project/bcm63xx/u-boot.git
7 years agopowerpc: mpc85xx: Implement CPU erratum A-007907 for secondary cores
York Sun [Tue, 17 Oct 2017 15:00:21 +0000 (08:00 -0700)]
powerpc: mpc85xx: Implement CPU erratum A-007907 for secondary cores

Commit 06ad970b53a3 ("powerpc: mpc85xx: Implemente workaround for CPU
erratum A-007907") clears L1CSR2 for the boot core, but other cores
don't run through the workaround. Add similar code for secondary
cores to clear DCSTASHID field in L1CSR2 register.

Signed-off-by: York Sun <york.sun@nxp.com>
7 years agoMerge git://git.denx.de/u-boot-uniphier
Tom Rini [Sun, 22 Oct 2017 23:21:04 +0000 (19:21 -0400)]
Merge git://git.denx.de/u-boot-uniphier

7 years agoARM: uniphier: use pr_*() more where appropriate
Masahiro Yamada [Sun, 22 Oct 2017 15:19:36 +0000 (00:19 +0900)]
ARM: uniphier: use pr_*() more where appropriate

Commit dd74b945af2e ("ARM: uniphier: use pr_() instead of printf()
where appropriate"), but I missed to update this file for some reason.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agodoc: uniphier: add simple guide to Verified Boot
Masahiro Yamada [Sun, 22 Oct 2017 06:15:01 +0000 (15:15 +0900)]
doc: uniphier: add simple guide to Verified Boot

Add a simple documentation about how to use the Verified Boot on
UniPhier boards.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: uniphier: remove verify=n from environments
Masahiro Yamada [Sat, 21 Oct 2017 10:51:13 +0000 (19:51 +0900)]
ARM: uniphier: remove verify=n from environments

If the environment "verify" is set to n, the image verification
is entirely skipped.  Remove it as a preparation for verified boot.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: uniphier: increase CONFIG_SYS_BOOTM_LEN to 32MB
Masahiro Yamada [Thu, 19 Oct 2017 10:17:51 +0000 (19:17 +0900)]
ARM: uniphier: increase CONFIG_SYS_BOOTM_LEN to 32MB

The default value of CONFIG_SYS_BOOTM_LEN, 0x800000, causes error
when uncompressing Image.gz out of FIT image.

   Uncompressing Kernel Image ... Error: inflate() returned -5
Image too large: increase CONFIG_SYS_BOOTM_LEN

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: dts: uniphier: sync DT with Linux 4.14-rc5
Masahiro Yamada [Tue, 17 Oct 2017 12:19:43 +0000 (21:19 +0900)]
ARM: dts: uniphier: sync DT with Linux 4.14-rc5

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: uniphier: split u-boot,dm-pre-reloc out to uniphier-v7-u-boot.dtsi
Masahiro Yamada [Tue, 17 Oct 2017 12:19:42 +0000 (21:19 +0900)]
ARM: uniphier: split u-boot,dm-pre-reloc out to uniphier-v7-u-boot.dtsi

UniPhier 32-bit SoCs use CONFIG_SPL_OF_CONTROL.  So, many nodes must
be marked as dm-pre-reloc to prevent fdtgrep from stripping them off.

Sprinkling U-Boot-specific properties all over the place is painful
because DT files are synced with Linux from time to time.

Split u-boot,dm-pre-reloc out to uniphier-v7-u-boot.dtsi, which is
appended to UniPhier V7 DTS before the build.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Jagan Teki <jagan@openedev.com>
7 years agoARM: uniphier: remove CONFIG_UNIPHIER_ETH
Masahiro Yamada [Tue, 17 Oct 2017 12:19:41 +0000 (21:19 +0900)]
ARM: uniphier: remove CONFIG_UNIPHIER_ETH

The option is never enabled by anyone.  Remove the code surrounded
by its ifdef.  This should be handled by the clock/reset drivers.

CONFIG_UNIPHIER_ETH in scripts/config_whitelist.txt will be dropped
by the next resync.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: uniphier: enable DWC3 xHCI driver really
Masahiro Yamada [Tue, 17 Oct 2017 12:19:40 +0000 (21:19 +0900)]
ARM: uniphier: enable DWC3 xHCI driver really

I thought commit d37d31849c6a ("ARM: uniphier: enable DWC3 xHCI
driver") enabled CONFIG_USB_DWC3_UNIPHIER, but CONFIG_USB_XHCI_DWC3
was missing in uniphier_v7_defconfig.  Re-add.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agouniphier_ld4_sld8: Re-add SMC911X_BASE address
Tom Rini [Sat, 21 Oct 2017 14:26:34 +0000 (10:26 -0400)]
uniphier_ld4_sld8: Re-add SMC911X_BASE address

This was dropped by accident in the Kconfig conversion.

Signed-off-by: Tom Rini <trini@konsulko.com>
7 years agonet: Add SMC911X driver to Kconfig, convert
Adam Ford [Tue, 5 Sep 2017 20:20:44 +0000 (15:20 -0500)]
net: Add SMC911X driver to Kconfig, convert

We add the various SMC91XX symbols to drivers/net/Kconfig and then this
converts the following to Kconfig:
   CONFIG_SMC911X
   CONFIG_SMC911X_BASE
   CONFIG_SMC911X_16_BIT
   CONFIG_SMC911X_32_BIT

Signed-off-by: Adam Ford <aford173@gmail.com>
[trini: Apply to the rest of the tree, re-squash old and new patch]
Signed-off-by: Tom Rini <trini@konsulko.com>
7 years agoConvert CONFIG_NAND_OMAP_GPMC et al and CONFIG_NAND_MXC to Kconfig
Adam Ford [Mon, 16 Oct 2017 19:08:26 +0000 (14:08 -0500)]
Convert CONFIG_NAND_OMAP_GPMC et al and CONFIG_NAND_MXC to Kconfig

This converts the following to Kconfig:
   CONFIG_NAND_MXC
   CONFIG_NAND_OMAP_GPMC
   CONFIG_NAND_OMAP_GPMC_PREFETCH
   CONFIG_NAND_OMAP_ELM
   CONFIG_SPL_NAND_AM33XX_BCH
   CONFIG_SPL_NAND_SIMPLE
   CONFIG_SYS_NAND_BUSWIDTH_16BIT

Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
[trini: Finish migration of CONFIG_SPL_NAND_SIMPLE, fix some build issues,
        add CONFIG_NAND_MXC so we can do CONFIG_SYS_NAND_BUSWIDTH_16BIT]
Signed-off-by: Tom Rini <trini@konsulko.com>
7 years agoMerge git://git.denx.de/u-boot-x86
Tom Rini [Thu, 19 Oct 2017 15:19:38 +0000 (11:19 -0400)]
Merge git://git.denx.de/u-boot-x86

7 years agox86: conga-qeval20-qa3-e3845-internal-uart_defconfig: Add ACPI resume support
Stefan Roese [Tue, 17 Oct 2017 06:09:14 +0000 (08:09 +0200)]
x86: conga-qeval20-qa3-e3845-internal-uart_defconfig: Add ACPI resume support

I've missed to add the ACPI resume support to this x86 build target.
This patch adds the ACPI resume support enabling S3 suspend /
resume.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
7 years agox86: provide CONFIG_BUILD_ROM
Heinrich Schuchardt [Mon, 16 Oct 2017 18:08:38 +0000 (18:08 +0000)]
x86: provide CONFIG_BUILD_ROM

Up to now we depended on an exported variable to build u-boot.rom.
We should be able to specify it in the configuration file, too.

With this patch this becomes possible using the new Kconfig option
CONFIG_BUILD_ROM.

This option depends on CONFIG_X86 and is selected in
qemu-x86_defconfig and qemu-x86_64_defconfig.

Cc: Simon Glass <sjg@chromium.org>
Cc: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
7 years agox86: baytrail: fsp: Move Azalia update codes to board
Bin Meng [Fri, 13 Oct 2017 08:30:06 +0000 (01:30 -0700)]
x86: baytrail: fsp: Move Azalia update codes to board

Azalia configuration may be different across boards, hence it's not
appropriate to do that in the SoC level. Instead, let's make the
SoC update_fsp_azalia_configs() routine as a weak version, and do
the actual work in the board codes.

So far it seems only som-db5800-som-6867 board enables the Azalia.
Move the original codes into som-db5800-som-6867.c.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
7 years agox86: baytrail: fsp: Use a function to update the Azalia config pointer
Bin Meng [Fri, 13 Oct 2017 08:30:05 +0000 (01:30 -0700)]
x86: baytrail: fsp: Use a function to update the Azalia config pointer

At present we directly pass the Azalia config pointer to the FSP UPD.
This updates to use a function to do the stuff, like Braswell does.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
7 years agox86: fsp: Consolidate Azalia header file
Bin Meng [Fri, 13 Oct 2017 08:30:04 +0000 (01:30 -0700)]
x86: fsp: Consolidate Azalia header file

So far there are two copies of Azalia struct defines with one in
baytrail and the other one in braswell. This consolidates these
two into one, put it in the common place, and remove the prefix
pch_ to these structs to make their names more generic.

This also corrects reset_wait_timer from us to ms.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
7 years agox86: Turn off running VGA ROM during S3 resume
Bin Meng [Thu, 12 Oct 2017 12:07:58 +0000 (05:07 -0700)]
x86: Turn off running VGA ROM during S3 resume

This is only needed when graphics console is used. For kernel with
native graphics driver, this can be turned off to speed up.

Change this option's default to n in the Kconfig.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
7 years agox86: baytrail: Fix unstable ACPI S3 resume
Bin Meng [Thu, 12 Oct 2017 12:07:57 +0000 (05:07 -0700)]
x86: baytrail: Fix unstable ACPI S3 resume

It was observed that when booting a Ubuntu 16.04 kernel, doing ACPI
S3 suspend/resume sometimes causes the Ubuntu kernel hang forever.
The issue is however not reproduced with a kernel built from i386/
x86_64 defconfig configuration.

The unstability is actually caused by unexpected interrupts being
generated during the S3 resume. For some unknown reason, FSP (gold4)
for BayTrail configures the GPIO DFX5 PAD to enable level interrupt
(bit 24 and 25). As this pin keeps generating interrupts during an
S3 resume, and there is no IRQ requester in the kernel to handle it,
the kernel seems to hang and does not continue resuming.

Clear the mysterious interrupt bits for this pin.

Reported-by: Stefan Roese <sr@denx.de>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Stefan Roese <sr@denx.de>
Reviewed-by: Stefan Roese <sr@denx.de>
7 years agox86: minnowmax: Adjust VGA rom address
Bin Meng [Sat, 7 Oct 2017 09:43:52 +0000 (02:43 -0700)]
x86: minnowmax: Adjust VGA rom address

Adjust VGA rom address to 0xfffb0000 so that u-boot.rom image
can be built again.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
7 years agoMerge git://git.denx.de/u-boot-i2c
Tom Rini [Wed, 18 Oct 2017 13:32:35 +0000 (09:32 -0400)]
Merge git://git.denx.de/u-boot-i2c

7 years agoMerge git://git.denx.de/u-boot-spi
Tom Rini [Wed, 18 Oct 2017 13:32:21 +0000 (09:32 -0400)]
Merge git://git.denx.de/u-boot-spi

7 years agoi2c: stm32f7_i2c: fix usage of useless local variable
Patrice Chotard [Tue, 17 Oct 2017 09:21:33 +0000 (11:21 +0200)]
i2c: stm32f7_i2c: fix usage of useless local variable

Remove useless local variable "s" and use directly
function's parameter "output"

Signed-off-by: Patrice Chotard <patrice.chotard@st.com>
7 years agoi2c: stm32f7_i2c: fix data abort
Christophe Kerello [Tue, 17 Oct 2017 09:21:32 +0000 (11:21 +0200)]
i2c: stm32f7_i2c: fix data abort

As "v" is a local variable in stm32_i2c_choose_solution()
"v" has to be copied into "s" to avoid data abort in
stm32_i2c_compute_timing().

