpct) {
+		asm volatile("mrs %0, cntpct_el0" : "=r" (cntpct));
+		asm volatile("mrs %0, cntpct_el0" : "=r" (temp));
+	}
+#endif
 	return cntpct;
 }
diff --git a/arch/arm/cpu/armv8/u-boot-spl.lds b/arch/arm/cpu/armv8/u-boot-spl.lds
new file mode 100644
index 0000000000..4df339c84a
--- /dev/null
+++ b/arch/arm/cpu/armv8/u-boot-spl.lds
@@ -0,0 +1,77 @@
+/*
+ * (C) Copyright 2013
+ * David Feng <fenghua@phytium.com.cn>
+ *
+ * (C) Copyright 2002
+ * Gary Jennejohn, DENX Software Engineering, <garyj@denx.de>
+ *
+ * (C) Copyright 2010
+ * Texas Instruments, <www.ti.com>
+ *	Aneesh V <aneesh@ti.com>
+ *
+ * SPDX-License-Identifier:	GPL-2.0+
+ */
+
+MEMORY { .sram : ORIGIN = CONFIG_SPL_TEXT_BASE,
+		LENGTH = CONFIG_SPL_MAX_SIZE }
+MEMORY { .sdram : ORIGIN = CONFIG_SPL_BSS_START_ADDR,
+		LENGTH = CONFIG_SPL_BSS_MAX_SIZE }
+
+OUTPUT_FORMAT("elf64-littleaarch64", "elf64-littleaarch64", "elf64-littleaarch64")
+OUTPUT_ARCH(aarch64)
+ENTRY(_start)
+SECTIONS
+{
+	.text : {
+		. = ALIGN(8);
+		*(.__image_copy_start)
+		CPUDIR/start.o (.text*)
+		*(.text*)
+	} >.sram
+
+	.rodata : {
+		. = ALIGN(8);
+		*(SORT_BY_ALIGNMENT(SORT_BY_NAME(.rodata*)))
+	} >.sram
+
+	.data : {
+		. = ALIGN(8);
+		*(.data*)
+	} >.sram
+
+	.u_boot_list : {
+		. = ALIGN(8);
+		KEEP(*(SORT(.u_boot_list*)));
+	} >.sram
+
+	.image_copy_end : {
+		. = ALIGN(8);
+		*(.__image_copy_end)
+	} >.sram
+
+	.end : {
+		. = ALIGN(8);
+		*(.__end)
+	} >.sram
+
+	.bss_start : {
+		. = ALIGN(8);
+		KEEP(*(.__bss_start));
+	} >.sdram
+
+	.bss : {
+		*(.bss*)
+		 . = ALIGN(8);
+	} >.sdram
+
+	.bss_end : {
+		KEEP(*(.__bss_end));
+	} >.sdram
+
+	/DISCARD/ : { *(.dynsym) }
+	/DISCARD/ : { *(.dynstr*) }
+	/DISCARD/ : { *(.dynamic*) }
+	/DISCARD/ : { *(.plt*) }
+	/DISCARD/ : { *(.interp*) }
+	/DISCARD/ : { *(.gnu*) }
+}
diff --git a/arch/arm/cpu/armv8/zynqmp/Makefile b/arch/arm/cpu/armv8/zynqmp/Makefile
index a997e045bd..efab5eabc9 100644
--- a/arch/arm/cpu/armv8/zynqmp/Makefile
+++ b/arch/arm/cpu/armv8/zynqmp/Makefile
@@ -7,3 +7,4 @@
 
 obj-y	+= clk.o
 obj-y	+= cpu.o
+obj-$(CONFIG_MP)	+= mp.o
diff --git a/arch/arm/cpu/armv8/zynqmp/clk.c b/arch/arm/cpu/armv8/zynqmp/clk.c
index 0af619d92f..9218586e94 100644
--- a/arch/arm/cpu/armv8/zynqmp/clk.c
+++ b/arch/arm/cpu/armv8/zynqmp/clk.c
@@ -6,6 +6,7 @@
  */
 
 #include <common.h>
