In the RK3399 DRAM driver, the function set_ds_odt() supports operating
in two different modes, selected by the ctl_phy_reg argument: when true,
the function reads and writes directly from the DRAM registers, accessed
through "chan->pctl->denali_*"; when false, the function reads and
writes from an array, accessed through "params->pctl_regs.denali_*",
which is written to DRAM registers at a later time.
However, phy_config_io(), which is called by set_ds_odt() to do a subset
of its register operations, operates directly on DRAM registers at all
times. This means that it reads incorrect values (and writes new values
prematurely) when ctl_phy_reg in set_ds_odt() is false. Fix this by
passing in the address of the registers to work with.
This prevents an "Invalid DRV value" error in the SPL debug log and
(presumably) results in a more correct end state. See the following logs
from a RK3399 NanoPi M4 board (4GB LPDDR3):
Before:
sdram_init() Starting SDRAM initialization...
phy_io_config() Invalid DRV value.
phy_io_config() Invalid DRV value.
sdram_init() sdram_init: data trained for rank 2, ch 0
phy_io_config() Invalid DRV value.
phy_io_config() Invalid DRV value.
sdram_init() sdram_init: data trained for rank 2, ch 1
Channel 0: LPDDR3, 933MHz
BW=32 Col=10 Bk=8 CS0 Row=15 CS1 Row=15 CS=2 Die BW=16 Size=2048MB
Channel 1: LPDDR3, 933MHz
BW=32 Col=10 Bk=8 CS0 Row=15 CS1 Row=15 CS=2 Die BW=16 Size=2048MB
256B stride
256B stride
sdram_init() Finish SDRAM initialization...
After:
sdram_init() Starting SDRAM initialization...
sdram_init() sdram_init: data trained for rank 2, ch 0
sdram_init() sdram_init: data trained for rank 2, ch 1
Channel 0: LPDDR3, 933MHz
BW=32 Col=10 Bk=8 CS0 Row=15 CS1 Row=15 CS=2 Die BW=16 Size=2048MB
Channel 1: LPDDR3, 933MHz
BW=32 Col=10 Bk=8 CS0 Row=15 CS1 Row=15 CS=2 Die BW=16 Size=2048MB
256B stride
256B stride
sdram_init() Finish SDRAM initialization...
Signed-off-by: Thomas Hebb <tommyhebb@gmail.com>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
There are some code different with rockchip vendor code which may lead
to different bugs, including:
1) Fix setting error about LPDDR3 dram size ODT.
2) Set phy io speed to 0x2.
3) Fix setting error about phy_pad_fdbk_drive.
4) Fix setting error about PI_WDQLVL_VREF_EN
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Update the calculation of the stride to support all the DRAM case.
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
The io setting are updated after some bugfix in different rk3399 boards,
sync the code from vendor.
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Add capacity detect for rk3399 so that the driver able to detect the
capacity automatically.
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Clean up the sdram_init to keep sync with rockchip source code.
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Clean up rk3399 dram driver source code for more readable.
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
For there are some structures and functions are common for all rockchip SoCs,
migrate to use the common code so that we can clean up reduandent codes.
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
rename sdram_common.c in arch/arm/mach-rockchip to sdram.c;
so that we can use the file name sdram_common.c in dram driver for
better understand the code;
clean the related file who has use the header file at the same time.
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
We need to store all the ram related cap/map info back to register
for each channel after all the init has been done in case some of register
was reset during the process.
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Unlike rest of dram type chips, LPDDR4 initialization start
with at board selected frequency (say 50MHz) and then it
switches into 400MHz and 800MHz simultaneously to make the
proper sequence work on each channel with associated training.
The lpddr4 set rate sequnce will follow by setting lpddr4
- dq out
- ca odt
- MR3
- MR12
- MR14
registers sets in sequential order.