Signed-off-by: Christophe Kerello <christophe.kerello@st.com>
Reviewed-by: Patrick DELAUNAY <patrick.delaunay@st.com>
Signed-off-by: Patrice Chotard <patrice.chotard@st.com>
7 years agoi2c: remove DECLARE_GLOBAL_DATA_PTR from i2c-uclass
Masahiro Yamada [Fri, 13 Oct 2017 10:29:03 +0000 (19:29 +0900)]
i2c: remove DECLARE_GLOBAL_DATA_PTR from i2c-uclass

No global pointer is used in this file.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
7 years agoPrepare v2017.11-rc2
Tom Rini [Tue, 17 Oct 2017 01:27:55 +0000 (21:27 -0400)]
Prepare v2017.11-rc2

Signed-off-by: Tom Rini <trini@konsulko.com>
7 years agoconfigs: Resync with savedefconfig
Tom Rini [Tue, 17 Oct 2017 01:35:43 +0000 (21:35 -0400)]
configs: Resync with savedefconfig

Rsync all defconfig files using moveconfig.py

Signed-off-by: Tom Rini <trini@konsulko.com>
7 years agotest/py/tests/test_sleep.py: test time approximately
Heinrich Schuchardt [Fri, 13 Oct 2017 20:28:31 +0000 (22:28 +0200)]
test/py/tests/test_sleep.py: test time approximately

On qemu errors like
assert 2.999650001525879 >= 3
occur.

According to the comment in the code the test is meant to be
approximate. So we should accept some milliseconds less.

Cc: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Stephen Warren <swarren@nvidia.com>
7 years agoscripts/get_maintainer.pl: update to current version
Heinrich Schuchardt [Fri, 13 Oct 2017 17:31:20 +0000 (19:31 +0200)]
scripts/get_maintainer.pl: update to current version

Update the script to version 0.26 (as of Linux v4.14-rc1)
Keep our "penguin_chief".
Keep our top_of_kernel_tree.

The negative forms of the command line parameters are described
when using --help.

New options are

 --git-blame-signatures => when used with --git-blame,
                           also include all commit signers
 --r => include reviewer(s) if any
 --letters => print all matching 'letter' types
              from all matching sections

File .get_maintainer.ignore can be used to specify
email addressees that shall be ignored.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
7 years agoarm: dra76: fastboot: extend cpu type for getvar command
Praneeth Bajjuri [Fri, 13 Oct 2017 03:47:05 +0000 (22:47 -0500)]
arm: dra76: fastboot: extend cpu type for getvar command

'commit fa24eca1f20a ("omap: Add routine for setting fastboot variables")'
adds initial support and usage of "fastboot getvar" command
for DRA75x and DRA72x devices.

and
'commit 0f9e6aee9dbc ("arm: dra76: Add support for ES1.0 detection")'
adds initial dra76 device definition

This patch is to extend usage of "fastboot getvar" for DRA76 device.

Signed-off-by: Praneeth Bajjuri <praneeth@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
7 years agoomap-common: fastboot: extend cpu type for DRA71x rev 2.1
Vishal Mahaveer [Fri, 13 Oct 2017 03:47:04 +0000 (22:47 -0500)]
omap-common: fastboot: extend cpu type for DRA71x rev 2.1

DRA71x processors are reduced pin and software compatible
derivative of DRA72 processors. Extend support for this
revision in "getvar cpu" command.

Signed-off-by: Vishal Mahaveer <vishalm@ti.com>
[praneeth@ti.com: rebase to u-boot master]
Signed-off-by: Praneeth Bajjuri <praneeth@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
7 years agoarm: dra7xx: Add vendor partition to Android GPT table for eMMC
Vishal Mahaveer [Fri, 13 Oct 2017 03:47:03 +0000 (22:47 -0500)]
arm: dra7xx: Add vendor partition to Android GPT table for eMMC

Add vendor partition to Android GPT table for eMMC.

A Vendor image contains SoC-specific code and configuration.

Prior to Android 8.0, the vendor partition was optional ;
files belonging to these images were placed in boot.img or system.img
with symlinks (such as /vendor >/system/vendor ) when absent.

Android 8.0 makes the vendor partition mandatory

The goal is to modularize Android partitions with standard interface between
the Android Platform (on system.img ) and vendor-provided code(on vendor.img).

This standard interface enables the Android Platform to be updated
without affecting the SoC partitions. This makes it possible to upgrade a
device system.img from Android 8.0 to Android P while other images (such as
vendor.img) remain at Android 8.0. This modularity enables timely
Android platform upgrades (such as monthly security updates )
without requiring SoC/ODM partners to update SoC- and device-specific code.

Signed-off-by: Vishal Mahaveer <vishalm@ti.com>
Signed-off-by: Praneeth Bajjuri <praneeth@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
7 years agoarm: am57xx: Add vendor partition to Android GPT table for eMMC
Praneeth Bajjuri [Fri, 13 Oct 2017 03:47:02 +0000 (22:47 -0500)]
arm: am57xx: Add vendor partition to Android GPT table for eMMC

Add vendor partition to Android GPT table for eMMC.

A Vendor image contains SoC-specific code and configuration.

Prior to Android 8.0, the vendor partition was optional ;
files belonging to these images were placed in boot.img or system.img
with symlinks (such as /vendor >/system/vendor ) when absent.

Android 8.0 makes the vendor partition mandatory

The goal is to modularize Android partitions with standard interface between
the Android Platform (on system.img ) and vendor-provided code(on vendor.img).

This standard interface enables the Android Platform to be updated
without affecting the SoC partitions. This makes it possible to upgrade a
device system.img from Android 8.0 to Android P while other images (such as
vendor.img) remain at Android 8.0. This modularity enables timely
Android platform upgrades (such as monthly security updates )
without requiring SoC/ODM partners to update SoC- and device-specific code.

