Mirrors/u-boot
2
0
Fork 0
mirror of https://github.com/AsahiLinux/u-boot synced 2024-11-06 13:14:27 +00:00
Code Issues Projects Releases Packages Wiki Activity
u-boot/arch/arm/mach-tegra/tegra210/clock.c

1267 lines
35 KiB
C
Raw Normal View History

SPDX: Convert all of our single license tags to Linux Kernel style When U-Boot started using SPDX tags we were among the early adopters and there weren't a lot of other examples to borrow from. So we picked the area of the file that usually had a full license text and replaced it with an appropriate SPDX-License-Identifier: entry. Since then, the Linux Kernel has adopted SPDX tags and they place it as the very first line in a file (except where shebangs are used, then it's second line) and with slightly different comment styles than us. In part due to community overlap, in part due to better tag visibility and in part for other minor reasons, switch over to that style. This commit changes all instances where we have a single declared license in the tag as both the before and after are identical in tag contents. There's also a few places where I found we did not have a tag and have introduced one. Signed-off-by: Tom Rini <trini@konsulko.com>
2018-05-06 21:58:06 +00:00
// SPDX-License-Identifier: GPL-2.0+
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
/*
i2c: t210: Add VI_I2C clock source support Fix VI_I2C clock source type. Will be needed by VI_I2C driver. Also added use of INTERNAL_ID macro in two places, needed to keep the id returned to 8 bits. Signed-off-by: Tom Warren <twarren@nvidia.com>
2020-03-27 17:24:31 +00:00
* (C) Copyright 2013-2020
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
* NVIDIA Corporation <www.nvidia.com>
*/
/* Tegra210 Clock control functions */
#include <common.h>
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#include <errno.h>
common: Drop init.h from common header Move this uncommon header out of the common header. Signed-off-by: Simon Glass <sjg@chromium.org>
2020-05-10 17:40:02 +00:00
#include <init.h>
common: Drop log.h from common header Move this header out of the common header. Signed-off-by: Simon Glass <sjg@chromium.org>
2020-05-10 17:40:05 +00:00
#include <log.h>
common: Drop net.h from common header Move this header out of the common header. Network support is used in quite a few places but it still does not warrant blanket inclusion. Note that this net.h header itself has quite a lot in it. It could be split into the driver-mode support, functions, structures, checksumming, etc. Signed-off-by: Simon Glass <sjg@chromium.org>
2020-05-10 17:39:56 +00:00
#include <asm/cache.h>
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
#include <asm/io.h>
#include <asm/arch/clock.h>
#include <asm/arch/sysctr.h>
#include <asm/arch/tegra.h>
#include <asm/arch-tegra/clk_rst.h>
#include <asm/arch-tegra/timer.h>
#include <div64.h>
#include <fdtdec.h>
common: Drop linux/bitops.h from common header Move this uncommon header out of the common header. Signed-off-by: Simon Glass <sjg@chromium.org>
2020-05-10 17:40:13 +00:00
#include <linux/bitops.h>
common: Drop linux/delay.h from common header Move this uncommon header out of the common header. Signed-off-by: Simon Glass <sjg@chromium.org>
2020-05-10 17:40:11 +00:00
#include <linux/delay.h>
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
/*
* Clock types that we can use as a source. The Tegra210 has muxes for the
* peripheral clocks, and in most cases there are four options for the clock
* source. This gives us a clock 'type' and exploits what commonality exists
* in the device.
*
* Letters are obvious, except for T which means CLK_M, and S which means the
* clock derived from 32KHz. Beware that CLK_M (also called OSC in the
* datasheet) and PLL_M are different things. The former is the basic
* clock supplied to the SOC from an external oscillator. The latter is the
* memory clock PLL.
*
* See definitions in clock_id in the header file.
*/
enum clock_type_id {
CLOCK_TYPE_AXPT, /* PLL_A, PLL_X, PLL_P, CLK_M */
CLOCK_TYPE_MCPA, /* and so on */
CLOCK_TYPE_MCPT,
CLOCK_TYPE_PCM,
CLOCK_TYPE_PCMT,
CLOCK_TYPE_PDCT,
CLOCK_TYPE_ACPT,
CLOCK_TYPE_ASPTE,
ARM: tegra: Fix mux type for disp1 and disp2 clocks on Tegra210 On Tegra210 the parents for the disp1 and disp2 clocks are slightly different from earlier chips. Only pll_p, pll_d_out0, pll_d2_out0 and clk_m are valid parents (technically pll_d_out is as well, but U-Boot doesn't know anything about it). Fix up the type name and the mux definition. Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2019-04-15 09:32:15 +00:00
CLOCK_TYPE_PDD2T,
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
CLOCK_TYPE_PCST,
tegra: Correct logic for reading pll_misc in clock_start_pll() The logic for simple PLLs on T124 was broken by this commit: 722e000c Tegra: PLL: use per-SoC pllinfo table instead of PLL_DIVM/N/P, etc. Correct it by reading from the same pll_misc register that it writes to and adding an entry for the DP PLL in the pllinfo table. Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-08-10 13:14:36 +00:00
CLOCK_TYPE_DP,
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
CLOCK_TYPE_PC2CC3M,
CLOCK_TYPE_PC2CC3S_T,
CLOCK_TYPE_PC2CC3M_T,
CLOCK_TYPE_PC2CC3M_T16, /* PC2CC3M_T, but w/16-bit divisor (I2C) */
CLOCK_TYPE_MC2CC3P_A,
CLOCK_TYPE_M,
CLOCK_TYPE_MCPTM2C2C3,
CLOCK_TYPE_PC2CC3T_S,
CLOCK_TYPE_AC2CC3P_TS2,
CLOCK_TYPE_PC01C00_C42C41TC40,
CLOCK_TYPE_COUNT,
CLOCK_TYPE_NONE = -1, /* invalid clock type */
};
enum {
CLOCK_MAX_MUX = 8 /* number of source options for each clock */
};
/*
* Clock source mux for each clock type. This just converts our enum into
* a list of mux sources for use by the code.
*
* Note:
* The extra column in each clock source array is used to store the mask
* bits in its register for the source.
*/
#define CLK(x) CLOCK_ID_ ## x
static enum clock_id clock_source[CLOCK_TYPE_COUNT][CLOCK_MAX_MUX+1] = {
{ CLK(AUDIO), CLK(XCPU), CLK(PERIPH), CLK(OSC),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_30},
{ CLK(MEMORY), CLK(CGENERAL), CLK(PERIPH), CLK(AUDIO),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_30},
{ CLK(MEMORY), CLK(CGENERAL), CLK(PERIPH), CLK(OSC),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_30},
{ CLK(PERIPH), CLK(CGENERAL), CLK(MEMORY), CLK(NONE),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_30},
{ CLK(PERIPH), CLK(CGENERAL), CLK(MEMORY), CLK(OSC),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_30},
{ CLK(PERIPH), CLK(DISPLAY), CLK(CGENERAL), CLK(OSC),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_30},
