oweals/u-boot.git
3 years agoMerge tag 'u-boot-imx-20200502' of https://gitlab.denx.de/u-boot/custodians/u-boot-imx
Tom Rini [Mon, 4 May 2020 13:29:42 +0000 (09:29 -0400)]
Merge tag 'u-boot-imx-20200502' of https://gitlab.denx.de/u-boot/custodians/u-boot-imx

i.MX for 2020.07
----------------

- imxrt: fix LCD clock, fix doc
- new board: Coral Dev
- imx8: enable Cache in SPL. SNVS, update SCFW API
- imx8MM: fix reset, 8MQ quand and QuadLite, CPU speed grading
- MX6ULL : is_imx6ull to include i.MX6ULZ
- Net: add config to enable TXC delay

Travis: https://travis-ci.org/github/sbabic/u-boot-imx/builds/682033914

3 years agonet: pcnet: fix I/O primitives for memory access
Daniel Schwierzeck [Sun, 3 May 2020 17:43:32 +0000 (19:43 +0200)]
net: pcnet: fix I/O primitives for memory access

Commit 69529c912059 ("net: pcnet: Switch to PCI memory access")
switched from PCI IO access to PCI memory access without updating
the I/O primitives. Contrary to SH, the primitives for memory
access and IO access are implemented differently. Thus doing
memory access with IO port primitives breaks the driver on
MIPS Malta board.

Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
3 years agoMerge tag 'u-boot-rockchip-20200501' of https://gitlab.denx.de/u-boot/custodians...
Tom Rini [Mon, 4 May 2020 11:28:14 +0000 (07:28 -0400)]
Merge tag 'u-boot-rockchip-20200501' of https://gitlab.denx.de/u-boot/custodians/u-boot-rockchip

- dts clean up to use -u-boot for px30, rk3399 boards
- dts sycn from upstream kernel for rk3328, rk3399
- add rockchip rng driver
- new board support: rk3328-roc-cc, rk3399-roc-pc,Nanopi M4 2GB

3 years agoMerge branch 'next' of git://git.denx.de/u-boot-sh
Tom Rini [Fri, 1 May 2020 21:58:31 +0000 (17:58 -0400)]
Merge branch 'next' of git://git.denx.de/u-boot-sh

This is part 1 of big network cleanup / DM conversion.
The dc2114x/rtl8139/pcnet/tulip PCI adapter drivers received checkpatch
cleanups in preparation for DM conversion. The smc911x is converted to
DM completely. The dwc_eth_qos cache fixes are applied.

3 years agomkimage: fit: Do not tail-pad fitImage with external data
Marek Vasut [Fri, 1 May 2020 15:40:25 +0000 (17:40 +0200)]
mkimage: fit: Do not tail-pad fitImage with external data

There is no reason to tail-pad fitImage with external data to 4-bytes,
while fitImage without external data does not have any such padding and
is often unaligned. DT spec also does not mandate any such padding.

Moreover, the tail-pad fills the last few bytes with uninitialized data,
which could lead to a potential information leak.

$ echo -n xy > /tmp/data ; \
./tools/mkimage -E -f auto -d /tmp/data /tmp/fitImage ; \
hexdump -vC /tmp/fitImage | tail -n 3

before:
00000260  61 2d 6f 66 66 73 65 74  00 64 61 74 61 2d 73 69  |a-offset.data-si|
00000270  7a 65 00 00 78 79 64 64                           |ze..xydd|
                   ^^       ^^ ^^
after:
00000260  61 2d 6f 66 66 73 65 74  00 64 61 74 61 2d 73 69  |a-offset.data-si|
00000270  7a 65 00 78 79                                    |ze.xy|

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>
Cc: Tom Rini <trini@konsulko.com>
3 years agoMerge branch '2020-05-01-master-imports'
Tom Rini [Fri, 1 May 2020 20:43:15 +0000 (16:43 -0400)]
Merge branch '2020-05-01-master-imports'

- Assorted bug fixes
- Framework for enabling D-CACHE in SPL on ARM

3 years ago.readthedocs.yml: fix type docs/
Heinrich Schuchardt [Fri, 1 May 2020 20:06:17 +0000 (22:06 +0200)]
.readthedocs.yml: fix type docs/

Out documentation directory is doc/ and not docs/.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
3 years agoactions: Fix syntax for enabling SYS_RELOC_GD_ENV_ADDR
Tom Rini [Fri, 1 May 2020 14:52:11 +0000 (10:52 -0400)]
actions: Fix syntax for enabling SYS_RELOC_GD_ENV_ADDR

The correct syntax is 'select SYS_...' and not 'select CONFIG_SYS...'

Fixes: d5c819b885c2 ("actions: Move defconfig options to Kconfig")
Signed-off-by: Tom Rini <trini@konsulko.com>
3 years agoclk: imx: clk-imxrt1050: fix lcdif clock gate
Giulio Benetti [Mon, 27 Apr 2020 15:53:06 +0000 (17:53 +0200)]
clk: imx: clk-imxrt1050: fix lcdif clock gate

LCDIF clock gate was wrong so set it according to RM.

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
3 years agovideo: mxsfb: add clk_enable()
Giulio Benetti [Mon, 27 Apr 2020 15:53:05 +0000 (17:53 +0200)]
video: mxsfb: add clk_enable()

BROM doesn't enable lcdif by default so add clk_enable() after
clk_set_rate().

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
3 years agoimxrt1020-evk: README: change dd command destination
Giulio Benetti [Mon, 27 Apr 2020 15:53:04 +0000 (17:53 +0200)]
imxrt1020-evk: README: change dd command destination

Set dd "of=" to "of=/dev/sdX" to be generic and prevent host hard drive
damage.

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
3 years agoimxrt1050-evk: README: change dd command destination
Giulio Benetti [Mon, 27 Apr 2020 15:53:03 +0000 (17:53 +0200)]
imxrt1050-evk: README: change dd command destination

Set dd "of=" to "of=/dev/sdX" to be generic and prevent host hard drive
damage.

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
3 years agoimx: spl: Fix use of removed SPL_FAT_SUPPORT config
Harald Seiler [Thu, 23 Apr 2020 11:07:53 +0000 (13:07 +0200)]
imx: spl: Fix use of removed SPL_FAT_SUPPORT config

CONFIG_SPL_FAT_SUPPORT was removed in commit 0c3a9ed409a5
("spl: Kconfig: Replace CONFIG_SPL_FAT_SUPPORT with CONFIG_SPL_FS_FAT").
Fixup a leftover use of the symbol.