Here is sameple log about LPDDR4-100 init sequence in Rockpro64:
Channel 0: LPDDR4, 50MHz
BW=32 Col=10 Bk=8 CS0 Row=15 CS1 Row=15 CS=2 Die BW=16 Size=2048MB
Channel 1: LPDDR4, 50MHz
BW=32 Col=10 Bk=8 CS0 Row=15 CS1 Row=15 CS=2 Die BW=16 Size=2048MB
256B stride
channel 0 training pass
channel 1 training pass
change freq to 400 MHz 0, 1
channel 0 training pass
channel 1 training pass
change freq to 800 MHz 1, 0
This patch add support to this init sequence via lpddr4 set rate
by taking sdram timing parameters from 400, 800 .inc files.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
(Fix travis error, use one ret instead of ret[2] in set_ctrl)
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
DDR set rate can be even required for lpddr4 and we
need to keep the lpddr4 code to compile only for relevant
boards which do support lpddr4.
For this requirement, and for code readability handle
data training via sdram_rk3399_ops with .set_rate and
same will update in future while supporting lpddr4 code.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Like data training in other sdram types, mr detection need
to taken care for lpddr4 with looped rank and associated
channel to make sure the proper configuration held.
Once the mr detection successful for active and configured
rank with channel number, the same can later reused during
actual LPDDR4 initialization.
So, add code to support for it.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
data training can be even required for lpddr4 and we
need to keep the lpddr4 code to compile only for relevant
boards which do support lpddr4.
For this requirement, and for code readability handle
data training via sdram_rk3399_ops and same will update
in future while supporting lpddr4 code.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
data training is using chan_info as first argument with
channel number as second argument instead of that use
dram_info as first argument so-that we can get the
chan_info at data training definition.
This was the argument handling is meaningful, readable
and it would help to add similar data training for
lpddr4 in future.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Update vref_mode_ac for lpddr4 based on VDDQ/3/2=16.8%
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
The mode_sel on lpddr4 value is depending on IO settings
of rd_vref.
Add support for it.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
The vref_mode_dq, vref_value_dq on lpddr4 value is depending
on IO settings of rd_vref.
Add support for it.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
For base.odt 1 the lpddr4 tsel_rd_en value is depending
on IO settings of rd_odt_en.
Add support for it.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
CTL 145, 146, 159, 160 registers are used to configure
soc odt on rk3399.
These soc odt values are updated from CS0_MR22_VAL and
CS1_MR22_VAL and for lpddr4 these values ORed with
tsel_rd_select_n.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
tsel contrl clock drives are required to configure PHY
929, 939 controls drive settings.
Add support for these control clock for all dramtype
sdrams.
Thse control clock drives are configure via tsel_ckcs_select_p
and tsel_ckcs_select_n variables.
tsel_ckcs_select_n is PHY_DRV_ODT_34_3 value where as
tsel_ckcs_select_p is retrived from IO settings for lpddr4
and rest uses PHY_DRV_ODT_34_3.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Now we have IO settings available for all supported sdram
frequencies, so retrieve these IO settings and make used
for LPDDR4 ds odt configuration.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Add IO settings for dram ctl and phy.
IO settings are useful for configuring ctl, phy odt, vref,
mr5, mode select and other needed input output operations
for lpddr4 or any other dramtype sdram.
Right now, this patch added IO setting for all supported
sdram frequencies.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
The hardware for LPDDR4 with
- CLK0P/N connect to lower 16-bits
- CLK1P/N connect to higher 16-bits
and usually dfi dram clk is configured via CLK1P/N, so
disabling dfi dram clk will disable the CLK1P/N as well.
So, add patch to not to disable dfi dram clk for lpddr4,
with rank 1.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
tsel write ca_p and ca_n values need to write on PHY 544, 672
and 800 to configure ds odt.
Configure the same PHY register for lpddr4 would require a mask
value of (300 << 8).
Add support for it.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Assign desired cs_map values for lpddr4 during set memory map.
Initial cs_map values is based on the sdram parameters, so
the same will adjusted based dramtype as LPDDR4.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
PHY_898, PHY_919 would require to configure PHY LP4 boot
pll control and ca for lpddr4.
So, configure the same in pctl_cfg for LPDDR4.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
It is possible in lpddr4 dram, where both the channels would
start at same time with ZQ Cal Start. If it uses ZQ Call start
then it will use RZQ.
For example LPDDR4 366 Dual-Die, Quad-Channel Package, RZQ maybe
connect to both channel. If ZQ Cal Start at the same time,
it will use the same RZQ.
It is not a problem of using RZQ in both the channels, but can not
use at the same time.