{ CLK(AUDIO), CLK(CGENERAL), CLK(PERIPH), CLK(OSC),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_30},
{ CLK(AUDIO), CLK(SFROM32KHZ), CLK(PERIPH), CLK(OSC),
CLK(EPCI), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_29},
ARM: tegra: Fix mux type for disp1 and disp2 clocks on Tegra210 On Tegra210 the parents for the disp1 and disp2 clocks are slightly different from earlier chips. Only pll_p, pll_d_out0, pll_d2_out0 and clk_m are valid parents (technically pll_d_out is as well, but U-Boot doesn't know anything about it). Fix up the type name and the mux definition. Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2019-04-15 09:32:15 +00:00
{ CLK(PERIPH), CLK(NONE), CLK(DISPLAY), CLK(NONE),
CLK(NONE), CLK(DISPLAY2), CLK(OSC), CLK(NONE),
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
MASK_BITS_31_29},
{ CLK(PERIPH), CLK(CGENERAL), CLK(SFROM32KHZ), CLK(OSC),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_28},
tegra: Correct logic for reading pll_misc in clock_start_pll() The logic for simple PLLs on T124 was broken by this commit: 722e000c Tegra: PLL: use per-SoC pllinfo table instead of PLL_DIVM/N/P, etc. Correct it by reading from the same pll_misc register that it writes to and adding an entry for the DP PLL in the pllinfo table. Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-08-10 13:14:36 +00:00
/* CLOCK_TYPE_DP */
{ CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_28},
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
/* Additional clock types on Tegra114+ */
/* CLOCK_TYPE_PC2CC3M */
{ CLK(PERIPH), CLK(CGENERAL2), CLK(CGENERAL), CLK(CGENERAL3),
CLK(MEMORY), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_29},
/* CLOCK_TYPE_PC2CC3S_T */
{ CLK(PERIPH), CLK(CGENERAL2), CLK(CGENERAL), CLK(CGENERAL3),
CLK(SFROM32KHZ), CLK(NONE), CLK(OSC), CLK(NONE),
MASK_BITS_31_29},
/* CLOCK_TYPE_PC2CC3M_T */
{ CLK(PERIPH), CLK(CGENERAL2), CLK(CGENERAL), CLK(CGENERAL3),
CLK(MEMORY), CLK(NONE), CLK(OSC), CLK(NONE),
MASK_BITS_31_29},
/* CLOCK_TYPE_PC2CC3M_T, w/16-bit divisor (I2C) */
{ CLK(PERIPH), CLK(CGENERAL2), CLK(CGENERAL), CLK(CGENERAL3),
CLK(MEMORY), CLK(NONE), CLK(OSC), CLK(NONE),
MASK_BITS_31_29},
/* CLOCK_TYPE_MC2CC3P_A */
{ CLK(MEMORY), CLK(CGENERAL2), CLK(CGENERAL), CLK(CGENERAL3),
CLK(PERIPH), CLK(NONE), CLK(AUDIO), CLK(NONE),
MASK_BITS_31_29},
/* CLOCK_TYPE_M */
{ CLK(MEMORY), CLK(NONE), CLK(NONE), CLK(NONE),
CLK(NONE), CLK(NONE), CLK(NONE), CLK(NONE),
MASK_BITS_31_30},
/* CLOCK_TYPE_MCPTM2C2C3 */
{ CLK(MEMORY), CLK(CGENERAL), CLK(PERIPH), CLK(OSC),
CLK(MEMORY2), CLK(CGENERAL2), CLK(CGENERAL3), CLK(NONE),
MASK_BITS_31_29},
/* CLOCK_TYPE_PC2CC3T_S */
{ CLK(PERIPH), CLK(CGENERAL2), CLK(CGENERAL), CLK(CGENERAL3),
CLK(OSC), CLK(NONE), CLK(SFROM32KHZ), CLK(NONE),
MASK_BITS_31_29},
/* CLOCK_TYPE_AC2CC3P_TS2 */
{ CLK(AUDIO), CLK(CGENERAL2), CLK(CGENERAL), CLK(CGENERAL3),
CLK(PERIPH), CLK(NONE), CLK(OSC), CLK(SRC2),
MASK_BITS_31_29},
/* CLOCK_TYPE_PC01C00_C42C41TC40 */
{ CLK(PERIPH), CLK(CGENERAL_1), CLK(CGENERAL_0), CLK(NONE),
CLK(CGENERAL4_2), CLK(CGENERAL4_1), CLK(OSC), CLK(CGENERAL4_0),
MASK_BITS_31_29},
};
/*
* Clock type for each peripheral clock source. We put the name in each
* record just so it is easy to match things up
*/
#define TYPE(name, type) type
static enum clock_type_id clock_periph_type[PERIPHC_COUNT] = {
/* 0x00 */
TYPE(PERIPHC_I2S2, CLOCK_TYPE_AXPT),
TYPE(PERIPHC_I2S3, CLOCK_TYPE_AXPT),
TYPE(PERIPHC_SPDIF_OUT, CLOCK_TYPE_AXPT),
TYPE(PERIPHC_SPDIF_IN, CLOCK_TYPE_PC2CC3M),
TYPE(PERIPHC_PWM, CLOCK_TYPE_PC2CC3S_T),
TYPE(PERIPHC_05h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_SBC2, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_SBC3, CLOCK_TYPE_PC2CC3M_T),
/* 0x08 */
TYPE(PERIPHC_08h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_I2C1, CLOCK_TYPE_PC2CC3M_T16),
TYPE(PERIPHC_I2C5, CLOCK_TYPE_PC2CC3M_T16),
TYPE(PERIPHC_0bh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_0ch, CLOCK_TYPE_NONE),
TYPE(PERIPHC_SBC1, CLOCK_TYPE_PC2CC3M_T),
ARM: tegra: Fix mux type for disp1 and disp2 clocks on Tegra210 On Tegra210 the parents for the disp1 and disp2 clocks are slightly different from earlier chips. Only pll_p, pll_d_out0, pll_d2_out0 and clk_m are valid parents (technically pll_d_out is as well, but U-Boot doesn't know anything about it). Fix up the type name and the mux definition. Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2019-04-15 09:32:15 +00:00
TYPE(PERIPHC_DISP1, CLOCK_TYPE_PDD2T),
TYPE(PERIPHC_DISP2, CLOCK_TYPE_PDD2T),
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
/* 0x10 */
TYPE(PERIPHC_10h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_11h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_VI, CLOCK_TYPE_MC2CC3P_A),
TYPE(PERIPHC_13h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_SDMMC1, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_SDMMC2, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_16h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_17h, CLOCK_TYPE_NONE),
/* 0x18 */
TYPE(PERIPHC_18h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_SDMMC4, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_VFIR, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_1Bh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_1Ch, CLOCK_TYPE_NONE),
TYPE(PERIPHC_HSI, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_UART1, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_UART2, CLOCK_TYPE_PC2CC3M_T),
/* 0x20 */
TYPE(PERIPHC_HOST1X, CLOCK_TYPE_MC2CC3P_A),
TYPE(PERIPHC_21h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_22h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_23h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_24h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_25h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_I2C2, CLOCK_TYPE_PC2CC3M_T16),
TYPE(PERIPHC_EMC, CLOCK_TYPE_MCPTM2C2C3),
/* 0x28 */
TYPE(PERIPHC_UART3, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_29h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_VI_SENSOR, CLOCK_TYPE_MC2CC3P_A),
TYPE(PERIPHC_2bh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_2ch, CLOCK_TYPE_NONE),
TYPE(PERIPHC_SBC4, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_I2C3, CLOCK_TYPE_PC2CC3M_T16),
TYPE(PERIPHC_SDMMC3, CLOCK_TYPE_PC2CC3M_T),
/* 0x30 */
TYPE(PERIPHC_UART4, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_UART5, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_VDE, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_OWR, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_NOR, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_CSITE, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_I2S1, CLOCK_TYPE_AXPT),