Signed-off-by: Praneeth Bajjuri <praneeth@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
7 years agousbtty: fix typos
Heinrich Schuchardt [Thu, 12 Oct 2017 21:37:37 +0000 (23:37 +0200)]
usbtty: fix typos

Fix typos in USB tty driver.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
7 years agotools: env: Add embedded.c to .gitignore
Bin Meng [Thu, 12 Oct 2017 03:15:04 +0000 (20:15 -0700)]
tools: env: Add embedded.c to .gitignore

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
7 years agocmd/pxe.c: Rework bootargs construction to clarify string checks
Tom Rini [Wed, 11 Oct 2017 19:34:33 +0000 (15:34 -0400)]
cmd/pxe.c: Rework bootargs construction to clarify string checks

As the code currently stands, we first check that the length of the
given command line, along with ip_str/mac_str along with an additional 1
for the NULL termination will fit within the buffer we have, and if not,
we return an error.  The way this code was originally written however
left Coverity "unhappy" due to using strcat rather than strncat.
Switching this to strncat however causes clang to be unhappy that we
aren't enforcing the "1" portion within strncat.  Rather than further
re-work the code to include a "- 1" in this case as well, make the
strcat code only be done within the else side of the length test.  This
keeps both clang and Coverity happy.

Fixes: 48ee0a87bc46 ("cmd/pxe.c: Rework initrd and bootargs handling slightly")
Signed-off-by: Tom Rini <trini@konsulko.com>
7 years agokbuild: fix dependency of DT build
Masahiro Yamada [Wed, 11 Oct 2017 14:56:10 +0000 (23:56 +0900)]
kbuild: fix dependency of DT build

I saw a DT build issue report some time before [1].  I was able to
reproduce the bug, and figure out the root cause.

Since commit 6d427c6b1fa0 ("binman: Automatically include a U-Boot
.dtsi file"), invalid .*.cmd files are generated.

Since that commit, DTS files are put into sed and piped to CPP.
Because CPP reads the stream from stdin, -Wp,-MD,$(depfile).pre.tmp
option generates a depfile with the target name "-".  This is not
the format expected by fixdep.

Use one more temporary file instead of using pipe.  With this, deps_
in the .*.cmd will be fixed.  Having a temp file name in source_ is
odd, but it is unsed in the build system.  Not a big deal.

[1] https://lists.denx.de/pipermail/u-boot/2017-June/294451.html

Fiexes: 6d427c6b1fa0 ("binman: Automatically include a U-Boot .dtsi file")
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agocommon: Remove genimg_get_image()
Tuomas Tynkkynen [Tue, 10 Oct 2017 18:59:43 +0000 (21:59 +0300)]
common: Remove genimg_get_image()

Removal of the legacy DataFlash code turned genimg_get_image() into a
no-op. Drop all calls to it and the function itself.

Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
7 years agoDrop CONFIG_HAS_DATAFLASH
Tuomas Tynkkynen [Tue, 10 Oct 2017 18:59:42 +0000 (21:59 +0300)]
Drop CONFIG_HAS_DATAFLASH

Last user of this option went away in commit:

fdc7718999 ("board: usb_a9263: Update to support DT and DM")

Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
7 years agospi: Drop CONFIG_ATMEL_DATAFLASH_SPI
Tuomas Tynkkynen [Tue, 10 Oct 2017 18:59:41 +0000 (21:59 +0300)]
spi: Drop CONFIG_ATMEL_DATAFLASH_SPI

Last user of this option went away in commit:

fdc7718999 ("board: usb_a9263: Update to support DT and DM")

Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
Acked-by: Wenyou Yang <wenyou.yang@microchip.com>
7 years agoenv: Drop CONFIG_ENV_IS_IN_DATAFLASH
Tuomas Tynkkynen [Tue, 10 Oct 2017 18:59:40 +0000 (21:59 +0300)]
env: Drop CONFIG_ENV_IS_IN_DATAFLASH

Last user of this option went away in commit:

fdc7718999 ("board: usb_a9263: Update to support DT and DM")

Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
7 years agofs/jffs2/mini_inflate.c: use "static inline" instead of "inline"
Vagrant Cascadian [Tue, 10 Oct 2017 16:42:39 +0000 (09:42 -0700)]
fs/jffs2/mini_inflate.c: use "static inline" instead of "inline"

This makes gcc no longer expect an out-of-line version of the
functions being present elsewhere.

This fixes a failure to build on several marvell targets with gcc-7 on
Debian:

  https://bugs.debian.org/877963

Signed-off-by: Adrian Bunk <bunk@debian.org>
Signed-off-by: Vagrant Cascadian <vagrant@debian.org>
7 years agojffs2: ECC corrected nand data is ignored
Engling, Uwe [Tue, 10 Oct 2017 14:20:55 +0000 (14:20 +0000)]
jffs2: ECC corrected nand data is ignored

Hello,

I ran into a problem with the JFFS2 filesystem driver implemented in U-Boot.

I've got a NAND device that has correctable ECC errors (corrected somewhere in mtd/nand/nand_base.c).
The NAND driver tells the filesystem layer (jffs2_1pass.c) above that there occurred correctable ECC errors and returns with a "value > 0".
The JFFS2 driver recognizes the corrected ECC errors as real error and skips this block because the only accepts a "return value == 0" as correct.

This problem exists for over 8 years (I checked version 2010.09) so I'm a little bit worried that I interpreted something wrong or didn't get the whole context.

Can someone confirm this bug (and the bugfix) in the u-boot jffs2 driver?

There was a mail in 2012 that mentioned the same problem, but there was no patch:
http://u-boot.10912.n7.nabble.com/JFFS2-seems-to-drop-nand-data-with-ECC-corrections-td142008.html
Sometime after this discussion the return value of nand_read() changed from -EUCLEAN as correctable ECC error to a positive value with the count of ECC corrected errors.

With kind reguards,
Uwe Engling

7 years agoARM: Add documentation for the qemu-arm board
Tuomas Tynkkynen [Mon, 9 Oct 2017 16:42:41 +0000 (19:42 +0300)]
ARM: Add documentation for the qemu-arm board

Add brief documentation for the recently merged qemu-arm board.

Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
7 years agommc: stm32_sdmmc2: increase polling status register delay
Christophe Kerello [Mon, 9 Oct 2017 15:02:28 +0000 (17:02 +0200)]
mmc: stm32_sdmmc2: increase polling status register delay

MMC commands like MMC_CMD_ALL_SEND_CID or MMC_CMD_SEND_CSD can reach
500 us. This patch increases the polling status register delay to avoid
a timeout on a command.

