sunxi: Add selective DRAM type and timing

DRAM chip varies, and one code cannot satisfy all DRAMs.

Add options to select a timing set.

Currently only DDR3-1333 (the original set) is added into it.

Signed-off-by: Icenowy Zheng <icenowy@aosc.xyz>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Tested-by: Jagan Teki <jagan@amarulasolutions.com>
diff --git a/arch/arm/include/asm/arch-sunxi/dram_sunxi_dw.h b/arch/arm/include/asm/arch-sunxi/dram_sunxi_dw.h
index d301ac9..03fd46b 100644
--- a/arch/arm/include/asm/arch-sunxi/dram_sunxi_dw.h
+++ b/arch/arm/include/asm/arch-sunxi/dram_sunxi_dw.h
@@ -205,4 +205,34 @@
 #define DXBDLR_WRITE_DELAY(x)	((x) << 8)
 #define DXBDLR_READ_DELAY(x)	((x) << 0)
 
+/*
+ * The delay parameters below allow to allegedly specify delay times of some
+ * unknown unit for each individual bit trace in each of the four data bytes
+ * the 32-bit wide access consists of. Also three control signals can be
+ * adjusted individually.
+ */
+#define BITS_PER_BYTE		8
+#define NR_OF_BYTE_LANES	(32 / BITS_PER_BYTE)
+/* The eight data lines (DQn) plus DM, DQS and DQSN */
+#define LINES_PER_BYTE_LANE	(BITS_PER_BYTE + 3)
+struct dram_para {
+	u16 page_size;
+	u8 bus_full_width;
+	u8 dual_rank;
+	u8 row_bits;
+	u8 bank_bits;
+	const u8 dx_read_delays[NR_OF_BYTE_LANES][LINES_PER_BYTE_LANE];
+	const u8 dx_write_delays[NR_OF_BYTE_LANES][LINES_PER_BYTE_LANE];
+	const u8 ac_delays[31];
+};
+
+static inline int ns_to_t(int nanoseconds)
+{
+	const unsigned int ctrl_freq = CONFIG_DRAM_CLK / 2;
+
+	return DIV_ROUND_UP(ctrl_freq * nanoseconds, 1000);
+}
+
+void mctl_set_timing_params(uint16_t socid, struct dram_para *para);
+
 #endif /* _SUNXI_DRAM_SUN8I_H3_H */
diff --git a/arch/arm/mach-sunxi/Kconfig b/arch/arm/mach-sunxi/Kconfig
index 8b8fc20..3e52947 100644
--- a/arch/arm/mach-sunxi/Kconfig
+++ b/arch/arm/mach-sunxi/Kconfig
@@ -216,6 +216,24 @@
 	This allows both the SPL and the U-Boot proper to be entered in
 	either mode and switch to AArch64 if needed.
 
+if SUNXI_DRAM_DW
+config SUNXI_DRAM_DDR3
+	bool
+
+choice
+	prompt "DRAM Type and Timing"
+	default SUNXI_DRAM_DDR3_1333
+
+config SUNXI_DRAM_DDR3_1333
+	bool "DDR3 1333"
+	select SUNXI_DRAM_DDR3
+	---help---
+	This option is the original only supported memory type, which suits
+	many H3/H5/A64 boards available now.
+
+endchoice
+endif
+
 config DRAM_TYPE
 	int "sunxi dram type"
 	depends on MACH_SUN8I_A83T
diff --git a/arch/arm/mach-sunxi/Makefile b/arch/arm/mach-sunxi/Makefile
index 41cee26..2a3c379 100644
--- a/arch/arm/mach-sunxi/Makefile
+++ b/arch/arm/mach-sunxi/Makefile
@@ -49,5 +49,6 @@
 obj-$(CONFIG_MACH_SUN8I_A33)	+= dram_sun8i_a33.o
 obj-$(CONFIG_MACH_SUN8I_A83T)	+= dram_sun8i_a83t.o
 obj-$(CONFIG_SUNXI_DRAM_DW)	+= dram_sunxi_dw.o
+obj-$(CONFIG_SUNXI_DRAM_DW)	+= dram_timings/
 obj-$(CONFIG_MACH_SUN9I)	+= dram_sun9i.o
 endif
diff --git a/arch/arm/mach-sunxi/dram_sunxi_dw.c b/arch/arm/mach-sunxi/dram_sunxi_dw.c
index 3f54c8e..a570642 100644
--- a/arch/arm/mach-sunxi/dram_sunxi_dw.c
+++ b/arch/arm/mach-sunxi/dram_sunxi_dw.c
@@ -16,34 +16,6 @@
 #include <asm/arch/cpu.h>
 #include <linux/kconfig.h>
 