TYPE(PERIPHC_DTV, CLOCK_TYPE_NONE),
/* 0x38 */
TYPE(PERIPHC_38h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_39h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_3ah, CLOCK_TYPE_NONE),
TYPE(PERIPHC_3bh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_MSENC, CLOCK_TYPE_MC2CC3P_A),
TYPE(PERIPHC_TSEC, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_3eh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_OSC, CLOCK_TYPE_NONE),
/* 0x40 */
TYPE(PERIPHC_40h, CLOCK_TYPE_NONE), /* start with 0x3b0 */
TYPE(PERIPHC_MSELECT, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_TSENSOR, CLOCK_TYPE_PC2CC3T_S),
TYPE(PERIPHC_I2S4, CLOCK_TYPE_AXPT),
TYPE(PERIPHC_I2S5, CLOCK_TYPE_AXPT),
TYPE(PERIPHC_I2C4, CLOCK_TYPE_PC2CC3M_T16),
TYPE(PERIPHC_SBC5, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_SBC6, CLOCK_TYPE_PC2CC3M_T),
/* 0x48 */
TYPE(PERIPHC_AUDIO, CLOCK_TYPE_AC2CC3P_TS2),
TYPE(PERIPHC_49h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_4ah, CLOCK_TYPE_NONE),
TYPE(PERIPHC_4bh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_4ch, CLOCK_TYPE_NONE),
TYPE(PERIPHC_HDA2CODEC2X, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_ACTMON, CLOCK_TYPE_PC2CC3S_T),
TYPE(PERIPHC_EXTPERIPH1, CLOCK_TYPE_ASPTE),
/* 0x50 */
TYPE(PERIPHC_EXTPERIPH2, CLOCK_TYPE_ASPTE),
TYPE(PERIPHC_EXTPERIPH3, CLOCK_TYPE_ASPTE),
TYPE(PERIPHC_52h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_I2CSLOW, CLOCK_TYPE_PC2CC3S_T),
TYPE(PERIPHC_SYS, CLOCK_TYPE_NONE),
TYPE(PERIPHC_55h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_56h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_57h, CLOCK_TYPE_NONE),
/* 0x58 */
TYPE(PERIPHC_58h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_59h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_5ah, CLOCK_TYPE_NONE),
TYPE(PERIPHC_5bh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_SATAOOB, CLOCK_TYPE_PCMT),
TYPE(PERIPHC_SATA, CLOCK_TYPE_PCMT),
TYPE(PERIPHC_HDA, CLOCK_TYPE_PC2CC3M_T),
TYPE(PERIPHC_5fh, CLOCK_TYPE_NONE),
/* 0x60 */
TYPE(PERIPHC_XUSB_CORE_HOST, CLOCK_TYPE_NONE),
TYPE(PERIPHC_XUSB_FALCON, CLOCK_TYPE_NONE),
TYPE(PERIPHC_XUSB_FS, CLOCK_TYPE_NONE),
TYPE(PERIPHC_XUSB_CORE_DEV, CLOCK_TYPE_NONE),
TYPE(PERIPHC_XUSB_SS, CLOCK_TYPE_NONE),
TYPE(PERIPHC_CILAB, CLOCK_TYPE_NONE),
TYPE(PERIPHC_CILCD, CLOCK_TYPE_NONE),
TYPE(PERIPHC_CILE, CLOCK_TYPE_NONE),
/* 0x68 */
TYPE(PERIPHC_DSIA_LP, CLOCK_TYPE_NONE),
TYPE(PERIPHC_DSIB_LP, CLOCK_TYPE_NONE),
TYPE(PERIPHC_ENTROPY, CLOCK_TYPE_NONE),
TYPE(PERIPHC_DVFS_REF, CLOCK_TYPE_NONE),
TYPE(PERIPHC_DVFS_SOC, CLOCK_TYPE_NONE),
TYPE(PERIPHC_TRACECLKIN, CLOCK_TYPE_NONE),
TYPE(PERIPHC_6eh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_6fh, CLOCK_TYPE_NONE),
/* 0x70 */
TYPE(PERIPHC_EMC_LATENCY, CLOCK_TYPE_NONE),
TYPE(PERIPHC_SOC_THERM, CLOCK_TYPE_NONE),
TYPE(PERIPHC_72h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_73h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_74h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_75h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_VI_SENSOR2, CLOCK_TYPE_NONE),
TYPE(PERIPHC_I2C6, CLOCK_TYPE_PC2CC3M_T16),
/* 0x78 */
TYPE(PERIPHC_78h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_EMC_DLL, CLOCK_TYPE_MCPTM2C2C3),
TYPE(PERIPHC_7ah, CLOCK_TYPE_NONE),
TYPE(PERIPHC_CLK72MHZ, CLOCK_TYPE_NONE),
TYPE(PERIPHC_7ch, CLOCK_TYPE_NONE),
TYPE(PERIPHC_7dh, CLOCK_TYPE_NONE),
TYPE(PERIPHC_VIC, CLOCK_TYPE_NONE),
TYPE(PERIPHC_7Fh, CLOCK_TYPE_NONE),
/* 0x80 */
TYPE(PERIPHC_SDMMC_LEGACY_TM, CLOCK_TYPE_NONE),
TYPE(PERIPHC_NVDEC, CLOCK_TYPE_NONE),
TYPE(PERIPHC_NVJPG, CLOCK_TYPE_NONE),
TYPE(PERIPHC_NVENC, CLOCK_TYPE_NONE),
TYPE(PERIPHC_84h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_85h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_86h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_87h, CLOCK_TYPE_NONE),
/* 0x88 */
TYPE(PERIPHC_88h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_89h, CLOCK_TYPE_NONE),
TYPE(PERIPHC_DMIC3, CLOCK_TYPE_NONE),
TYPE(PERIPHC_APE, CLOCK_TYPE_NONE),
TYPE(PERIPHC_QSPI, CLOCK_TYPE_PC01C00_C42C41TC40),
i2c: t210: Add VI_I2C clock source support Fix VI_I2C clock source type. Will be needed by VI_I2C driver. Also added use of INTERNAL_ID macro in two places, needed to keep the id returned to 8 bits. Signed-off-by: Tom Warren <twarren@nvidia.com>
2020-03-27 17:24:31 +00:00
TYPE(PERIPHC_VI_I2C, CLOCK_TYPE_PC2CC3M_T16),
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
TYPE(PERIPHC_USB2_HSIC_TRK, CLOCK_TYPE_NONE),
TYPE(PERIPHC_PEX_SATA_USB_RX_BYP, CLOCK_TYPE_NONE),
/* 0x90 */
TYPE(PERIPHC_MAUD, CLOCK_TYPE_NONE),
TYPE(PERIPHC_TSECB, CLOCK_TYPE_NONE),
};
/*
* This array translates a periph_id to a periphc_internal_id
*
* Not present/matched up:
* uint vi_sensor; _VI_SENSOR_0, 0x1A8
* SPDIF - which is both 0x08 and 0x0c
*
*/
#define NONE(name) (-1)
#define OFFSET(name, value) PERIPHC_ ## name
#define INTERNAL_ID(id) (id & 0x000000ff)
static s8 periph_id_to_internal_id[PERIPH_ID_COUNT] = {
/* Low word: 31:0 */
NONE(CPU),
NONE(COP),
NONE(TRIGSYS),
NONE(ISPB),
NONE(RESERVED4),
NONE(TMR),
PERIPHC_UART1,
PERIPHC_UART2, /* and vfir 0x68 */
/* 8 */
NONE(GPIO),
PERIPHC_SDMMC2,
PERIPHC_SPDIF_IN,
PERIPHC_I2S2,
PERIPHC_I2C1,
NONE(RESERVED13),
PERIPHC_SDMMC1,
PERIPHC_SDMMC4,
/* 16 */
NONE(TCW),
PERIPHC_PWM,
PERIPHC_I2S3,
NONE(RESERVED19),
PERIPHC_VI,
NONE(RESERVED21),
NONE(USBD),
NONE(ISP),
/* 24 */
NONE(RESERVED24),
NONE(RESERVED25),
PERIPHC_DISP2,
PERIPHC_DISP1,
PERIPHC_HOST1X,
NONE(VCP),
PERIPHC_I2S1,
NONE(CACHE2),
/* Middle word: 63:32 */
NONE(MEM),
NONE(AHBDMA),
NONE(APBDMA),
NONE(RESERVED35),
NONE(RESERVED36),
NONE(STAT_MON),
NONE(RESERVED38),
NONE(FUSE),
/* 40 */
NONE(KFUSE),
PERIPHC_SBC1, /* SBCx = SPIx */
PERIPHC_NOR,
NONE(RESERVED43),
PERIPHC_SBC2,
NONE(XIO),
PERIPHC_SBC3,
PERIPHC_I2C5,
/* 48 */
NONE(DSI),
NONE(RESERVED49),
PERIPHC_HSI,
NONE(RESERVED51),
NONE(CSI),
NONE(RESERVED53),
PERIPHC_I2C2,
PERIPHC_UART3,
/* 56 */
NONE(MIPI_CAL),
PERIPHC_EMC,
NONE(USB2),
NONE(USB3),
NONE(RESERVED60),
PERIPHC_VDE,
NONE(BSEA),
NONE(BSEV),
/* Upper word 95:64 */
NONE(RESERVED64),
PERIPHC_UART4,
PERIPHC_UART5,
PERIPHC_I2C3,
PERIPHC_SBC4,
PERIPHC_SDMMC3,
NONE(PCIE),
PERIPHC_OWR,
/* 72 */
NONE(AFI),
PERIPHC_CSITE,
NONE(PCIEXCLK),
NONE(AVPUCQ),
NONE(LA),
NONE(TRACECLKIN),
NONE(SOC_THERM),
NONE(DTV),
/* 80 */
NONE(RESERVED80),
PERIPHC_I2CSLOW,
NONE(DSIB),
PERIPHC_TSEC,
NONE(RESERVED84),
NONE(RESERVED85),
NONE(RESERVED86),
NONE(EMUCIF),
/* 88 */
NONE(RESERVED88),
NONE(XUSB_HOST),
NONE(RESERVED90),
PERIPHC_MSENC,
NONE(RESERVED92),