Fixes: 9d86dbd9cf9d ("imx: spl: implement spl_boot_mode for i.MX7/8/8M")
Signed-off-by: Harald Seiler <hws@denx.de>
3 years agoimx: spl: Remove ifdefs in spl_mmc_boot_mode()
Harald Seiler [Thu, 23 Apr 2020 11:07:52 +0000 (13:07 +0200)]
imx: spl: Remove ifdefs in spl_mmc_boot_mode()

It is hard to read code which contains nested ifdef blocks.  Replace
them with normal if-blocks and the IS_ENABLED() macro.  This is not only
more readable but also helps as both arms are validated by the compiler
in all cases.

Signed-off-by: Harald Seiler <hws@denx.de>
3 years agoRevert "imx: defconfig: Enable CONFIG_SPL_FORCE_MMC_BOOT"
Harald Seiler [Thu, 23 Apr 2020 11:07:51 +0000 (13:07 +0200)]
Revert "imx: defconfig: Enable CONFIG_SPL_FORCE_MMC_BOOT"

CONFIG_SPL_FORCE_MMC_BOOT was removed in a previous patch as its
behavior is the correct one in all cases.  Remove all uses of it from
defconfigs.

This reverts commit 3201e5b444ae3a13aa31e8b5101ad38d7ff0640d and removes
CONFIG_SPL_FORCE_MMC_BOOT from the imx28_xea defconfig.

Signed-off-by: Harald Seiler <hws@denx.de>
3 years agoRevert "imx: Introduce CONFIG_SPL_FORCE_MMC_BOOT to force MMC boot on falcon mode"
Harald Seiler [Thu, 23 Apr 2020 11:07:50 +0000 (13:07 +0200)]
Revert "imx: Introduce CONFIG_SPL_FORCE_MMC_BOOT to force MMC boot on falcon mode"

The CONFIG_SPL_FORCE_MMC_BOOT config flag is not needed as its behavior
is the correct one in all cases;  using spl_boot_device() instead of the
boot_device parameter will lead to inconsistency issues, for example,
when a board_boot_order() is defined.  In fact, this is the reason the
parameter was introduced in the first place, in commit 2b1cdafa9fdd
("common: Pass the boot device into spl_boot_mode()").

This reverts commit 772b55723bcbe8ebe84f579d9cdc831d8e18579d.

Link: https://lists.denx.de/pipermail/u-boot/2020-April/405979.html
Signed-off-by: Harald Seiler <hws@denx.de>
3 years agoimx: spl: return boot mode for asked MMC device in spl_mmc_boot_mode()
Anatolij Gustschin [Thu, 23 Apr 2020 11:07:49 +0000 (13:07 +0200)]
imx: spl: return boot mode for asked MMC device in spl_mmc_boot_mode()

Boards may extend or re-define the boot list in their board_boot_order()
function by modifying spl_boot_list. E.g. a board might boot SPL from a
slow SPI NOR flash and then load the U-Boot from an eMMC or SD-card.
Or it might use additional MMC boot device in spl_boot_list for cases
when the image in SPI NOR flash is not found, so it could fall back to
eMMC, SD-card or another boot device.

Getting the MMC boot mode in spl_mmc will fail when we are trying to
boot from an MMC device in the spl_boot_list and the original board
boot mode (as returned by spl_boot_device()) is not an MMC boot mode.
Fix it by checking the asked MMC boot device from the spl_mmc_boot_mode()
argument.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
3 years agocommon/board_r: Drop initr_bedbug wrapper
Ovidiu Panait [Mon, 20 Apr 2020 07:31:46 +0000 (10:31 +0300)]
common/board_r: Drop initr_bedbug wrapper

Drop initr_bedbug wrapper and call bedbug_init directly during the init
sequence.

Signed-off-by: Ovidiu Panait <ovpanait@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
3 years agocmd/bedbug.c: Make bedbug_init have a return value
Ovidiu Panait [Mon, 20 Apr 2020 07:31:45 +0000 (10:31 +0300)]
cmd/bedbug.c: Make bedbug_init have a return value

Do this as a preparation for removing initr_bedbug wrapper from
common/board_r.c.

Signed-off-by: Ovidiu Panait <ovpanait@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
3 years agocommon/board_r: arm: Merge initr_enable_interrupts into interrupts_init
Ovidiu Panait [Mon, 20 Apr 2020 07:31:44 +0000 (10:31 +0300)]
common/board_r: arm: Merge initr_enable_interrupts into interrupts_init

initr_enable_interrupts() is an ARM-specific wrapper over
enable_interrupts(), which is run during the common init sequence. It can
be eliminated by moving the enable_interrupts() call to the end of
interrupt_init() function, in arch/arm/lib/interrupts*.c.

Signed-off-by: Ovidiu Panait <ovpanait@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
3 years agomalloc: dlmalloc: add an ability for the malloc to be re-init/init multiple times
Marek Bykowski [Wed, 29 Apr 2020 16:23:07 +0000 (18:23 +0200)]
malloc: dlmalloc: add an ability for the malloc to be re-init/init multiple times

Malloc gets initialized with a call to mem_malloc_init() with the address
the allocation starts to and its size. Currently it is not possible to
move the malloc from one memory area to another as the malloc would eventually
fail.

This patch adds in the ability to re-init the malloc with the updated
start address and the size.

One of the use cases of this feature is SPL U-Boot running from within
the static memory and calling to malloc init from within board_init_f():

arch/arm/cpu/armv8/start.S:reset vector
arch/arm/cpu/armv8/start.S:main()
arch/arm/lib/crt0_64.S:board_init_f()
board/<my_board>/common/spl.c:board_init_f()
        board/<my_board>/common/spl.c:mem_malloc_init((ulong)CONFIG_SYS_SPL_MALLOC_START,
                        CONFIG_SYS_SPL_MALLOC_SIZE);

Shortly after the DDR (main) memory is init and ready we call to malloc init
again but this time with the start address in the DDR memory and a much greater
size for moving the allocation off the static to the DDR memory:

board/<my_board>/common/spl.c:mem_malloc_init((ulong)CONFIG_SPL_MALLOC_OFFSET,
CONFIG_SPL_MALLOC_SIZE);

Where CONFIG_SYS_SPL_MALLOC_START and CONFIG_SPL_MALLOC_OFFSET are the start
addresses of the malloc in the static and DDR memories respectively and
CONFIG_SYS_SPL_MALLOC_SIZE=SZ_16K and CONFIG_SPL_MALLOC_SIZE=SZ_2M are
the sizes of the mallocs in these memories. Note, now we have a much greater
memory, enlarging from 16K to 2M, available for allocation.

There is an alternative approach already existing in U-Boot with the use of
an early (simplified) malloc and the proper (dlamalloc) malloc however
necessitating managing the two mallocs whereas this approach proposes using
a single dlmalloc.