+#include <asm/arch/clk.h>
 #include <asm/arch/hardware.h>
 #include <asm/arch/sys_proto.h>
 
@@ -16,6 +17,8 @@ unsigned long get_uart_clk(int dev_id)
 	u32 ver = zynqmp_get_silicon_version();
 
 	switch (ver) {
+	case ZYNQMP_CSU_VERSION_VELOCE:
+		return 48000;
 	case ZYNQMP_CSU_VERSION_EP108:
 		return 25000000;
 	}
diff --git a/arch/arm/cpu/armv8/zynqmp/cpu.c b/arch/arm/cpu/armv8/zynqmp/cpu.c
index 6fae03c1af..60d7d20e17 100644
--- a/arch/arm/cpu/armv8/zynqmp/cpu.c
+++ b/arch/arm/cpu/armv8/zynqmp/cpu.c
@@ -20,9 +20,175 @@ unsigned int zynqmp_get_silicon_version(void)
 	gd->cpu_clk = get_tbclk();
 
 	switch (gd->cpu_clk) {
+	case 0 ... 1000000:
+		return ZYNQMP_CSU_VERSION_VELOCE;
 	case 50000000:
 		return ZYNQMP_CSU_VERSION_QEMU;
 	}
 
 	return ZYNQMP_CSU_VERSION_EP108;
 }
+
+#ifndef CONFIG_SYS_DCACHE_OFF
+#include <asm/armv8/mmu.h>
+
+#define SECTION_SHIFT_L1	30UL
+#define SECTION_SHIFT_L2	21UL
+#define BLOCK_SIZE_L0		0x8000000000UL
+#define BLOCK_SIZE_L1		(1 << SECTION_SHIFT_L1)
+#define BLOCK_SIZE_L2		(1 << SECTION_SHIFT_L2)
+
+#define TCR_TG1_4K		(1 << 31)
+#define TCR_EPD1_DISABLE	(1 << 23)
+#define ZYNQMO_VA_BITS		40
+#define ZYNQMP_TCR		TCR_TG1_4K | \
+				TCR_EPD1_DISABLE | \
+				TCR_SHARED_OUTER | \
+				TCR_SHARED_INNER | \
+				TCR_IRGN_WBWA | \
+				TCR_ORGN_WBWA | \
+				TCR_T0SZ(ZYNQMO_VA_BITS)
+
+#define MEMORY_ATTR	PMD_SECT_AF | PMD_SECT_INNER_SHARE |	\
+			PMD_ATTRINDX(MT_NORMAL) |	\
+			PMD_TYPE_SECT
+#define DEVICE_ATTR	PMD_SECT_AF | PMD_SECT_PXN |	\
+			PMD_SECT_UXN | PMD_ATTRINDX(MT_DEVICE_NGNRNE) |	\
+			PMD_TYPE_SECT
+
+/* 4K size is required to place 512 entries in each level */
+#define TLB_TABLE_SIZE	0x1000
+
+struct attr_tbl {
+	u32 num;
+	u64 attr;
+};
+
+static struct attr_tbl attr_tbll1t0[4] = { {16, 0x0},
+					   {8, DEVICE_ATTR},
+					   {32, MEMORY_ATTR},
+					   {456, DEVICE_ATTR}
+					 };
+static struct attr_tbl attr_tbll2t3[4] = { {0x180, DEVICE_ATTR},
+					   {0x40, 0x0},
+					   {0x3F, DEVICE_ATTR},
+					   {0x1, MEMORY_ATTR}
+					 };
+
+/*
+ * This mmu table looks as below
+ * Level 0 table contains two entries to 512GB sizes. One is Level1 Table 0
+ * and other Level1 Table1.
+ * Level1 Table0 contains entries for each 1GB from 0 to 511GB.
+ * Level1 Table1 contains entries for each 1GB from 512GB to 1TB.