NONE(RESERVED93),
NONE(RESERVED94),
NONE(XUSB_DEV),
/* V word: 31:0 */
NONE(CPUG),
NONE(CPULP),
NONE(V_RESERVED2),
PERIPHC_MSELECT,
NONE(V_RESERVED4),
PERIPHC_I2S4,
PERIPHC_I2S5,
PERIPHC_I2C4,
/* 104 */
PERIPHC_SBC5,
PERIPHC_SBC6,
PERIPHC_AUDIO,
NONE(APBIF),
NONE(V_RESERVED12),
NONE(V_RESERVED13),
NONE(V_RESERVED14),
PERIPHC_HDA2CODEC2X,
/* 112 */
NONE(ATOMICS),
NONE(V_RESERVED17),
NONE(V_RESERVED18),
NONE(V_RESERVED19),
NONE(V_RESERVED20),
NONE(V_RESERVED21),
NONE(V_RESERVED22),
PERIPHC_ACTMON,
/* 120 */
NONE(EXTPERIPH1),
NONE(EXTPERIPH2),
NONE(EXTPERIPH3),
NONE(OOB),
PERIPHC_SATA,
PERIPHC_HDA,
NONE(TZRAM),
NONE(SE),
/* W word: 31:0 */
NONE(HDA2HDMICODEC),
NONE(SATACOLD),
NONE(W_RESERVED2),
NONE(W_RESERVED3),
NONE(W_RESERVED4),
NONE(W_RESERVED5),
NONE(W_RESERVED6),
NONE(W_RESERVED7),
/* 136 */
NONE(CEC),
NONE(W_RESERVED9),
NONE(W_RESERVED10),
NONE(W_RESERVED11),
NONE(W_RESERVED12),
NONE(W_RESERVED13),
NONE(XUSB_PADCTL),
NONE(W_RESERVED15),
/* 144 */
NONE(W_RESERVED16),
NONE(W_RESERVED17),
NONE(W_RESERVED18),
NONE(W_RESERVED19),
NONE(W_RESERVED20),
NONE(ENTROPY),
NONE(DDS),
NONE(W_RESERVED23),
/* 152 */
NONE(W_RESERVED24),
NONE(W_RESERVED25),
NONE(W_RESERVED26),
NONE(DVFS),
NONE(XUSB_SS),
NONE(W_RESERVED29),
NONE(W_RESERVED30),
NONE(W_RESERVED31),
/* X word: 31:0 */
NONE(SPARE),
NONE(X_RESERVED1),
NONE(X_RESERVED2),
NONE(X_RESERVED3),
NONE(CAM_MCLK),
NONE(CAM_MCLK2),
PERIPHC_I2C6,
NONE(X_RESERVED7),
/* 168 */
NONE(X_RESERVED8),
NONE(X_RESERVED9),
NONE(X_RESERVED10),
NONE(VIM2_CLK),
NONE(X_RESERVED12),
NONE(X_RESERVED13),
NONE(EMC_DLL),
NONE(X_RESERVED15),
/* 176 */
NONE(X_RESERVED16),
NONE(CLK72MHZ),
NONE(VIC),
NONE(X_RESERVED19),
NONE(X_RESERVED20),
NONE(DPAUX),
NONE(SOR0),
NONE(X_RESERVED23),
/* 184 */
NONE(GPU),
NONE(X_RESERVED25),
NONE(X_RESERVED26),
NONE(X_RESERVED27),
NONE(X_RESERVED28),
NONE(X_RESERVED29),
NONE(X_RESERVED30),
NONE(X_RESERVED31),
/* Y: 192 (192 - 223) */
NONE(Y_RESERVED0),
PERIPHC_SDMMC_LEGACY_TM,
PERIPHC_NVDEC,
PERIPHC_NVJPG,
NONE(Y_RESERVED4),
PERIPHC_DMIC3, /* 197 */
PERIPHC_APE, /* 198 */
NONE(Y_RESERVED7),
/* 200 */
NONE(Y_RESERVED8),
NONE(Y_RESERVED9),
NONE(Y_RESERVED10),
NONE(Y_RESERVED11),
NONE(Y_RESERVED12),
NONE(Y_RESERVED13),
NONE(Y_RESERVED14),
NONE(Y_RESERVED15),
/* 208 */
PERIPHC_VI_I2C, /* 208 */
NONE(Y_RESERVED17),
NONE(Y_RESERVED18),
PERIPHC_QSPI, /* 211 */
NONE(Y_RESERVED20),
NONE(Y_RESERVED21),
NONE(Y_RESERVED22),
NONE(Y_RESERVED23),
/* 216 */
NONE(Y_RESERVED24),
NONE(Y_RESERVED25),
NONE(Y_RESERVED26),
PERIPHC_NVENC, /* 219 */
NONE(Y_RESERVED28),
NONE(Y_RESERVED29),
NONE(Y_RESERVED30),
NONE(Y_RESERVED31),
};
Tegra: PLL: use per-SoC pllinfo table instead of PLL_DIVM/N/P, etc. Added PLL variables (dividers mask/shift, lock enable/detect, etc.) to new pllinfo struct for each Soc/PLL. PLLA/C/D/E/M/P/U/X. Used pllinfo struct in all clock functions, validated on T210. Should be equivalent to prior code on T124/114/30/20. Thanks to Marcel Ziswiler for corrections to the T20/T30 values. Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-06-25 16:50:44 +00:00
/*
* PLL divider shift/mask tables for all PLL IDs.
*/
struct clk_pll_info tegra_pll_info_table[CLOCK_ID_PLL_COUNT] = {
/*
* NOTE: If kcp_mask/kvco_mask == 0, they're not used in that PLL (PLLC, etc.)
* If lock_ena or lock_det are >31, they're not used in that PLL (PLLC, etc.)
*/
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 10, .n_mask = 0xFF, .p_shift = 20, .p_mask = 0x1F,
.lock_ena = 32, .lock_det = 27, .kcp_shift = 0, .kcp_mask = 0, .kvco_shift = 0, .kvco_mask = 0 }, /* PLLC */
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 8, .n_mask = 0xFF, .p_shift = 20, .p_mask = 0x1F,
.lock_ena = 4, .lock_det = 27, .kcp_shift = 1, .kcp_mask = 3, .kvco_shift = 0, .kvco_mask = 1 }, /* PLLM */
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 10, .n_mask = 0xFF, .p_shift = 20, .p_mask = 0x1F,
.lock_ena = 18, .lock_det = 27, .kcp_shift = 0, .kcp_mask = 3, .kvco_shift = 2, .kvco_mask = 1 }, /* PLLP */
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 8, .n_mask = 0xFF, .p_shift = 20, .p_mask = 0x1F,
.lock_ena = 28, .lock_det = 27, .kcp_shift = 25, .kcp_mask = 3, .kvco_shift = 24, .kvco_mask = 1 }, /* PLLA */
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 8, .n_mask = 0xFF, .p_shift = 16, .p_mask = 0x1F,
.lock_ena = 29, .lock_det = 27, .kcp_shift = 25, .kcp_mask = 3, .kvco_shift = 24, .kvco_mask = 1 }, /* PLLU */
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 11, .n_mask = 0xFF, .p_shift = 20, .p_mask = 0x07,
.lock_ena = 18, .lock_det = 27, .kcp_shift = 23, .kcp_mask = 3, .kvco_shift = 22, .kvco_mask = 1 }, /* PLLD */
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 8, .n_mask = 0xFF, .p_shift = 20, .p_mask = 0x1F,
.lock_ena = 18, .lock_det = 27, .kcp_shift = 1, .kcp_mask = 3, .kvco_shift = 0, .kvco_mask = 1 }, /* PLLX */
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 8, .n_mask = 0xFF, .p_shift = 0, .p_mask = 0,
.lock_ena = 9, .lock_det = 11, .kcp_shift = 6, .kcp_mask = 3, .kvco_shift = 0, .kvco_mask = 1 }, /* PLLE */
{ .m_shift = 0, .m_mask = 0, .n_shift = 0, .n_mask = 0, .p_shift = 0, .p_mask = 0,
.lock_ena = 0, .lock_det = 0, .kcp_shift = 0, .kcp_mask = 0, .kvco_shift = 0, .kvco_mask = 0 }, /* PLLS (gone)*/
tegra: Correct logic for reading pll_misc in clock_start_pll() The logic for simple PLLs on T124 was broken by this commit: 722e000c Tegra: PLL: use per-SoC pllinfo table instead of PLL_DIVM/N/P, etc. Correct it by reading from the same pll_misc register that it writes to and adding an entry for the DP PLL in the pllinfo table. Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-08-10 13:14:36 +00:00
{ .m_shift = 0, .m_mask = 0xFF, .n_shift = 8, .n_mask = 0xFF, .p_shift = 19, .p_mask = 0x1F,
.lock_ena = 30, .lock_det = 27, .kcp_shift = 25, .kcp_mask = 3, .kvco_shift = 24, .kvco_mask = 1 }, /* PLLDP */
Tegra: PLL: use per-SoC pllinfo table instead of PLL_DIVM/N/P, etc. Added PLL variables (dividers mask/shift, lock enable/detect, etc.) to new pllinfo struct for each Soc/PLL. PLLA/C/D/E/M/P/U/X. Used pllinfo struct in all clock functions, validated on T210. Should be equivalent to prior code on T124/114/30/20. Thanks to Marcel Ziswiler for corrections to the T20/T30 values. Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-06-25 16:50:44 +00:00
};
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
/*
* Get the oscillator frequency, from the corresponding hardware configuration
* field. Note that Tegra30+ support 3 new higher freqs, but we map back
* to the old T20 freqs. Support for the higher oscillators is TBD.
