Jagan Teki [Thu, 17 Mar 2016 06:53:18 +0000 (12:23 +0530)]
doc: driver-model: Update dm tests run using test.py
Since all the tests are implemented in pytest infrastructure,
So update the dm tests with the same instead of ./test/dm/test-dm.sh
Cc: Tom Rini <trini@konsulko.com> Cc: Simon Glass <sjg@chromium.org> Acked-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Jagan Teki <jteki@openedev.com>
Nishanth Menon [Tue, 15 Mar 2016 21:15:32 +0000 (16:15 -0500)]
dm: part: fix missing driver name in debug print
Fixes the following warning with PART_DEBUG enabled:
disk/part.c: In function ‘get_partition_info’:
disk/part.c:372:3: warning: format ‘%s’ expects a matching ‘char *’ argument [-Wformat]
Michal Simek [Mon, 11 Apr 2016 11:51:26 +0000 (13:51 +0200)]
GPIO: pca953x: Remove compilation warnings on arm64
Warnings:
w+../drivers/gpio/pca953x.c: In function ‘do_pca953x’:
w+../drivers/gpio/pca953x.c:220:5: warning: cast from pointer to integer
of different size [-Wpointer-to-int-cast]
w+../drivers/gpio/pca953x.c:233:10: warning: cast from pointer to
integer of different size [-Wpointer-to-int-cast]
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Michal Simek [Mon, 11 Apr 2016 06:07:07 +0000 (08:07 +0200)]
ARM64: zynqmp: Add defconfig for zcu102 revB board
Support natively revB board. Till now support for revB was done via
zcu102 defconfig where device-tree was changed to revB.
This patch is adding direct defconfig for RevB.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Michal Simek [Mon, 11 Apr 2016 09:44:40 +0000 (11:44 +0200)]
ARM64: zynqmp: Remove netdev.h from board file
Including netdev.h is causing compilation warning:
+ int fecmxc_register_mii_postcall(struct eth_device *dev, int
(*cb)(int));
+ ^
w+In file included from ../board/xilinx/zynqmp/zynqmp.c:9:0:
w+../include/netdev.h:204:41: warning: ‘struct eth_device’ declared
inside parameter list [enabled by default]
w+../include/netdev.h:204:41: warning: its scope is only this definition
or declaration, which is probably not what you want [enabled by default]
This patch removes it.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Michal Simek [Fri, 27 Nov 2015 12:22:58 +0000 (13:22 +0100)]
ARM64: zynqmp: Add backward compatible string for uart
Mainline kernel has no r1p12 compatible string that's why console stops
to work with the latest DTS files. Append generic compatible string.
Keep in your mind that using this generic compatible string not all uart
features will be available.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
This patch adds ddrc memory controller node in dts.
size mentioned in dts is 0x30000, because we need to access DDR_QOS
INTR registers located at fd090208 from this driver.
Signed-off-by: Naga Sureshkumar Relli <nagasure@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Michal Simek [Thu, 7 Apr 2016 13:28:33 +0000 (15:28 +0200)]
ARM64: zynqmp: Add missing interrupt-parent to PMU node
ZynqMP is not using global interrupt-parent setting that's why
it has to be listed in every node separately. PMU node missed it and
this patch is adding it.
Reported-by: John Linn <John.Linn@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Soren Brinkmann [Mon, 11 Jan 2016 23:34:42 +0000 (15:34 -0800)]
ARM64: zynqmp: DT: Add power domains
Add power-domains to the DT and attach devices to them.
The power-domains are all logical domains as understood by firmware.
Each PD is identified by a unique identifier that the platform firmware
understands.
Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Alistair Francis [Thu, 25 Feb 2016 17:30:03 +0000 (09:30 -0800)]
ARM64: zynqmp: Use C pre-processor for includes in dts
Change the dtsi include code to use the C pre-processor #include instead
of the device tree /include/. This brings all ZynqMP device trees inline
with each other.
Signed-off-by: Alistair Francis <alistair.francis@xilinx.com> Reviewed-by: Sören Brinkmann <soren.brinkmann@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Moritz Fischer <moritz.fischer@ettus.com> Signed-off-by: Olof Johansson <olof@lixom.net> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Stefan Roese [Fri, 8 Apr 2016 13:58:30 +0000 (15:58 +0200)]
arm: mvebu: theadorable: Add PEX-switch detection and reset code
Sometimes the PCIe link for the PEX-switch will not come-up. In this case,
the board is not in a usable state. This patch makes sure that in this
case a soft-reset is issued. If this soft-reset does not result in the
PEX-switch being detected after some soft-reset cycles, an I2C message
is sent to the uC to issue a complete power-cycle of the board.