Signed-off-by: Marek Bykowski <marek.bykowski@gmail.com>
3 years agorsa: sig: fix config signature check for fit with padding
Philippe Reynes [Wed, 29 Apr 2020 13:26:17 +0000 (15:26 +0200)]
rsa: sig: fix config signature check for fit with padding

The signature check on config node is broken on fit with padding.
To compute the signature for config node, U-Boot compute the
signature on all properties of requested node for this config,
except for the property "data". But, when padding is used for
binary in a fit, there isn't a property "data" but two properties:
"data-offset" and "data-size". So to fix the check of signature,
we also don't use the properties "data-offset" and "data-size"
when checking the signature on config node.

Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
3 years agotest/py: vboot: add a test to check fit signature on fit with padding
Philippe Reynes [Wed, 29 Apr 2020 13:26:16 +0000 (15:26 +0200)]
test/py: vboot: add a test to check fit signature on fit with padding

The pytest vboot does all his tests on fit without padding.
We add the same tests on fit with padding.

Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
3 years agohush: avoid NULL check before free()
Heinrich Schuchardt [Tue, 28 Apr 2020 19:56:10 +0000 (21:56 +0200)]
hush: avoid NULL check before free()

free() checks if its argument is NULL. Don't duplicate this in the calling
code.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
3 years agofs: ext4: avoid NULL check before free()
Heinrich Schuchardt [Tue, 28 Apr 2020 19:50:02 +0000 (21:50 +0200)]
fs: ext4: avoid NULL check before free()

free() checks if its argument is NULL. Don't duplicate this in the calling
code.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
3 years agocmd/gpt: avoid NULL check before free()
Heinrich Schuchardt [Tue, 28 Apr 2020 19:40:13 +0000 (21:40 +0200)]
cmd/gpt: avoid NULL check before free()

free() checks if its argument is NULL. Do not duplicate this in the calling
code.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Tom Rini <trini@konsulko.com>
3 years agotools/fit-image: print a warning when cmd-line for dtc might be truncated
Sven Roederer [Mon, 27 Apr 2020 00:08:39 +0000 (02:08 +0200)]
tools/fit-image: print a warning when cmd-line for dtc might be truncated

Signed-off-by: Sven Roederer <devel-sven@geroedel.de>
3 years agotools/mkimage: fix handling long filenames
Sven Roederer [Mon, 27 Apr 2020 00:08:38 +0000 (02:08 +0200)]
tools/mkimage: fix handling long filenames

The cmdline for calling the dtc was cut-off when using long filenames (e.g.
245 bytes) for output-file and datafile of "-f" parameter.
For FIT-images cmd[MKIMAGE_MAX_DTC_CMDLINE_LEN] is declared (hardcoded 512 bytes),
and contains some static values, the path of a tmpfile and a datafile. tmpfile is
max MKIMAGE_MAX_TMPFILE_LEN (256) and datafile might be also this size. Having two
very long pathname results in a truncation os the executed shell command, as the
truncated datafile path will not be found.
Redefine MKIMAGE_MAX_DTC_CMDLINE_LEN to "2 * MKIMAGE_MAX_TMPFILE_LEN + 35 for the
parameters.
This likely applies to the "-d" parameter, too.

Signed-off-by: Sven Roederer <devel-sven@geroedel.de>
3 years agoarm: enable distro boot for bananapi-r2
Matthias Brugger [Sat, 25 Apr 2020 23:17:46 +0000 (01:17 +0200)]
arm: enable distro boot for bananapi-r2

This patch enables distro boot for the bananapi-r2, based on
a MediaTek mt7623n.

Signed-off-by: Matthias Brugger <mbrugger@suse.com>
3 years agowatchdog MediaTek add upstream compatible
Matthias Brugger [Sat, 25 Apr 2020 23:17:45 +0000 (01:17 +0200)]
watchdog MediaTek add upstream compatible

The upstream compatible is called mt6589-wdt.
Add this compatible to the driver.

Signed-off-by: Matthias Brugger <mbrugger@suse.com>
3 years agotimer MediaTek use upstream compatible
Matthias Brugger [Sat, 25 Apr 2020 23:17:44 +0000 (01:17 +0200)]
timer MediaTek use upstream compatible

The timers compatible string in upstream is called
mt6577-timer. Add this compatible to the driver.

Signed-off-by: Matthias Brugger <mbrugger@suse.com>
3 years agodoc: describe the analysis of crash dumps
Heinrich Schuchardt [Fri, 24 Apr 2020 21:31:20 +0000 (23:31 +0200)]
doc: describe the analysis of crash dumps

Provide an overview of the analysis of U-Boot crash dumps.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
3 years agortc: pcf2127: don't add/subtract 1 to tm_mon
Rasmus Villemoes [Fri, 1 May 2020 13:24:50 +0000 (15:24 +0200)]
rtc: pcf2127: don't add/subtract 1 to tm_mon

As noted in rtc_def.h, the tm_mon field in struct rtc_time is 1-12,
unlike in struct tm where it is 0-11. Currently, running "date" prints
the wrong

Date: 2020-04-01 (Friday)    Time: 13:05:30

and setting the RTC via the date command is also broken.

Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
3 years agoarm: caches: manage phys_addr_t overflow in mmu_set_region_dcache_behaviour
Patrick Delaunay [Fri, 24 Apr 2020 18:20:17 +0000 (20:20 +0200)]
arm: caches: manage phys_addr_t overflow in mmu_set_region_dcache_behaviour

Solved the overflow on phys_addr_t type for start + size in
mmu_set_region_dcache_behaviour() function.

This overflow is avoided by dividing start and end by 2 before addition,
and we only expecting that start and size are even.

This patch doesn't change the current function behavior if the
parameters (start or size) are not aligned on MMU_SECTION_SIZE.

For example, this overflow occurs on ARM32 with:
start = 0xC0000000 and size = 0x40000000
then start + size = 0x100000000 and end = 0x0.

For information the function behavior change with risk of regression,
if we just shift start and size before the addition.
Example with 2MB section size:
  MMU_SECTION_SIZE 0x200000 and MMU_SECTION_SHIFT = 21
  with start = 0x1000000, size = 0x1000000,
  - with the proposed patch, start = 0 and end = 0x1 as previously
  - with the more simple patch:
    end = (start >> MMU_SECTION_SHIFT) + (size >> MMU_SECTION_SHIFT)
    the value of end change:
    start >> 21 = 0, size >> 21 = 0 and end = 0x0 !!!

Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
3 years agoarm: caches: add DCACHE_DEFAULT_OPTION
Patrick Delaunay [Fri, 24 Apr 2020 18:20:16 +0000 (20:20 +0200)]
arm: caches: add DCACHE_DEFAULT_OPTION

Add the new flags DCACHE_DEFAULT_OPTION to define the default
option to use according the compilation flags
CONFIG_SYS_ARM_CACHE_*.

This new compilation flag allows to simplify dram_bank_mmu_setup()
and can be used as third parameter (option=dcache option to select)
of mmu_set_region_dcache_behaviour function.

Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
3 years agoarm: caches: protect dram_bank_mmu_setup access to bi_dram
Patrick Delaunay [Fri, 24 Apr 2020 18:20:15 +0000 (20:20 +0200)]
arm: caches: protect dram_bank_mmu_setup access to bi_dram

Add protection in dram_bank_mmu_setup() to avoid access to bd->bi_dram
before relocation.

This patch allow to use the generic weak function dram_bank_mmu_setup
to activate the MMU and the data cache in SPL or in U-Boot before
relocation, when bd->bi_dram is not yet initialized.

In this cases, the MMU must be initialized explicitly with
mmu_set_region_dcache_behaviour function.

Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
3 years agoconfigs: migrate CONFIG_SYS_ARM_CACHE_* in Kconfig
Patrick Delaunay [Fri, 10 Apr 2020 14:02:02 +0000 (16:02 +0200)]
configs: migrate CONFIG_SYS_ARM_CACHE_* in Kconfig

Move CONFIG_SYS_ARM_CACHE_WRITETHROUGH and
CONFIG_SYS_ARM_CACHE_WRITEALLOC into Kconfig done by moveconfig.py.

Kconfig uses a choice between the 3 values supported in U-Boot,
including the new configuration CONFIG_SYS_ARM_CACHE_WRITEBACK
(the default configuration).

The patch also avoids to select simultaneously 2 configurations.

Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
3 years agotiny-printf: Support %i
Marek Vasut [Fri, 10 Apr 2020 18:54:49 +0000 (20:54 +0200)]
tiny-printf: Support %i

The most basic printf("%i", value) formating string was missing,
add it for the sake of convenience.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Stefan Roese <sr@denx.de>
3 years agonet: smc911x: Add DM support
Marek Vasut [Sun, 15 Mar 2020 16:39:01 +0000 (17:39 +0100)]
net: smc911x: Add DM support

Add support for U-Boot DM and DT probing. Furthermore, build the
SMC911x standalone EEPROM example only for the non-DM case, as it
is not converted yet.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
3 years agoarm: imx: Add support for Google's Coral Dev Board
Alifer Moraes [Fri, 6 Mar 2020 10:46:33 +0000 (07:46 -0300)]
arm: imx: Add support for Google's Coral Dev Board

Add initial support for Google's Coral Dev Board based on i.MX8MQ.

https://coral.ai/products/dev-board

The Phanbell naming has been used here to match the naming convention
used in Google's U-Boot source tree:

https://coral.googlesource.com/uboot-imx/

Co-developed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Alifer Moraes <alifer.wsdm@gmail.com>
Tested-by: Marco Franchi <marcofrk@gmail.com>
3 years agoARM: imx8m: Fix reset in SPL on Toradex iMX8MM Verdin
Marek Vasut [Wed, 29 Apr 2020 13:04:27 +0000 (15:04 +0200)]
ARM: imx8m: Fix reset in SPL on Toradex iMX8MM Verdin

Board files should not re-implement do_reset() to work around this
function not being defined in for specific configurations. Rather,
the fix is to compile in drivers which implement this properly.
This patch enables sysreset and watchdog drivers in SPL and ties
them together to implement the same as the do_reset() hack in the
board file, except correctly in the DM/DT framework.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Flavio Suligoi <f.suligoi@asem.it>
Cc: Harald Seiler <hws@denx.de>
Cc: Igor Opaniuk <igor.opaniuk@toradex.com>
Cc: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Cc: Oleksandr Suvorov <oleksandr.suvorov@toradex.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Stefano Babic <sbabic@denx.de>
Reviewed-by: Oleksandr Suvorov <oleksandr.suvorov@toradex.com>
Acked-by: Igor Opaniuk <igor.opaniuk@toradex.com>
Acked-by: Igor Opaniuk <igor.opaniuk@toradex.com>
3 years agoARM: imx8m: Fix reset in SPL on NXP iMX8MP EVK
Marek Vasut [Wed, 29 Apr 2020 13:04:26 +0000 (15:04 +0200)]
ARM: imx8m: Fix reset in SPL on NXP iMX8MP EVK

Board files should not re-implement do_reset() to work around this
function not being defined in for specific configurations. Rather,
the fix is to compile in drivers which implement this properly.
This patch enables sysreset and watchdog drivers in SPL and ties
them together to implement the same as the do_reset() hack in the
board file, except correctly in the DM/DT framework.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Flavio Suligoi <f.suligoi@asem.it>
Cc: Harald Seiler <hws@denx.de>
Cc: Igor Opaniuk <igor.opaniuk@toradex.com>
Cc: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Cc: Oleksandr Suvorov <oleksandr.suvorov@toradex.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Stefano Babic <sbabic@denx.de>
3 years agoARM: imx8m: Fix reset in SPL on NXP iMX8MN EVK
Marek Vasut [Wed, 29 Apr 2020 13:04:25 +0000 (15:04 +0200)]
ARM: imx8m: Fix reset in SPL on NXP iMX8MN EVK

Board files should not re-implement do_reset() to work around this
function not being defined in for specific configurations. Rather,
the fix is to compile in drivers which implement this properly.
This patch enables sysreset and watchdog drivers in SPL and ties
them together to implement the same as the do_reset() hack in the
board file, except correctly in the DM/DT framework.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Flavio Suligoi <f.suligoi@asem.it>
Cc: Harald Seiler <hws@denx.de>
Cc: Igor Opaniuk <igor.opaniuk@toradex.com>
Cc: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Cc: Oleksandr Suvorov <oleksandr.suvorov@toradex.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Stefano Babic <sbabic@denx.de>
3 years agoARM: imx8m: Fix reset in SPL on NXP iMX8MM EVK
Marek Vasut [Wed, 29 Apr 2020 13:04:24 +0000 (15:04 +0200)]
ARM: imx8m: Fix reset in SPL on NXP iMX8MM EVK

Board files should not re-implement do_reset() to work around this
function not being defined in for specific configurations. Rather,
the fix is to compile in drivers which implement this properly.
This patch enables sysreset and watchdog drivers in SPL and ties
them together to implement the same as the do_reset() hack in the
board file, except correctly in the DM/DT framework.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Flavio Suligoi <f.suligoi@asem.it>
Cc: Harald Seiler <hws@denx.de>
Cc: Igor Opaniuk <igor.opaniuk@toradex.com>
Cc: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Cc: Oleksandr Suvorov <oleksandr.suvorov@toradex.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Stefano Babic <sbabic@denx.de>
3 years agoARM: imx8m: Don't use the addr parameter of reset_cpu()
Claudius Heine [Wed, 29 Apr 2020 13:04:23 +0000 (15:04 +0200)]
ARM: imx8m: Don't use the addr parameter of reset_cpu()

imx8m has the only implementation of reset_cpu() which does not ignore
the addr parameter and instead gives it some meaning as the base address
of watchdog registers.  This breaks convention with the rest of U-Boot
where the parameter is ignored and callers are passing in 0.