*/
enum clock_osc_freq clock_get_osc_freq(void)
{
struct clk_rst_ctlr *clkrst =
(struct clk_rst_ctlr *)NV_PA_CLK_RST_BASE;
u32 reg;
reg = readl(&clkrst->crc_osc_ctrl);
reg = (reg & OSC_FREQ_MASK) >> OSC_FREQ_SHIFT;
/*
* 0 = 13MHz, 1 = 16.8MHz, 4 = 19.2MHz, 5 = 38.4MHz,
* 8 = 12MHz, 9 = 48MHz, 12 = 26MHz
*/
if (reg == 5) {
debug("OSC_FREQ is 38.4MHz (%d) ...\n", reg);
Tegra: clocks: Add 38.4MHz OSC support for T210 use Added 38.4MHz/48MHz entries to pll_x_table for CPU PLL. Needs to be measured - should be close to 700MHz (1.4G/2). Note that some freqs aren't in the PLLU table in T210 TRM (13, 26MHz), so I used the 12MHz table entry for them. They shouldn't be selected since they're not viable T210 OSC freqs. Since there are now 2 new OSC defines, all tables (pll_x_table, PLLU) had to increase by two entries, but since 38.4/48MHz are not viable osc freqs on T20/30/114, etc, they're just set to 0. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-06-22 20:03:44 +00:00
/* Map it to the 5th CLOCK_OSC_ enum, i.e. 4 */
return 4;
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
}
/*
* Map to most common (T20) freqs (except 38.4, handled above):
* 13/16.8 = 0, 19.2 = 1, 12/48 = 2, 26 = 3
*/
return reg >> 2;
}
/* Returns a pointer to the clock source register for a peripheral */
u32 *get_periph_source_reg(enum periph_id periph_id)
{
struct clk_rst_ctlr *clkrst =
(struct clk_rst_ctlr *)NV_PA_CLK_RST_BASE;
enum periphc_internal_id internal_id;
/* Coresight is a special case */
if (periph_id == PERIPH_ID_CSI)
return &clkrst->crc_clk_src[PERIPH_ID_CSI+1];
assert(periph_id >= PERIPH_ID_FIRST && periph_id < PERIPH_ID_COUNT);
internal_id = INTERNAL_ID(periph_id_to_internal_id[periph_id]);
assert(internal_id != -1);
if (internal_id < PERIPHC_VW_FIRST)
/* L, H, U */
return &clkrst->crc_clk_src[internal_id];
if (internal_id < PERIPHC_X_FIRST) {
/* VW */
internal_id -= PERIPHC_VW_FIRST;
return &clkrst->crc_clk_src_vw[internal_id];
}
if (internal_id < PERIPHC_Y_FIRST) {
/* X */
internal_id -= PERIPHC_X_FIRST;
return &clkrst->crc_clk_src_x[internal_id];
}
/* Y */
internal_id -= PERIPHC_Y_FIRST;
return &clkrst->crc_clk_src_y[internal_id];
}
ARM: tegra: add APIs the clock uclass driver will need A future patch will implement a clock uclass driver for Tegra. That driver will call into Tegra's existing clock code to simplify the transition; this avoids tieing the clock uclass patches into significant refactoring of the existing custom clock API implementation. Some of the Tegra clock APIs that manipulate peripheral clocks require both the peripheral clock ID and parent clock ID to be passed in together. However, the clock uclass API does not require any such "parent" parameter, so the clock driver must determine this information itself. This patch implements new Tegra- specific clock API clock_get_periph_parent() for this purpose. The new API is implemented in the core Tegra clock code rather than SoC- specific clock code. The implementation uses various SoC-/clock-specific data. That data is only available in SoC-specific clock code. Consequently, two new internal APIs are added that enable the core clock code to retrieve this information from the SoC-specific clock code. Due to the structure of the Tegra clock code, this leads to some unfortunate code duplication. However, this situation predates this patch. Ideally, future work will de-duplicate the Tegra clock code, and migrate it into drivers/clk/tegra. However, such refactoring is kept separate from this series. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2016-09-13 16:45:56 +00:00
int get_periph_clock_info(enum periph_id periph_id, int *mux_bits,
int *divider_bits, int *type)
{
enum periphc_internal_id internal_id;
if (!clock_periph_id_isvalid(periph_id))
return -1;
i2c: t210: Add VI_I2C clock source support Fix VI_I2C clock source type. Will be needed by VI_I2C driver. Also added use of INTERNAL_ID macro in two places, needed to keep the id returned to 8 bits. Signed-off-by: Tom Warren <twarren@nvidia.com>
2020-03-27 17:24:31 +00:00
internal_id = INTERNAL_ID(periph_id_to_internal_id[periph_id]);
ARM: tegra: add APIs the clock uclass driver will need A future patch will implement a clock uclass driver for Tegra. That driver will call into Tegra's existing clock code to simplify the transition; this avoids tieing the clock uclass patches into significant refactoring of the existing custom clock API implementation. Some of the Tegra clock APIs that manipulate peripheral clocks require both the peripheral clock ID and parent clock ID to be passed in together. However, the clock uclass API does not require any such "parent" parameter, so the clock driver must determine this information itself. This patch implements new Tegra- specific clock API clock_get_periph_parent() for this purpose. The new API is implemented in the core Tegra clock code rather than SoC- specific clock code. The implementation uses various SoC-/clock-specific data. That data is only available in SoC-specific clock code. Consequently, two new internal APIs are added that enable the core clock code to retrieve this information from the SoC-specific clock code. Due to the structure of the Tegra clock code, this leads to some unfortunate code duplication. However, this situation predates this patch. Ideally, future work will de-duplicate the Tegra clock code, and migrate it into drivers/clk/tegra. However, such refactoring is kept separate from this series. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2016-09-13 16:45:56 +00:00
if (!periphc_internal_id_isvalid(internal_id))
return -1;
*type = clock_periph_type[internal_id];
if (!clock_type_id_isvalid(*type))
return -1;
*mux_bits = clock_source[*type][CLOCK_MAX_MUX];
if (*type == CLOCK_TYPE_PC2CC3M_T16)
*divider_bits = 16;
else
*divider_bits = 8;
return 0;
}
enum clock_id get_periph_clock_id(enum periph_id periph_id, int source)
{
enum periphc_internal_id internal_id;
int type;
if (!clock_periph_id_isvalid(periph_id))
return CLOCK_ID_NONE;
i2c: t210: Add VI_I2C clock source support Fix VI_I2C clock source type. Will be needed by VI_I2C driver. Also added use of INTERNAL_ID macro in two places, needed to keep the id returned to 8 bits. Signed-off-by: Tom Warren <twarren@nvidia.com>
2020-03-27 17:24:31 +00:00
internal_id = INTERNAL_ID(periph_id_to_internal_id[periph_id]);
ARM: tegra: add APIs the clock uclass driver will need A future patch will implement a clock uclass driver for Tegra. That driver will call into Tegra's existing clock code to simplify the transition; this avoids tieing the clock uclass patches into significant refactoring of the existing custom clock API implementation. Some of the Tegra clock APIs that manipulate peripheral clocks require both the peripheral clock ID and parent clock ID to be passed in together. However, the clock uclass API does not require any such "parent" parameter, so the clock driver must determine this information itself. This patch implements new Tegra- specific clock API clock_get_periph_parent() for this purpose. The new API is implemented in the core Tegra clock code rather than SoC- specific clock code. The implementation uses various SoC-/clock-specific data. That data is only available in SoC-specific clock code. Consequently, two new internal APIs are added that enable the core clock code to retrieve this information from the SoC-specific clock code. Due to the structure of the Tegra clock code, this leads to some unfortunate code duplication. However, this situation predates this patch. Ideally, future work will de-duplicate the Tegra clock code, and migrate it into drivers/clk/tegra. However, such refactoring is kept separate from this series. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2016-09-13 16:45:56 +00:00
if (!periphc_internal_id_isvalid(internal_id))
return CLOCK_ID_NONE;
type = clock_periph_type[internal_id];
if (!clock_type_id_isvalid(type))
return CLOCK_ID_NONE;
return clock_source[type][source];
}
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
/**
* Given a peripheral ID and the required source clock, this returns which
* value should be programmed into the source mux for that peripheral.
*
* There is special code here to handle the one source type with 5 sources.
