oweals/u-boot.git
5 years agoarmv8: ls1028aqds: Remove the definition of CONFIG_SYS_I2C_EARLY_INIT
Chuanhua Han [Wed, 10 Jul 2019 07:48:37 +0000 (15:48 +0800)]
armv8: ls1028aqds: Remove the definition of CONFIG_SYS_I2C_EARLY_INIT

Since i2c uses dm mode, i2c controller will be initialized when reading
and writing devices on i2c bus. So there is no need for the original
non-dm mode i2c early initialization function call, this patch removed
the definition of CONFIG_SYS_I2C_EARLY_INIT.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Alex Marginean <alexm.osslist@gmail.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoarmv8: dts: ls1028ardb: Add slave nodes under the i2c0 controller
Chuanhua Han [Wed, 10 Jul 2019 13:16:52 +0000 (21:16 +0800)]
armv8: dts: ls1028ardb: Add slave nodes under the i2c0 controller

Add some slave nodes to support the i2c dm on the device side under the i2c0.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Alex Marginean <alexm.osslist@gmail.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agogpio: do not include <asm/arch/gpio.h> on ARCH_LS1028A
Chuanhua Han [Wed, 10 Jul 2019 13:16:50 +0000 (21:16 +0800)]
gpio: do not include <asm/arch/gpio.h> on ARCH_LS1028A

As no gpio.h is defined for this architecture, to avoid
compilation failure, do not include <asm/arch/gpio.h> for
arch ls1028a.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Alex Marginean <alexm.osslist@gmail.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoboards: ls1028a: Add support of I2C driver model
Chuanhua Han [Wed, 10 Jul 2019 13:16:49 +0000 (21:16 +0800)]
boards: ls1028a: Add support of I2C driver model

DM_I2C_COMPAT is a compatibility layer that allows using the non-DM
I2C API when DM_I2C is used.When DM_I2C_COMPAT is not enabled for
compilation, a compilation error will be generated. This patch
solves the problem that the i2c-related api of the ls1028a platform
does not support dm.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Alex Marginean <alexm.osslist@gmail.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoconfigs: lx2160: enable DM support for pcf2127 rtc
Chuanhua Han [Wed, 10 Jul 2019 13:00:21 +0000 (21:00 +0800)]
configs: lx2160: enable DM support for pcf2127 rtc

Enable related configs on all lx2160ardb boards to support pcf2127
rtc DM feature.

Also remove SYS_I2C_MXC_I2Cx, where x is from 1 to 8 from
Kconfig.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Signed-off-by: Biwen Li <biwen.li@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoarmv8: dts: lx2160aqds : Add pcf2127 node
Chuanhua Han [Wed, 10 Jul 2019 13:05:13 +0000 (21:05 +0800)]
armv8: dts: lx2160aqds : Add pcf2127 node

Add the pcf2127-rtc node under the i2c0->i2c-mux@77->i2c@3 in dts for
lx2160aqds boards.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoarmv8: dts: lx2160ardb : Add the "u-boot, dm-pre-reloc" for i2c0
Chuanhua Han [Wed, 10 Jul 2019 13:00:26 +0000 (21:00 +0800)]
armv8: dts: lx2160ardb : Add the "u-boot, dm-pre-reloc" for i2c0

Lx2160ardb need to use i2c0 before relocation, so we also need to set
u-boot, dm-pre-reloc to initialize node before relocation.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoarmv8: dts: Add pcf2127 node for lx2160ardb
Chuanhua Han [Wed, 10 Jul 2019 13:00:25 +0000 (21:00 +0800)]
armv8: dts: Add pcf2127 node for lx2160ardb

Adds the pcf2127-rtc node under the i2c4 node dts of lx2160ardb boards.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoarmv8: dts: fsl-lx2160a: add i2c controller and gpio DT nodes
Chuanhua Han [Wed, 10 Jul 2019 13:00:24 +0000 (21:00 +0800)]
armv8: dts: fsl-lx2160a: add i2c controller and gpio DT nodes

In lx2160a soc, there are eight i2c controllers, this patch adds i2c
nodes for lx2160a, and the gpio2 nodes on which the i2c4 controller
depends.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agogpio: do not include <asm/arch/gpio.h> on ARCH_LX2160A
Chuanhua Han [Wed, 10 Jul 2019 13:00:23 +0000 (21:00 +0800)]
gpio: do not include <asm/arch/gpio.h> on ARCH_LX2160A

As no gpio.h is defined for this architecture, to avoid
compilation failure, do not include <asm/arch/gpio.h> for
arch ls2160a.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agodrivers: i2c: mxc: Fix compiler error when using i2c dm mode
Chuanhua Han [Wed, 10 Jul 2019 13:00:22 +0000 (21:00 +0800)]
drivers: i2c: mxc: Fix compiler error when using i2c dm mode

I2C dm mode enablemenet causes below compilation errors:

In file included from include/config.h:8:0,
                 from include/common.h:20:
include/config_fallbacks.h:51:4: error: #error "Cannot define
CONFIG_SYS_I2C when CONFIG_DM_I2C is used"
 #  error "Cannot define CONFIG_SYS_I2C when CONFIG_DM_I2C is used"
    ^~~~~
In file included from include/config.h:8:0,
                 from include/common.h:20:
include/config_fallbacks.h:51:4: error: #error "Cannot define
CONFIG_SYS_I2C when CONFIG_DM_I2C is used"
 #  error "Cannot define CONFIG_SYS_I2C when CONFIG_DM_I2C is used"
    ^~~~~

board/freescale/lx2160a/lx2160a.c: In function 'board_early_init_f':
board/freescale/lx2160a/lx2160a.c:108:2: warning: implicit declaration
of function 'i2c_early_init_f'; did you mean 'arch_early_init_r'?
[-Wimplicit-function-declaration]
  i2c_early_init_f();
  ^~~~~~~~~~~~~~~~
  arch_early_init_r

 drivers/i2c/mxc_i2c.c: In function 'mxc_i2c_probe':
  drivers/i2c/mxc_i2c.c:824:8: warning: implicit declaration of function
'enable_i2c_clk';
  did you mean 'enable_irq_wake'? [-Wimplicit-function-declaration]
  ret = enable_i2c_clk(1, bus->seq);
        ^~~~~~~~~~~~~~
        enable_irq_wake

So fix these compilation errors.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoboards: lx2160a: Add support of I2C driver model
Chuanhua Han [Wed, 10 Jul 2019 13:00:20 +0000 (21:00 +0800)]
boards: lx2160a: Add support of I2C driver model

DM_I2C_COMPAT is a compatibility layer that allows using the non-DM I2C
API when DM_I2C is used. When DM_I2C_COMPAT is not enabled for
compilation, a compilation error will be generated. This patch solves
the problem that the i2c-related api of the lx2160a platform does not
support dm.

Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoconfigs: ls1043aqds: Move CONFIG_FSL_QSPI to defconfig
Ashish Kumar [Tue, 9 Jul 2019 09:20:15 +0000 (14:50 +0530)]
configs: ls1043aqds: Move CONFIG_FSL_QSPI to defconfig

Move CONFIG_FSL_QSPI from header file to defconfigs,
consequently unset imply config(CONFIG_SPI_FLASH_BAR) which
is not valid for LS series.

Signed-off-by: Ashish Kumar <Ashish.Kumar@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoarmv8: ls1028a: select BOARD_LATE_INIT config
Yuantian Tang [Tue, 2 Jul 2019 08:16:22 +0000 (16:16 +0800)]
armv8: ls1028a: select BOARD_LATE_INIT config

Select BOARD_LATE_INIT for ls1028ardb and ls1028aqds targets
so that late init work can be done.

Signed-off-by: Yuantian Tang <andy.tang@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agocommon: qixis: make the qixis compatible with new soc
Yuantian Tang [Wed, 19 Jun 2019 06:39:28 +0000 (14:39 +0800)]
common: qixis: make the qixis compatible with new soc

This driver needs modification to work with new soc,
like ls1028, since bitmap of RCFG is changed to
RESV[7:5] LIVE[4] WDEN[3] RESV[2:1] GO[0]
   000      1       0        00      0

Also the RCW location is moved to only dutcfg0.
RESV[7:4] RCWSRC[3:0]
   1111   configurable

Following commands are functional now
qixis_reset
qixis_reset sd
qixis_reset qspi
qixis_reset emmc

Signed-off-by: Ashish Kumar <Ashish.Kumar@nxp.com>
Signed-off-by: Yuantian Tang <andy.tang@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agodrivers: net: mc: Report extra memory to Linux
Meenakshi Aggarwal [Thu, 23 May 2019 09:43:43 +0000 (15:13 +0530)]
drivers: net: mc: Report extra memory to Linux

MC firmware need to be aligned to 512M, so minimum 512MB DDR is reserved.
But MC support to work with 128MB or 256MB DDR memory also, in this
case, rest of the memory is not usable.
So reporting this extra memory to Linux through dtb memory fixup.

Signed-off-by: Meenakshi Aggarwal <meenakshi.aggarwal@nxp.com>
Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
5 years agoMerge branch '2019-08-20-master-imports'
Tom Rini [Wed, 21 Aug 2019 01:40:12 +0000 (21:40 -0400)]
Merge branch '2019-08-20-master-imports'

- Assorted bugfixes

5 years agoMerge branch '2019-08-20-ti-imports'
Tom Rini [Wed, 21 Aug 2019 01:39:40 +0000 (21:39 -0400)]
Merge branch '2019-08-20-ti-imports'

- More DaVinci cleanups
- Other minor omap2plus cleanups

5 years agotest/py: Add cmd_memory dependency back to test_mmc_wr
Michal Simek [Mon, 19 Aug 2019 09:06:13 +0000 (11:06 +0200)]
test/py: Add cmd_memory dependency back to test_mmc_wr

Based on discussion with Stephen Warren there was recommendation to list
both memory and random command dependencies just in case that dependency is
not properly handled by Kconfig.

Fixes: a09c1f7e1c1b ("test/py: Fix MMC/SD block write test dependency")
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
5 years agoeasylogo: avoid buffer overrun
Heinrich Schuchardt [Thu, 15 Aug 2019 21:54:15 +0000 (23:54 +0200)]
easylogo: avoid buffer overrun

Building easylogo with `HOST_TOOLS_ALL=y make tools` results in a build
warning due to a possible buffer overrun:

tools/easylogo/easylogo.c:453:4: note: ‘sprintf’ output between 7 and
262 bytes into a destination of size 256
    sprintf (str, "%s, 0x%02x", app, *dataptr++);
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Truncate the output to fit into the destination buffer.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
5 years agojffs2: remove unused code files
Uwe Kleine-König [Wed, 14 Aug 2019 20:07:29 +0000 (22:07 +0200)]
jffs2: remove unused code files

I failed to find where these two files are used and a few test compile
runs with JFFS2 enabled succeeded also without these.

5 years agommc: mtk-sd: Add MT8183 SoC support
Fabien Parent [Mon, 12 Aug 2019 18:26:58 +0000 (20:26 +0200)]
mmc: mtk-sd: Add MT8183 SoC support

Add support for the MT8183 in the MediaTek MMC driver.

Signed-off-by: Fabien Parent <fparent@baylibre.com>
5 years agomailmap: provide usage instruction
Heinrich Schuchardt [Sat, 10 Aug 2019 11:17:21 +0000 (13:17 +0200)]
mailmap: provide usage instruction

Looking at the contents of file .mailmap it seems that some editors assumed
that translation is done by entering multiple lines into the file and the
last one replaces the others. This is not how it works. The translation
occurs according to entries in single lines as described in the
git-check-mailmap man-page.

Add a description of the file format.

Add an entry for Alexander Graf as his old email address is not valid
anymore.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
5 years agoREADME: Clarify use of BSS during SPL board_init_f()
Andreas Dannenberg [Thu, 8 Aug 2019 17:54:49 +0000 (12:54 -0500)]
README: Clarify use of BSS during SPL board_init_f()

The earlier commit....

commit a5a5d997b41a ("spl: Allow performing BSS init early before board_init_f()")

...introduced the ability to use BSS from SPL's board_init_f() as it may
be required in certain exceptional use cases so go ahead and update the
README to reflect this change. Note that as highlighted with the changes
the use of the associated CONFIG option is generally not recommended.

Signed-off-by: Andreas Dannenberg <dannenberg@ti.com>
Reviewed-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
5 years agoarm: omap2: am43xx: Enable CONFIG_BLK
Suniel Mahesh [Mon, 19 Aug 2019 06:27:39 +0000 (11:57 +0530)]
arm: omap2: am43xx: Enable CONFIG_BLK

With DM_MMC enabled, enable CONFIG_BLK to remove this
compile warning for am43xx based targets:

===================== WARNING ======================
This board does not use CONFIG_DM_MMC. Please update
the board to use CONFIG_DM_MMC before the v2019.04 release.
Failure to update by the deadline may result in board removal.
See doc/driver-model/MIGRATION.txt for more info.
====================================================

Targets were compile tested, build was clean.

Signed-off-by: Suniel Mahesh <sunil.m@techveda.org>
5 years agoarm: dts: Makefile: clean *dtb_HS
Suniel Mahesh [Fri, 16 Aug 2019 08:46:36 +0000 (14:16 +0530)]
arm: dts: Makefile: clean *dtb_HS

TI HS platforms generate *dtb_HS binary blobs and there is no
rule for cleanup. Added entry for cleanup in clean-files target.