Stefan Roese [Thu, 7 Apr 2016 08:48:14 +0000 (10:48 +0200)]
arm: mvebu: theadorable: Add bootcounter support
This patch adds bootcount support to the MVEBU theadorable board. Since
no reset-safe registers seem to be available, it uses the last 4KiB
of SDRAM for the bootcounter location.
Stefan Roese [Thu, 7 Apr 2016 08:48:13 +0000 (10:48 +0200)]
arm: mvebu: theadorable: Add USB power toggle and tweak USB PHY register
This patch adds an USB power toggle for theadorable. Additionally, the
USB PHY RX Channel Control 0 Register is changed to fix some issues
noticed while accessing some specific USB sticks.
Stefan Roese [Thu, 7 Apr 2016 05:41:00 +0000 (07:41 +0200)]
gpio: mvebu_gpio: Add missing out value set to gpio_direction_output()
This patch adds the missing configuration of the output value to the
gpio_direction_output() function. Without this, calling
gpio_direction_output() does not set the out-value at all and only
configures the gpio as output.
Signed-off-by: Stefan Roese <sr@denx.de> Cc: Kevin Smith <kevin.smith@elecsyscorp.com> Reviewed-by: Kevin Smith <kevin.smith@elecsyscorp.com>
Hans de Goede [Sat, 9 Apr 2016 13:02:24 +0000 (15:02 +0200)]
sunxi: Add defconfig and dts file for Orange Pi One SBC
The Orange Pi One SBC, is a stripped down version of the popular
Orange Pi PC. The one is a H3 based SBC, with 512M of RAM,
micro-sd slot, 1 host usb, 1 otg usb, hdmi and 100Mbit ethernet.
The dts is identical to the dts submitted to the upstream kernel.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Sat, 9 Apr 2016 13:00:52 +0000 (15:00 +0200)]
sunxi: H3: Do not clear usb companion clk-gate / reset on remove
On the H3 we need to enable the clk and de-assert the reset of the
companion to be able to talk to the actual usb host controller.
Before this commit we were also disabling the companion clk-gate /
asserting its reset on remove, causing the later remove callback of
the companion itself to (sometimes) fail with:
ERROR: USB HC reset timed out!
This commit fixes this by not disabling the companion's clk-gate nor
asserting its reset on remove.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Fri, 1 Apr 2016 10:55:08 +0000 (12:55 +0200)]
sunxi: Enable usb ports on the Sinovoip BPI M3
DLDO3 is used to provide Port-D power and PD is used for the
usb-hub / sata-5v enable pins. The 2.5V comes from the schematic
and matches the factory image fex file.
The dts changes are the minimal changes needed for u-boot to
pick-up the usb host controllers. The upstream kernel does not
(yet) have usb host support.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Fri, 1 Apr 2016 10:53:20 +0000 (12:53 +0200)]
sunxi: Set DCDC1 to 3.3V on the Sinovoip BPI M3
This is the value used in the fex file of the manufacturer images,
and also the DCDC1 default. Sometimes lower values are used to save
battery power, but that does not apply to a SBC.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
Hans de Goede [Fri, 1 Apr 2016 20:39:26 +0000 (22:39 +0200)]
sunxi: Add INITIAL_USB_SCAN_DELAY Kconfig option
Some boards have on board usb devices which need longer than the USB
spec's 1 second to connect from board powerup. Add a config option which
when non 0 adds an extra delay before the first usb bus scan.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Ian Campbell <ijc@hellion.org.uk>
spl: mmc: raw: Try to load u-boot if Linux image is not found
If CONFIG_SPL_OS_BOOT is enabled and Linux image is not flashed at
RAW_MODE_KERNEL_SECTOR in MMC, spl still assumes that Linux is
available and tries to boot it and hangs. In order to avoid this,
adding a check to verify if parsed image header is of type IH_OS_LINUX.
If it fails then fall back to load u-boot image.
Hans de Goede [Sat, 9 Apr 2016 11:53:48 +0000 (13:53 +0200)]
arm: Replace v7_maint_dcache_all(ARMV7_DCACHE_CLEAN_INVAL_ALL) with asm code
v7_maint_dcache_all() does not work reliable when build with gcc6,
see: https://bugzilla.redhat.com/show_bug.cgi?id=1318788
While debugging this I learned that v7_maint_dcache_all() is unreliable
when build with gcc5 too when it is marked as noinline.
