Mirrors/u-boot
2
0
Fork 0
mirror of https://github.com/AsahiLinux/u-boot synced 2024-11-06 21:24:29 +00:00
Code Issues Projects Releases Packages Wiki Activity
u-boot/drivers/ddr/marvell/a38x/ddr3_training_pbs.c

1013 lines
30 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: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/*
* Copyright (C) Marvell International Ltd. and its affiliates
*/
#include "ddr3_init.h"
ARM: mvebu: a38x: sync ddr training code with mv_ddr-armada-18.09.02 This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-18.09 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. Specifically this syncs with commit 99d772547314 ("Bump mv_ddr to release armada-18.09.2"). The complete log of changes is best obtained from the mv-ddr-marvell.git repository but some relevant highlights are: ddr3: add missing txsdll parameter ddr3: fix tfaw timimg parameter ddr3: fix trrd timimg parameter merge ddr3 topology header file with mv_ddr_topology one mv_ddr: a38x: fix zero memory size scrubbing issue The upstream code is incorporated omitting the portions not relevant to Armada-38x and DDR3. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT -UCONFIG_A3700 -UA3900 -UA80X0 \ -UA70X0 Signed-off-by: Chris Packham <judge.packham@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Baruch Siach <baruch@tkos.co.il> Signed-off-by: Stefan Roese <sr@denx.de>
2018-12-03 01:26:49 +00:00
#include "mv_ddr_training_db.h"
#include "mv_ddr_common.h"
#include "mv_ddr_regs.h"
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
#define TYPICAL_PBS_VALUE 12
u32 nominal_adll[MAX_INTERFACE_NUM * MAX_BUS_NUM];
enum hws_training_ip_stat train_status[MAX_INTERFACE_NUM];
u8 result_mat[MAX_INTERFACE_NUM][MAX_BUS_NUM][BUS_WIDTH_IN_BITS];
u8 result_mat_rx_dqs[MAX_INTERFACE_NUM][MAX_BUS_NUM][MAX_CS_NUM];
/* 4-EEWA, 3-EWA, 2-SWA, 1-Fail, 0-Pass */
u8 result_all_bit[MAX_BUS_NUM * BUS_WIDTH_IN_BITS * MAX_INTERFACE_NUM];
u8 max_pbs_per_pup[MAX_INTERFACE_NUM][MAX_BUS_NUM];
u8 min_pbs_per_pup[MAX_INTERFACE_NUM][MAX_BUS_NUM];
u8 max_adll_per_pup[MAX_INTERFACE_NUM][MAX_BUS_NUM];
u8 min_adll_per_pup[MAX_INTERFACE_NUM][MAX_BUS_NUM];
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
u32 pbsdelay_per_pup[NUM_OF_PBS_MODES][MAX_INTERFACE_NUM][MAX_BUS_NUM][MAX_CS_NUM];
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
u8 adll_shift_lock[MAX_INTERFACE_NUM][MAX_BUS_NUM];
u8 adll_shift_val[MAX_INTERFACE_NUM][MAX_BUS_NUM];
enum hws_pattern pbs_pattern = PATTERN_VREF;
static u8 pup_state[MAX_INTERFACE_NUM][MAX_BUS_NUM];
/*
* Name: ddr3_tip_pbs
* Desc: PBS
* Args: TBD
* Notes:
* Returns: OK if success, other error code if fail.