+ * Level2 Table0, Level2 Table1, Level2 Table2 and Level2 Table3 contains
+ * entries for each 2MB starting from 0GB, 1GB, 2GB and 3GB respectively.
+ */
+static void zynqmp_mmu_setup(void)
+{
+	int el;
+	u32 index_attr;
+	u64 i, section_l1t0, section_l1t1;
+	u64 section_l2t0, section_l2t1, section_l2t2, section_l2t3;
+	u64 *level0_table = (u64 *)gd->arch.tlb_addr;
+	u64 *level1_table_0 = (u64 *)(gd->arch.tlb_addr + TLB_TABLE_SIZE);
+	u64 *level1_table_1 = (u64 *)(gd->arch.tlb_addr + (2 * TLB_TABLE_SIZE));
+	u64 *level2_table_0 = (u64 *)(gd->arch.tlb_addr + (3 * TLB_TABLE_SIZE));
+	u64 *level2_table_1 = (u64 *)(gd->arch.tlb_addr + (4 * TLB_TABLE_SIZE));
+	u64 *level2_table_2 = (u64 *)(gd->arch.tlb_addr + (5 * TLB_TABLE_SIZE));
+	u64 *level2_table_3 = (u64 *)(gd->arch.tlb_addr + (6 * TLB_TABLE_SIZE));
+
+	level0_table[0] =
+		(u64)level1_table_0 | PMD_TYPE_TABLE;
+	level0_table[1] =
+		(u64)level1_table_1 | PMD_TYPE_TABLE;
+
+	/*
+	 * set level 1 table 0, covering 0 to 512GB
+	 * set level 1 table 1, covering 512GB to 1TB
+	 */
+	section_l1t0 = 0;
+	section_l1t1 = BLOCK_SIZE_L0;
+
+	index_attr = 0;
+	for (i = 0; i < 512; i++) {
+		level1_table_0[i] = section_l1t0;
+		level1_table_0[i] |= attr_tbll1t0[index_attr].attr;
+		attr_tbll1t0[index_attr].num--;
+		if (attr_tbll1t0[index_attr].num == 0)
+			index_attr++;
+		level1_table_1[i] = section_l1t1;
+		level1_table_1[i] |= DEVICE_ATTR;
+		section_l1t0 += BLOCK_SIZE_L1;
+		section_l1t1 += BLOCK_SIZE_L1;
+	}
+
+	level1_table_0[0] =
+		(u64)level2_table_0 | PMD_TYPE_TABLE;
+	level1_table_0[1] =
+		(u64)level2_table_1 | PMD_TYPE_TABLE;
+	level1_table_0[2] =
+		(u64)level2_table_2 | PMD_TYPE_TABLE;
+	level1_table_0[3] =
+		(u64)level2_table_3 | PMD_TYPE_TABLE;
+
+	section_l2t0 = 0;
+	section_l2t1 = section_l2t0 + BLOCK_SIZE_L1; /* 1GB */
+	section_l2t2 = section_l2t1 + BLOCK_SIZE_L1; /* 2GB */
+	section_l2t3 = section_l2t2 + BLOCK_SIZE_L1; /* 3GB */
+
+	index_attr = 0;
+
+	for (i = 0; i < 512; i++) {
+		level2_table_0[i] = section_l2t0 | MEMORY_ATTR;
+		level2_table_1[i] = section_l2t1 | MEMORY_ATTR;
+		level2_table_2[i] = section_l2t2 | DEVICE_ATTR;
+		level2_table_3[i] = section_l2t3 |
+				    attr_tbll2t3[index_attr].attr;
+		attr_tbll2t3[index_attr].num--;
+		if (attr_tbll2t3[index_attr].num == 0)
+			index_attr++;
+		section_l2t0 += BLOCK_SIZE_L2;
+		section_l2t1 += BLOCK_SIZE_L2;
+		section_l2t2 += BLOCK_SIZE_L2;
+		section_l2t3 += BLOCK_SIZE_L2;
+	}
+
+	/* flush new MMU table */
+	flush_dcache_range(gd->arch.tlb_addr,
+			   gd->arch.tlb_addr + gd->arch.tlb_size);
+
+	/* point TTBR to the new table */
+	el = current_el();
+	set_ttbr_tcr_mair(el, gd->arch.tlb_addr,
+			  ZYNQMP_TCR, MEMORY_ATTRIBUTES);
+
+	set_sctlr(get_sctlr() | CR_M);
+}
+
+int arch_cpu_init(void)
+{
+	icache_enable();
+	__asm_invalidate_dcache_all();
+	__asm_invalidate_tlb_all();
+	return 0;
+}
+
+/*
+ * This function is called from lib/board.c.