Signed-off-by: Suniel Mahesh <sunil.m@techveda.org>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
5 years agoARM: DRA7: Fixup DPLL clock rate fixup logic for newer kernels
Suman Anna [Fri, 16 Aug 2019 22:30:16 +0000 (17:30 -0500)]
ARM: DRA7: Fixup DPLL clock rate fixup logic for newer kernels

The commit 1b42ab3eda8a ("ARM: DRA7: Fixup DSPEVE, IVA and GPU clock
frequencies based on OPP") updates the kernel device-tree blob to adjust
the DSP, IVA and GPU DPLL clocks based on a one-time OPP choice selected
in U-Boot. All these DPLL clocks are children of the cm_core_aon clocks
DT node.

The hierarchy of this clocks DT node has changed in newer Linux kernels
starting from v5.0, and this results in a failure in ft_fixup_clocks()
function to update the clock rates on these newer kernels. Fix this by
updating the lookup logic to look through both the newer and older
DT hierarchy paths for the cm_core_aon clocks node.

Signed-off-by: Suman Anna <s-anna@ti.com>
5 years agoarm: baltos: switch to driver model for the watchdog timer
Yegor Yefremov [Thu, 15 Aug 2019 09:08:04 +0000 (11:08 +0200)]
arm: baltos: switch to driver model for the watchdog timer

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
5 years agodavinci: omapl138-lcdk: Remove empty compiler directives
Adam Ford [Wed, 14 Aug 2019 13:11:09 +0000 (08:11 -0500)]
davinci: omapl138-lcdk: Remove empty compiler directives

There is an #ifdef with nothing inside it any longer.  This patch
removes this empty #ifdef

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoARM: davinci: Remove duplicated references
Adam Ford [Tue, 13 Aug 2019 19:38:11 +0000 (14:38 -0500)]
ARM: davinci: Remove duplicated references

The Kconfig file calls to ti/common/Kconfig twice which makes
several of the menu items repeat themselves.  In an effort to
clean this up, this patch removes the second call which eliminates
the duplicate menu items.

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoARM: da850evm: Remove dead code
Adam Ford [Tue, 13 Aug 2019 13:28:28 +0000 (08:28 -0500)]
ARM: da850evm: Remove dead code

Now that SPL supports DM_SERIAL and the direct NOR boot supports
DM_SERIAL, the check to see if DM_SERIAL is defined can go away,
because all da850evm variants now support DM_SERIAL.  This patch
simply removes some dead precompiler defines.

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoARM: da850evm: Remove dead SPI Code
Adam Ford [Tue, 13 Aug 2019 13:27:28 +0000 (08:27 -0500)]
ARM: da850evm: Remove dead SPI Code

With SPL now supporting DM_SPI, the need for compiler directives
and hard-coded addresses is obsolete.  This patch removes some
dead legacy code defining the SPI base address

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoRevert "ARM: da850-evm: Enable SPI Flash and NAND Flash when booting NOR"
Adam Ford [Tue, 13 Aug 2019 19:04:02 +0000 (14:04 -0500)]
Revert "ARM: da850-evm: Enable SPI Flash and NAND Flash when booting NOR"

Sorry for the noise, but there appears to be a regression with older
hardware.  Since it broke the direct_nor boot option, it should be
reverted until a better solution is available.

This reverts commit 51cd1e2373274dc3167dabba628dcfc25828d36d.

5 years agoARM: omapl138_lcdk: Enable USB
Adam Ford [Mon, 12 Aug 2019 21:45:21 +0000 (16:45 -0500)]
ARM: omapl138_lcdk: Enable USB

The OMAPL138-lcdk has two USB controllers which are currently
disabled.  This patch enables them.

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoconfigs: Rename environment variable fit_bootfile to name_fit
Andrew F. Davis [Mon, 12 Aug 2019 19:59:55 +0000 (15:59 -0400)]
configs: Rename environment variable fit_bootfile to name_fit

Like we did with 'fit_loadaddr' to 'addr_fit', the variable
'fit_bootfile' contains a name and so should be prefixed with
name_. Make this change here.

Signed-off-by: Andrew F. Davis <afd@ti.com>
5 years agoconfigs: Rename environment variable fit_loadaddr to addr_fit
Andrew F. Davis [Mon, 12 Aug 2019 19:59:54 +0000 (15:59 -0400)]
configs: Rename environment variable fit_loadaddr to addr_fit

This is the first part of a larger effort I would like to propose to
unify and simplify the default set of environment variables.

When many early environment variables were named there were fewer images
being loaded, usually just a kernel. At this time names like 'loadaddr'
would suffice. Now we have more images and many more commands that act on
them, often re-using the same variable for several different uses. The
contents of a variable are also not immediately known causing one to have
to look up a chain of variables to understand what a command is actually
doing. I suggest the following.

To start, all variables containing names should be prefixed with name_
and addresses with addr_. This is like how K2 already does things and
allows for simple universal commands like:

get_fdt_nfs=nfs ${addr_fdt} /boot/${name_fdt}

Which is very clear on what is intended here and would work across all
board that using the this naming convention.

We can do this one variable at a time, start here with addr_fit.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Acked-by: Andreas Dannenberg <dannenberg@ti.com>
5 years agoconfigs: Remove unneeded overlay_files environment variable
Andrew F. Davis [Mon, 12 Aug 2019 19:59:53 +0000 (15:59 -0400)]
configs: Remove unneeded overlay_files environment variable

The variable 'name_overlays' serves the same purpose. Remove
'overlay_files' and use 'name_overlays' everywhere.

Signed-off-by: Andrew F. Davis <afd@ti.com>
5 years agoarm: baltos: use device tree alias to access Ethernet slave
Yegor Yefremov [Fri, 9 Aug 2019 05:21:57 +0000 (07:21 +0200)]
arm: baltos: use device tree alias to access Ethernet slave

The full path has changed in the recent kernels so that it is
not possible to load them. Aliases "ethernet0" and "ethernet1"
are still present in both legacy and new kernels.

Also, fix error messages to correspond to the taken actions.