-/*
- * The delay parameters below allow to allegedly specify delay times of some
- * unknown unit for each individual bit trace in each of the four data bytes
- * the 32-bit wide access consists of. Also three control signals can be
- * adjusted individually.
- */
-#define BITS_PER_BYTE		8
-#define NR_OF_BYTE_LANES	(32 / BITS_PER_BYTE)
-/* The eight data lines (DQn) plus DM, DQS and DQSN */
-#define LINES_PER_BYTE_LANE	(BITS_PER_BYTE + 3)
-struct dram_para {
-	u16 page_size;
-	u8 bus_full_width;
-	u8 dual_rank;
-	u8 row_bits;
-	u8 bank_bits;
-	const u8 dx_read_delays[NR_OF_BYTE_LANES][LINES_PER_BYTE_LANE];
-	const u8 dx_write_delays[NR_OF_BYTE_LANES][LINES_PER_BYTE_LANE];
-	const u8 ac_delays[31];
-};
-
-static inline int ns_to_t(int nanoseconds)
-{
-	const unsigned int ctrl_freq = CONFIG_DRAM_CLK / 2;
-
-	return DIV_ROUND_UP(ctrl_freq * nanoseconds, 1000);
-}
-
 static void mctl_phy_init(u32 val)
 {
 	struct sunxi_mctl_ctl_reg * const mctl_ctl =
@@ -269,90 +241,6 @@
 	}
 }
 
-static void mctl_set_timing_params(uint16_t socid, struct dram_para *para)
-{
-	struct sunxi_mctl_ctl_reg * const mctl_ctl =
-			(struct sunxi_mctl_ctl_reg *)SUNXI_DRAM_CTL0_BASE;
-
-	u8 tccd		= 2;
-	u8 tfaw		= ns_to_t(50);
-	u8 trrd		= max(ns_to_t(10), 4);
-	u8 trcd		= ns_to_t(15);
-	u8 trc		= ns_to_t(53);
-	u8 txp		= max(ns_to_t(8), 3);
-	u8 twtr		= max(ns_to_t(8), 4);
-	u8 trtp		= max(ns_to_t(8), 4);
-	u8 twr		= max(ns_to_t(15), 3);
-	u8 trp		= ns_to_t(15);
-	u8 tras		= ns_to_t(38);
-	u16 trefi	= ns_to_t(7800) / 32;
-	u16 trfc	= ns_to_t(350);
-
-	u8 tmrw		= 0;
-	u8 tmrd		= 4;
-	u8 tmod		= 12;
-	u8 tcke		= 3;
-	u8 tcksrx	= 5;
-	u8 tcksre	= 5;
-	u8 tckesr	= 4;
-	u8 trasmax	= 24;
-
-	u8 tcl		= 6; /* CL 12 */
-	u8 tcwl		= 4; /* CWL 8 */
-	u8 t_rdata_en	= 4;
-	u8 wr_latency	= 2;
-
-	u32 tdinit0	= (500 * CONFIG_DRAM_CLK) + 1;		/* 500us */
-	u32 tdinit1	= (360 * CONFIG_DRAM_CLK) / 1000 + 1;	/* 360ns */
-	u32 tdinit2	= (200 * CONFIG_DRAM_CLK) + 1;		/* 200us */
-	u32 tdinit3	= (1 * CONFIG_DRAM_CLK) + 1;		/* 1us */
-
-	u8 twtp		= tcwl + 2 + twr;	/* WL + BL / 2 + tWR */