+ * It recreates MMU table in main memory. MMU and d-cache are enabled earlier.
+ * There is no need to disable d-cache for this operation.
+ */
+void enable_caches(void)
+{
+	/* The data cache is not active unless the mmu is enabled */
+	if (!(get_sctlr() & CR_M)) {
+		invalidate_dcache_all();
+		__asm_invalidate_tlb_all();
+		zynqmp_mmu_setup();
+	}
+	puts("Enabling Caches...\n");
+
+	set_sctlr(get_sctlr() | CR_C);
+}
+#endif
diff --git a/arch/arm/cpu/armv8/zynqmp/mp.c b/arch/arm/cpu/armv8/zynqmp/mp.c
new file mode 100644
index 0000000000..17e32a7b7c
--- /dev/null
+++ b/arch/arm/cpu/armv8/zynqmp/mp.c
@@ -0,0 +1,242 @@
+/*
+ * (C) Copyright 2014 - 2015 Xilinx, Inc.
+ * Michal Simek <michal.simek@xilinx.com>
+ *
+ * SPDX-License-Identifier:	GPL-2.0+
+ */
+
+#include <common.h>
+#include <asm/arch/hardware.h>
+#include <asm/arch/sys_proto.h>
+#include <asm/io.h>
+
+#define LOCK		0
+#define SPLIT		1
+
+#define HALT		0
+#define RELEASE		1
+
+#define ZYNQMP_BOOTADDR_HIGH_MASK		0xFFFFFFFF
+#define ZYNQMP_R5_HIVEC_ADDR			0xFFFF0000
+#define ZYNQMP_R5_LOVEC_ADDR			0x0
+#define ZYNQMP_RPU_CFG_CPU_HALT_MASK		0x01
+#define ZYNQMP_RPU_CFG_HIVEC_MASK		0x04
+#define ZYNQMP_RPU_GLBL_CTRL_SPLIT_LOCK_MASK	0x08
+#define ZYNQMP_RPU_GLBL_CTRL_TCM_COMB_MASK	0x40
+#define ZYNQMP_RPU_GLBL_CTRL_SLCLAMP_MASK	0x10
+
+#define ZYNQMP_CRLAPB_RST_LPD_AMBA_RST_MASK	0x04
+#define ZYNQMP_CRLAPB_RST_LPD_R50_RST_MASK	0x01
+#define ZYNQMP_CRLAPB_RST_LPD_R51_RST_MASK	0x02
+#define ZYNQMP_CRLAPB_CPU_R5_CTRL_CLKACT_MASK	0x1000000
+
+#define ZYNQMP_TCM_START_ADDRESS		0xFFE00000
+#define ZYNQMP_TCM_BOTH_SIZE			0x40000
+
+#define ZYNQMP_CORE_APU0	0
+#define ZYNQMP_CORE_APU3	3
+
+#define ZYNQMP_MAX_CORES	6
+
+int is_core_valid(unsigned int core)
+{
+	if (core < ZYNQMP_MAX_CORES)
+		return 1;
+
+	return 0;
+}
+
+int cpu_reset(int nr)
+{
+	puts("Feature is not implemented.\n");
+	return 0;
+}
+
+static void set_r5_halt_mode(u8 halt, u8 mode)
+{
+	u32 tmp;
+
+	tmp = readl(&rpu_base->rpu0_cfg);
+	if (halt == HALT)
+		tmp &= ~ZYNQMP_RPU_CFG_CPU_HALT_MASK;
+	else
+		tmp |= ZYNQMP_RPU_CFG_CPU_HALT_MASK;
+	writel(tmp, &rpu_base->rpu0_cfg);
+
+	if (mode == LOCK) {
