Add "max-frequency" property to MMC dt nodes and set the
maximum frequency to 192MHz for MMC1/MMC2, 64MHz for MMC3
and 192MHz for MMC4. pdata quirks must be utilized to
detect presence of rev 1.1/1.0 of silicon and adjust
maximum frequencies as per restrictions documented in i843.
Signed-off-by: NKishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

On dra7, as per TRM, the HW shutdown (TSHUT) temperature is hardcoded
to 123C and cannot be modified by SW. This means when the temperature
reaches 123C HW asserts TSHUT output which signals a warm reset.
This reset is held until the temperature goes below the TSHUT low (105C).

While in SW, the thermal driver continuously monitors current temperature
and takes decisions based on whether it reached an alert or a critical point.
The intention of setting a SW critical point is to prevent force reset by HW
and instead do an orderly_poweroff(). But if the SW critical temperature is
greater than or equal to that of HW then it defeats the purpose. To address
this and let SW take action before HW does keep the SW critical temperature
less than HW TSHUT value.

The value for SW critical temperature was chosen as 120C just to ensure
we give SW sometime before HW catches up.

Document reference
SPRUI30C – DRA75x, DRA74x Technical Reference Manual - November 2016
SPRUHZ6H - AM572x Technical Reference Manual - November 2016

Tested on:
DRA75x PG 2.0 Rev H EVM
Signed-off-by: NRavikumar Kattekola <rk@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Compiling the DT file with W=1, DTC warns like follows:

Warning (unit_address_vs_reg): Node /opp_table0/opp@1000000000 has a
unit name, but no reg property

Fix this by replacing '@' with '-' as the OPP nodes will never have a
"reg" property.
Reported-by: NKrzysztof Kozlowski <krzk@kernel.org>
Reported-by: NMasahiro Yamada <yamada.masahiro@socionext.com>
Suggested-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NViresh Kumar <viresh.kumar@linaro.org>
Acked-by: NRob Herring <robh@kernel.org>
Acked-by: NTony Lindgren <tony@atomide.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Currently the slope and offset values for calculating the
hot spot temperature of a particular thermal zone is part
of driver data. Pass them here instead and obtain the values
while of node parsing.
Signed-off-by: NKeerthy <j-keerthy@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

After the ti-cpufreq driver has been added, we can now drop the
operating-points table present in dra7.dtsi for the cpu and add an
operating-points-v2 table with all OPPs available for all silicon
revisions. Also add necessary data for use by ti-cpufreq to selectively
enable the appropriate OPPs at runtime as part of the operating-points
table.

As we now need to define voltage ranges for each OPP, we define the
minimum and maximum voltage to match the ranges possible for AVS class0
voltage as defined by the DRA7/AM57 Data Manual, with the exception of
using a range for OPP_OD based on historical data to ensure that SoCs
from older lots still continue to boot, even though more optimal voltages
are now the standard. Once an AVS Class0 driver is in place it will be
possible for these OPP voltages to be adjusted to any voltage within the
provided range.

Information from SPRS953, Revised December 2015.
Signed-off-by: NDave Gerlach <d-gerlach@ti.com>
eviewed-by: NLukasz Majewski <lukma@denx.de>
Acked-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NTony Lindgren <tony@atomide.com>

dtc recently added PCI bus checks. Fix these warnings.
Signed-off-by: NRob Herring <robh@kernel.org>
Cc: "Benoît Cousson" <bcousson@baylibre.com>
Cc: Tony Lindgren <tony@atomide.com>
Cc: linux-omap@vger.kernel.org
Signed-off-by: NTony Lindgren <tony@atomide.com>

Since everybody copied my own mistake from the DT binding example,
let's address all the offenders in one swift go.

Most of them got the CPU interface size wrong (4kB, while it should
be 8kB), except for both keystone platforms which got the control
interface wrong (4kB instead of 8kB).