Fixes: d2041725e84b ("imx8m: restrict reset_cpu")
Co-developed-by: Harald Seiler <hws@denx.de>
Signed-off-by: Harald Seiler <hws@denx.de>
Signed-off-by: Claudius Heine <ch@denx.de>
3 years agoARM: imx8m: Fix indentation of reset_cpu() function
Harald Seiler [Wed, 29 Apr 2020 13:04:22 +0000 (15:04 +0200)]
ARM: imx8m: Fix indentation of reset_cpu() function

Use proper code-style, tabs instead of spaces for indentation.

Signed-off-by: Harald Seiler <hws@denx.de>
3 years agoARM: imx8m: Do not define do_reset() if sysreset is enabled
Marek Vasut [Wed, 29 Apr 2020 13:04:21 +0000 (15:04 +0200)]
ARM: imx8m: Do not define do_reset() if sysreset is enabled

The SPL can also be compiled with sysreset drivers just fine, so
update the condition to cater for that option.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Flavio Suligoi <f.suligoi@asem.it>
Cc: Harald Seiler <hws@denx.de>
Cc: Igor Opaniuk <igor.opaniuk@toradex.com>
Cc: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Cc: Oleksandr Suvorov <oleksandr.suvorov@toradex.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Stefano Babic <sbabic@denx.de>
3 years agodt-bindings: pinctrl: imxrt1020: remove useless comment
Giulio Benetti [Mon, 20 Apr 2020 15:48:12 +0000 (17:48 +0200)]
dt-bindings: pinctrl: imxrt1020: remove useless comment

A comment note has been left after completing pinctrl listing, so let's
remove it since it's useless.

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
3 years agoARM: imx: imx8m: Do not warn about cpu-idle-states if missing
Marek Vasut [Fri, 24 Apr 2020 19:37:33 +0000 (21:37 +0200)]
ARM: imx: imx8m: Do not warn about cpu-idle-states if missing

If the cpu-idle-states is missing from the DT in the first place, do
not fail on removing in. Just move on and do not even print an error,
since not being able to remove something which is not there in the
first place is not an error and surely does not justify failing to
boot.

Turn the surrounding prints into debugs to reduce the useless noise.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Frieder Schrempf <frieder.schrempf@kontron.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Stefano Babic <sbabic@denx.de>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
3 years agoarm: dts: imx8mm: sync dts from Linux Kernel 5.6.7
Adam Ford [Mon, 27 Apr 2020 12:11:19 +0000 (07:11 -0500)]
arm: dts: imx8mm: sync dts from Linux Kernel 5.6.7

This patch synchronizes the device tree with that from 5.6.7.

This also adds nodes for crypto and ddrc, which makes keeping
the device tree files from individual boards in sync with the
Linux kernel.  This is helpful when boads reference those nodes.

Signed-off-by: Adam Ford <aford173@gmail.com>
3 years agotreewide: Remove unused FSL QSPI config options for IMX platforms
Kuldeep Singh [Thu, 20 Feb 2020 17:27:54 +0000 (22:57 +0530)]
treewide: Remove unused FSL QSPI config options for IMX platforms

Some of these options are not used by the driver anymore and some of them
are obsolete as the information is gathered from the dt.
So, remove the unused config options now.

Signed-off-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Signed-off-by: Ashish Kumar <Ashish.Kumar@nxp.com>
Signed-off-by: Kuldeep Singh <kuldeep.singh@nxp.com>
3 years agoimx8qxp_mek: Run with caches enabled
Fabio Estevam [Mon, 20 Apr 2020 18:55:29 +0000 (15:55 -0300)]
imx8qxp_mek: Run with caches enabled

There is no need to run with caches disabled.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8qxp_mek: Remove SPL watchdog option
Fabio Estevam [Mon, 20 Apr 2020 12:55:47 +0000 (09:55 -0300)]
imx8qxp_mek: Remove SPL watchdog option

Currently the following watchdog error is seen in SPL:

U-Boot SPL 2020.04-00407-g8d5d3bcf3c (Apr 20 2020 - 09:48:09 -0300)
Normal Boot
WDT:   Not found!
...

There is no watchdog driver for i.MX8 at the moment, nor code for
configuring the watchdog in SPL, so remove the CONFIG_SPL_WATCHDOG_SUPPORT
option for now.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Stefano Babic <sbabic@denx.de>
3 years agoimx8: Run SPL with caches enabled
Fabio Estevam [Tue, 21 Apr 2020 15:02:45 +0000 (12:02 -0300)]
imx8: Run SPL with caches enabled

It is safe to run SPL with caches enabled on i.MX8, so
remove such restriction.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Acked-by: Peng Fan <peng.fan@nxp.com>
Acked-by: Oliver Graute <oliver.graute@kococonnector.com>
Acked-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Igor Opaniuk <igor.opaniuk@toradex.com>
Acked-by: Igor Opaniuk <igor.opaniuk@toradex.com>
3 years agoconfig: apalis-imx8: Enable FEC TXC delay
Philippe Schenker [Wed, 11 Mar 2020 10:52:59 +0000 (11:52 +0100)]
config: apalis-imx8: Enable FEC TXC delay

define FEC_ENET_ENABLE_TXC_DELAY to enable the delay on TXC line on
the MAC. This has to be done in order to meet RGMII specs.
According to RGMII specs the clock should get delayed so the edges of
the clock are preferrably in the middle of the edges of data-lines so
they can be sampled properly.
Our PHY expects the MAC to delay TXC line, as it is also written
in the spec. This patch makes sure the TXC delay on the FEC is enabled

Signed-off-by: Philippe Schenker <philippe.schenker@toradex.com>
Reviewed-by: Oleksandr Suvorov <oleksandr.suvorov@toradex.com>
3 years agonet: fec: Add possibility to enable TXC delay
Philippe Schenker [Wed, 11 Mar 2020 10:52:58 +0000 (11:52 +0100)]
net: fec: Add possibility to enable TXC delay

This patch enables the possibility to set FEC_ENET_ENABLE_TXC_DELAY or
FEC_ENET_ENABLE_RXC_DELAY so one can via a define enable the RXC or TXC
delay in the MAC.

Signed-off-by: Philippe Schenker <philippe.schenker@toradex.com>
Reviewed-by: Oleksandr Suvorov <oleksandr.suvorov@toradex.com>
3 years agoboard: apalis_imx6: Add KSZ9131 phy skew settings
Philippe Schenker [Wed, 11 Mar 2020 10:59:26 +0000 (11:59 +0100)]
board: apalis_imx6: Add KSZ9131 phy skew settings

This patch adds skew register settings for KSZ9131. It checks first
which phy is on the board and then applies the correct skew settings.