*/
int ddr3_tip_pbs(u32 dev_num, enum pbs_dir pbs_mode)
{
u32 res0[MAX_INTERFACE_NUM];
ARM: mvebu: a38x: sync ddr training code with mv_ddr-armada-18.09.02 This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-18.09 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. Specifically this syncs with commit 99d772547314 ("Bump mv_ddr to release armada-18.09.2"). The complete log of changes is best obtained from the mv-ddr-marvell.git repository but some relevant highlights are: ddr3: add missing txsdll parameter ddr3: fix tfaw timimg parameter ddr3: fix trrd timimg parameter merge ddr3 topology header file with mv_ddr_topology one mv_ddr: a38x: fix zero memory size scrubbing issue The upstream code is incorporated omitting the portions not relevant to Armada-38x and DDR3. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT -UCONFIG_A3700 -UA3900 -UA80X0 \ -UA70X0 Signed-off-by: Chris Packham <judge.packham@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Baruch Siach <baruch@tkos.co.il> Signed-off-by: Stefan Roese <sr@denx.de>
2018-12-03 01:26:49 +00:00
int adll_tap = MEGA / mv_ddr_freq_get(medium_freq) / 64;
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
int pad_num = 0;
enum hws_search_dir search_dir =
(pbs_mode == PBS_RX_MODE) ? HWS_HIGH2LOW : HWS_LOW2HIGH;
enum hws_dir dir = (pbs_mode == PBS_RX_MODE) ? OPER_READ : OPER_WRITE;
int iterations = (pbs_mode == PBS_RX_MODE) ? 31 : 63;
u32 res_valid_mask = (pbs_mode == PBS_RX_MODE) ? 0x1f : 0x3f;
int init_val = (search_dir == HWS_LOW2HIGH) ? 0 : iterations;
enum hws_edge_compare search_edge = EDGE_FP;
u32 pup = 0, bit = 0, if_id = 0, all_lock = 0, cs_num = 0;
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
u32 reg_addr = 0;
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
u32 validation_val = 0;
u32 cs_enable_reg_val[MAX_INTERFACE_NUM];
u16 *mask_results_dq_reg_map = ddr3_tip_get_mask_results_dq_reg();
u8 temp = 0;
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
u32 octets_per_if_num = ddr3_tip_dev_attr_get(dev_num, MV_ATTR_OCTET_PER_INTERFACE);
struct mv_ddr_topology_map *tm = mv_ddr_topology_map_get();
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/* save current cs enable reg val */
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/* save current cs enable reg val */
CHECK_STATUS(ddr3_tip_if_read
(dev_num, ACCESS_TYPE_UNICAST, if_id,
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
DUAL_DUNIT_CFG_REG, cs_enable_reg_val, MASK_ALL_BITS));
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/* enable single cs */
CHECK_STATUS(ddr3_tip_if_write
(dev_num, ACCESS_TYPE_UNICAST, if_id,
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
DUAL_DUNIT_CFG_REG, (1 << 3), (1 << 3)));
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
}
reg_addr = (pbs_mode == PBS_RX_MODE) ?
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
CRX_PHY_REG(effective_cs) :
CTX_PHY_REG(effective_cs);
ddr3_tip_read_adll_value(dev_num, nominal_adll, reg_addr, MASK_ALL_BITS);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/* stage 1 shift ADLL */
ddr3_tip_ip_training(dev_num, ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE, ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE, RESULT_PER_BIT,
HWS_CONTROL_ELEMENT_ADLL, search_dir, dir,
tm->if_act_mask, init_val, iterations,
pbs_pattern, search_edge, CS_SINGLE, cs_num,
train_status);
validation_val = (pbs_mode == PBS_RX_MODE) ? 0x1f : 0;
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
min_adll_per_pup[if_id][pup] =
(pbs_mode == PBS_RX_MODE) ? 0x1f : 0x3f;
pup_state[if_id][pup] = 0x3;
adll_shift_lock[if_id][pup] = 1;
max_adll_per_pup[if_id][pup] = 0x0;
}
}
/* EBA */
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
CHECK_STATUS(ddr3_tip_if_read
(dev_num, ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE,
mask_results_dq_reg_map[
bit + pup * BUS_WIDTH_IN_BITS],
res0, MASK_ALL_BITS));
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1;
if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
("FP I/F %d, bit:%d, pup:%d res0 0x%x\n",
if_id, bit, pup,
res0[if_id]));
if (pup_state[if_id][pup] != 3)
continue;
/* if not EBA state than move to next pup */
if ((res0[if_id] & 0x2000000) == 0) {
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
("-- Fail Training IP\n"));
/* training machine failed */
pup_state[if_id][pup] = 1;
adll_shift_lock[if_id][pup] = 0;
continue;
}
else if ((res0[if_id] & res_valid_mask) ==
validation_val) {
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
("-- FAIL EBA %d %d %d %d\n",
if_id, bit, pup,
res0[if_id]));
pup_state[if_id][pup] = 4;
/* this pup move to EEBA */
adll_shift_lock[if_id][pup] = 0;
continue;
} else {
/*
* The search ended in Pass we need
* Fail
*/
res0[if_id] =
(pbs_mode == PBS_RX_MODE) ?
((res0[if_id] &
res_valid_mask) + 1) :
((res0[if_id] &
res_valid_mask) - 1);
max_adll_per_pup[if_id][pup] =
(max_adll_per_pup[if_id][pup] <
res0[if_id]) ?