*
* @param periph_id peripheral to start
* @param source PLL id of required parent clock
* @param mux_bits Set to number of bits in mux register: 2 or 4
* @param divider_bits Set to number of divider bits (8 or 16)
* @return mux value (0-4, or -1 if not found)
*/
int get_periph_clock_source(enum periph_id periph_id,
enum clock_id parent, int *mux_bits, int *divider_bits)
{
enum clock_type_id type;
ARM: tegra: add APIs the clock uclass driver will need A future patch will implement a clock uclass driver for Tegra. That driver will call into Tegra's existing clock code to simplify the transition; this avoids tieing the clock uclass patches into significant refactoring of the existing custom clock API implementation. Some of the Tegra clock APIs that manipulate peripheral clocks require both the peripheral clock ID and parent clock ID to be passed in together. However, the clock uclass API does not require any such "parent" parameter, so the clock driver must determine this information itself. This patch implements new Tegra- specific clock API clock_get_periph_parent() for this purpose. The new API is implemented in the core Tegra clock code rather than SoC- specific clock code. The implementation uses various SoC-/clock-specific data. That data is only available in SoC-specific clock code. Consequently, two new internal APIs are added that enable the core clock code to retrieve this information from the SoC-specific clock code. Due to the structure of the Tegra clock code, this leads to some unfortunate code duplication. However, this situation predates this patch. Ideally, future work will de-duplicate the Tegra clock code, and migrate it into drivers/clk/tegra. However, such refactoring is kept separate from this series. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2016-09-13 16:45:56 +00:00
int mux, err;
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
ARM: tegra: add APIs the clock uclass driver will need A future patch will implement a clock uclass driver for Tegra. That driver will call into Tegra's existing clock code to simplify the transition; this avoids tieing the clock uclass patches into significant refactoring of the existing custom clock API implementation. Some of the Tegra clock APIs that manipulate peripheral clocks require both the peripheral clock ID and parent clock ID to be passed in together. However, the clock uclass API does not require any such "parent" parameter, so the clock driver must determine this information itself. This patch implements new Tegra- specific clock API clock_get_periph_parent() for this purpose. The new API is implemented in the core Tegra clock code rather than SoC- specific clock code. The implementation uses various SoC-/clock-specific data. That data is only available in SoC-specific clock code. Consequently, two new internal APIs are added that enable the core clock code to retrieve this information from the SoC-specific clock code. Due to the structure of the Tegra clock code, this leads to some unfortunate code duplication. However, this situation predates this patch. Ideally, future work will de-duplicate the Tegra clock code, and migrate it into drivers/clk/tegra. However, such refactoring is kept separate from this series. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2016-09-13 16:45:56 +00:00
err = get_periph_clock_info(periph_id, mux_bits, divider_bits, &type);
assert(!err);
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
for (mux = 0; mux < CLOCK_MAX_MUX; mux++)
if (clock_source[type][mux] == parent)
return mux;
/* if we get here, either us or the caller has made a mistake */
printf("Caller requested bad clock: periph=%d, parent=%d\n", periph_id,
parent);
return -1;
}
void clock_set_enable(enum periph_id periph_id, int enable)
{
struct clk_rst_ctlr *clkrst =
(struct clk_rst_ctlr *)NV_PA_CLK_RST_BASE;
u32 *clk;
u32 reg;
/* Enable/disable the clock to this peripheral */
assert(clock_periph_id_isvalid(periph_id));
if ((int)periph_id < (int)PERIPH_ID_VW_FIRST)
clk = &clkrst->crc_clk_out_enb[PERIPH_REG(periph_id)];
else if ((int)periph_id < (int)PERIPH_ID_X_FIRST)
clk = &clkrst->crc_clk_out_enb_vw[PERIPH_REG(periph_id)];
else if ((int)periph_id < (int)PERIPH_ID_Y_FIRST)
clk = &clkrst->crc_clk_out_enb_x;
else
clk = &clkrst->crc_clk_out_enb_y;
reg = readl(clk);
if (enable)
reg |= PERIPH_MASK(periph_id);
else
reg &= ~PERIPH_MASK(periph_id);
writel(reg, clk);
}
void reset_set_enable(enum periph_id periph_id, int enable)
{
struct clk_rst_ctlr *clkrst =
(struct clk_rst_ctlr *)NV_PA_CLK_RST_BASE;
u32 *reset;
u32 reg;
/* Enable/disable reset to the peripheral */
assert(clock_periph_id_isvalid(periph_id));
if (periph_id < PERIPH_ID_VW_FIRST)
reset = &clkrst->crc_rst_dev[PERIPH_REG(periph_id)];
else if ((int)periph_id < (int)PERIPH_ID_X_FIRST)
reset = &clkrst->crc_rst_dev_vw[PERIPH_REG(periph_id)];
else if ((int)periph_id < (int)PERIPH_ID_Y_FIRST)
reset = &clkrst->crc_rst_devices_x;
else
reset = &clkrst->crc_rst_devices_y;
reg = readl(reset);
if (enable)
reg |= PERIPH_MASK(periph_id);
else
reg &= ~PERIPH_MASK(periph_id);
writel(reg, reset);
}
#ifdef CONFIG_OF_CONTROL
/*
* Convert a device tree clock ID to our peripheral ID. They are mostly
* the same but we are very cautious so we check that a valid clock ID is
* provided.
*
* @param clk_id Clock ID according to tegra210 device tree binding
* @return peripheral ID, or PERIPH_ID_NONE if the clock ID is invalid
*/
enum periph_id clk_id_to_periph_id(int clk_id)
{
if (clk_id > PERIPH_ID_COUNT)
return PERIPH_ID_NONE;
switch (clk_id) {
case PERIPH_ID_RESERVED4:
case PERIPH_ID_RESERVED25:
case PERIPH_ID_RESERVED35:
case PERIPH_ID_RESERVED36:
case PERIPH_ID_RESERVED38:
case PERIPH_ID_RESERVED43:
case PERIPH_ID_RESERVED49:
case PERIPH_ID_RESERVED53:
case PERIPH_ID_RESERVED64:
case PERIPH_ID_RESERVED84:
case PERIPH_ID_RESERVED85:
case PERIPH_ID_RESERVED86:
case PERIPH_ID_RESERVED88:
case PERIPH_ID_RESERVED90:
case PERIPH_ID_RESERVED92:
case PERIPH_ID_RESERVED93:
case PERIPH_ID_RESERVED94:
case PERIPH_ID_V_RESERVED2:
case PERIPH_ID_V_RESERVED4:
case PERIPH_ID_V_RESERVED17:
case PERIPH_ID_V_RESERVED18:
case PERIPH_ID_V_RESERVED19:
case PERIPH_ID_V_RESERVED20:
case PERIPH_ID_V_RESERVED21:
case PERIPH_ID_V_RESERVED22:
case PERIPH_ID_W_RESERVED2:
case PERIPH_ID_W_RESERVED3:
case PERIPH_ID_W_RESERVED4:
case PERIPH_ID_W_RESERVED5:
case PERIPH_ID_W_RESERVED6:
case PERIPH_ID_W_RESERVED7:
case PERIPH_ID_W_RESERVED9:
case PERIPH_ID_W_RESERVED10:
case PERIPH_ID_W_RESERVED11:
case PERIPH_ID_W_RESERVED12:
case PERIPH_ID_W_RESERVED13:
case PERIPH_ID_W_RESERVED15:
case PERIPH_ID_W_RESERVED16:
case PERIPH_ID_W_RESERVED17:
case PERIPH_ID_W_RESERVED18:
case PERIPH_ID_W_RESERVED19:
case PERIPH_ID_W_RESERVED20:
case PERIPH_ID_W_RESERVED23:
case PERIPH_ID_W_RESERVED29:
case PERIPH_ID_W_RESERVED30:
case PERIPH_ID_W_RESERVED31:
return PERIPH_ID_NONE;
default:
return clk_id;
}
}
#endif /* CONFIG_OF_CONTROL */
/*
* T210 redefines PLLP_OUT2 as PLLP_VCO/DIVP, so do different OUT1-4 setup here.
* PLLP_BASE/MISC/etc. is already set up for 408MHz in the BootROM.
*/
void tegra210_setup_pllp(void)
{
struct clk_rst_ctlr *clkrst = (struct clk_rst_ctlr *)NV_PA_CLK_RST_BASE;
u32 reg;
/* Set PLLP_OUT1, 3 & 4 freqs to 9.6, 102 & 204MHz */
/* OUT1 */
/* Assert RSTN before enable */
reg = PLLP_OUT1_RSTN_EN;
writel(reg, &clkrst->crc_pll[CLOCK_ID_PERIPH].pll_out[0]);
/* Set divisor and reenable */
reg = (IN_408_OUT_9_6_DIVISOR << PLLP_OUT1_RATIO)
| PLLP_OUT1_OVR | PLLP_OUT1_CLKEN | PLLP_OUT1_RSTN_DIS;
writel(reg, &clkrst->crc_pll[CLOCK_ID_PERIPH].pll_out[0]);
/* OUT3, 4 */
/* Assert RSTN before enable */
reg = PLLP_OUT4_RSTN_EN | PLLP_OUT3_RSTN_EN;
writel(reg, &clkrst->crc_pll[CLOCK_ID_PERIPH].pll_out[1]);
/* Set divisor and reenable */
reg = (IN_408_OUT_204_DIVISOR << PLLP_OUT4_RATIO)
| PLLP_OUT4_OVR | PLLP_OUT4_CLKEN | PLLP_OUT4_RSTN_DIS
| (IN_408_OUT_102_DIVISOR << PLLP_OUT3_RATIO)
| PLLP_OUT3_OVR | PLLP_OUT3_CLKEN | PLLP_OUT3_RSTN_DIS;
writel(reg, &clkrst->crc_pll[CLOCK_ID_PERIPH].pll_out[1]);
/*
* NOTE: If you want to change PLLP_OUT2 away from 204MHz,
* you can change PLLP_BASE DIVP here. Currently defaults
* to 1, which is 2^1, or 2, so PLLP_OUT2 is 204MHz.
* See Table 13 in section 5.1.4 in T210 TRM for more info.