This commit fixes the reliability issues by replacing the C-code with
the ready to use asm implementation from the kernel.
Given that this code when written as C-code clearly is quite fragile
(also see the existing comments about the C-code being the way it is
to get optimal assembly) and that we have a proven asm alternative,
I believe that this is the best solution.
Note that we actually already had a copy of the kernel's
v7_flush_dcache_all() before this commit in
arch/arm/mach-uniphier/arm32/lowlevel_init.S.
This commit moves that code arch/arm/cpu/armv7/cache_v7_asm.S, renames
it to __v7_flush_dcache_all(), and adds a v7_flush_dcache_all() wrapper
which saves / restores the clobbered registers for use from C-code.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Stefan Roese [Fri, 8 Apr 2016 13:56:29 +0000 (15:56 +0200)]
lib/crc8: Add crc start value
To make the usage of this function more flexible, lets add the CRC start
value as parameter to this function. This way it can be used by other
functions requiring different start values than 0 as well.
For non-zero CRC start values to work, I've reworked the function a bit.
The new implementation is copied from the Linux version in
drivers/i2c/i2c-core.c / i2c_smbus_pec(). Which supports non-zero
CRC stating values.
I've double-checked that the results for zero starting values are
identical to the results from the original version of this function.
Signed-off-by: Stefan Roese <sr@denx.de> Cc: Simon Glass <sjg@chromium.org> Reviewed-by: Simon Glass <sjg@chromium.org>
Stephen Warren [Wed, 6 Apr 2016 17:46:59 +0000 (11:46 -0600)]
test/py: README: link to example hook scripts
When implementing test/py hook scripts, it's helpful to read some working
examples. Provide a link to some. The link was mentioned in the commit
message which first added test/py, but not in any documentation file.
Suggested-by: Lukasz Majewski <l.majewski@samsung.com> Signed-off-by: Stephen Warren <swarren@nvidia.com> Reviewed-by: Tom Rini <trini@konsulko.com>
Stefan Roese [Wed, 6 Apr 2016 09:02:22 +0000 (11:02 +0200)]
bootcounter_ram: Flush dcache after data is written into SDRAM
This patch adds a call to flush_dcache_range() to bootcount_store() to
make sure, that the bootcounter data (including the patterns) is
written to memory. Without this, platforms with dcache enabled may not
have the bootcounter updated upon reset.
Signed-off-by: Stefan Roese <sr@denx.de> Cc: Valentin Longchamp <valentin.longchamp@keymile.com>
Michal Simek [Tue, 5 Apr 2016 11:34:20 +0000 (13:34 +0200)]
disk: part: Enable get_info pointer for CONFIG_SPL_FAT_SUPPORT
The patch "dm: part: Convert partition API use to linker lists"
(sha1: 96e5b03c8ab749b6547f6a3ceb4d4b9f274211aa) is adding new
dependency for enabling SPL_EXT_SUPPORT to be able to get
information about DOS partition.
get_info is also required for FAT support only which is used on Xilinx
Zynq boards.
Reported-by: Nathan Rossi <nathan@nathanrossi.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com> Reviewed-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Stephen Warren [Tue, 5 Apr 2016 02:00:41 +0000 (20:00 -0600)]
rpi: remove redundant board files
Now that rpi_*defconfig and Kconfig (rather than the config header file)
provide the identity of the build, we don't need to separate config
headers and board directories for each RPi variant. Set CONFIG_SYS_BOARD
and CONFIG_SYS_CONFIG_NAME so that we can get rid of the duplication. This
requires a tiny number of extra ifdefs in the config header.
The only disadvantage of this approach is that the $board/$board_name
environment variables aren't as descriptive as they used to be. This isn't
really an issue because those only exist to allow scripts to create DTB
filenames at runtime. However, the RPi board code already sets $fdtfile to
something more accurate based on FW-reported board ID anyway.
While at it, unify some Kconfig select options, and add a MAINTAINERS
entry for bcm283x too.
Partially-suggested-by: Tom Rini <trini@konsulko.com> Signed-off-by: Stephen Warren <swarren@wwwdotorg.org> Reviewed-by: Tom Rini <trini@konsulko.com>
stm32: stm32_flash: add memory barrier during flash write
After writing data to flash space, next instruction is checking if flash
controller is busy writing to the flash memory. Memory barrier is required here
to avoid transaction re-ordering for data write and busy status check.