(u8)res0[if_id] :
max_adll_per_pup[if_id][pup];
min_adll_per_pup[if_id][pup] =
(res0[if_id] >
min_adll_per_pup[if_id][pup]) ?
min_adll_per_pup[if_id][pup] :
(u8)
res0[if_id];
/*
* vs the Rx we are searching for the
* smallest value of DQ shift so all
* Bus would fail
*/
adll_shift_val[if_id][pup] =
(pbs_mode == PBS_RX_MODE) ?
max_adll_per_pup[if_id][pup] :
min_adll_per_pup[if_id][pup];
}
}
}
}
/* EEBA */
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
if (pup_state[if_id][pup] != 4)
continue;
/*
* if pup state different from EEBA than move to
* next pup
*/
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x54 + effective_cs * 0x10) :
(0x14 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST, if_id,
ACCESS_TYPE_UNICAST, pup, DDR_PHY_DATA,
reg_addr, 0x1f));
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x55 + effective_cs * 0x10) :
(0x15 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST, if_id,
ACCESS_TYPE_UNICAST, pup, DDR_PHY_DATA,
reg_addr, 0x1f));
/* initialize the Edge2 Max. */
adll_shift_val[if_id][pup] = 0;
min_adll_per_pup[if_id][pup] =
(pbs_mode == PBS_RX_MODE) ? 0x1f : 0x3f;
max_adll_per_pup[if_id][pup] = 0x0;
ddr3_tip_ip_training(dev_num, ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE,
ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE, RESULT_PER_BIT,
HWS_CONTROL_ELEMENT_ADLL,
search_dir, dir,
tm->if_act_mask, init_val,
iterations, pbs_pattern,
search_edge, CS_SINGLE, cs_num,
train_status);
DEBUG_PBS_ENGINE(DEBUG_LEVEL_INFO,
("ADLL shift results:\n"));
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
CHECK_STATUS(ddr3_tip_if_read
(dev_num, ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE,
mask_results_dq_reg_map[
bit + pup *
BUS_WIDTH_IN_BITS],
res0, MASK_ALL_BITS));
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
("FP I/F %d, bit:%d, pup:%d res0 0x%x\n",
if_id, bit, pup,
res0[if_id]));
if ((res0[if_id] & 0x2000000) == 0) {
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
(" -- EEBA Fail\n"));
bit = BUS_WIDTH_IN_BITS;
/* exit bit loop */
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
("-- EEBA Fail Training IP\n"));
/*
* training machine failed but pass
* before in the EBA so maybe the DQS
* shift change env.
*/
pup_state[if_id][pup] = 2;
adll_shift_lock[if_id][pup] = 0;
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x54 + effective_cs * 0x10) :
(0x14 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num,
ACCESS_TYPE_UNICAST,
if_id,
ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
0x0));
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x55 + effective_cs * 0x10) :
(0x15 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num,
ACCESS_TYPE_UNICAST,
if_id,
ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
0x0));
continue;
} else if ((res0[if_id] & res_valid_mask) ==
validation_val) {
/* exit bit loop */
bit = BUS_WIDTH_IN_BITS;
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
("-- FAIL EEBA\n"));
/* this pup move to SBA */
pup_state[if_id][pup] = 2;
adll_shift_lock[if_id][pup] = 0;
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x54 + effective_cs * 0x10) :
(0x14 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num,
ACCESS_TYPE_UNICAST,
if_id,
ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
0x0));
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x55 + effective_cs * 0x10) :
(0x15 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num,
ACCESS_TYPE_UNICAST,
if_id,
ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
0x0));
continue;
} else {
adll_shift_lock[if_id][pup] = 1;
/*
* The search ended in Pass we need
* Fail
*/
res0[if_id] =
(pbs_mode == PBS_RX_MODE) ?
((res0[if_id] &
res_valid_mask) + 1) :
((res0[if_id] &
res_valid_mask) - 1);
max_adll_per_pup[if_id][pup] =
(max_adll_per_pup[if_id][pup] <
res0[if_id]) ?
(u8)res0[if_id] :
max_adll_per_pup[if_id][pup];
min_adll_per_pup[if_id][pup] =
(res0[if_id] >
min_adll_per_pup[if_id][pup]) ?
min_adll_per_pup[if_id][pup] :
(u8)res0[if_id];
/*
* vs the Rx we are searching for the
* smallest value of DQ shift so all Bus
* would fail
*/
adll_shift_val[if_id][pup] =
(pbs_mode == PBS_RX_MODE) ?