-	u8 twr2rd	= tcwl + 2 + twtr;	/* WL + BL / 2 + tWTR */
-	u8 trd2wr	= tcl + 2 + 1 - tcwl;	/* RL + BL / 2 + 2 - WL */
-
-	/* set mode register */
-	writel(0x1c70, &mctl_ctl->mr[0]);	/* CL=11, WR=12 */
-	writel(0x40, &mctl_ctl->mr[1]);
-	writel(0x18, &mctl_ctl->mr[2]);		/* CWL=8 */
-	writel(0x0, &mctl_ctl->mr[3]);
-
-	if (socid == SOCID_R40)
-		writel(0x3, &mctl_ctl->lp3mr11);	/* odt_en[7:4] */
-
-	/* set DRAM timing */
-	writel(DRAMTMG0_TWTP(twtp) | DRAMTMG0_TFAW(tfaw) |
-	       DRAMTMG0_TRAS_MAX(trasmax) | DRAMTMG0_TRAS(tras),
-	       &mctl_ctl->dramtmg[0]);
-	writel(DRAMTMG1_TXP(txp) | DRAMTMG1_TRTP(trtp) | DRAMTMG1_TRC(trc),
-	       &mctl_ctl->dramtmg[1]);
-	writel(DRAMTMG2_TCWL(tcwl) | DRAMTMG2_TCL(tcl) |
-	       DRAMTMG2_TRD2WR(trd2wr) | DRAMTMG2_TWR2RD(twr2rd),
-	       &mctl_ctl->dramtmg[2]);
-	writel(DRAMTMG3_TMRW(tmrw) | DRAMTMG3_TMRD(tmrd) | DRAMTMG3_TMOD(tmod),
-	       &mctl_ctl->dramtmg[3]);
-	writel(DRAMTMG4_TRCD(trcd) | DRAMTMG4_TCCD(tccd) | DRAMTMG4_TRRD(trrd) |
-	       DRAMTMG4_TRP(trp), &mctl_ctl->dramtmg[4]);
-	writel(DRAMTMG5_TCKSRX(tcksrx) | DRAMTMG5_TCKSRE(tcksre) |
-	       DRAMTMG5_TCKESR(tckesr) | DRAMTMG5_TCKE(tcke),
-	       &mctl_ctl->dramtmg[5]);
-
-	/* set two rank timing */
-	clrsetbits_le32(&mctl_ctl->dramtmg[8], (0xff << 8) | (0xff << 0),
-			((socid == SOCID_H5 ? 0x33 : 0x66) << 8) | (0x10 << 0));
-
-	/* set PHY interface timing, write latency and read latency configure */
-	writel((0x2 << 24) | (t_rdata_en << 16) | (0x1 << 8) |
-	       (wr_latency << 0), &mctl_ctl->pitmg[0]);
-
-	/* set PHY timing, PTR0-2 use default */
-	writel(PTR3_TDINIT0(tdinit0) | PTR3_TDINIT1(tdinit1), &mctl_ctl->ptr[3]);
-	writel(PTR4_TDINIT2(tdinit2) | PTR4_TDINIT3(tdinit3), &mctl_ctl->ptr[4]);
-
-	/* set refresh timing */
-	writel(RFSHTMG_TREFI(trefi) | RFSHTMG_TRFC(trfc), &mctl_ctl->rfshtmg);
-}
-
 static u32 bin_to_mgray(int val)
 {
 	static const u8 lookup_table[32] = {
@@ -449,7 +337,12 @@
 	struct sunxi_mctl_com_reg * const mctl_com =
 			(struct sunxi_mctl_com_reg *)SUNXI_DRAM_COM_BASE;
 