+		tmp = readl(&rpu_base->rpu1_cfg);
+		if (halt == HALT)
+			tmp &= ~ZYNQMP_RPU_CFG_CPU_HALT_MASK;
+		else
+			tmp |= ZYNQMP_RPU_CFG_CPU_HALT_MASK;
+		writel(tmp, &rpu_base->rpu1_cfg);
+	}
+}
+
+static void set_r5_tcm_mode(u8 mode)
+{
+	u32 tmp;
+
+	tmp = readl(&rpu_base->rpu_glbl_ctrl);
+	if (mode == LOCK) {
+		tmp &= ~ZYNQMP_RPU_GLBL_CTRL_SPLIT_LOCK_MASK;
+		tmp |= ZYNQMP_RPU_GLBL_CTRL_TCM_COMB_MASK |
+		       ZYNQMP_RPU_GLBL_CTRL_SLCLAMP_MASK;
+	} else {
+		tmp |= ZYNQMP_RPU_GLBL_CTRL_SPLIT_LOCK_MASK;
+		tmp &= ~(ZYNQMP_RPU_GLBL_CTRL_TCM_COMB_MASK |
+		       ZYNQMP_RPU_GLBL_CTRL_SLCLAMP_MASK);
+	}
+
+	writel(tmp, &rpu_base->rpu_glbl_ctrl);
+}
+
+static void set_r5_reset(u8 mode)
+{
+	u32 tmp;
+
+	tmp = readl(&crlapb_base->rst_lpd_top);
+	tmp |= (ZYNQMP_CRLAPB_RST_LPD_AMBA_RST_MASK |
+	       ZYNQMP_CRLAPB_RST_LPD_R50_RST_MASK);
+
+	if (mode == LOCK)
+		tmp |= ZYNQMP_CRLAPB_RST_LPD_R51_RST_MASK;
+
+	writel(tmp, &crlapb_base->rst_lpd_top);
+}
+
+static void release_r5_reset(u8 mode)
+{
+	u32 tmp;
+
+	tmp = readl(&crlapb_base->rst_lpd_top);
+	tmp &= ~(ZYNQMP_CRLAPB_RST_LPD_AMBA_RST_MASK |
+	       ZYNQMP_CRLAPB_RST_LPD_R50_RST_MASK);
+
+	if (mode == LOCK)
+		tmp &= ~ZYNQMP_CRLAPB_RST_LPD_R51_RST_MASK;
+
+	writel(tmp, &crlapb_base->rst_lpd_top);
+}
+
+static void enable_clock_r5(void)
+{
+	u32 tmp;
+
+	tmp = readl(&crlapb_base->cpu_r5_ctrl);
+	tmp |= ZYNQMP_CRLAPB_CPU_R5_CTRL_CLKACT_MASK;
+	writel(tmp, &crlapb_base->cpu_r5_ctrl);
+
+	/* Give some delay for clock
+	 * to propogate */
+	udelay(0x500);
+}
+
+int cpu_disable(int nr)
+{
+	if (nr >= ZYNQMP_CORE_APU0 && nr <= ZYNQMP_CORE_APU3) {
+		u32 val = readl(&crfapb_base->rst_fpd_apu);
+		val |= 1 << nr;
+		writel(val, &crfapb_base->rst_fpd_apu);
+	} else {
+		set_r5_reset(LOCK);
+	}
+
+	return 0;
+}
+
+int cpu_status(int nr)
+{
+	if (nr >= ZYNQMP_CORE_APU0 && nr <= ZYNQMP_CORE_APU3) {
+		u32 addr_low = readl(((u8 *)&apu_base->rvbar_addr0_l) + nr * 8);
+		u32 addr_high = readl(((u8 *)&apu_base->rvbar_addr0_h) +
+				      nr * 8);
+		u32 val = readl(&crfapb_base->rst_fpd_apu);
+		val &= 1 << nr;
+		printf("APU CPU%d %s - starting address HI: %x, LOW: %x\n",