*/
}
void clock_early_init(void)
{
struct clk_rst_ctlr *clkrst =
(struct clk_rst_ctlr *)NV_PA_CLK_RST_BASE;
Tegra: PLL: use per-SoC pllinfo table instead of PLL_DIVM/N/P, etc. Added PLL variables (dividers mask/shift, lock enable/detect, etc.) to new pllinfo struct for each Soc/PLL. PLLA/C/D/E/M/P/U/X. Used pllinfo struct in all clock functions, validated on T210. Should be equivalent to prior code on T124/114/30/20. Thanks to Marcel Ziswiler for corrections to the T20/T30 values. Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-06-25 16:50:44 +00:00
struct clk_pll_info *pllinfo = &tegra_pll_info_table[CLOCK_ID_DISPLAY];
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
u32 data;
tegra210_setup_pllp();
/*
* PLLC output frequency set to 600Mhz
* PLLD output frequency set to 925Mhz
*/
switch (clock_get_osc_freq()) {
case CLOCK_OSC_FREQ_12_0: /* OSC is 12Mhz */
clock_set_rate(CLOCK_ID_CGENERAL, 600, 12, 0, 8);
clock_set_rate(CLOCK_ID_DISPLAY, 925, 12, 0, 12);
break;
case CLOCK_OSC_FREQ_26_0: /* OSC is 26Mhz */
clock_set_rate(CLOCK_ID_CGENERAL, 600, 26, 0, 8);
clock_set_rate(CLOCK_ID_DISPLAY, 925, 26, 0, 12);
break;
case CLOCK_OSC_FREQ_13_0: /* OSC is 13Mhz */
clock_set_rate(CLOCK_ID_CGENERAL, 600, 13, 0, 8);
clock_set_rate(CLOCK_ID_DISPLAY, 925, 13, 0, 12);
break;
case CLOCK_OSC_FREQ_19_2:
clock_set_rate(CLOCK_ID_CGENERAL, 125, 4, 0, 0);
clock_set_rate(CLOCK_ID_DISPLAY, 96, 2, 0, 12);
break;
Tegra: clocks: Add 38.4MHz OSC support for T210 use Added 38.4MHz/48MHz entries to pll_x_table for CPU PLL. Needs to be measured - should be close to 700MHz (1.4G/2). Note that some freqs aren't in the PLLU table in T210 TRM (13, 26MHz), so I used the 12MHz table entry for them. They shouldn't be selected since they're not viable T210 OSC freqs. Since there are now 2 new OSC defines, all tables (pll_x_table, PLLU) had to increase by two entries, but since 38.4/48MHz are not viable osc freqs on T20/30/114, etc, they're just set to 0. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-06-22 20:03:44 +00:00
case CLOCK_OSC_FREQ_38_4:
clock_set_rate(CLOCK_ID_CGENERAL, 125, 8, 0, 0);
clock_set_rate(CLOCK_ID_DISPLAY, 96, 4, 0, 0);
break;
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
default:
/*
* These are not supported. It is too early to print a
* message and the UART likely won't work anyway due to the
* oscillator being wrong.
*/
break;
}
/* PLLC_MISC1: Turn IDDQ off. NOTE: T210 PLLC_MISC_1 maps to pll_misc */
clrbits_le32(&clkrst->crc_pll[CLOCK_ID_CGENERAL].pll_misc,
(1 << PLLC_IDDQ));
udelay(2);
/*
* PLLC_MISC: Take PLLC out of reset. NOTE: T210 PLLC_MISC maps
* to pll_out[1]
*/
clrbits_le32(&clkrst->crc_pll[CLOCK_ID_CGENERAL].pll_out[1],
(1 << PLLC_RESET));
udelay(2);
/* PLLD_MISC: Set CLKENABLE and LOCK_DETECT bits */
Tegra: PLL: use per-SoC pllinfo table instead of PLL_DIVM/N/P, etc. Added PLL variables (dividers mask/shift, lock enable/detect, etc.) to new pllinfo struct for each Soc/PLL. PLLA/C/D/E/M/P/U/X. Used pllinfo struct in all clock functions, validated on T210. Should be equivalent to prior code on T124/114/30/20. Thanks to Marcel Ziswiler for corrections to the T20/T30 values. Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-06-25 16:50:44 +00:00
data = (1 << PLLD_ENABLE_CLK) | (1 << pllinfo->lock_ena);
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
writel(data, &clkrst->crc_pll[CLOCK_ID_DISPLAY].pll_misc);
udelay(2);
}
ARM: tegra: Implement clk_m On currently supported SoCs, clk_m always runs at the same frequency as the oscillator input. However newer SoC generations such as Tegra210 no longer have that restriction. Prepare for that by separating clk_m from the oscillator clock and allow SoC code to override the clk_m rate. Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-08-20 09:42:19 +00:00
unsigned int clk_m_get_rate(unsigned parent_rate)
{
struct clk_rst_ctlr *clkrst = (struct clk_rst_ctlr *)NV_PA_CLK_RST_BASE;
u32 value, div;
value = readl(&clkrst->crc_spare_reg0);
div = ((value >> 2) & 0x3) + 1;
return parent_rate / div;
}
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
void arch_timer_init(void)
{
struct sysctr_ctlr *sysctr = (struct sysctr_ctlr *)NV_PA_TSC_BASE;
u32 freq, val;
ARM: tegra: clk_m is the architected timer source clock While clk_m and the oscillator run at the same frequencies on Tegra114 and Tegra124, clk_m is the proper source for the architected timer. On more recent Tegra generations, Tegra210 and later, both the oscillator and clk_m can run at different frequencies. clk_m will be divided down from the oscillator. Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-08-20 09:42:20 +00:00
freq = clock_get_rate(CLOCK_ID_CLK_M);
debug("%s: clk_m freq is %dHz [0x%08X]\n", __func__, freq, freq);
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
ARM: tegra: clk_m is the architected timer source clock While clk_m and the oscillator run at the same frequencies on Tegra114 and Tegra124, clk_m is the proper source for the architected timer. On more recent Tegra generations, Tegra210 and later, both the oscillator and clk_m can run at different frequencies. clk_m will be divided down from the oscillator. Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-08-20 09:42:20 +00:00
if (current_el() == 3)
asm("msr cntfrq_el0, %0\n" : : "r" (freq));
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
/* Only Tegra114+ has the System Counter regs */
debug("%s: setting CNTFID0 to 0x%08X\n", __func__, freq);
writel(freq, &sysctr->cntfid0);
val = readl(&sysctr->cntcr);
val |= TSC_CNTCR_ENABLE | TSC_CNTCR_HDBG;
writel(val, &sysctr->cntcr);
debug("%s: TSC CNTCR = 0x%08X\n", __func__, val);
}
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#define PLLREFE_MISC 0x4c8
#define PLLREFE_MISC_LOCK BIT(27)
#define PLLREFE_MISC_IDDQ BIT(24)
#define PLLREFE_BASE 0x4c4
#define PLLREFE_BASE_BYPASS BIT(31)
#define PLLREFE_BASE_ENABLE BIT(30)
#define PLLREFE_BASE_REF_DIS BIT(29)
#define PLLREFE_BASE_KCP(kcp) (((kcp) & 0x3) << 27)
#define PLLREFE_BASE_KVCO BIT(26)
#define PLLREFE_BASE_DIVP(p) (((p) & 0x1f) << 16)
#define PLLREFE_BASE_DIVN(n) (((n) & 0xff) << 8)
#define PLLREFE_BASE_DIVM(m) (((m) & 0xff) << 0)
static int tegra_pllref_enable(void)
{
u32 value;
unsigned long start;
/*
* This sequence comes from Tegra X1 TRM section "Cold Boot, with no
* Recovery Mode or Boot from USB", sub-section "PLLREFE".
*/
value = readl(NV_PA_CLK_RST_BASE + PLLREFE_MISC);
value &= ~PLLREFE_MISC_IDDQ;
writel(value, NV_PA_CLK_RST_BASE + PLLREFE_MISC);
udelay(5);
value = PLLREFE_BASE_ENABLE |
PLLREFE_BASE_KCP(0) |
PLLREFE_BASE_DIVP(0) |
PLLREFE_BASE_DIVN(0x41) |
PLLREFE_BASE_DIVM(4);
writel(value, NV_PA_CLK_RST_BASE + PLLREFE_BASE);
debug("waiting for pllrefe lock\n");
start = get_timer(0);
while (get_timer(start) < 250) {
value = readl(NV_PA_CLK_RST_BASE + PLLREFE_MISC);
if (value & PLLREFE_MISC_LOCK)
break;
}
if (!(value & PLLREFE_MISC_LOCK)) {
debug(" timeout\n");
return -ETIMEDOUT;
}
debug(" done\n");
return 0;
}
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
#define PLLE_SS_CNTL 0x68
#define PLLE_SS_CNTL_SSCINCINTR(x) (((x) & 0x3f) << 24)
#define PLLE_SS_CNTL_SSCINC(x) (((x) & 0xff) << 16)
#define PLLE_SS_CNTL_SSCINVERT (1 << 15)
#define PLLE_SS_CNTL_SSCCENTER (1 << 14)
#define PLLE_SS_CNTL_SSCBYP (1 << 12)
#define PLLE_SS_CNTL_INTERP_RESET (1 << 11)
#define PLLE_SS_CNTL_BYPASS_SS (1 << 10)
#define PLLE_SS_CNTL_SSCMAX(x) (((x) & 0x1ff) << 0)
#define PLLE_BASE 0x0e8
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#define PLLE_BASE_ENABLE (1 << 31)
#define PLLE_BASE_PLDIV_CML(x) (((x) & 0x1f) << 24)
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
#define PLLE_BASE_NDIV(x) (((x) & 0xff) << 8)
#define PLLE_BASE_MDIV(x) (((x) & 0xff) << 0)
#define PLLE_MISC 0x0ec
#define PLLE_MISC_IDDQ_SWCTL (1 << 14)
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#define PLLE_MISC_IDDQ_OVERRIDE_VALUE (1 << 13)
#define PLLE_MISC_LOCK (1 << 11)
ARM: tegra210: set PLLE_PTS bit when enabling PLLE This bit needs to be set for system suspend/resume to work. This setting will be documented in an updated TRM at some time in the future. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2016-03-22 15:45:36 +00:00
#define PLLE_PTS (1 << 8)
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#define PLLE_MISC_KCP(x) (((x) & 0x3) << 6)
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
#define PLLE_MISC_VREG_CTRL(x) (((x) & 0x3) << 2)
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#define PLLE_MISC_KVCO (1 << 0)
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
#define PLLE_AUX 0x48c
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#define PLLE_AUX_SS_SEQ_INCLUDE (1 << 31)
#define PLLE_AUX_REF_SEL_PLLREFE (1 << 28)
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
#define PLLE_AUX_SEQ_ENABLE (1 << 24)
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#define PLLE_AUX_SS_SWCTL (1 << 6)
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
#define PLLE_AUX_ENABLE_SWCTL (1 << 4)
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
#define PLLE_AUX_USE_LOCKDET (1 << 3)
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
int tegra_plle_enable(void)
{
u32 value;
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
unsigned long start;
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
/* PLLREF feeds PLLE */
tegra_pllref_enable();
/*
* This sequence comes from Tegra X1 TRM section "Cold Boot, with no
* Recovery Mode or Boot from USB", sub-section "PLLEs".