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
5 years agoMerge tag 'u-boot-rockchip-20190819' of https://gitlab.denx.de/u-boot/custodians...
Tom Rini [Mon, 19 Aug 2019 13:22:57 +0000 (09:22 -0400)]
Merge tag 'u-boot-rockchip-20190819' of https://gitlab.denx.de/u-boot/custodians/u-boot-rockchip

- Add ROC-RK3399-PC board support
- Move CONFIG_SPI_FLASH_GIGADEVICE and CONFIG_CMD_USB_MASS_STORAGE to
  Kconfig
- using SYSRESET_POWER_OFF for poweroff
  (Note that patch for rk8xx pmic is droped for it can not pass Travis
  build)
- fix ofnode_get_name() assert

5 years agoMerge https://gitlab.denx.de/u-boot/custodians/u-boot-x86
Tom Rini [Mon, 19 Aug 2019 13:21:46 +0000 (09:21 -0400)]
Merge https://gitlab.denx.de/u-boot/custodians/u-boot-x86

- QEMU build warning fix when CONFIG_DISTRO_DEFAULTS=n
- Small fixes on x86 reST docs
- Allow CBFS to be used in SPL
- Remove x86 specific GD flags

5 years agosysreset: move stm32mp sysreset poweroff implementation to sysreset uclass
Urja Rannikko [Thu, 16 May 2019 21:48:42 +0000 (21:48 +0000)]
sysreset: move stm32mp sysreset poweroff implementation to sysreset uclass

This is a generic implementation. Add CONFIG_SYSRESET_CMD_POWEROFF
to signal when we need it. Enable it from the STPMIC1 config and in
sandbox.

The config flag is transitionary, that is it can be removed after all
poweroff implementations use sysreset, and just have CMD_POWEROFF depend
on sysreset.

Signed-off-by: Urja Rannikko <urjaman@gmail.com>
Reviewed-by: Patrice Chotard <patrice.chotard@st.com>
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com>
Tested-by: Patrick Delaunay <patrick.delaunay@st.com>
5 years agosysreset: switch to using SYSRESET_POWER_OFF for poweroff
Urja Rannikko [Thu, 16 May 2019 21:48:41 +0000 (21:48 +0000)]
sysreset: switch to using SYSRESET_POWER_OFF for poweroff

It seems that SYSRESET_POWER_OFF was added recently, and all previous code
used SYSRESET_POWER for poweroff. SYSRESET_POWER is supposed to be a
PMIC-level power cycle, not a poweroff.

(Comment by Simon Glass)
SYSRESET_POWER means to do a power reset (removing and reinstating all power)
SYSRESET_POWER_OFF means to turn the device off and leave it off

Signed-off-by: Urja Rannikko <urjaman@gmail.com>
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com>
(Update comment to help understand the patch)
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
5 years agoconfigs: update rk3288 veyron defconfigs
Urja Rannikko [Mon, 13 May 2019 13:51:05 +0000 (13:51 +0000)]
configs: update rk3288 veyron defconfigs

Updates jerry, mickey, minnie and speedy defconfigs to:
- fit the SPL in 32k
- boot from SPI (only)
- remove gadget support (these have no OTG port)

Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Urja Rannikko <urjaman@gmail.com>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
(Rebase on top of tree)
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
5 years agoconfigs: Move CONFIG_CMD_USB_MASS_STORAGE properly into Kconfig
Urja Rannikko [Mon, 13 May 2019 13:51:04 +0000 (13:51 +0000)]
configs: Move CONFIG_CMD_USB_MASS_STORAGE properly into Kconfig

This affects RK3036, RK322X and RK3288 - the defconfig changes done by
moveconfig.py for the veyrons were left out on purpose because they dont
have an OTG port, and will get their config updated in the next commit.

Signed-off-by: Urja Rannikko <urjaman@gmail.com>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
5 years agoconfigs: Move CONFIG_SPI_FLASH_GIGADEVICE properly into Kconfig
Urja Rannikko [Mon, 13 May 2019 13:51:03 +0000 (13:51 +0000)]
configs: Move CONFIG_SPI_FLASH_GIGADEVICE properly into Kconfig

Affects rk3288 veyrons and rk3036, this was mostly done by
moveconfig.py.

Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Urja Rannikko <urjaman@gmail.com>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
5 years agocore: ofnode: do not assert if node not valid in ofnode_get_name()
Kever Yang [Fri, 19 Jul 2019 03:23:47 +0000 (11:23 +0800)]
core: ofnode: do not assert if node not valid in ofnode_get_name()

In some case with LIVE DT, some node always not valid, or not have
a valid name, eg. blk driver add by mmc.
Return fail instead of Assert for this kind of ofnode, and this
help with assert happen from time to time when of_live is enabled
and DEBUG is enabled.

Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
5 years agorockchip: rk3399: Add ROC-RK3399-PC support
Levin Du [Fri, 26 Jul 2019 07:43:54 +0000 (15:43 +0800)]
rockchip: rk3399: Add ROC-RK3399-PC support

Add initial support for ROC-RK3399-PC board.

Specification
- Rockchip RK3399
- LPDDR4 4GiB
- eMMC slot
- SD card slot
- RTL8211E 1Gbps
- HDMI Out, DP, MIPI DSI/CSI, EDP
- PCIe M.2
- USB 2.0, USB-3.0
- USB C Type

Commit details of rk3399-roc-pc.dts sync from Linux v5.2:
"arm64: dts: rockchip: add support for ROC-RK3399-PC board"
(sha1: 8bb878cf20ae10809c36db96993bfce7026d062b)

Signed-off-by: Levin Du <djw@t-chip.com.cn>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
5 years agoglobal_data: Remove comment of reserved arch-specific GD flags
Stefan Roese [Fri, 16 Aug 2019 12:45:29 +0000 (14:45 +0200)]
global_data: Remove comment of reserved arch-specific GD flags

With the removal of the x86 specific GD flags, there are no arch-
specific GD flags any more. Let's remove the comment about reserving the
upper 16 bits for arch-specific flags in the common header. This gives
us more flexibility with the usage of the GD flags.

As a matter of fact, we are already using more than 16 bits for common
GD flags (with the addition of GD_FLG_WDT_READY).

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Tom Rini <trini@konsulko.com>
Cc: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
5 years agox86: Remove x86 specific GD flags as they are not referenced at all
Stefan Roese [Fri, 16 Aug 2019 12:45:28 +0000 (14:45 +0200)]
x86: Remove x86 specific GD flags as they are not referenced at all

This patch removes the x86 architecture specific GD flags
(GD_FLG_COLD_BOOT & GD_FLG_WARM_BOOT), as they are not used. Only
GD_FLG_COLD_BOOT is referenced in coreboot.c but assigned in start16.S.
But the coreboot target does not use start16.S at all and boots directly
from the 32-bit start code.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Tom Rini <trini@konsulko.com>
Cc: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
5 years agocbfs: Rename camel-case variables
Simon Glass [Thu, 15 Aug 2019 01:56:15 +0000 (19:56 -0600)]
cbfs: Rename camel-case variables

Rename some camel-case variables to match U-Boot style.