Signed-off-by: Christophe Kerello <christophe.kerello@st.com>
Signed-off-by: Patrice Chotard <patrice.chotard@st.com>
7 years agodm: clk: fix PWR_CR3 register's bit 2 name
Patrice Chotard [Mon, 9 Oct 2017 09:41:24 +0000 (11:41 +0200)]
dm: clk: fix PWR_CR3 register's bit 2 name

Fix bit 2 name of PWR_CR3 register to match with the
last STM32H7 reference manual available here :

http://www.st.com/content/st_com/en/support/resources/
resource-selector.html?querycriteria=productId=SS1951$$
resourceCategory=technical_literature$$resourceType=reference_manual

Update also comment about voltage scaling 1 values

Signed-off-by: Patrice Chotard <patrice.chotard@st.com>
Reviewed-by: Vikas Manocha <vikas.manocha@st.com>
7 years agodm: clk: remove CLK() macro for clk_stm32h7
Patrice Chotard [Mon, 9 Oct 2017 09:41:23 +0000 (11:41 +0200)]
dm: clk: remove CLK() macro for clk_stm32h7

CLK() macro is a residue of a previously reworked patch,
remove it.

Signed-off-by: Patrice Chotard <patrice.chotard@st.com>
Reviewed-by: Vikas Manocha <vikas.manocha@st.com>
7 years agoARM: tegra: Remove unused define CONFIG_ARMCORTEXA9
Tuomas Tynkkynen [Sun, 8 Oct 2017 19:44:05 +0000 (22:44 +0300)]
ARM: tegra: Remove unused define CONFIG_ARMCORTEXA9

This macro isn't used by anything, and in fact hasn't ever been used
according to the Git logs.

Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
7 years agofs: Set .opendir to fs_opendir_unsupported for btrfs
Marek Behún [Fri, 6 Oct 2017 14:56:07 +0000 (16:56 +0200)]
fs: Set .opendir to fs_opendir_unsupported for btrfs

Otherwise fs_opendir will fault.

Signed-off-by: Marek Behun <marek.behun@nic.cz>
7 years agofs: btrfs: Remove a foreign language note
Marek Behún [Fri, 6 Oct 2017 13:04:59 +0000 (15:04 +0200)]
fs: btrfs: Remove a foreign language note

I accidentaly left a foreign language note in the code from development.

Signed-off-by: Marek Behun <marek.behun@nic.cz>
7 years agofs: btrfs: Fix usage of uninitialized variables
Marek Behún [Fri, 6 Oct 2017 13:04:57 +0000 (15:04 +0200)]
fs: btrfs: Fix usage of uninitialized variables

The variable res should be initialized to 0 in these functions,
because if the searched key is not found, the variable is used
uninitialized.

Reported-by: Coverity (CID: 167335)
Reported-by: Coverity (CID: 167336)
Reported-by: Coverity (CID: 167337)
Signed-off-by: Marek Behun <marek.behun@nic.cz>
7 years agolinux/time.h: Remove dead code
Marek Behún [Fri, 6 Oct 2017 13:04:56 +0000 (15:04 +0200)]
linux/time.h: Remove dead code

Since
  rem = ((long) *tim_p) % SECSPERDAY;
the second while cycle
  while (rem >= SECSPERDAY)
is dead.

Reported-by: Coverity (CID: 167334)
Signed-off-by: Marek Behun <marek.behun@nic.cz>
7 years agoiso: Reduce verbosity on test and info calls
Alexander Graf [Fri, 6 Oct 2017 11:35:07 +0000 (13:35 +0200)]
iso: Reduce verbosity on test and info calls

The test and info callbacks into the partition callback struct are
used by the "part list" command on the command line. That command
is used by the distro script.

With verb=1 set, "part list" thus throws a lot of warnings about
partitions it can't find when an upper layer searches for partitions.

So let's reduce verbosity to bring it to the same level of noise
as the other partition targets.

Signed-off-by: Alexander Graf <agraf@suse.de>
7 years agofixdep: fix dependency on options surrounded by CONFIG_VAL()
Masahiro Yamada [Fri, 6 Oct 2017 01:24:43 +0000 (10:24 +0900)]
fixdep: fix dependency on options surrounded by CONFIG_VAL()

CONFIG options surrounded by

  CONFIG_IS_ENABLED(...)
  CONFIG_IS_BUILTIN(...)
  CONFIG_IS_MODULE(...)
  CONFIG_VAL(...)

need special care for proper dependency tracking.

I do not remember why, but I missed to add CONFIG_VAL(...) handling.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agotools: bmp_logo: correctly interpret BMP files with larger headers
Jon Smith [Thu, 5 Oct 2017 12:35:02 +0000 (08:35 -0400)]
tools: bmp_logo: correctly interpret BMP files with larger headers

All BMP files were being treated as though they had a 40 byte header.
There are several BMP header formats consisting of additional data.
This was causing some of the header to be read as color information,
skewing the color palette.

Signed-off-by: Jon Smith <jtsmith@pdiarm.com>
7 years agocommon: Add a prototype for s_init()
Diego Dorta [Thu, 5 Oct 2017 12:13:38 +0000 (09:13 -0300)]
common: Add a prototype for s_init()

When compiling with W=1 the following warning is observed:

arch/arm/mach-imx/mx6/soc.c:590:6: warning: no previous prototype for ‘s_init’ [-Wmissing-prototypes] void s_init(void)

Remove this warning by adding the function prototype into include/common.h file.

Signed-off-by: Diego Dorta <diego.dorta@nxp.com>
7 years agotools/mkimage: Fix DTC run command to handle file names with space
Mirza, Taimoor [Wed, 4 Oct 2017 15:28:03 +0000 (20:28 +0500)]
tools/mkimage: Fix DTC run command to handle file names with space

fit_handle_file function does not quote input and output files while preparing
command to run DTC to convert .its to .itb. This results in a failure if input
or output files contain spaces in their names. Quote input and output files in
DTC command to avoid this failure.