Skew settings calculation for the KSZ9131:
The i.MX6 SoC has an output skew tolerance of -100ps to 900ps. All
PCB traces where routed exactly the same length so we can calculate
the skew settings without taking the length into consideration. The
traces are all length matched.

RXC skew (PHY to MAC):
- We use the 2ns DLL controlled delay on the PHY
- We do not use the skew registers

This results in the following values:

RXC
PHY fixed Delay 2000ps
PHY Added Delay 0ps
T_setup_R min 2.00ns
T_setup_R typ 2.00ns
T_setup_R max 2.00ns
T_hold_R min 1.60ns
T_hold_R typ 2.00ns
T_hold_R max 2.40ns

That means we are well within RGMII specs.

TXC skew (MAC to PHY):
- We use the 2ns DLL controlled delay on the PHY
- We then subtract ~0.6ns with TXD[0:3] and TXC clock pad skew
  register in a resulting ~1.4ns delay.

This results in the following values under consideration of the
tolerances:

TXC min TXC typ TXC max
MAC min -100ps -100ps -100ps
MAC max 900ps 900ps 900ps
PHY fixed Delay 2000ps 2000ps 2000ps
PHY added Delay -340ps -600ps -859ps
T_setup_T min 1.56ns 1.30ns 1.04ns
T_setup_T typ 2.06ns 1.80ns 1.54ns
T_setup_T max 2.56ns 2.30ns 2.04ns
T_hold_T min 1.04ns 1.30ns 1.56ns
T_hold_T typ 1.94ns 2.20ns 2.46ns
T_hold_T max 2.84ns 3.10ns 3.36ns

This shows that T_hold_T min and T_setup_T min times are out of spec
for RGMII timing. However the KSZ9131 has a minimal value for this time
of 0.8ns which is met under all circumstances.

Signed-off-by: Philippe Schenker <philippe.schenker@toradex.com>
3 years agoboard: verdin-imx8mm: Add KSZ9131 phy skew settings
Philippe Schenker [Wed, 11 Mar 2020 10:59:25 +0000 (11:59 +0100)]
board: verdin-imx8mm: Add KSZ9131 phy skew settings

This patch determines which phy is placed on the board with the PHY ID
then it sets the same settings for KSZ9031 as before but for KSZ9131
it enables both RXC and TXC delay lines in the PHY.
This will compensate the missing delay from the MAC. Other skew
settings are not needed as the traces on board are routed exactly the
same length

Signed-off-by: Philippe Schenker <philippe.schenker@toradex.com>
Tested-by: Igor Opaniuk <igor.opaniuk@toradex.com>
Acked-by: Igor Opaniuk <igor.opaniuk@toradex.com>
3 years agoverdin-imx8mm: Change board phy skew values for our ksz9031
Philippe Schenker [Wed, 11 Mar 2020 10:59:24 +0000 (11:59 +0100)]
verdin-imx8mm: Change board phy skew values for our ksz9031

This patches uses the existing functions for interacting with the
KSZ9031 and uses the values appropriate for our board.

Signed-off-by: Philippe Schenker <philippe.schenker@toradex.com>
Tested-by: Igor Opaniuk <igor.opaniuk@toradex.com>
Acked-by: Igor Opaniuk <igor.opaniuk@toradex.com>
3 years agonet: phy: micrel: Add basic support for KSZ9131
Philippe Schenker [Wed, 11 Mar 2020 10:59:23 +0000 (11:59 +0100)]
net: phy: micrel: Add basic support for KSZ9131

This adds basic support for the new Micrel KSZ9131 phy.

Signed-off-by: Philippe Schenker <philippe.schenker@toradex.com>
3 years agonet: phy: micrel: Use defines for PHY_IDs and MASK
Philippe Schenker [Wed, 11 Mar 2020 10:59:22 +0000 (11:59 +0100)]
net: phy: micrel: Use defines for PHY_IDs and MASK

Signed-off-by: Philippe Schenker <philippe.schenker@toradex.com>
3 years agomx8mq_evk: Remove unrelated comment
Fabio Estevam [Mon, 20 Apr 2020 15:38:51 +0000 (12:38 -0300)]
mx8mq_evk: Remove unrelated comment

The comment does not relate to the setup_i2c() function, so just
remove it.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
3 years agoimx8mp_evk: Remove unrelated comment
Fabio Estevam [Mon, 20 Apr 2020 15:38:50 +0000 (12:38 -0300)]
imx8mp_evk: Remove unrelated comment

The comment does not relate to the setup_i2c() function, so just
remove it.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
3 years agoimx: imx8m: Don't use the addr parameter of reset_cpu
Claudius Heine [Wed, 4 Mar 2020 14:23:12 +0000 (15:23 +0100)]
imx: imx8m: Don't use the addr parameter of reset_cpu

imx8m has the only implementation of `reset_cpu` which does not ignore
the addr parameter and instead gives it some meaning as the base address
of watchdog registers.  This breaks convention with the rest of U-Boot
where the parameter is ignored and callers are passing in 0.

Fixes: d2041725e84b ("imx8m: restrict reset_cpu")
Co-Authored-by: Harald Seiler <hws@denx.de>
Signed-off-by: Claudius Heine <ch@denx.de>
Signed-off-by: Harald Seiler <hws@denx.de>
Reviewed-by: Marek Vasut <marex@denx.de>
3 years agoimx: imx8m*: Remove do_reset from board files
Claudius Heine [Wed, 4 Mar 2020 14:23:11 +0000 (15:23 +0100)]
imx: imx8m*: Remove do_reset from board files

Use the `do_reset` implementation of `arch/arm/lib/reset.c` in SPL
instead.  It is very close to what is done here, anyway, and plays
more nicely with the rest of U-Boot than adding a custom `do_reset`
implementation into board files.

`do_reset` from `arch/arm/lib/reset.c` calls `reset_cpu` with 0 as the
addr parameter while the boards are passing WDOG1_BASE_ADDR.  This is
ok because the `reset_cpu` implementation uses WDOG1_BASE_ADDR by
default if 0 is passed in.

Co-Authored-by: Harald Seiler <hws@denx.de>
Signed-off-by: Claudius Heine <ch@denx.de>
Signed-off-by: Harald Seiler <hws@denx.de>
Reviewed-by: Marek Vasut <marex@denx.de>
3 years agoARM: reset: use do_reset in SPL/TPL if SYSRESET was not enabled for them
Claudius Heine [Wed, 4 Mar 2020 14:23:10 +0000 (15:23 +0100)]
ARM: reset: use do_reset in SPL/TPL if SYSRESET was not enabled for them

In case CONFIG_SYSRESET is set, do_reset from reset.c will not be available
anywere, even if SYSRESET is disabled for SPL/TPL.