Camel case is not generally allowed in U-Boot. Rename this variable to fit
in with the style.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
5 years agocbfs: Add functions to support multiple CBFSs
Simon Glass [Thu, 15 Aug 2019 01:56:14 +0000 (19:56 -0600)]
cbfs: Add functions to support multiple CBFSs

Sometimes an image has multiple CBFS. The current CBFS API is limited to
handling only one at time. Also it keeps track of the CBFS internally in
BSS, which does not work before relocation, for example.

Add a few new functions to overcome these limitations.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
5 years agocbfs: Move result variable into the struct
Simon Glass [Thu, 15 Aug 2019 01:56:13 +0000 (19:56 -0600)]
cbfs: Move result variable into the struct

Move the result variable into the struct also, so that it can be used when
BSS is not available. Add a function to read it.

Note that all functions sill use the BSS version of the data.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
5 years agocbfs: Move static variables into a struct
Simon Glass [Thu, 15 Aug 2019 01:56:12 +0000 (19:56 -0600)]
cbfs: Move static variables into a struct

At present there are a number of static variables in BSS. This cannot work
with SPL, at least until BSS is available in board_init_r().

Move the variables into a struct, so it is possible to malloc() it and use
it before BSS is available.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
5 years agocbfs: Move declarations above functions
Simon Glass [Thu, 15 Aug 2019 01:56:11 +0000 (19:56 -0600)]
cbfs: Move declarations above functions

At present this file has a function at the top, above declarations. This
is normally avoided, so fix it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
5 years agocbfs: Allow CBFS to be used in SPL
Simon Glass [Thu, 15 Aug 2019 01:56:10 +0000 (19:56 -0600)]
cbfs: Allow CBFS to be used in SPL

Add a new Kconfig option to enable CBFS in SPL. This can be useful when
the memory-init code is in CBFS.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
5 years agodoc: arch: correct links in x86.rst
Heinrich Schuchardt [Thu, 15 Aug 2019 19:20:56 +0000 (21:20 +0200)]
doc: arch: correct links in x86.rst

Correctly reference uefi/uefi.rst and uefi/u-boot_on_efi.rst.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
5 years agodoc: formatting slimbootloader.rst
Heinrich Schuchardt [Thu, 15 Aug 2019 19:02:30 +0000 (21:02 +0200)]
doc: formatting slimbootloader.rst

Avoid a warning when building the 'make htmldocs' target:

doc/board/intel/slimbootloader.rst:90: WARNING: Title underline too short.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
5 years agox86: qemu: Fix build warnings with CONFIG_DISTRO_DEFAULTS=n
Bin Meng [Wed, 14 Aug 2019 08:23:05 +0000 (01:23 -0700)]
x86: qemu: Fix build warnings with CONFIG_DISTRO_DEFAULTS=n

Use DISTRO_BOOTENV to decouple BOOTENV from CONFIG_DISTRO_DEFAULTS.

Reported-by: Heinrich Schuchardt <xypron.debian@gmx.de>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
5 years agoMerge tag 'efi-2019-10-rc3' of https://gitlab.denx.de/u-boot/custodians/u-boot-efi
Tom Rini [Sat, 17 Aug 2019 14:31:25 +0000 (10:31 -0400)]
Merge tag 'efi-2019-10-rc3' of https://gitlab.denx.de/u-boot/custodians/u-boot-efi

Pull request for UEFI sub-system for v2019.10-rc3

This pull request provides corrections for the SetVirtualAddress runtime
service and avoids possible calls to NULL by consumers of the
EFI_PXE_BASE_CODE_PROTOCOL.

5 years agoMerge branch 'master' of git://git.denx.de/u-boot-socfpga
Tom Rini [Sat, 17 Aug 2019 14:30:56 +0000 (10:30 -0400)]
Merge branch 'master' of git://git.denx.de/u-boot-socfpga

- Misc gen5 fixes

5 years agoMerge https://gitlab.denx.de/u-boot/custodians/u-boot-riscv
Tom Rini [Fri, 16 Aug 2019 11:22:21 +0000 (07:22 -0400)]
Merge https://gitlab.denx.de/u-boot/custodians/u-boot-riscv

- Fix sifive serial y-modem transfer.
- Access CSRs using CSR numbers.
- Update doc sifive-fu540
- Support big endian hosts and target.

5 years agoefi_loader: do not call efi_runtime_detach twice
Heinrich Schuchardt [Wed, 14 Aug 2019 04:49:09 +0000 (06:49 +0200)]
efi_loader: do not call efi_runtime_detach twice

Commit 7f95104d91cc ("efi_loader: detach runtime in ExitBootServices()")
added a call to efi_runtime_detach() to ExitBootServices() but did not
remove the call in SetVirtualAddressMap().

Remove the superfluous function call.

Correct a comment referring to efi_runtime_detach().

Fixes: 7f95104d91cc ("efi_loader: detach runtime in ExitBootServices()")
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
5 years agoefi_loader: parameter check in SetVirtualAddressMap
Heinrich Schuchardt [Wed, 14 Aug 2019 03:19:37 +0000 (05:19 +0200)]
efi_loader: parameter check in SetVirtualAddressMap

Check the parameters DescriptorSize and DescriptiorVersion of
SetVirtualAddressMap() as prescribed by the UEFI specification.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
5 years agoefi_loader: EFI_PXE_BASE_CODE_PROTOCOL stub
Heinrich Schuchardt [Tue, 6 Aug 2019 06:13:33 +0000 (08:13 +0200)]
efi_loader: EFI_PXE_BASE_CODE_PROTOCOL stub

U-Boot implements the EFI_PXE_BASE_CODE_PROTOCOL because GRUB uses the mode
information for booting via PXE. All function pointers in the protocol were
NULL up to now which will cause immediate crashes when the services of the
protocol are called.

Create function stubs for all services of the protocol returning
EFI_UNSUPPORTED.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
5 years agoarm: socfpga: gen5: don't zero bss in board_init_f()
Simon Goldschmidt [Fri, 12 Jul 2019 18:03:09 +0000 (20:03 +0200)]
arm: socfpga: gen5: don't zero bss in board_init_f()

The socfpga gen5 SPL manually zeroed bss in board_init_f(). Now that the
DDR driver does not use bss any more, bss is not used before board_init_r()
and we can remove this hack.

bss is normally zeroed by crt0.S, but after board_init_f(), before
board_init_r(). socfpga just had this double-zeroing because it invalidly
used bss in board_init_f() already (during DDR initialization).

Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
5 years agoARM: socfpga: update CONFIG_SPL_FS_LOAD_PAYLOAD_NAME to u-boot.img
Dalon Westergreen [Wed, 7 Aug 2019 17:37:36 +0000 (10:37 -0700)]
ARM: socfpga: update CONFIG_SPL_FS_LOAD_PAYLOAD_NAME to u-boot.img

Bring cyclone5 / arria5 / arria10 in line with convention and use
u-boot.img as CONFIG_SPL_FS_LOAD_PAYLOAD_NAME.