Signed-off-by: Mirza, Taimoor <Taimoor_Mirza@mentor.com>
7 years agofdt: update bcm283x device tree sources to Linux 4.14 state
Alexander Graf [Wed, 4 Oct 2017 12:39:16 +0000 (14:39 +0200)]
fdt: update bcm283x device tree sources to Linux 4.14 state

Upstream Linux has received a few device tree updates to the RPi
which we should propagate into the builtin U-Boot one as well to
gain hardware support.

This patch bumps the dts files to their 4.14 Linux counterparts
with the exception of sdhost on 32bit RPi versions. There we stay
with iproc as the sdhost driver is missing in U-Boot.

Signed-off-by: Alexander Graf <agraf@suse.de>
7 years agopart_efi: In is_gpt_valid() check argument validity before allocation
Tom Rini [Tue, 3 Oct 2017 13:38:44 +0000 (09:38 -0400)]
part_efi: In is_gpt_valid() check argument validity before allocation

While this goes somewhat against normal coding style we should ensure
that dev_desc is not NULL before we dereference it in allocation of
legacy_mbr.

Reported-by: Coverity (CID: 167292)
Signed-off-by: Tom Rini <trini@konsulko.com>
7 years agoSPL: FIT: Kconfig: Change SPL_FIT_IMAGE_POST_PROCESS entry
ext-vasily.gurevich@vaisala.com [Mon, 2 Oct 2017 13:42:55 +0000 (13:42 +0000)]
SPL: FIT: Kconfig: Change SPL_FIT_IMAGE_POST_PROCESS entry

Remove depends on TI_SECURE_DEVICE for other platforms.

Signed-off-by: Vasily Gurevich <ext-vasily.gurevich@vaisala.com>
7 years agoclk: uniphier: add NAND controller clock
Masahiro Yamada [Fri, 13 Oct 2017 17:21:19 +0000 (02:21 +0900)]
clk: uniphier: add NAND controller clock

This allows the NAND driver to enable clock and get its clock rate.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agomtd: nand: denali_dt: add clock support
Masahiro Yamada [Fri, 13 Oct 2017 17:21:18 +0000 (02:21 +0900)]
mtd: nand: denali_dt: add clock support

Enable clock in the probe hook.  The clock rate will be necessary
when setup_data_interface hook is supported.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: uniphier: adjust ifdefs for new UniPhier DWC3 CONFIG
Masahiro Yamada [Fri, 13 Oct 2017 17:21:17 +0000 (02:21 +0900)]
ARM: uniphier: adjust ifdefs for new UniPhier DWC3 CONFIG

Now USB 3.0 feature is enabled/disabled by CONFIG_USB_DWC3_UNIPHIER.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: uniphier: enable DWC3 xHCI driver
Masahiro Yamada [Fri, 13 Oct 2017 17:21:16 +0000 (02:21 +0900)]
ARM: uniphier: enable DWC3 xHCI driver

Enable CONFIGs for the DWC3 core and the UniPhier specific glue layer.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: uniphier: change the default of SoC select to UNIPHIER_V7_MULTI
Masahiro Yamada [Fri, 13 Oct 2017 10:22:09 +0000 (19:22 +0900)]
ARM: uniphier: change the default of SoC select to UNIPHIER_V7_MULTI

ARCH_UNIPHIER_V8_MULTI depends on !SPL, so the default may be hidden.
Use a clearer default.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoreset: uniphier: fix the first argument passed to dev_err()
Masahiro Yamada [Fri, 13 Oct 2017 10:22:02 +0000 (19:22 +0900)]
reset: uniphier: fix the first argument passed to dev_err()

priv->dev does not exist.  Pass the correct pointer to udevice.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoclk: uniphier: add PXs3 clock data
Masahiro Yamada [Fri, 13 Oct 2017 10:22:00 +0000 (19:22 +0900)]
clk: uniphier: add PXs3 clock data

Add basic clock data for Socionext's new SoC PXs3.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoclk: uniphier: rework for better clock tree structure
Masahiro Yamada [Fri, 13 Oct 2017 10:21:59 +0000 (19:21 +0900)]
clk: uniphier: rework for better clock tree structure

U-Boot does not support fancy clock tree structures like the Linux
common clock framework.  Implement a simple clock tree model at the
driver level.  With this, the clock data will be simplified.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoi2c: uniphier-f: replace debug() with dev_dbg()
Masahiro Yamada [Fri, 13 Oct 2017 10:21:58 +0000 (19:21 +0900)]
i2c: uniphier-f: replace debug() with dev_dbg()

Use dev_dbg() functions.  It will be helpful to prefix log messages
with the corresponding device name when the core framework is ready.

While I am here, I renamed "dev", which was actually private data,
into "priv" because dev->dev looks confusing.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoi2c: uniphier: replace debug() with dev_dbg()
Masahiro Yamada [Fri, 13 Oct 2017 10:21:57 +0000 (19:21 +0900)]
i2c: uniphier: replace debug() with dev_dbg()

Use dev_dbg() functions.  It will be helpful to prefix log messages
with the corresponding device name when the core framework is ready.

While I am here, I renamed "dev", which was actually private data,
into "priv" because dev->dev looks confusing.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agousb: dwc3-uniphier: replace <common.h> with <linux/bitops.h>
Masahiro Yamada [Fri, 13 Oct 2017 10:21:56 +0000 (19:21 +0900)]
usb: dwc3-uniphier: replace <common.h> with <linux/bitops.h>

Including <common.h> pulls in a lot of bloat.  What this driver needs
is BIT(), so replace it with <linux/bitops.h>

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Marek Vasut <marex@denx.de>
7 years agoARM: uniphier: use pr_() instead of printf() where appropriate
Masahiro Yamada [Fri, 13 Oct 2017 10:21:55 +0000 (19:21 +0900)]
ARM: uniphier: use pr_() instead of printf() where appropriate

Replace printf() with pr_() to specify proper loglevel.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agopinctrl: uniphier: simplify input enable and delete pin arrays
Masahiro Yamada [Fri, 13 Oct 2017 10:21:54 +0000 (19:21 +0900)]
pinctrl: uniphier: simplify input enable and delete pin arrays