In a few cases where I knew for sure what implementation was used,
I've added the "arm,gic-400" compatible string. I'm 99% sure that
this is what everyone is using, but short of having the TRM for
all the other SoCs, I've left them alone.
Acked-by: NShawn Guo <shawnguo@kernel.org>
Acked-by: NTony Lindgren <tony@atomide.com>
Acked-by: NSantosh Shilimkar <ssantosh@kernel.org>
Acked-by: NKrzysztof Kozlowski <krzk@kernel.org>
Acked-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NAntoine Tenart <antoine.tenart@free-electrons.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NMatthias Brugger <matthias.bgg@gmail.com>
Acked-by: NHeiko Stuebner <heiko@sntech.de>
Reviewed-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: NMarc Zyngier <marc.zyngier@arm.com>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

AHCI provides the register PORTS_IMPL to let the software know which port
is supported. The register must be initialized by the bootloader. However
in some cases u-boot doesn't properly initialize this value (if it is not
compiled with SATA support for example or if the SATA initialization fails).
The DTS entry "ports-implemented" can be used to override the value in
PORTS_IMPL.

Without this patch the SATA will not work in the following two cases:

* if there has been a failure to initialize SATA in u-boot.

* if ahci_platform module has been removed and re-inserted. The reason is
  that the content of PORTS_IMPL is lost after the module is removed.
  I suspect that it's because the controller is reset by the hwmod.

Cc: <stable@vger.kernel.org> # v4.6+
Signed-off-by: NJean-Jacques Hiblot <jjhiblot@ti.com>
Acked-by: NRoger Quadros <rogerq@ti.com>
[tony@atomide.com: updated comments with what goes wrong]
Signed-off-by: NTony Lindgren <tony@atomide.com>

Even if no_bd_ram property is described in TI CPSW bindings the support for
it has never been introduced in CPSW driver, so there are no real users of
it. Hence, remove no_bd_ram property from documentation and DT files.

Cc: 'Rob Herring <robh+dt@kernel.org>'
Signed-off-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 55871eb6 ("ARM: dts: dra7: Remove skeleton.dtsi usage")
removed the skeleton.dtsi usage since we want to get rid of it.

But this can cause issues when booting a kernel with a boot-loader
that doesn't create a chosen node if this isn't present in the DTB
since the decompressor relies on a pre-existing chosen node to be
available to insert the command line and merge other ATAGS info.

Fixes: 55871eb6 ("ARM: dts: dra7: Remove skeleton.dtsi usage")
Reported-by: NPali Rohar <pali.rohar@gmail.com>
Signed-off-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Drivers using pinctrl-single,pins have #pinctrl-cells = <1>, while
pinctrl-single,bits need #pinctrl-cells = <2>.

Note that this patch can be optionally applied separately from the
driver changes as the driver supports also the legacy binding without
#pinctrl-cells.
Acked-by: NRob Herring <robh@kernel.org>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NTony Lindgren <tony@atomide.com>

This reverts commit f80bc97f.

The original commit updated the cpufreq operating points tables for
dra7xx but was merged before the driver making use of the node was
merged, which breaks the existing cpufreq implementation on the system,
so revert the patch until the ti-cpufreq driver is merged.
Signed-off-by: NDave Gerlach <d-gerlach@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

The skeleton.dtsi file was removed in ARM64 for different reasons as
explained in commit ("3ebee5a2 arm64: dts: kill skeleton.dtsi").

These also applies to ARM and it will also allow to get rid of the
following DTC warnings in the future:

"Node /memory has a reg or ranges property, but no unit name"

The disassembled DTB are almost the same besides an empty chosen
node being removed and nodes reordered, so it should not have
functional changes.
Signed-off-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Current clocks tree definition for CPSW/CPTS doesn't
correspond TRM for dra7/am57 SoCs.

CPTS: has to be sourced from gmac_rft_clk_mux clock
CPSW: DPLL_GMAC -> CLKOUT_M2 -> GMAC_250M_CLK -> 1/2 ->
      -> GMAC_MAIN_CLK (125 MHZ)

Hence, correct clock tree for GMAC_MAIN_CLK and use proper
clock for CPTS. This also require updating of CPTS clock
multiplier.
Signed-off-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Reviewed-by: NMugunthan V N <mugunthanvnm@ti.com>
Acked-by: NTero Kristo <t-kristo@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Since DRA7 has multiple PCIe Rootcomplex, add "linux,pci-domain"
property to assign a PCI domain number to each of the host
bridges.
Signed-off-by: NKishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Silicon limitation i845 documents how to cope with false
disconnection condition on USB2 PHY. Reference: AM572x
silicon errata document SPRZ429H, revised January 2016.