Signed-off-by: Dalon Westergreen <dalon.westergreen@intel.com>
5 years agoarm: socfpga: Fix SYSRESET_SOCFPGA_S10 config name
Ley Foon Tan [Wed, 7 Aug 2019 08:30:53 +0000 (16:30 +0800)]
arm: socfpga: Fix SYSRESET_SOCFPGA_S10 config name

The CONFIG name should be SYSRESET_SOCFPGA_S10 instead of
SYSRESET_SOCFPGA_STRATIX10.

Signed-off-by: Ley Foon Tan <ley.foon.tan@intel.com>
5 years agoriscv: tools: Add big endian target support to prelink-riscv
Marcus Comstedt [Fri, 2 Aug 2019 17:45:16 +0000 (19:45 +0200)]
riscv: tools: Add big endian target support to prelink-riscv

Signed-off-by: Marcus Comstedt <marcus@mc.pp.se>
Cc: Rick Chen <rick@andestech.com>
Reviewed-by: Rick Chen <rick@andestech.com>
5 years agoriscv: tools: Fix prelink-riscv to work on big endian hosts
Marcus Comstedt [Fri, 2 Aug 2019 17:45:15 +0000 (19:45 +0200)]
riscv: tools: Fix prelink-riscv to work on big endian hosts

All ELF fields whose values are inspected by the code are converted to
CPU byteorder first.  Values which are copied verbatim (relocation
fixups) are not swapped to CPU byteorder and back as it is not needed.

Signed-off-by: Marcus Comstedt <marcus@mc.pp.se>
Cc: Rick Chen <rick@andestech.com>
Reviewed-by: Rick Chen <rick@andestech.com>
5 years agodoc: sifive-fu540: Update README to explicitly load DTB for Linux
Anup Patel [Fri, 26 Jul 2019 04:24:30 +0000 (04:24 +0000)]
doc: sifive-fu540: Update README to explicitly load DTB for Linux

We should explicitly load DTB from TFTP server or MMC/SD card
for Linux booting. This will allow us:
1. To use different Linux DTB for SiFive Unleashed board with
   expansion board connected.
2. Avoid re-flashing OpenSBI firmware whenever board connections
   change.

This patch updates reference bootlog in SiFive FU540 README
as-per above.

Signed-off-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
5 years agoriscv: Access CSRs using CSR numbers
Bin Meng [Thu, 11 Jul 2019 06:43:13 +0000 (23:43 -0700)]
riscv: Access CSRs using CSR numbers

We should prefer accessing CSRs using their CSR numbers
because:
1. It compiles fine with older toolchains.
2. We can use latest CSR names in #define macro names of CSR
   numbers as-per RISC-V spec.
3. We can access newly added CSRs even if toolchain does not
   recognize newly added CSRs by name.

This commit is inspired from Linux kernel commit a3182c91ef4e
("RISC-V: Access CSRs using CSR numbers").

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Rick Chen <rick@andestech.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Lukas Auer <lukas.auer@aisec.fraunhofer.de>
5 years agoriscv: Sync csr.h with Linux kernel v5.2
Bin Meng [Thu, 11 Jul 2019 06:43:12 +0000 (23:43 -0700)]
riscv: Sync csr.h with Linux kernel v5.2

This syncs csr.h with Linux kernel 5.2, and imports asm.h that
is required by csr.h.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Rick Chen <rick@andestech.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Lukas Auer <lukas.auer@aisec.fraunhofer.de>
5 years agoriscv : serial: use rx watermark to indicate rx data is present
Sagar Shrikant Kadam [Tue, 9 Jul 2019 12:23:44 +0000 (05:23 -0700)]
riscv : serial: use rx watermark to indicate rx data is present

In y-modem transfer mode, tstc/getc fail to check if there is any
data available / received in RX FIFO, and so y-modem transfer never
succeeds. Using receive watermark bit within ip register fixes the
issue.

This patch is based on commit c7392b7bc4e1 ("Use the RX watermark
interrupt pending bit for TSTC") available at[1]

[1] https://github.com/sifive/HiFive_U-Boot/tree/regression

Signed-off-by: Sagar Shrikant Kadam <sagar.kadam@sifive.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Tested-by: Anup Patel <anup.patel@wdc.com>
Reviewed-by: Padmarao Begari <padmarao.begari@microchip.com>
Tested-by: Padmarao Begari <padmarao.begari@microchip.com>
5 years agoconfigs: Resync with savedefconfig
Tom Rini [Wed, 14 Aug 2019 12:11:27 +0000 (08:11 -0400)]
configs: Resync with savedefconfig

Rsync all defconfig files using moveconfig.py

Signed-off-by: Tom Rini <trini@konsulko.com>
5 years agoPrepare v2019.10-rc2 v2019.10-rc2
Tom Rini [Tue, 13 Aug 2019 22:23:35 +0000 (18:23 -0400)]
Prepare v2019.10-rc2

Signed-off-by: Tom Rini <trini@konsulko.com>
5 years agoMerge https://gitlab.denx.de/u-boot/custodians/u-boot-marvell
Tom Rini [Tue, 13 Aug 2019 03:03:44 +0000 (23:03 -0400)]
Merge https://gitlab.denx.de/u-boot/custodians/u-boot-marvell

- Various improvements to Keymile boards - mostly DT conversation
  (Pascal & Holger)
- Removal of now unsupported Keymile boards (Pascal & Holger)
- Small MVEBU PCI fix (Marek)
- Turris Omnia defconfig update (Marek)
- Misc Allied Telesis defconfig updates (Chris)

5 years agoMerge tag 'u-boot-amlogic-20190812' of https://gitlab.denx.de/u-boot/custodians/u...
Tom Rini [Mon, 12 Aug 2019 22:48:34 +0000 (18:48 -0400)]
Merge tag 'u-boot-amlogic-20190812' of https://gitlab.denx.de/u-boot/custodians/u-boot-amlogic

- amlogic: add support for the SEI Robotic SEI510

5 years agoMerge branch '2019-08-11-ti-imports'
Tom Rini [Mon, 12 Aug 2019 22:47:53 +0000 (18:47 -0400)]
Merge branch '2019-08-11-ti-imports'

- More DaVinci updates and fixes
- PCIe support on am65x
- Watchdog converted to DM
- Assorted other bugfixes

5 years agoMerge branch '2019-08-11-master-imports'
Tom Rini [Mon, 12 Aug 2019 22:46:07 +0000 (18:46 -0400)]
Merge branch '2019-08-11-master-imports'

- environment cleanup
- HiKey 960 support
- Some PCI fixes

5 years agogitlab-ci: Consolidate some testsuite jobs
Tom Rini [Mon, 12 Aug 2019 14:09:08 +0000 (10:09 -0400)]
gitlab-ci: Consolidate some testsuite jobs

We do not need to split binman, buildman, dtoc and patman test suite
runs into 3 jobs.  Instead, run them as a single job.