'do_reset' is called from SPL for instance from the panic handler and
PANIC_HANG is not set

Signed-off-by: Claudius Heine <ch@denx.de>
Reviewed-by: Marek Vasut <marex@denx.de>
3 years agoimx8: Configure SNVS
Franck LENORMAND [Wed, 9 Oct 2019 08:27:43 +0000 (10:27 +0200)]
imx8: Configure SNVS

Add a module to configure the tamper and secure violation of
the SNVS using the SCU API.

The module also adds some commands:
 - snvs_cfg: Configure the SNVS HP and LP registers
 - snvs_dgo_cfg: Configure the SNVS DGO bloc if present (8QXP)
 - tamper_pin_cfg: Change the configuration of the tamper pins
 - snvs_clear_status: Allow to write to LPSR and LPTDSR to clear
   status bits

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Franck LENORMAND <franck.lenormand@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8: Update SCFW API to version 1.5
Ye Li [Thu, 14 Nov 2019 05:20:43 +0000 (21:20 -0800)]
imx8: Update SCFW API to version 1.5

Sync the latest SCFW API with below commit
6dcd0242ae7a53ac ("SCF-105: Revert accidental change") to add
interfaces for PM resource reset and read/write SNVS security
violation and tamper DGO registers.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8: scu api: Add support for SECO manufacturing protection APIs
Breno Lima [Wed, 16 Oct 2019 21:10:54 +0000 (18:10 -0300)]
imx8: scu api: Add support for SECO manufacturing protection APIs

SECO provides APIs to support CAAM manufacturing protection:

- sc_seco_get_mp_key()
- sc_seco_get_mp_sign()
- sc_seco_update_mpmr()

Add SCFW APIs support.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Breno Lima <breno.lima@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agodoc: ahab: Add encrypted boot documentation for i.MX8/8x devices
Breno Lima [Mon, 11 Nov 2019 21:54:09 +0000 (18:54 -0300)]
doc: ahab: Add encrypted boot documentation for i.MX8/8x devices

Add AHAB encrypted boot documentation for i.MX8/8x family devices
covering the following topics:

- How to encrypt and sign the 2nd container in flash.bin image.
- How to encrypt and sign a standalone container image.

Include a CSF example to encrypt 2nd container in flash.bin image.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Catia Han <yaqian.han@nxp.com>
Signed-off-by: Breno Lima <breno.lima@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8m: Enable WDOG_B for timeout
Ye Li [Mon, 9 Dec 2019 08:47:18 +0000 (00:47 -0800)]
imx8m: Enable WDOG_B for timeout

When doing reset_cpu, in normal case the WDOG_B outputs immediately
after we clean WDA bit. But on mscale, the WDOG_B may be later than
internal reset, and cause PMIC not reset. As we enabled the SD3.0
support, the PMIC must be reset to reset SD card.

Change the reset_cpu to enable the WDOG_B for timeout as well, and set
WDOG timeout to 1s.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Acked-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8mm: clock: fix fracpll decode issue
Ye Li [Tue, 24 Mar 2020 02:54:29 +0000 (19:54 -0700)]
imx8mm: clock: fix fracpll decode issue

The fracpll decoding is using the bit definitions for int pll. Most of
them are same, but the CLKE bit is different. Fix the wrong CLKE_MASK
for fracpll and correct all bit definitions in fracpll decoding.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8: parser: fix some bad debug message formating
Peng Fan [Wed, 22 Apr 2020 07:45:42 +0000 (15:45 +0800)]
imx8: parser: fix some bad debug message formating

In SPL build, the formatting '%llx' in debug() is not supported.
Also, fix some misplaced parameters in printf.

Modified from Seb Fagard's downstream patch

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8: ahab: fix some bad debug message formating
Seb Fagard [Mon, 30 Mar 2020 15:13:31 +0000 (17:13 +0200)]
imx8: ahab: fix some bad debug message formating

In SPL build, the formatting '%llx' in debug() is not supported.
Also, fix some misplaced parameters in printf.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Seb Fagard <sebastien.fagard@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8: ahab: fix 'end address' parameter of rm_find_memreg
Peng Fan [Wed, 22 Apr 2020 07:36:34 +0000 (15:36 +0800)]
imx8: ahab: fix 'end address' parameter of rm_find_memreg

parameter 'end address' must be inclusive of address range.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8: Change to use new SECO API commands
Ye Li [Tue, 15 Oct 2019 06:08:46 +0000 (23:08 -0700)]
imx8: Change to use new SECO API commands

Latest SCFW has removed old MISC SECO commands. So update the codes
to use new SECO commands.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8: parser: fix 'end address' parameter of rm_find_memreg
Peng Fan [Wed, 22 Apr 2020 07:25:31 +0000 (15:25 +0800)]
imx8: parser: fix 'end address' parameter of rm_find_memreg

parameter 'end address' must be inclusive of address range.

Modified from Seb's downstream patch.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8m: Dump DRAM PLL rate by clocks command
Ye Li [Tue, 24 Mar 2020 02:57:55 +0000 (19:57 -0700)]
imx8m: Dump DRAM PLL rate by clocks command

Add the dump of DRAM PLL into "clocks" command

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
3 years agoimx8mq: Set ARM core clock directly from ARM PLL
Peng Fan [Wed, 22 Apr 2020 03:18:23 +0000 (11:18 +0800)]
imx8mq: Set ARM core clock directly from ARM PLL

For ARM core clock, there are two input branches, and can select via mux:
one from ARM PLL directly, second from CCM A53 clock root.

Currently we are using second branch. But IC confirmed the CCM A53 root
signoff timing is 1Ghz, so we should switch to input from ARM PLL directly.

This patch fixes the CORE SEL slice configuration and switch ARM clock
to ARM PLL.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8mq: Enable eMMC HS400 and SD UHS mode on EVK
Ye Li [Fri, 25 Oct 2019 08:28:48 +0000 (01:28 -0700)]
imx8mq: Enable eMMC HS400 and SD UHS mode on EVK

iMX8MQ EVK board has a eMMC5.0 chip and supports SD3.0, so enable the UHS
and HS400 configs to enhance the eMMC/SD access.