max_adll_per_pup[if_id][pup] :
min_adll_per_pup[if_id][pup];
}
}
}
}
/* Print Stage result */
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
("FP I/F %d, ADLL Shift for EBA: pup[%d] Lock status = %d Lock Val = %d,%d\n",
if_id, pup,
adll_shift_lock[if_id][pup],
max_adll_per_pup[if_id][pup],
min_adll_per_pup[if_id][pup]));
}
}
DEBUG_PBS_ENGINE(DEBUG_LEVEL_INFO,
("Update ADLL Shift of all pups:\n"));
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
if (adll_shift_lock[if_id][pup] != 1)
continue;
/* if pup not locked continue to next pup */
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x3 + effective_cs * 4) :
(0x1 + effective_cs * 4);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST, if_id,
ACCESS_TYPE_UNICAST, pup, DDR_PHY_DATA,
reg_addr, adll_shift_val[if_id][pup]));
DEBUG_PBS_ENGINE(DEBUG_LEVEL_TRACE,
("FP I/F %d, Pup[%d] = %d\n", if_id,
pup, adll_shift_val[if_id][pup]));
}
}
/* PBS EEBA&EBA */
/* Start the Per Bit Skew search */
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
max_pbs_per_pup[if_id][pup] = 0x0;
min_pbs_per_pup[if_id][pup] = 0x1f;
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
/* reset result for PBS */
result_all_bit[bit + pup * BUS_WIDTH_IN_BITS +
if_id * MAX_BUS_NUM *
BUS_WIDTH_IN_BITS] = 0;
}
}
}
iterations = 31;
search_dir = HWS_LOW2HIGH;
/* !!!!! ran sh (search_dir == HWS_LOW2HIGH)?0:iterations; */
init_val = 0;
ddr3_tip_ip_training(dev_num, ACCESS_TYPE_MULTICAST, PARAM_NOT_CARE,
ACCESS_TYPE_MULTICAST, PARAM_NOT_CARE,
RESULT_PER_BIT, HWS_CONTROL_ELEMENT_DQ_SKEW,
search_dir, dir, tm->if_act_mask, init_val,
iterations, pbs_pattern, search_edge,
CS_SINGLE, cs_num, train_status);
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
if (adll_shift_lock[if_id][pup] != 1) {
/* if pup not lock continue to next pup */
continue;
}
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
CHECK_STATUS(ddr3_tip_if_read
(dev_num, ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE,
mask_results_dq_reg_map[
bit +
pup * BUS_WIDTH_IN_BITS],
res0, MASK_ALL_BITS));
DEBUG_PBS_ENGINE(DEBUG_LEVEL_INFO,
("Per Bit Skew search, FP I/F %d, bit:%d, pup:%d res0 0x%x\n",
if_id, bit, pup,
res0[if_id]));
if ((res0[if_id] & 0x2000000) == 0) {
DEBUG_PBS_ENGINE(DEBUG_LEVEL_INFO,
("--EBA PBS Fail - Training IP machine\n"));
/* exit the bit loop */
bit = BUS_WIDTH_IN_BITS;
/*
* ADLL is no long in lock need new
* search
*/
adll_shift_lock[if_id][pup] = 0;
/* Move to SBA */
pup_state[if_id][pup] = 2;
max_pbs_per_pup[if_id][pup] = 0x0;
min_pbs_per_pup[if_id][pup] = 0x1f;
continue;
} else {
temp = (u8)(res0[if_id] &
res_valid_mask);
max_pbs_per_pup[if_id][pup] =
(temp >
max_pbs_per_pup[if_id][pup]) ?
temp :
max_pbs_per_pup[if_id][pup];
min_pbs_per_pup[if_id][pup] =
(temp <
min_pbs_per_pup[if_id][pup]) ?
temp :
min_pbs_per_pup[if_id][pup];
result_all_bit[bit +
pup * BUS_WIDTH_IN_BITS +
if_id * MAX_BUS_NUM *
BUS_WIDTH_IN_BITS] =
temp;
}
}
}
}
/* Check all Pup lock */
all_lock = 1;
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
all_lock = all_lock * adll_shift_lock[if_id][pup];
}
}
/* Only if not all Pups Lock */
if (all_lock == 0) {
DEBUG_PBS_ENGINE(DEBUG_LEVEL_INFO,
("##########ADLL shift for SBA###########\n"));
/* ADLL shift for SBA */
search_dir = (pbs_mode == PBS_RX_MODE) ? HWS_LOW2HIGH :
HWS_HIGH2LOW;
init_val = (search_dir == HWS_LOW2HIGH) ? 0 : iterations;
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1;
if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
if (adll_shift_lock[if_id][pup] == 1) {
/*if pup lock continue to next pup */
continue;
}
/*init the var altogth init before */
adll_shift_lock[if_id][pup] = 0;
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x54 + effective_cs * 0x10) :
(0x14 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr, 0));
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x55 + effective_cs * 0x10) :
(0x15 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr, 0));
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x5f + effective_cs * 0x10) :
(0x1f + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr, 0));
/* initilaze the Edge2 Max. */
adll_shift_val[if_id][pup] = 0;
min_adll_per_pup[if_id][pup] = 0x1f;
max_adll_per_pup[if_id][pup] = 0x0;
ddr3_tip_ip_training(dev_num,
ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE,
ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE,
RESULT_PER_BIT,
HWS_CONTROL_ELEMENT_ADLL,
search_dir, dir,
tm->if_act_mask,
init_val, iterations,
pbs_pattern,
search_edge, CS_SINGLE,
cs_num, train_status);
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
CHECK_STATUS(ddr3_tip_if_read
(dev_num,
ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE,
mask_results_dq_reg_map
[bit +
pup *
BUS_WIDTH_IN_BITS],
res0, MASK_ALL_BITS));
DEBUG_PBS_ENGINE(
DEBUG_LEVEL_INFO,
("FP I/F %d, bit:%d, pup:%d res0 0x%x\n",
if_id, bit, pup, res0[if_id]));
if ((res0[if_id] & 0x2000000) == 0) {
/* exit the bit loop */
bit = BUS_WIDTH_IN_BITS;
/* Fail SBA --> Fail PBS */
pup_state[if_id][pup] = 1;
DEBUG_PBS_ENGINE
(DEBUG_LEVEL_INFO,
(" SBA Fail\n"));
continue;
} else {
/*
* - increment to get all
* 8 bit lock.