Using compatible "ti,dra7x-usb2" enables the recommended
software workaround for this issue. Use it for USB1 PHY.

The workaround is already in place for USB2 PHY.
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NRoger Quadros <rogerq@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Add "ti,cpsw-mdio" for am335x/am437x/dra7 SoCs where MDIO is
implemented as part of TI CPSW and, this way, enable PM runtime auto
suspend for Davinci MDIO driver on these paltforms.
Signed-off-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adding dt node for hardware random number generator IP.
Signed-off-by: NLokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

DRA7 SoC has the same SHA IP as OMAP5. Add DT entry for the same.
Signed-off-by: NLokesh Vutla <lokeshvutla@ti.com>
[t-kristo@ti.com: changed SHA to use EDMA instead of SDMA]
Signed-off-by: NTero Kristo <t-kristo@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

DRA7 SoC has the same AES IP as OMAP4. Add DT entries for both AES cores.
Signed-off-by: NJoel Fernandes <joelf@ti.com>
Signed-off-by: NLokesh Vutla <lokeshvutla@ti.com>
[t-kristo@ti.com: squashed in the change to use EDMA, squashed in
                  support for two AES cores]
Signed-off-by: NTero Kristo <t-kristo@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

DRA7xx SoCs have a DES3DES IP. Add DT data for the same.
Signed-off-by: NJoel Fernandes <joelf@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

There is no reason to hold s/w dependent parameter in device tree.
Even more, there is no reason in this parameter because davinici_cpdma
driver splits pool of descriptors equally between tx and rx channels
anyway.
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The unit address of the second PCIe instance
is set to be same as that of the first instance
(copy-paste error).

Fix it.
Signed-off-by: NKishon Vijay Abraham I <kishon@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Correct misspelling, "emda3" -> "edma3".
Reported-by: NAdam J Allison <adamj.allison@gmail.com>
Signed-off-by: NRobert P. J. Day <rpjday@crashcourse.ca>
Acked-by: NRob Herring <robh@kernel.org>
Acked-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Add an operating-points-v2 table with all OPPs available for all silicon
revisions along with necessary data for use by ti-opp driver to selectively
enable the appropriate OPPs at runtime and handle voltage transitions

As we now need to define voltage ranges for each OPP, we define the
minimum and maximum voltage to match the ranges possible for AVS class0
voltage as defined by the DRA7/AM57 Data Manual, with the exception of
using a range for OPP_OD based on historical data to ensure that SoCs
from older lots still continue to boot, even though more optimal voltages
are now the standard. Once an AVS Class0 driver is in place it will be
possible for these OPP voltages to be adjusted to any voltage within the
provided range.

Information from SPRS953, Revised December 2015.
Signed-off-by: NDave Gerlach <d-gerlach@ti.com>
Acked-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Nearly all of the information in the cpus node, especially for cpu0, is
the same between dra74x and dra72x so move the common information to
the parent dra7.dtsi to avoid duplication of data.
Signed-off-by: NDave Gerlach <d-gerlach@ti.com>
Acked-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Create a system control module node for the control module portion that
resides under l4_wkup.
Signed-off-by: NDave Gerlach <d-gerlach@ti.com>
Acked-by: NViresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Secure variants of DRA7xx and AM57xx SoCs may need to reserve a region
of the SRAM for use by secure software. To account for this, add a child
node to the ocmcram1 node that will act as a placeholder at the start
of the SRAM for the reserved region of memory that may be required
by secure services. The node is added with size 0 so that by default
parts will have the full space available but the bootloader or board dts
file is able to resize the node as needed depending on how much reserved
space is needed, if any, so end users of the ocmcram1 region on HS parts
must be aware that a smaller amount of SRAM than expected may be available.
Signed-off-by: NDave Gerlach <d-gerlach@ti.com>
Reviewed-by: NAndreas Dannenberg <dannenberg@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Add all ocmcram nodes to dra7.dtsi using the generic mmio-sram driver.
DRA7xx and AM57xx families of SoCs can contain three ocmcram regions of
SRAM, one of 512kb and also an optional two additional of 1Mb each. Mark
the two additional 1MB regions of SRAM as disabled as only ocmcmram1 is
on all variants of the SoCs, then depending on which specific variant
is in use the ocmcram2 and ocmcram3 nodes can be enabled in the board
dts file if the data manual for that part number indicates the ocmcram
region is available.
Signed-off-by: NDave Gerlach <d-gerlach@ti.com>
Reviewed-by: NAndreas Dannenberg <dannenberg@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Add PWMSS device tree nodes for DRA7 SoC family and add documentation
for dt bindings.
Signed-off-by: NVignesh R <vigneshr@ti.com>
[fcooper@ti.com: Add eCAP and use updated bindings for PWMSS and ePWM]
Signed-off-by: NFranklin S Cooper Jr <fcooper@ti.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Add dma channel information to the gpmc.
Signed-off-by: NFranklin S Cooper Jr <fcooper@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