The pin data are implemented for old SoCs to specify the bit shift of
the IECTRL register.  They are not wortwhile given the required memory
footprint.  Delete all the pin data and enable all bits of the IECTRL
register.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agopinctrl: uniphier: set PUPD_SIMPLE cap flag for PXs3
Masahiro Yamada [Fri, 13 Oct 2017 10:21:53 +0000 (19:21 +0900)]
pinctrl: uniphier: set PUPD_SIMPLE cap flag for PXs3

Like other recenct UniPhier SoCs, the pupdctrl number of PXs3
matches to the pin number.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: dts: uniphier: update GPIO nodes
Masahiro Yamada [Fri, 13 Oct 2017 10:21:52 +0000 (19:21 +0900)]
ARM: dts: uniphier: update GPIO nodes

Switch to the single node design.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agogpio: uniphier: rework single device node model
Masahiro Yamada [Fri, 13 Oct 2017 10:21:51 +0000 (19:21 +0900)]
gpio: uniphier: rework single device node model

First, I implemented this driver as per-bank model, but it was
a design mistake.

  - There are 31 banks in the maximum case.  It is painful to add
    so many nodes to DT.

  - The IRQ control registers are shared between banks.  Per-bank
    design is a problem for Linux.  The counterpart for Linux turned
    around to the single node model.

Rework based on the driver for Linux.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoARM: uniphier: switch to CONFIG_ENV_IS_NOWHERE
Masahiro Yamada [Fri, 13 Oct 2017 10:21:50 +0000 (19:21 +0900)]
ARM: uniphier: switch to CONFIG_ENV_IS_NOWHERE

The non-volatile storage varies board by board.  The default should
be NOWHERE.  Please choose a proper device via Kconfig.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agomtd: nand: do not check R/B# for CMD_SET_FEATURES in nand_command(_lp)
Masahiro Yamada [Fri, 15 Sep 2017 12:44:59 +0000 (21:44 +0900)]
mtd: nand: do not check R/B# for CMD_SET_FEATURES in nand_command(_lp)

Set Features (0xEF) command toggles the R/B# pin after 4 sub feature
parameters are written.

Currently, nand_command(_lp) calls chip->dev_ready immediately after
the address cycle because NAND_CMD_SET_FEATURES falls into default:
label.  No wait is needed at this point.

If you see nand_onfi_set_features(), R/B# is already cared by the
chip->waitfunc call.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
[ Linux commit: c5d664aa5a4c4b257a54eb35045031630d105f49 ]

7 years agomtd: nand: do not check R/B# for CMD_READID in nand_command(_lp)
Masahiro Yamada [Fri, 15 Sep 2017 12:44:58 +0000 (21:44 +0900)]
mtd: nand: do not check R/B# for CMD_READID in nand_command(_lp)

Read ID (0x90) command does not toggle the R/B# pin.  Without this
patch, NAND_CMD_READID falls into the default: label, then R/B# is
checked by chip->dev_ready().

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
[ Linux commit: 3158fa0e739615769cc047d2428f30f4c3b6640e ]

7 years agomoveconfig: fix error message in do_autoconf()
Chris Packham [Sun, 27 Aug 2017 08:00:51 +0000 (20:00 +1200)]
moveconfig: fix error message in do_autoconf()

Move the % arch outside the double quote so that the missing toolchain
message is displayed correctly.

Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
7 years agoMerge tag 'signed-efi-next' of git://github.com/agraf/u-boot
Tom Rini [Fri, 13 Oct 2017 13:53:58 +0000 (09:53 -0400)]
Merge tag 'signed-efi-next' of git://github.com/agraf/u-boot

Patch queue for efi - 2017-10-13

This is the second batch of amazing improvements for efi_loader in 2017.11:

  - New self tests to verify our own code
  - A few bug fixes
  - colored text support
  - event and SNP improvements, should get us close to iPXE working

7 years agoefi_selftest: correctly check return values
Heinrich Schuchardt [Thu, 12 Oct 2017 23:00:05 +0000 (01:00 +0200)]
efi_selftest: correctly check return values

When cancelling the timer we should check the return
value provided by the set_timer service.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
7 years agoMerge git://git.denx.de/u-boot-fsl-qoriq
Tom Rini [Thu, 12 Oct 2017 17:36:58 +0000 (13:36 -0400)]
Merge git://git.denx.de/u-boot-fsl-qoriq

7 years agoMerge git://www.denx.de/git/u-boot-imx
Tom Rini [Thu, 12 Oct 2017 17:36:37 +0000 (13:36 -0400)]
Merge git://www.denx.de/git/u-boot-imx

7 years agoARM: imx6: Add DHCOM i.MX6 PDK board support
Marek Vasut [Mon, 9 Oct 2017 19:51:10 +0000 (21:51 +0200)]
ARM: imx6: Add DHCOM i.MX6 PDK board support

Add support for the DHCOM i.MX6 PDK board. This board has:
- FEC ethernet
- EHCI USB host
- 3x SDMMC

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
7 years agoimx: mx7: spl: remove redundant SATA definitionas
Uri Mashiach [Tue, 10 Oct 2017 06:13:10 +0000 (09:13 +0300)]
imx: mx7: spl: remove redundant SATA definitionas

The i.MX7 SOC doesn't include the SATA interface.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
7 years agomx6: sys_proto: Add prototypes for imx6_pcie_toggle() functions
Diego Dorta [Thu, 5 Oct 2017 12:15:58 +0000 (09:15 -0300)]
mx6: sys_proto: Add prototypes for imx6_pcie_toggle() functions

When compiling with W=1 errors are observed:

drivers/pci/pcie_imx.c:517:12: warning: no previous prototype for ‘imx6_pcie_toggle_power’ [-Wmissing-prototypes] __weak int imx6_pcie_toggle_power(void)

drivers/pci/pcie_imx.c:528:12: warning: no previous prototype for ‘imx6_pcie_toggle_reset’ [-Wmissing-prototypes] __weak int imx6_pcie_toggle_reset(void)

Remove these warnings by adding the functions prototypes on arch-mx6/sys_proto.

