17 Jul, 2015
1 commit
-
All the mux configurations needs to be done as part of the IODelay
sequence to avoid glitch.
Adding all the mux configuration for DRA72-evm.
MANUAL/VIRTUAL mode settings will be added once the data is available.Signed-off-by: Lokesh Vutla
15 Jul, 2015
6 commits
-
On dra72 evm cpsw slave1 is muxed with VIN, hence switch cpsw to use
slave0 using pcf driver. Also use phy#3 instead of phy#2 when using
slave0.Note: With this change, the cpsw driver in kernel needs to use phy#3
on dra72.Signed-off-by: Vignesh R
-
On DRA7, pcf chip present at address 0x21 on i2c1, is used to
switch between cpsw slave0 and slave1. Hence, enable pcf
driver for the same.Signed-off-by: Vignesh R
-
TI's pcf8575 is a 16-bit I2C based GPIO expander.The device features a
16-bit quasi-bidirectional I/O ports. Each quasi-bidirectional I/O can
be used as an input or output without the use of a data-direction
control signal. The I/Os should be high before being used as inputs.
This driver is based on pcf857x driver available in Linux 4.0 kernel.
It supports basic reading and writing of gpio pins.Signed-off-by: Vignesh R
-
DRA72-evm supports DDR at 666MHz. SDRAM_TIM2 is programmed for 532MHz.
Fixing it to 666MHz.Signed-off-by: Lokesh Vutla
-
Since buildman now includes most of the features of MAKEALL it is probably
time to talk about deprecating MAKEALL.Comments welcome.
Signed-off-by: Simon Glass
-
Signed-off-by: Tom Rini
11 Jul, 2015
1 commit
-
Commit 2b42c9317db ("ahci: support LBA48 data reads for 2+TB drives")
introduced conditional code which triggers a warning when compiled
with DEBUG enabled:In file included from common/cmd_scsi.c:12:0:
common/cmd_scsi.c: In function 'scsi_read':
include/common.h:109:4: warning: 'smallblks' may be used uninitialized in this function [-Wmaybe-uninitialized]
...Since this is for debug only, take the easy way and initialize the
variable explicitly on declaration to avoid the warning.
(Fix a nearby whitespace error on the way.)Tested-by: Bin Meng
Signed-off-by: Andre Przywara
10 Jul, 2015
17 commits
-
Since commit 09c3280754f8 (mtd, nand: Move common functions from
cmd_nand.c to common place), NAND commands would not work at all
on large devices.=> nand read 80000000 10000 10000
NAND read: Offset exceeds device limit
=> nand erase 100000 100000NAND erase: Offset exceeds device limit
The type of the "size" of "struct mtd_info" is uint64_t, while
mtd_arg_off_size() and mtd_arg_off() treat chipsize as int type.
The chipsize is wrapped around if the argument is given with 2GB
or larger.Acked-by: Heiko Schocher
Acked-by: Scott Wood
Signed-off-by: Masahiro Yamada -
This patch enabled the USB/EHCI support for the Marvell
DB-88F6820-GP eval board.Signed-off-by: Stefan Roese
Cc: Marek Vasut
Cc: Luka Perkov -
This patch adds USB EHCI host support for the common mvebu platform.
Including the Armada 38x.Tested on DB-88F6280-GP eval board.
Signed-off-by: Stefan Roese
Reviewed-by: Marek Vasut
Cc: Luka Perkov -
Configure and enable the SATA/SCSI (AHCI) support for the Marvell
DB-88F6820-GP eval board.Signed-off-by: Stefan Roese
Cc: Luka Perkov -
This patch adds support for the common AHCI controller on the Marvell
Armada 38x.Tested on the Marvell DB-88F6820-GP eval board.
Signed-off-by: Stefan Roese
Cc: Luka Perkov -
This patch changes the initialization of the AHCI controller to not
enable the default interrupts (DEF_PORT_IRQ). As interrupts are
not used in U-Boot in general, this should not break the common AHCI
driver operation.This change is needed to support the Marvell Armada 38x AHCI
controller. With interrupts enabled, this results in timeouts in
ahci_device_data_io(). Not enabling these interrupts fixes this
problem and the common AHCI driver works fine.Signed-off-by: Stefan Roese
Reviewed-by: Simon Glass
Cc: Tom Rini
Cc: Hans de Goede
Cc: Luka Perkov -
This patch adds MMC/SDIO support to the Marvell DB-88F6820-GP board
configuration. Including support for the common partitions and
filesystems.Signed-off-by: Stefan Roese
Cc: Pantelis Antoniou
Cc: Luka Perkov -
Armada A38x implements an SDHCI compatible SDIO controller. This patch
enables the Marvell driver to support this SoC. And enables the
SDIO controller if selected by the board configuration.Tested on Marvell DB-88F6820-GP board.