*/
adll_shift_lock[if_id][pup]++;
/*
* The search ended in Pass
* we need Fail
*/
res0[if_id] =
(pbs_mode == PBS_RX_MODE) ?
((res0[if_id] & res_valid_mask) + 1) :
((res0[if_id] & res_valid_mask) - 1);
max_adll_per_pup[if_id][pup] =
(max_adll_per_pup[if_id]
[pup] < res0[if_id]) ?
(u8)res0[if_id] :
max_adll_per_pup[if_id][pup];
min_adll_per_pup[if_id][pup] =
(res0[if_id] >
min_adll_per_pup[if_id]
[pup]) ?
min_adll_per_pup[if_id][pup] :
(u8)res0[if_id];
/*
* vs the Rx we are searching for
* the smallest value of DQ shift
* so all Bus would fail
*/
adll_shift_val[if_id][pup] =
(pbs_mode == PBS_RX_MODE) ?
max_adll_per_pup[if_id][pup] :
min_adll_per_pup[if_id][pup];
}
}
/* 1 is lock */
adll_shift_lock[if_id][pup] =
(adll_shift_lock[if_id][pup] == 8) ?
1 : 0;
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x3 + effective_cs * 4) :
(0x1 + effective_cs * 4);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
adll_shift_val[if_id][pup]));
DEBUG_PBS_ENGINE(
DEBUG_LEVEL_INFO,
("adll_shift_lock[%x][%x] = %x\n",
if_id, pup,
adll_shift_lock[if_id][pup]));
}
}
/* End ADLL Shift for SBA */
/* Start the Per Bit Skew search */
/* The ADLL shift finished with a Pass */
search_edge = (pbs_mode == PBS_RX_MODE) ? EDGE_PF : EDGE_FP;
search_dir = (pbs_mode == PBS_RX_MODE) ?
HWS_LOW2HIGH : HWS_HIGH2LOW;
iterations = 0x1f;
/* - The initial value is different in Rx and Tx mode */
init_val = (pbs_mode == PBS_RX_MODE) ? 0 : iterations;
ddr3_tip_ip_training(dev_num, ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE, ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE, RESULT_PER_BIT,
HWS_CONTROL_ELEMENT_DQ_SKEW,
search_dir, dir, tm->if_act_mask,
init_val, iterations, pbs_pattern,
search_edge, CS_SINGLE, cs_num,
train_status);
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1;
if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
CHECK_STATUS(ddr3_tip_if_read
(dev_num,
ACCESS_TYPE_MULTICAST,
PARAM_NOT_CARE,
mask_results_dq_reg_map
[bit +
pup *
BUS_WIDTH_IN_BITS],
res0, MASK_ALL_BITS));
if (pup_state[if_id][pup] != 2) {
/*
* if pup is not SBA continue
* to next pup
*/
bit = BUS_WIDTH_IN_BITS;
continue;
}
DEBUG_PBS_ENGINE(
DEBUG_LEVEL_INFO,
("Per Bit Skew search, PF I/F %d, bit:%d, pup:%d res0 0x%x\n",
if_id, bit, pup, res0[if_id]));
if ((res0[if_id] & 0x2000000) == 0) {
DEBUG_PBS_ENGINE
(DEBUG_LEVEL_INFO,
("SBA Fail\n"));
max_pbs_per_pup[if_id][pup] =
0x1f;
result_all_bit[
bit + pup *
BUS_WIDTH_IN_BITS +
if_id * MAX_BUS_NUM *
BUS_WIDTH_IN_BITS] =
0x1f;
} else {
temp = (u8)(res0[if_id] &
res_valid_mask);
max_pbs_per_pup[if_id][pup] =
(temp >
max_pbs_per_pup[if_id]
[pup]) ? temp :
max_pbs_per_pup
[if_id][pup];
min_pbs_per_pup[if_id][pup] =
(temp <
min_pbs_per_pup[if_id]
[pup]) ? temp :
min_pbs_per_pup
[if_id][pup];
result_all_bit[
bit + pup *
BUS_WIDTH_IN_BITS +
if_id * MAX_BUS_NUM *
BUS_WIDTH_IN_BITS] =
temp;
adll_shift_lock[if_id][pup] = 1;
}
}
}
}
/* Check all Pup state */
all_lock = 1;
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/*
* DEBUG_PBS_ENGINE(DEBUG_LEVEL_INFO,
* ("pup_state[%d][%d] = %d\n",if_id,pup,pup_state
* [if_id][pup]));
*/
}
}
/* END OF SBA */
/* Norm */
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1;
if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/* if pup not lock continue to next pup */
if (adll_shift_lock[if_id][pup] != 1) {
DEBUG_PBS_ENGINE(
DEBUG_LEVEL_ERROR,
("PBS failed for IF #%d\n",
if_id));