Signed-off-by: Tom Rini <trini@konsulko.com>
5 years agogitlab-ci: Add USER environment variable to buildman/patman tests
Tom Rini [Mon, 12 Aug 2019 14:02:18 +0000 (10:02 -0400)]
gitlab-ci: Add USER environment variable to buildman/patman tests

The way that some of the tests here are designed, they expect USER to be
set in the environment.  This is not the case in the docker images, so
set a reasonable value.

Signed-off-by: Tom Rini <trini@konsulko.com>
5 years agoarm: K3: sysfw-loader: Do not require full printf() for version info
Andreas Dannenberg [Mon, 5 Aug 2019 18:46:23 +0000 (13:46 -0500)]
arm: K3: sysfw-loader: Do not require full printf() for version info

A previous commit...

commit 2a51e16bd57a ("configs: Make USE_TINY_PRINTF depend on SPL||TPL and be default")

...causes the System Firmware version string during SPL boot to no longer
getting printed to the console as expected. To fix this issue rework the
handling of that string to only use basic printf() syntax rather than
for example disabling CONFIG_USE_TINY_PRINTF on affected devices, this
way maintaining most of the memory size benefit the initial patch brings
when it comes to SPL.

Signed-off-by: Andreas Dannenberg <dannenberg@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
5 years agofirmware: ti_sci: Parse all resource ranges even if some is not available
Vignesh Raghavendra [Mon, 5 Aug 2019 17:26:44 +0000 (12:26 -0500)]
firmware: ti_sci: Parse all resource ranges even if some is not available

Do not fail if any of the requested subtypes are not available, but set the
number of resources to 0 and continue parsing the resource ranges.

Based on Linux kernel patch by Peter Ujfalusi <peter.ujfalusi@ti.com>

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
5 years agoARM: da850-evm: Enable SPI Flash and NAND Flash when booting NOR
Adam Ford [Sat, 3 Aug 2019 12:56:10 +0000 (07:56 -0500)]
ARM: da850-evm: Enable SPI Flash and NAND Flash when booting NOR

Historically there have been various boot options, SPI flash,
NAND or NOR. The NOR flash is mutually exclusive with MMC, but
it isn't mutually exclusive with NAND or SPI Flash, so this patch
enables both NAND flash and SPI Flash when booting from NOR.

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoARM: da850-evm: Enable filesystem commands when booting from NOR
Adam Ford [Fri, 2 Aug 2019 21:21:55 +0000 (16:21 -0500)]
ARM: da850-evm: Enable filesystem commands when booting from NOR

The configuration with NOR is mutually exclusive with MMC, and as
such, the filesystem commands were disabled.  With the USB host
enabled, this patch enables the file system command which can
be executed on storage devices attached to the USB.

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoARM: da850-evm: Increase environment NOR partition offset
Adam Ford [Fri, 2 Aug 2019 12:03:39 +0000 (07:03 -0500)]
ARM: da850-evm: Increase environment NOR partition offset

The current size allocated to U-Boot is 384k, but U-Boot has grown
to 436K which means that saving the environmental variables wipes
out part of the U-Boot source and the board ceases to function.
Due to the sector and erase size for the NOR part and a desire to
not have to change partition sizes often, this patch moves the
U-Boot environmental variables to an offset of 1M so saveenv
does not brick the board.  This patch also sets up MTDIDS and
MTDPARTS to clearly show where U-Boot and U-Boot's environmental
variables are located.

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoARM: da850-evm: Fix Lowlevel init
Adam Ford [Thu, 1 Aug 2019 13:47:55 +0000 (08:47 -0500)]
ARM: da850-evm: Fix Lowlevel init

A previous patch for enabling the NAND config set a flag called
CONFIG_SKIP_LOWLEVEL_INIT when it should have been called
CONFIG_SKIP_LOWLEVEL_INIT_ONLY.  The affect this had was creating
a delay on startup for the NOR version which is XIP and doesn't have
SPL, so the lowlevel initialization functions need to operate.
This delay was not really noticeable at first, but the delays have been
getting longer, finally reached the point of nearly seven seconds
before the board would appear to start.

This patch sets the CONFIG_SKIP_LOWLEVEL_INIT_ONLY instead which means
"The normal CP15 init (such as enabling the instruction cache) is still
performed" per the README.  It doesn't appear to have any adverse
behavior on the SPI Flash or the NAND flash boards which use SPL.

Fixes: 93f3362762f0 ("ARM: configs: Add da850evm_nand to boot from NAND")

Signed-off-by: Adam Ford <aford173@gmail.com>
5 years agoconfigs: am65x_evm_a53: enable support for PCIe ethernet cards
Sekhar Nori [Thu, 1 Aug 2019 13:43:01 +0000 (19:13 +0530)]
configs: am65x_evm_a53: enable support for PCIe ethernet cards

Enable support for Intel E1000 based PCIe ethernet cards that
can be used to test PCIe RC functionality on AM65x EVM.

Signed-off-by: Sekhar Nori <nsekhar@ti.com>
5 years agoarm: dts: k3-am65: add support for PCIe and SERDES
Sekhar Nori [Thu, 1 Aug 2019 13:43:00 +0000 (19:13 +0530)]
arm: dts: k3-am65: add support for PCIe and SERDES

Add needed device-tree nodes to support PCIe 0
and SERDES on AM65x SoC. The nodes are kept
disabled by default.

Signed-off-by: Sekhar Nori <nsekhar@ti.com>
5 years agoconfigs: am65x_evm_a53: enable PCIe support
Sekhar Nori [Thu, 1 Aug 2019 13:42:59 +0000 (19:12 +0530)]
configs: am65x_evm_a53: enable PCIe support

Enable support for PCIe and related configurations
on AM654 EVM platform.

Signed-off-by: Sekhar Nori <nsekhar@ti.com>
5 years agophy: add support for AM654x SERDES
Sekhar Nori [Thu, 1 Aug 2019 13:42:58 +0000 (19:12 +0530)]
phy: add support for AM654x SERDES

Add a new SERDES driver for TI's AM654x SoC which configures
the SERDES only for PCIe. Support fo USB3 can be added later.

SERDES in am654x has three input clocks (left input, external
reference clock and right input) and two output clocks (left
output and right output) in addition to a PLL mux clock which
the SERDES uses for Clock Multiplier Unit (CMU refclock).

The PLL mux clock can select from one of the three input
clocks. The right output can select between left input and
external reference clock while the left output can select
between the right input and external reference clock.