Signed-off-by: Diego Dorta <diego.dorta@nxp.com>
7 years agomx6: clock: Add a prototype for do_mx6_showclocks()
Diego Dorta [Thu, 5 Oct 2017 12:15:57 +0000 (09:15 -0300)]
mx6: clock: Add a prototype for do_mx6_showclocks()

When compiling with W=1 the following warning is observed:

arch/arm/mach-imx/mx6/clock.c:1268:5: warning: no previous prototype for ‘do_mx6_showclocks’ [-Wmissing-prototypes] int do_mx6_showclocks(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])

Remove this warning by adding the function prototype into arch-mx6/clock.h file.

Signed-off-by: Diego Dorta <diego.dorta@nxp.com>
7 years agoi.mx6ul: xpress: update UART init for current board revision
Anatolij Gustschin [Mon, 2 Oct 2017 19:32:55 +0000 (21:32 +0200)]
i.mx6ul: xpress: update UART init for current board revision

UART pinmux has been changed on the last board revision. Change
board pinmux accordingly. Console is on UART7 now, add pinmux,
base address and update console string in environment.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Stefano Babic <sbabic@denx.de>
7 years agomx6slevk: Use PARTUUID to specify the rootfs location
Fabio Estevam [Mon, 2 Oct 2017 13:11:37 +0000 (10:11 -0300)]
mx6slevk: Use PARTUUID to specify the rootfs location

mx6slevk can run different kernel versions, such as NXP 4.1 or mainline.

Currently the rootfs location is passed via mmcblk number and the
problem with this approach is that the mmcblk number for the SD
card changes depending on the kernel version.

In order to avoid such issue, use the UUID method to specify the
rootfs location.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
7 years agodrivers: pci: imx: fix imx_pcie_remove function
Sven-Ola Tuecke [Thu, 5 Oct 2017 11:46:42 +0000 (08:46 -0300)]
drivers: pci: imx: fix imx_pcie_remove function

We have at least a minor count of boards, that failed to re-initialize
PCI express in the Linux kernel. Typical failure rate is 20% on affected
boards. This is mitigated by commit 6ecbe1375671 ("drivers: pci: imx:
add imx_pcie_remove function").

However, at least on some i.MX6 custom boards, when calling
assert_core_reset() as part of the first-time PCIe init, read access
to PCIE_PL_PFLR simply hangs. Surround this readl() with
imx_pcie_fix_dabt_handler() does not help. For this reason, the forced
LTSSM detection is only used on the second assert_core_reset() that is
called shorly before starting the Linux kernel.

Signed-off-by: Sven-Ola Tuecke <sven-ola.tuecke@numberfour.eu>
Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Tested-by: David Müller <d.mueller@elsoft.ch>
7 years agowandboard: Add support for the latest revd1 revision
Fabio Estevam [Mon, 2 Oct 2017 18:47:29 +0000 (15:47 -0300)]
wandboard: Add support for the latest revd1 revision

Latest wandboard hardware revision is revd1, which brings the following
new features:

- PFUZE100 PMIC
- AR8035 Ethernet PHY
- Upgrade Wifi/BT chip to BCM4339/BCM43430.

The detection mechanism is to probe the PMIC and when it is
found, then the revision of the board is revd1.

As the detection is done via PMIC, we need to print the board version
at a later stage via CONFIG_DISPLAY_BOARDINFO_LATE and also need
to disable CONFIG_DISPLAY_BOARDINFO, which is done much earlier.

Make the necessary adjustments for the AR8035 PHY to work on revd1.

Based on Richard Hu's work from Technexion's U-Boot tree.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
7 years agodoc: update imx_usb_loader URL
Stefan Agner [Tue, 3 Oct 2017 14:43:26 +0000 (16:43 +0200)]
doc: update imx_usb_loader URL

The changes required to use U-Boot's Serial Download Protocol
implementation are now available in upstream imx_usb_loader
repository. Update the URL accordingly.

Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
7 years agoarm: imx7d: add support for Compulab cl-som-imx7
Ilya Ledvich [Sun, 24 Sep 2017 06:00:25 +0000 (09:00 +0300)]
arm: imx7d: add support for Compulab cl-som-imx7

Add initial support for Compulab cl-som-imx7 SoM.
The initial support includes:
 - MMC
 - eMMC
 - SPI flash
 - I2C
 - FEC
 - USB
 - Serial console

Signed-off-by: Ilya Ledvich <ilya@compulab.co.il>
7 years agoimx: mx7: SPL support for i.MX7
Uri Mashiach [Sun, 24 Sep 2017 06:00:24 +0000 (09:00 +0300)]
imx: mx7: SPL support for i.MX7

Add configuration file and spl_boot_device function for the i.MX7 SPL.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
7 years agoimx: mx7: DDR controller configuration for the i.MX7 architecture
Uri Mashiach [Sun, 24 Sep 2017 06:00:23 +0000 (09:00 +0300)]
imx: mx7: DDR controller configuration for the i.MX7 architecture

The configuration files imximage.cfg are used for the DDR controller
configuration.
Add DDR configuration function to replace the DDR controller
configuration in the imximage.cfg file. The function can be used for
DDR size detection.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
7 years agoimx: mx7: fix the CCM_ macros
Uri Mashiach [Sun, 24 Sep 2017 06:00:22 +0000 (09:00 +0300)]
imx: mx7: fix the CCM_ macros

The CCM_ macros use the CCM_BASE_ADDRESS macro, which doesn't exist.
Replace the CCM_BASE_ADDRESS macros with CCM_BASE_ADDR.

Signed-off-by: Uri Mashiach <uri.mashiach@compulab.co.il>
Acked-by: Igor Grinberg <grinberg@compulab.co.il>
7 years agoicorem6_rqs: env: Return mmc devno
Jagan Teki [Thu, 28 Sep 2017 17:59:14 +0000 (23:29 +0530)]
icorem6_rqs: env: Return mmc devno

Instead of changing mmc devno from dts nodes better
to return the detected devno so-that env trigger the same.

Cc: Stefano Babic <sbabic@denx.de>
Cc: Tom Rini <trini@konsulko.com>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>