training_result[training_stage][if_id]
= TEST_FAILED;
result_mat[if_id][pup][bit] = 0;
max_pbs_per_pup[if_id][pup] = 0;
min_pbs_per_pup[if_id][pup] = 0;
} else {
training_result[
training_stage][if_id] =
(training_result[training_stage]
[if_id] == TEST_FAILED) ?
TEST_FAILED : TEST_SUCCESS;
result_mat[if_id][pup][bit] =
result_all_bit[
bit + pup *
BUS_WIDTH_IN_BITS +
if_id * MAX_BUS_NUM *
BUS_WIDTH_IN_BITS] -
min_pbs_per_pup[if_id][pup];
}
DEBUG_PBS_ENGINE(
DEBUG_LEVEL_INFO,
("The abs min_pbs[%d][%d] = %d\n",
if_id, pup,
min_pbs_per_pup[if_id][pup]));
}
}
}
/* Clean all results */
ddr3_tip_clean_pbs_result(dev_num, pbs_mode);
/* DQ PBS register update with the final result */
for (if_id = 0; if_id < MAX_INTERFACE_NUM; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
DEBUG_PBS_ENGINE(
DEBUG_LEVEL_INFO,
("Final Results: if_id %d, pup %d, Pup State: %d\n",
if_id, pup, pup_state[if_id][pup]));
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
if (dq_map_table == NULL) {
DEBUG_PBS_ENGINE(
DEBUG_LEVEL_ERROR,
("dq_map_table not initialized\n"));
return MV_FAIL;
}
pad_num = dq_map_table[
bit + pup * BUS_WIDTH_IN_BITS +
if_id * BUS_WIDTH_IN_BITS *
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
MAX_BUS_NUM];
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
DEBUG_PBS_ENGINE(DEBUG_LEVEL_INFO,
("result_mat: %d ",
result_mat[if_id][pup]
[bit]));
reg_addr = (pbs_mode == PBS_RX_MODE) ?
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
PBS_RX_PHY_REG(effective_cs, 0) :
PBS_TX_PHY_REG(effective_cs, 0);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr + pad_num,
result_mat[if_id][pup][bit]));
}
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
if (max_pbs_per_pup[if_id][pup] == min_pbs_per_pup[if_id][pup]) {
temp = TYPICAL_PBS_VALUE;
} else {
temp = ((max_adll_per_pup[if_id][pup] -
min_adll_per_pup[if_id][pup]) *
adll_tap /
(max_pbs_per_pup[if_id][pup] -
min_pbs_per_pup[if_id][pup]));
}
pbsdelay_per_pup[pbs_mode]
[if_id][pup][effective_cs] = temp;
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/* RX results ready, write RX also */
if (pbs_mode == PBS_TX_MODE) {
/* Write TX results */
reg_addr = (0x14 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
(max_pbs_per_pup[if_id][pup] -
min_pbs_per_pup[if_id][pup]) /
2));
reg_addr = (0x15 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
(max_pbs_per_pup[if_id][pup] -
min_pbs_per_pup[if_id][pup]) /
2));
/* Write previously stored RX results */
reg_addr = (0x54 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
result_mat_rx_dqs[if_id][pup]
[effective_cs]));
reg_addr = (0x55 + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr,
result_mat_rx_dqs[if_id][pup]
[effective_cs]));
} else {
/*
* RX results may affect RL results correctess,
* so just store the results that will written
* in TX stage
*/
result_mat_rx_dqs[if_id][pup][effective_cs] =
(max_pbs_per_pup[if_id][pup] -
min_pbs_per_pup[if_id][pup]) / 2;
}
DEBUG_PBS_ENGINE(
DEBUG_LEVEL_INFO,
(", PBS tap=%d [psec] ==> skew observed = %d\n",
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
temp,
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
((max_pbs_per_pup[if_id][pup] -
min_pbs_per_pup[if_id][pup]) *
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
temp)));
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
}
}
/* Write back to the phy the default values */
reg_addr = (pbs_mode == PBS_RX_MODE) ?