GPMC driver provides GPI support for the GPMC_WAIT pins.
Mark it gpio controller capable.
Signed-off-by: NRoger Quadros <rogerq@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Add the DT node for Timer12 present on DRA7 family of
SoCs. Timer12 is present in PD_WKUPAON power domain, and
has the same capabilities as the other timers, except for
the fact that it serves as a secure timer on HS devices
and is clocked only from the secure 32K clock.
Signed-off-by: NSuman Anna <s-anna@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

The Timers 13 through 16 have been added previously in
disabled state. These timers are common timers that are
present on all DRA7 family of SoCs, so enable these
devices by default like the rest of the DMTimers.
Signed-off-by: NSuman Anna <s-anna@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Add nodes to represent all McASP ports in the dra7 family.
For system consistency use the eDMA for audio operations. sDMA would be
fine for 4/5/6/7/8 since their DAT port is not through L3 interconnect.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

McASP3 does not support constant addressing mode on the DAT
port, so increment transfers must be used instead.  This
restriction is also applicable for McASP1 and McASP2.

This DMA addressing constraint poses a major problem for sDMA
where constant addressing mode is used on the peripheral side.
Unfortunately, using increment transfers in sDMA comes with
important side effects.

The addressing mode used in eDMA is INC, so the silicon limitation
described above has no impact and the McASP3 DAT port can be
safely added by switching to eDMA instead of sDMA.
Signed-off-by: NMisael Lopez Cruz <misael.lopez@ti.com>
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

DRA7 family has eDMA available along with the sDMA and in some cases it is
better suited for servicing peripherals.

Add the needed nodes for eDMA to be usable:
edma-tpcc, edma-tptc0/1 and the edma-xbar.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Move the sDMA xbar nodes under the L4 interconnect node.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

This patch fixes the following DTC warnings:

"pbias_regulator has a reg or ranges property, but no unit name"
Signed-off-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: NTony Lindgren <tony@atomide.com>

Errata id: i877

Description:
------------
The RGMII 1000 Mbps Transmit timing is based on the output clock
(rgmiin_txc) being driven relative to the rising edge of an internal
clock and the output control/data (rgmiin_txctl/txd) being driven relative
to the falling edge of an internal clock source. If the internal clock
source is allowed to be static low (i.e., disabled) for an extended period
of time then when the clock is actually enabled the timing delta between
the rising edge and falling edge can change over the lifetime of the
device. This can result in the device switching characteristics degrading
over time, and eventually failing to meet the Data Manual Delay Time/Skew
specs.
To maintain RGMII 1000 Mbps IO Timings, SW should minimize the
duration that the Ethernet internal clock source is disabled. Note that
the device reset state for the Ethernet clock is "disabled".
Other RGMII modes (10 Mbps, 100Mbps) are not affected

Workaround:
-----------
If the SoC Ethernet interface(s) are used in RGMII mode at 1000 Mbps,
SW should minimize the time the Ethernet internal clock source is disabled
to a maximum of 200 hours in a device life cycle. This is done by enabling
the clock as early as possible in IPL (QNX) or SPL/u-boot (Linux/Android)
by setting the register CM_GMAC_CLKSTCTRL[1:0]CLKTRCTRL = 0x2:SW_WKUP.

So, do not allow to gate the cpsw clocks using ti,no-idle property in
cpsw node assuming 1000 Mbps is being used all the time. If someone does
not need 1000 Mbps and wants to gate clocks to cpsw, this property needs
to be deleted in their respective board files.
Signed-off-by: NMugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: NGrygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: NLokesh Vutla <lokeshvutla@ti.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NPaul Walmsley <paul@pwsan.com>