So, to avoid this, we have an option of dram tINIT3 value for
increasing the frequency for channel 1.
This patch increase the available tINIT3 with existing running
dram frequency.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
lpddr4 has PLL bypass mode during phy initialization phase,
which does all pll configurations.
So no need to wait explicitly during pctl config.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
mode_sel assignment is based on dram type.
In phy_io_config, already have vref setting based
on the dram type, so move this mode_sel assignment
on vref setting area.
No functionality change.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add rank_mask based on the rank number for lpddr4.
This would keep the wdql data training loop based on the
desired rank mask value instead of looping for all values.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add rank_mask based on the rank number for lpddr4.
This would keep the ca data training loop based on the
desired rank mask value instead of looping for all values.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Some dramtypes like lpddr4 initialization would required to
configure phy IO even after pctl_cfg and after set_ds_odt.
For those cases the set_ds_odt would be an initial call to
setup the phy.
To satisfy all the cases, trigger phy IO from set_ds_odt.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add DdrMode structure with associated bit fields.
These would help to reconfigure sdram capabilities during
lpddr4 setup related configs.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add DdrTimingC0 structure with associated bit fields.
These would help to reconfigure sdram capabilities during
lpddr4 setup related configs.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add dram config macro for handling ddr version number.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
cs0_row, cs1_row and cs1_col needs more bits to show its
correct value, update to make use of both sys_reg2,
sys_reg3.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
(Squash similar patches into one patch)
Signed-off-by: Kever Yang <Kever.yang@rock-chips.com>
Use dram config variable name as sys_reg2 instead of sys_reg
since the final variable value is to written into a pmugrf
register named as sys_reg2.
This reflect the both variable and associated register
names are same and also help to add next sys_reg's to
add it in future.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add simplified and meaningful macro for all setting.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
(Squash the similar patches into 1 patch)
Signed-off-by: Kever Yang <Kever.yang@rock-chips.com>
This would help to debug the sdram base parameters while
debugging existing chip or while supporting new sdram type.
It require explicit enablement of CONFIG_RAM_ROCKCHIP_DEBUG
for showing the debug prints.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Right now the rk3399 sdram driver assume that the board
has configured with 2 channels, so any possibility to
enable single channel on the same driver will encounter
channel #1 data training failure.
Log:
U-Boot TPL board init
sdram_init: data training failed
rk3399_dmc_init DRAM init failed -5
So, add an algorithm that can capable to compute the active
or configured rank with associated channel like
a) do rank loop to compute the active rank, with associated
channel numbers
b) then, succeed the data training only for configured channel
c) preserve the rank for given channel
d) do channel loop for setting the active channel
e) if given rank is zero or inactive on the specific channel,
clear the timings for the associated channel
f) finally, return error if number of channels is zero
Tested in NanoPI-NEO4 since it support single channel sdram
configuration.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
(add PI_READ_GATE_TRAINING for LPDDR3 to support rk3399-evb case)
Signed-off-by: Kever Yang <Kever.yang@rock-chips.com>
Add stride computation for the sdram which support
single channel a
This configuration available in NanoPi NEO4 and the
same can work with existing rk3399-sdram-ddr3-1866.dtsi
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
stride value from sdram timings can be computed dynamically
based on the determined capacity for the given channel.
Right now these stride values are taken as part of sdram timings
via dtsi, but it possible to use same timings dtsi for given
frequency even though the configured board sdram do support
single channel with different size by dynamically detect the
stride value.
Example, NanoPi NEO4 do have DDR3-1866, but with single channel
and 1GB size with dynamic stride detection it is possible to
use existing rk3399-sdram-ddr3-1866.dtsi whose stride,
number of channels and capacity it support is d efferent.
So, add initial support to calculate the stride value for
2 channels sdram, which is available by default on existing
boards.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add support for pctl start for both channel 0, 1 control
and phy registers.
This would also handle pwrup_srefresh_exit init based
on the channel number.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add pwrup_srefresh_exit to be part of dram_info so-that
the it can help to support pwrup_srefresh_exit in individual
channels while starting pctl in future.
No functionality change.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
Add support for phy pctrl reset support for both channel 0, 1.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: YouMin Chen <cym@rock-chips.com>
Reviewed-by: Kever Yang <Kever.yang@rock-chips.com>