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
CRX_PHY_REG(effective_cs) :
CTX_PHY_REG(effective_cs);
ddr3_tip_write_adll_value(dev_num, nominal_adll, reg_addr);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
reg_addr = (pbs_mode == PBS_RX_MODE) ?
(0x5a + effective_cs * 0x10) :
(0x1a + effective_cs * 0x10);
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST, if_id,
ACCESS_TYPE_UNICAST, pup, DDR_PHY_DATA, reg_addr,
0));
/* restore cs enable value */
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
CHECK_STATUS(ddr3_tip_if_write
(dev_num, ACCESS_TYPE_UNICAST, if_id,
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
DUAL_DUNIT_CFG_REG, cs_enable_reg_val[if_id],
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
MASK_ALL_BITS));
}
/* exit test mode */
CHECK_STATUS(ddr3_tip_if_write
(dev_num, ACCESS_TYPE_MULTICAST, PARAM_NOT_CARE,
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
ODPG_WR_RD_MODE_ENA_REG, 0xffff, MASK_ALL_BITS));
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
/*
* no valid window found
* (no lock at EBA ADLL shift at EBS)
*/
if (pup_state[if_id][pup] == 1)
return MV_FAIL;
}
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
}
return MV_OK;
}
/*
* Name: ddr3_tip_pbs_rx.
* Desc: PBS TX
* Args: TBD
* Notes:
* Returns: OK if success, other error code if fail.
*/
int ddr3_tip_pbs_rx(u32 uidev_num)
{
return ddr3_tip_pbs(uidev_num, PBS_RX_MODE);
}
/*
* Name: ddr3_tip_pbs_tx.
* Desc: PBS TX
* Args: TBD
* Notes:
* Returns: OK if success, other error code if fail.
*/
int ddr3_tip_pbs_tx(u32 uidev_num)
{
return ddr3_tip_pbs(uidev_num, PBS_TX_MODE);
}
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
#ifdef DDR_VIEWER_TOOL
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/*
* Print PBS Result
*/
int ddr3_tip_print_all_pbs_result(u32 dev_num)
{
u32 curr_cs;
ARM: mvebu: a38x: sync ddr training code with mv_ddr-armada-18.09.02 This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-18.09 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. Specifically this syncs with commit 99d772547314 ("Bump mv_ddr to release armada-18.09.2"). The complete log of changes is best obtained from the mv-ddr-marvell.git repository but some relevant highlights are: ddr3: add missing txsdll parameter ddr3: fix tfaw timimg parameter ddr3: fix trrd timimg parameter merge ddr3 topology header file with mv_ddr_topology one mv_ddr: a38x: fix zero memory size scrubbing issue The upstream code is incorporated omitting the portions not relevant to Armada-38x and DDR3. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT -UCONFIG_A3700 -UA3900 -UA80X0 \ -UA70X0 Signed-off-by: Chris Packham <judge.packham@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Baruch Siach <baruch@tkos.co.il> Signed-off-by: Stefan Roese <sr@denx.de>
2018-12-03 01:26:49 +00:00
unsigned int max_cs = mv_ddr_cs_num_get();
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (curr_cs = 0; curr_cs < max_cs; curr_cs++) {
ddr3_tip_print_pbs_result(dev_num, curr_cs, PBS_RX_MODE);
ddr3_tip_print_pbs_result(dev_num, curr_cs, PBS_TX_MODE);
}
return MV_OK;
}
/*
* Print PBS Result
*/
int ddr3_tip_print_pbs_result(u32 dev_num, u32 cs_num, enum pbs_dir pbs_mode)
{
u32 data_value = 0, bit = 0, if_id = 0, pup = 0;
u32 reg_addr = (pbs_mode == PBS_RX_MODE) ?