+		       nr, val ? "OFF" : "ON" , addr_high, addr_low);
+	} else {
+		u32 val = readl(&crlapb_base->rst_lpd_top);
+		val &= 1 << (nr - 4);
+		printf("RPU CPU%d %s\n", nr - 4, val ? "OFF" : "ON");
+	}
+
+	return 0;
+}
+
+static void set_r5_start(u8 high)
+{
+	u32 tmp;
+
+	tmp = readl(&rpu_base->rpu0_cfg);
+	if (high)
+		tmp |= ZYNQMP_RPU_CFG_HIVEC_MASK;
+	else
+		tmp &= ~ZYNQMP_RPU_CFG_HIVEC_MASK;
+	writel(tmp, &rpu_base->rpu0_cfg);
+
+	tmp = readl(&rpu_base->rpu1_cfg);
+	if (high)
+		tmp |= ZYNQMP_RPU_CFG_HIVEC_MASK;
+	else
+		tmp &= ~ZYNQMP_RPU_CFG_HIVEC_MASK;
+	writel(tmp, &rpu_base->rpu1_cfg);
+}
+
+int cpu_release(int nr, int argc, char * const argv[])
+{
+	if (nr >= ZYNQMP_CORE_APU0 && nr <= ZYNQMP_CORE_APU3) {
+		u64 boot_addr = simple_strtoull(argv[0], NULL, 16);
+		/* HIGH */
+		writel((u32)(boot_addr >> 32),
+		       ((u8 *)&apu_base->rvbar_addr0_h) + nr * 8);
+		/* LOW */
+		writel((u32)(boot_addr & ZYNQMP_BOOTADDR_HIGH_MASK),
+		       ((u8 *)&apu_base->rvbar_addr0_l) + nr * 8);
+
+		u32 val = readl(&crfapb_base->rst_fpd_apu);
+		val &= ~(1 << nr);
+		writel(val, &crfapb_base->rst_fpd_apu);
+	} else {
+		if (argc != 2) {
+			printf("Invalid number of arguments to release.\n");
+			printf("<addr> <mode>-Start addr lockstep or split\n");
+			return 1;
+		}
+
+		u32 boot_addr = simple_strtoul(argv[0], NULL, 16);
+		if (!(boot_addr == ZYNQMP_R5_LOVEC_ADDR ||
+		      boot_addr == ZYNQMP_R5_HIVEC_ADDR)) {
+			printf("Invalid starting address 0x%x\n", boot_addr);
+			printf("0 or 0xffff0000 are permitted\n");
+			return 1;
+		}
+
+		if (!strncmp(argv[1], "lockstep", 8)) {
+			printf("R5 lockstep mode\n");
+			set_r5_tcm_mode(LOCK);
+			set_r5_halt_mode(HALT, LOCK);
+
+			if (boot_addr == 0)
+				set_r5_start(0);
+			else
+				set_r5_start(1);
+
+			enable_clock_r5();
+			release_r5_reset(LOCK);
+			set_r5_halt_mode(RELEASE, LOCK);
+		} else if (!strncmp(argv[1], "split", 5)) {
+			printf("R5 split mode\n");
+			set_r5_tcm_mode(SPLIT);
+			set_r5_halt_mode(HALT, SPLIT);
+			enable_clock_r5();
+			release_r5_reset(SPLIT);
+			set_r5_halt_mode(RELEASE, SPLIT);
+		} else {
+			printf("Unsupported mode\n");
+			return 1;
+		}
+	}
+
+	return 0;
+}