*/
/* 1. Select XTAL as the source */
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
value = readl(NV_PA_CLK_RST_BASE + PLLE_AUX);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
value &= ~PLLE_AUX_REF_SEL_PLLREFE;
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
writel(value, NV_PA_CLK_RST_BASE + PLLE_AUX);
value = readl(NV_PA_CLK_RST_BASE + PLLE_MISC);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
value &= ~PLLE_MISC_IDDQ_OVERRIDE_VALUE;
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
writel(value, NV_PA_CLK_RST_BASE + PLLE_MISC);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
/* 2. Wait 5 us */
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
udelay(5);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
/*
* 3. Program the following registers to generate a low jitter 100MHz
* clock.
*/
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
value = readl(NV_PA_CLK_RST_BASE + PLLE_BASE);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
value &= ~PLLE_BASE_PLDIV_CML(0x1f);
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
value &= ~PLLE_BASE_NDIV(0xff);
value &= ~PLLE_BASE_MDIV(0xff);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
value |= PLLE_BASE_PLDIV_CML(0xe);
value |= PLLE_BASE_NDIV(0x7d);
value |= PLLE_BASE_MDIV(2);
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
writel(value, NV_PA_CLK_RST_BASE + PLLE_BASE);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
value = readl(NV_PA_CLK_RST_BASE + PLLE_MISC);
ARM: tegra210: set PLLE_PTS bit when enabling PLLE This bit needs to be set for system suspend/resume to work. This setting will be documented in an updated TRM at some time in the future. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2016-03-22 15:45:36 +00:00
value |= PLLE_PTS;
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
value &= ~PLLE_MISC_KCP(3);
value &= ~PLLE_MISC_VREG_CTRL(3);
value &= ~PLLE_MISC_KVCO;
writel(value, NV_PA_CLK_RST_BASE + PLLE_MISC);
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
value = readl(NV_PA_CLK_RST_BASE + PLLE_BASE);
value |= PLLE_BASE_ENABLE;
writel(value, NV_PA_CLK_RST_BASE + PLLE_BASE);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
/* 4. Wait for LOCK */
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
debug("waiting for plle lock\n");
start = get_timer(0);
while (get_timer(start) < 250) {
value = readl(NV_PA_CLK_RST_BASE + PLLE_MISC);
if (value & PLLE_MISC_LOCK)
break;
}
if (!(value & PLLE_MISC_LOCK)) {
debug(" timeout\n");
return -ETIMEDOUT;
}
debug(" done\n");
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
/* 5. Enable SSA */
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
value = readl(NV_PA_CLK_RST_BASE + PLLE_SS_CNTL);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
value &= ~PLLE_SS_CNTL_SSCINC(0xff);
value |= PLLE_SS_CNTL_SSCINC(1);
value &= ~PLLE_SS_CNTL_SSCINCINTR(0x3f);
value |= PLLE_SS_CNTL_SSCINCINTR(0x23);
value &= ~PLLE_SS_CNTL_SSCMAX(0x1fff);
value |= PLLE_SS_CNTL_SSCMAX(0x21);
value &= ~PLLE_SS_CNTL_SSCINVERT;
value &= ~PLLE_SS_CNTL_SSCCENTER;
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
value &= ~PLLE_SS_CNTL_BYPASS_SS;
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
value &= ~PLLE_SS_CNTL_SSCBYP;
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
writel(value, NV_PA_CLK_RST_BASE + PLLE_SS_CNTL);
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
/* 6. Wait 300 ns */
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
ARM: tegra210: implement PLLE init procedure from TRM Implement the procedure that the TRM mandates to initialize PLLREFE and PLLE. This makes the PLL actually lock. Note that this section of the TRM is being cleaned up to remove some confusion. The set of register accesses in this patch should be final, although the step numbers/descriptions might still change. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-10-05 22:58:52 +00:00
udelay(1);
ARM: Tegra210: Add SoC code/include files for T210 All based off of Tegra124. As a Tegra210 board is brought up, these may change a bit to match the HW more closely, but probably 90% of this is identical to T124. Note that since T210 is a 64-bit build, it has no SPL component, and hence no cpu.c for Tegra210. Signed-off-by: Tom Warren <twarren@nvidia.com>
2015-02-02 20:22:29 +00:00
value &= ~PLLE_SS_CNTL_INTERP_RESET;
writel(value, NV_PA_CLK_RST_BASE + PLLE_SS_CNTL);
return 0;
}
ARM: tegra: add peripheral clock init table Currently, Tegra peripheral drivers control two aspects of their HW module clock(s): 1) The clock enable/rate for the peripheral clock itself. 2) The system-level clock tree setup, i.e. the clock parent. Aspect 1 is reasonable, but aspect 2 is a system-level decision, not something that an individual peripheral driver should in general know about or influence. Such system-level knowledge ties the driver to a specific SoC implementation, even when they use generic APIs for clock manipulation, since they must have SoC-specific knowledge such as parent clock IDs. Limited exceptions exist, such as where peripheral HW is expected to dynamically switch between clock sources at run-time, such as CPU clock scaling or display clock conflict management in a multi-head scenario. This patch enhances the Tegra core code to perform system-level clock tree setup, in a similar fashion to the Linux kernel Tegra clock driver. This will allow future patches to simplify peripheral drivers by removing the clock parent setup logic. This change is required prior to converting peripheral drivers to use the standard clock APIs, since: 1) The clock uclass doesn't currently support a set_parent() operation. Adding one is possible, but not necessary at the moment. 2) The clock APIs retrieve all clock IDs from device tree, and the DT bindings for almost all peripherals only includes information about the relevant peripheral clocks, and not any potential parent clocks. Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Tom Warren <twarren@nvidia.com>
2016-09-13 16:45:55 +00:00
struct periph_clk_init periph_clk_init_table[] = {
{ PERIPH_ID_SBC1, CLOCK_ID_PERIPH },
{ PERIPH_ID_SBC2, CLOCK_ID_PERIPH },
{ PERIPH_ID_SBC3, CLOCK_ID_PERIPH },
{ PERIPH_ID_SBC4, CLOCK_ID_PERIPH },
{ PERIPH_ID_SBC5, CLOCK_ID_PERIPH },
{ PERIPH_ID_SBC6, CLOCK_ID_PERIPH },
{ PERIPH_ID_HOST1X, CLOCK_ID_PERIPH },
{ PERIPH_ID_SDMMC1, CLOCK_ID_PERIPH },
{ PERIPH_ID_SDMMC2, CLOCK_ID_PERIPH },
{ PERIPH_ID_SDMMC3, CLOCK_ID_PERIPH },
{ PERIPH_ID_SDMMC4, CLOCK_ID_PERIPH },
{ PERIPH_ID_PWM, CLOCK_ID_SFROM32KHZ },
{ PERIPH_ID_I2C1, CLOCK_ID_PERIPH },
{ PERIPH_ID_I2C2, CLOCK_ID_PERIPH },
{ PERIPH_ID_I2C3, CLOCK_ID_PERIPH },
{ PERIPH_ID_I2C4, CLOCK_ID_PERIPH },
{ PERIPH_ID_I2C5, CLOCK_ID_PERIPH },
{ PERIPH_ID_I2C6, CLOCK_ID_PERIPH },
{ -1, },
};
Reference in a new issue Copy permalink