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
PBS_RX_PHY_REG(cs_num, 0) :
PBS_TX_PHY_REG(cs_num , 0);
u32 octets_per_if_num = ddr3_tip_dev_attr_get(dev_num, MV_ATTR_OCTET_PER_INTERFACE);
struct mv_ddr_topology_map *tm = mv_ddr_topology_map_get();
printf("%s,CS%d,PBS,ADLLRATIO,,,",
(pbs_mode == PBS_RX_MODE) ? "Rx" : "Tx", cs_num);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (if_id = 0; if_id < MAX_INTERFACE_NUM; if_id++) {
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
for (pup = 0; pup < octets_per_if_num; pup++) {
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
printf("%d,",
pbsdelay_per_pup[pbs_mode][if_id][pup][cs_num]);
}
}
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
printf("CS%d, %s ,PBS\n", cs_num,
(pbs_mode == PBS_RX_MODE) ? "Rx" : "Tx");
for (bit = 0; bit < BUS_WIDTH_IN_BITS; bit++) {
printf("%s, DQ", (pbs_mode == PBS_RX_MODE) ? "Rx" : "Tx");
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
printf("%d ,PBS,,, ", bit);
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
for (pup = 0; pup <= octets_per_if_num;
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
pup++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_BUS_ACTIVE(tm->bus_act_mask, pup);
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
CHECK_STATUS(ddr3_tip_bus_read
(dev_num, if_id,
ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr + bit,
&data_value));
printf("%d , ", data_value);
}
}
printf("\n");
}
printf("\n");
return MV_OK;
}
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
#endif /* DDR_VIEWER_TOOL */
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
/*
* Fixup PBS Result
*/
int ddr3_tip_clean_pbs_result(u32 dev_num, enum pbs_dir pbs_mode)
{
u32 if_id, pup, bit;
u32 reg_addr = (pbs_mode == PBS_RX_MODE) ?
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
PBS_RX_PHY_REG(effective_cs, 0) :
PBS_TX_PHY_REG(effective_cs, 0);
u32 octets_per_if_num = ddr3_tip_dev_attr_get(dev_num, MV_ATTR_OCTET_PER_INTERFACE);
struct mv_ddr_topology_map *tm = mv_ddr_topology_map_get();
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (if_id = 0; if_id <= MAX_INTERFACE_NUM - 1; if_id++) {
ARM: mvebu: a38x: sync ddr training code with upstream This syncs drivers/ddr/marvell/a38x/ with the mv_ddr-armada-17.10 branch of https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell.git. The upstream code is incorporated omitting the ddr4 and apn806 and folding the nested a38x directory up one level. After that a semi-automated step is used to drop unused features with unifdef find drivers/ddr/marvell/a38x/ -name '*.[ch]' | \ xargs unifdef -m -UMV_DDR -UMV_DDR_ATF -UCONFIG_DDR4 \ -UCONFIG_APN806 -UCONFIG_MC_STATIC \ -UCONFIG_MC_STATIC_PRINT -UCONFIG_PHY_STATIC \ -UCONFIG_64BIT INTER_REGS_BASE is updated to be defined as SOC_REGS_PHY_BASE. Some now empty files are removed and the ternary license is replaced with a SPDX GPL-2.0+ identifier. Signed-off-by: Chris Packham <judge.packham@gmail.com> Signed-off-by: Stefan Roese <sr@denx.de>
2018-05-10 01:28:29 +00:00
VALIDATE_IF_ACTIVE(tm->if_act_mask, if_id);
for (pup = 0; pup <= octets_per_if_num; pup++) {
arm: mvebu: Add Armada 38x DDR3 training code from Marvell bin_hdr This patch adds the DDR3 setup and training code taken from the Marvell U-Boot repository. This code used to be included as a binary (bin_hdr) into the Armada A38x boot image. Not linked with the main U-Boot. With this code addition and the serdes/PHY setup code, the Armada A38x support in mainline U-Boot is finally self-contained. So the complete image for booting can be built from mainline U-Boot. Without any additional external inclusion. Note: This code has undergone many hours (days!) of coding-style cleanup and refactoring. It still is not checkpatch clean though, I'm afraid. As the factoring of the code has so many levels of indentation that many lines are longer than 80 chars. Signed-off-by: Stefan Roese <sr@denx.de>
2015-03-26 14:36:56 +00:00
for (bit = 0; bit <= BUS_WIDTH_IN_BITS + 3; bit++) {
CHECK_STATUS(ddr3_tip_bus_write
(dev_num, ACCESS_TYPE_UNICAST,
if_id, ACCESS_TYPE_UNICAST, pup,
DDR_PHY_DATA, reg_addr + bit, 0));
}
}
}
return MV_OK;
}
Reference in a new issue Copy permalink