-	writel(MCTL_CR_BL8 | MCTL_CR_2T | MCTL_CR_DDR3 | MCTL_CR_INTERLEAVED |
+	writel(MCTL_CR_BL8 | MCTL_CR_INTERLEAVED |
+#if defined CONFIG_SUNXI_DRAM_DDR3
+	       MCTL_CR_DDR3 | MCTL_CR_2T |
+#else
+#error Unsupported DRAM type!
+#endif
 	       (para->bank_bits == 3 ? MCTL_CR_EIGHT_BANKS : MCTL_CR_FOUR_BANKS) |
 	       MCTL_CR_BUS_FULL_WIDTH(para->bus_full_width) |
 	       (para->dual_rank ? MCTL_CR_DUAL_RANK : MCTL_CR_SINGLE_RANK) |
diff --git a/arch/arm/mach-sunxi/dram_timings/Makefile b/arch/arm/mach-sunxi/dram_timings/Makefile
new file mode 100644
index 0000000..7e71c76
--- /dev/null
+++ b/arch/arm/mach-sunxi/dram_timings/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_SUNXI_DRAM_DDR3_1333)	+= ddr3_1333.o
diff --git a/arch/arm/mach-sunxi/dram_timings/ddr3_1333.c b/arch/arm/mach-sunxi/dram_timings/ddr3_1333.c
new file mode 100644
index 0000000..0471e8a
--- /dev/null
+++ b/arch/arm/mach-sunxi/dram_timings/ddr3_1333.c
@@ -0,0 +1,87 @@
+#include <common.h>
+#include <asm/arch/dram.h>
+#include <asm/arch/cpu.h>
+
+void mctl_set_timing_params(uint16_t socid, struct dram_para *para)
+{
+	struct sunxi_mctl_ctl_reg * const mctl_ctl =
+			(struct sunxi_mctl_ctl_reg *)SUNXI_DRAM_CTL0_BASE;
+
+	u8 tccd		= 2;
+	u8 tfaw		= ns_to_t(50);
+	u8 trrd		= max(ns_to_t(10), 4);
+	u8 trcd		= ns_to_t(15);
+	u8 trc		= ns_to_t(53);
+	u8 txp		= max(ns_to_t(8), 3);
+	u8 twtr		= max(ns_to_t(8), 4);
+	u8 trtp		= max(ns_to_t(8), 4);
+	u8 twr		= max(ns_to_t(15), 3);
+	u8 trp		= ns_to_t(15);
+	u8 tras		= ns_to_t(38);
+	u16 trefi	= ns_to_t(7800) / 32;
+	u16 trfc	= ns_to_t(350);
+
+	u8 tmrw		= 0;
+	u8 tmrd		= 4;
+	u8 tmod		= 12;
+	u8 tcke		= 3;
+	u8 tcksrx	= 5;
+	u8 tcksre	= 5;
+	u8 tckesr	= 4;
+	u8 trasmax	= 24;
+
+	u8 tcl		= 6; /* CL 12 */
+	u8 tcwl		= 4; /* CWL 8 */
+	u8 t_rdata_en	= 4;
+	u8 wr_latency	= 2;
+
+	u32 tdinit0	= (500 * CONFIG_DRAM_CLK) + 1;		/* 500us */
+	u32 tdinit1	= (360 * CONFIG_DRAM_CLK) / 1000 + 1;	/* 360ns */
+	u32 tdinit2	= (200 * CONFIG_DRAM_CLK) + 1;		/* 200us */
+	u32 tdinit3	= (1 * CONFIG_DRAM_CLK) + 1;		/* 1us */
+
+	u8 twtp		= tcwl + 2 + twr;	/* WL + BL / 2 + tWR */
+	u8 twr2rd	= tcwl + 2 + twtr;	/* WL + BL / 2 + tWTR */
+	u8 trd2wr	= tcl + 2 + 1 - tcwl;	/* RL + BL / 2 + 2 - WL */
+
+	/* set mode register */
+	writel(0x1c70, &mctl_ctl->mr[0]);	/* CL=11, WR=12 */
+	writel(0x40, &mctl_ctl->mr[1]);
+	writel(0x18, &mctl_ctl->mr[2]);		/* CWL=8 */
+	writel(0x0, &mctl_ctl->mr[3]);
+
+	if (socid == SOCID_R40)
+		writel(0x3, &mctl_ctl->lp3mr11);	/* odt_en[7:4] */
+
+	/* set DRAM timing */
+	writel(DRAMTMG0_TWTP(twtp) | DRAMTMG0_TFAW(tfaw) |
+	       DRAMTMG0_TRAS_MAX(trasmax) | DRAMTMG0_TRAS(tras),
+	       &mctl_ctl->dramtmg[0]);
+	writel(DRAMTMG1_TXP(txp) | DRAMTMG1_TRTP(trtp) | DRAMTMG1_TRC(trc),
+	       &mctl_ctl->dramtmg[1]);
+	writel(DRAMTMG2_TCWL(tcwl) | DRAMTMG2_TCL(tcl) |
+	       DRAMTMG2_TRD2WR(trd2wr) | DRAMTMG2_TWR2RD(twr2rd),
+	       &mctl_ctl->dramtmg[2]);
+	writel(DRAMTMG3_TMRW(tmrw) | DRAMTMG3_TMRD(tmrd) | DRAMTMG3_TMOD(tmod),
+	       &mctl_ctl->dramtmg[3]);
+	writel(DRAMTMG4_TRCD(trcd) | DRAMTMG4_TCCD(tccd) | DRAMTMG4_TRRD(trrd) |
+	       DRAMTMG4_TRP(trp), &mctl_ctl->dramtmg[4]);
+	writel(DRAMTMG5_TCKSRX(tcksrx) | DRAMTMG5_TCKSRE(tcksre) |
+	       DRAMTMG5_TCKESR(tckesr) | DRAMTMG5_TCKE(tcke),
+	       &mctl_ctl->dramtmg[5]);
+
+	/* set two rank timing */
+	clrsetbits_le32(&mctl_ctl->dramtmg[8], (0xff << 8) | (0xff << 0),
+			((socid == SOCID_H5 ? 0x33 : 0x66) << 8) | (0x10 << 0));
+
+	/* set PHY interface timing, write latency and read latency configure */
+	writel((0x2 << 24) | (t_rdata_en << 16) | (0x1 << 8) |
+	       (wr_latency << 0), &mctl_ctl->pitmg[0]);
+
+	/* set PHY timing, PTR0-2 use default */
+	writel(PTR3_TDINIT0(tdinit0) | PTR3_TDINIT1(tdinit1), &mctl_ctl->ptr[3]);
+	writel(PTR4_TDINIT2(tdinit2) | PTR4_TDINIT3(tdinit3), &mctl_ctl->ptr[4]);
+
+	/* set refresh timing */
+	writel(RFSHTMG_TREFI(trefi) | RFSHTMG_TRFC(trfc), &mctl_ctl->rfshtmg);
+}