The driver has support to select PLL mux and left/right output
mux as specified in device tree.

Signed-off-by: Sekhar Nori <nsekhar@ti.com>
5 years agopcie: ti: add driver for AM65x PCIe RC
Sekhar Nori [Thu, 1 Aug 2019 13:42:57 +0000 (19:12 +0530)]
pcie: ti: add driver for AM65x PCIe RC

Add driver supporting PCIe root-complex available
on TI's AM65x SoC.

Signed-off-by: Sekhar Nori <nsekhar@ti.com>
5 years agodm: core: add support for getting register address and size
Sekhar Nori [Thu, 1 Aug 2019 13:42:56 +0000 (19:12 +0530)]
dm: core: add support for getting register address and size

Current dev_read_*() API lacks support to get address and size
of a "reg" property by name or index. Add support for the same.

Livetree support has been added but not tested on real hardware.

The existing unit tests testing reading address from device-tree
have been updated to test address as well as size.

Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
5 years agoclk: add support for clk_is_match()
Sekhar Nori [Thu, 1 Aug 2019 13:42:55 +0000 (19:12 +0530)]
clk: add support for clk_is_match()

Add support for clk_is_match() which is required to
know if two clock pointers point to the same exact
physical clock.

Also add a unit test for the new API.

Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
5 years agopower: domain: ti-sci-power-domain: Add device ID info to get/set_device failure
Nishanth Menon [Thu, 1 Aug 2019 13:38:24 +0000 (19:08 +0530)]
power: domain: ti-sci-power-domain: Add device ID info to get/set_device failure

Failure log for ti_sci_power_domain_on/off is as below:
"ti_sci_power_domain_on: get/set_device failed (-19)"

The above information is useless for debug without information
on what specific device access failed. So add that information as well.

Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
5 years agoboard/BuR/brsmarc1: initial commit
Hannes Schmelzer [Thu, 1 Aug 2019 05:04:46 +0000 (07:04 +0200)]
board/BuR/brsmarc1: initial commit

This commit adds support for the B&R brsmarc1 SoM.

The SoM is based on TI's AM335x SoC.
Mainly vxWorks 6.9.4.x is running on the board,
doing some PLC stuff on various carrier boards.

Signed-off-by: Hannes Schmelzer <hannes.schmelzer@br-automation.com>
5 years agowatchdog: omap_wdt: Disable DM watchdog support in SPL
Suniel Mahesh [Wed, 31 Jul 2019 16:24:07 +0000 (21:54 +0530)]
watchdog: omap_wdt: Disable DM watchdog support in SPL

This patch disables DM watchdog support for SPL builds and uses
the legacy omap watchdog driver on TI AM335x chipsets.

The following build error is reported if DM watchdog support was
enabled in SPL:

  CC      spl/drivers/usb/gadget/rndis.o
  LD      spl/drivers/usb/gadget/built-in.o
  LD      spl/drivers/usb/musb-new/built-in.o
  LD      spl/drivers/built-in.o
  LD      spl/u-boot-spl
arm-linux-ld.bfd: u-boot-spl section .u_boot_list will not fit in region .sram
arm-linux-ld.bfd: region .sram overflowed by 440 bytes
make[1]: *** [spl/u-boot-spl] Error 1
make: *** [spl/u-boot-spl] Error 2

Adjusted WATCHDOG_RESET macro accordingly. Earlier it was pointing
to hw_watchdog_reset. Since CONFIG_WATCHDOG replaces CONFIG_HW_WATCHDOG,
now WATCHDOG_RESET macro points to watchdog_reset. This watchdog_reset
is not defined anywhere for am33xx/omap2 and needs to be defined. Fixed
this by simply calling hw_watchdog_reset in watchdog_reset.

Built and tested on AM335x device (BeagleboneBlack), compile tested for
all other AM33xx/omap2 based boards.

Signed-off-by: Suniel Mahesh <sunil.m@techveda.org>
[trini: Fix watchdog.h logic]
Signed-off-by: Tom Rini <trini@konsulko.com>
5 years agopci: pci_mvebu: set BAR0 after memory space is set
Marek Behún [Wed, 7 Aug 2019 13:01:56 +0000 (15:01 +0200)]
pci: pci_mvebu: set BAR0 after memory space is set

The non-DM version of this driver used to set BAR0 register after the
calls to pci_set_region.
I found out that for some strange reason the ath10k driver in kernel
fails to work if this is done the other way around.
I know that Linux's driver should not depend on how U-Boot does things,
but for some strange reason it does and this seems to be the simplest
solution. Fix it since it caused regressions on Omnia.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Cc: Stefan Roese <sr@denx.de>
Cc: Dirk Eibach <dirk.eibach@gdsys.cc>
Cc: Mario Six <mario.six@gdsys.cc>
Cc: Chris Packham <chris.packham@alliedtelesis.co.nz>
Cc: Phil Sutter <phil@nwl.cc>
Cc: VlaoMao <vlaomao@gmail.com>
Tested-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
5 years agowatchdog: omap_wdt: Convert watchdog driver to use DT and DM
Suniel Mahesh [Wed, 31 Jul 2019 16:24:06 +0000 (21:54 +0530)]
watchdog: omap_wdt: Convert watchdog driver to use DT and DM

This patch adds device tree and driver model watchdog support,
converts the legacy omap watchdog driver to driver model for
TI AM335x chipsets. The following compile warning is removed:

===================== WARNING ======================
This board does not use CONFIG_WDT (DM watchdog support).
Please update the board to use CONFIG_WDT before the
v2019.10 release.
Failure to update by the deadline may result in board removal.
See doc/driver-model/MIGRATION.txt for more info.
====================================================

CONFIG_HW_WATCHDOG is no more a default option for AM33XX devices
after DT/DM conversion, adjusted kconfig accordingly.

DM watchdog support is enabled by default in SPL. The SPL image
doesn't fit into SRAM because of size constraints and build breaks
with an overflow. For this reason DM watchdog support should be
disabled in SPL, driver code should be adjusted accordingly to serve
this purpose.
Built and tested on AM335x device (BeagleboneBlack), compile tested
for all other AM33xx based boards.

Signed-off-by: Suniel Mahesh <sunil.m@techveda.org>
5 years agoARM: omap3: overo: Fix MMC init for SPL
Anselm Busse [Wed, 31 Jul 2019 15:30:28 +0000 (17:30 +0200)]
ARM: omap3: overo: Fix MMC init for SPL

The SPL for the Overo board does not initialise the MMC. Hence, it
cannot load the main boot loader from the SD card susequently. This
patch moves the initialisation code for the MMC so it gets included in
the SPL.

[trini: Add missing header]
Signed-off-by: Tom Rini <trini@konsulko.com>