The change also needs to set usdhc clock to 400Mhz, and add the
off-on-delay-us to SD reset pin, otherwise some SD cards will
fail to select UHS mode in re-initialization.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Ye Li <ye.li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8m: update clock root and fix core_sel
Peng Fan [Wed, 22 Apr 2020 02:55:56 +0000 (10:55 +0800)]
imx8m: update clock root and fix core_sel

Update clock root table to let it be easy to configure clock at
very early stage. Also the core_sel mux parent should be A53 CLK
root and ARM PLL.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8m: acquire ATF commit hash
Peng Fan [Wed, 22 Apr 2020 02:51:13 +0000 (10:51 +0800)]
imx8m: acquire ATF commit hash

Acquire ATF commit hash when booting U-Boot to make user easy
to know the ATF version.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8: move SIP macro to common header
Peng Fan [Wed, 22 Apr 2020 02:50:04 +0000 (10:50 +0800)]
imx8: move SIP macro to common header

Move the SIP macro to common header and unify the name to
make others could reuse them.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx: update is_imx6ull to include i.MX6ULZ
Peng Fan [Wed, 5 Feb 2020 12:17:17 +0000 (20:17 +0800)]
imx: update is_imx6ull to include i.MX6ULZ

Update is_imx6ull helper to include i.MX6ULZ SoC. i.MX6ULZ could
share same macro, then we no need to add is_imx6ulz in various drivers.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx: imx8m: add i.MX8MN variants support
Peng Fan [Wed, 5 Feb 2020 09:39:27 +0000 (17:39 +0800)]
imx: imx8m: add i.MX8MN variants support

Add i.MX8MN variants support

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx: imx8m: add i.MX8MQ Dual and QuadLite support
Peng Fan [Wed, 5 Feb 2020 09:34:54 +0000 (17:34 +0800)]
imx: imx8m: add i.MX8MQ Dual and QuadLite support

Add i.MX8MQ Dual and QuadLite variants.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx: cpu: support speed grade for i.MX8MP
Peng Fan [Fri, 17 Jan 2020 08:11:29 +0000 (16:11 +0800)]
imx: cpu: support speed grade for i.MX8MP

i.MX8MP speed grade use same layout as i.MX8MN, so reuse it for i.MX8MP

Reviewed-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx8mn: Update speed grade
Peng Fan [Wed, 12 Dec 2018 10:47:58 +0000 (02:47 -0800)]
imx8mn: Update speed grade

imx8mn speed grade fuse uses new definitions as below. So have
to update get_cpu_speed_grade_hz function to match it.

SPEED_GRADE[5:4]    SPEED_GRADE[3:0]    MHz
      xx                0000            2300
      xx                0001            2200
      xx                0010            2100
      xx                0011            2000
      xx                0100            1900
      xx                0101            1800
      xx                0110            1700
      xx                0111            1600
      xx                1000            1500
      xx                1001            1400
      xx                1010            1300
      xx                1011            1200
      xx                1100            1100
      xx                1101            1000
      xx                1110             900
      xx                1111             800

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
3 years agoimx8mm: Update CPU speed grading
Ye Li [Wed, 17 Oct 2018 06:12:37 +0000 (23:12 -0700)]
imx8mm: Update CPU speed grading

According to iMX8MM datasheet (IMX8MMIEC_Rev_D and IMX8MMCEC_Rev_D),
the speed grading for imx8mm is 800Mhz, 1.2Ghz, 1.6Ghz and 1.8Ghz.
Update them to get_cpu_speed_grade_hz function.

Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
3 years agoimx: fix cpu_type helper
Peng Fan [Wed, 5 Feb 2020 09:44:28 +0000 (17:44 +0800)]
imx: fix cpu_type helper

i.MX8MP use 0x182 as its ID, so 0xff is not valid to get the cpu type,
extend it to 0x1ff.

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
3 years agoARM: imx: imx8mm: Add missing clock entries for FEC clock
Marek Vasut [Fri, 24 Apr 2020 19:37:26 +0000 (21:37 +0200)]
ARM: imx: imx8mm: Add missing clock entries for FEC clock

All the FEC ethernet clock entries for iMX8MM are missing, while they
are already present on iMX8MQ. Fill in the nodes on iMX8MM, as the FEC
ethernet gets bogus clock information otherwise which makes ethernet
inoperable.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Stefano Babic <sbabic@denx.de>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
3 years agonet: smc911x: Split non-DM specific bits from common code
Marek Vasut [Sun, 15 Mar 2020 16:25:27 +0000 (17:25 +0100)]
net: smc911x: Split non-DM specific bits from common code

Split network handling functions into non-DM specific parts and
common code in preparation for conversion to DM.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
3 years agonet: smc911x: Clean up the status handling in smc911x_recv()
Marek Vasut [Sun, 15 Mar 2020 16:02:05 +0000 (17:02 +0100)]
net: smc911x: Clean up the status handling in smc911x_recv()

Invert the status handling logic in smc911x_recv(), to make the
function easier to read, no functional change.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
3 years agonet: smc911x: Pass around driver private data
Marek Vasut [Sun, 15 Mar 2020 14:36:09 +0000 (15:36 +0100)]
net: smc911x: Pass around driver private data

Introduce a private data structure for this driver with embedded
struct eth_device and pass it around. This prepares the driver to
work with both DM and non-DM systems.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
3 years agonet: smc911x: Convert IO accessors to {read,write}{w,l}()
Marek Vasut [Sun, 15 Mar 2020 13:42:23 +0000 (14:42 +0100)]
net: smc911x: Convert IO accessors to {read,write}{w,l}()

Convert the IO accessors to standard ones instead of using volatile
void pointers, as those do not cover all the bus access details.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
3 years agonet: smc911x: Drop weak alias from 32bit accessors
Marek Vasut [Sun, 15 Mar 2020 13:38:59 +0000 (14:38 +0100)]
net: smc911x: Drop weak alias from 32bit accessors

These accessors are not overridden by any board, and even if they were,
this is something which should be handled via DM now, so remove the
weak alias option. Moreover, drop the inline keyword, as the compiler
can decide better.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
3 years agonet: smc911x: Inline all functions from header file
Marek Vasut [Sun, 15 Mar 2020 14:03:07 +0000 (15:03 +0100)]
net: smc911x: Inline all functions from header file

Inline all the functions from the header file, as they are not used
outside of the driver or the standalone EEPROM example.

Note that this does introduce considerable amount of duplication in
the standalone EEPROM example, however that one has to be rewritten
anyway, roughly such that the SMC911x driver would expose DM EEPROM
interface and the standalone example would use that.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
3 years agonet: smc911x: Pull MII registration into separate function
Marek Vasut [Sat, 21 Mar 2020 16:25:41 +0000 (17:25 +0100)]
net: smc911x: Pull MII registration into separate function

Pull the MII interface registration into separate function to avoid the
ifdeffery in smc911x_initialize(). Moreover, adjust the fail path such
that we use goto labels.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
3 years agonet: smc911x: Fix potential memleak() in init fail path
Marek Vasut [Sun, 15 Mar 2020 14:57:14 +0000 (15:57 +0100)]
net: smc911x: Fix potential memleak() in init fail path

Fix memleak in the init fail path, where if allocation or registration
of MDIO bus fails, then ethernet interface is not unregistered and the
private data are not freed, yet the probe function reports a failure.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
3 years agonet: smc911x: Invert the logic in smc911x_miiphy_{read,write}()
Marek Vasut [Sun, 15 Mar 2020 14:43:20 +0000 (15:43 +0100)]
net: smc911x: Invert the logic in smc911x_miiphy_{read,write}()

Invert the logic in the aforementioned functions to reduce indent,
no functional change.

Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>