Signed-off-by: Stefan Roese
Cc: Pantelis Antoniou
Cc: Luka Perkov -
While implementing SDIO/MMC SPL booting for the Marvell Armada 38x, the
following problem occured. The SPL runs in internal SRAM which is
the L2 cache locked to memory. When the MMC buffers now are located
on the stack (or bss), the SDIO controller (SDHCI) can't write into
this L2 cache memory.This patch introduces a method to use a fixed buffer that will be
used for all transfers by defining CONFIG_FIXED_SDHCI_ALIGNED_BUFFER.
This way, the board can use this buffer address located in SDRAM
for all transfers. This solves this SPL problem on the A38x and
should only be used in the SPL U-Boot version.Tested for SPL booting on Marvell Armada 38x DB-88F6820-GP board.
Signed-off-by: Stefan Roese
Cc: Pantelis Antoniou
Cc: Luka Perkov -
The loop counter based timeout detection does not work on the Armada
38x based board (DB-88F6820-GP). At least with dcache enabled a
timeout is detected. Without dcache enabled, the timeout does not
occur. Increasing the loop counter solves this issue. But a better
solution is to use a timer based timeout detection instead. This
patch now implements this timer based detection.Signed-off-by: Stefan Roese
Cc: Pantelis Antoniou
Cc: Luka Perkov -
Signed-off-by: Stefan Roese
Cc: Luka Perkov
Cc: Albert Aribaud -
When driver model is not used the current code does not correctly select
the pinmux for the I2C bus. This bug was introduced by this commit:8dfcbaa dm: i2c: s3c24x0: adjust to dm-i2c api
Signed-off-by: Simon Glass
Acked-by: Heiko Schocher
Reviewed-by: Przemyslaw Marczak
Signed-off-by: Minkyu Kang -
Adds the 'F:' entry for the board's defconfig
Signed-off-by: Minkyu Kang
-
When compling under 64bit platforms, there are lots of warnings,
like:drivers/block/ahci.c:114:18: warning: cast to pointer from integer
of different size [-Wint-to-pointer-cast]
u8 *port_mmio = (u8 *)probe_ent->port[port].port_mmio;
^
drivers/block/ahci.c: In function ?.hci_host_init?.
drivers/block/ahci.c:218:49: warning: cast from pointer to integer
of different size [-Wpointer-to-int-cast]
probe_ent->port[i].port_mmio = ahci_port_base((u32) mmio, i);......
Reviewed-by: Simon Glass
Signed-off-by: Shaohui Xie
Signed-off-by: Tang Yuantian -
Baltos has USB0 connected to a USB hub and thus is host-only. USB1
is connected to microUSB connector and thus should use OTG mode.Signed-off-by: Yegor Yefremov
09 Jul, 2015
15 commits
-
Because it is possible for the MTD number to change, causing a
filesystem mount failure, we should use the volume name instead
of the MTD number and let Linux resolve the correct one.Signed-off-by: Mike Scherban
-
On STM32F429 gpio PC6/PC7 can be allocated for USART6, as
reported in the comment.
But current code in
drivers/serial/serial_stm32.c
uses a different gpio mapping (PG14/PG9) for USART6.Fix the comment to match current code in the driver.
Signed-off-by: Antonio Borneo
To: u-boot@lists.denx.de
To: Kamil Lulko
Cc: Tom Rini -
Signed-off-by: Antonio Borneo
To: u-boot@lists.denx.de
To: Kamil Lulko
Cc: Tom Rini -
When "scripts/get_maintainer.pl" parses "board/.../MAINTAINERS",
it uses the line containing board name as delimiter.
Without this line, the script happily mixes the lines from current
board MAINTAINERS file with lines from another file.Fix it by adding a reasonable board name.
Tested by comparing output of:
cat board/st/stm32f429-discovery/MAINTAINERS
./scripts/get_maintainer.pl -f board/st/stm32f429-discoverySigned-off-by: Antonio Borneo
To: u-boot@lists.denx.de
To: Kamil Lulko
Cc: Tom Rini -
Use one mtd partition for rootfs and configuration by
means of ubi volumes and get rid of configuration partition.
We can use partition layout for both 256MB and 512MB flash.Signed-off-by: Samuel Egli
Cc: Heiko Schocher
Cc: Roger Meier
Reviewed-by: Tom Rini
Acked-by: Heiko Schocher -
Specify proper U-Boot offset, enable prefetch mode,
increase bootm size and add FIT fallback, if board_name
is not present in kernel-fit.itb image.Signed-off-by: Yegor Yefremov
-
This action is need to make I2C communication with PMIC
stable for low temperature. Print current I2C speed in
SPL for visual control.Signed-off-by: Yegor Yefremov
-
Signed-off-by: Yegor Yefremov
Reviewed-by: Simon Glass -
Add CONFIG_SYS_GENERIC_BOARD to board's config header.
Boot-tested on am3517_evm board.
Signed-off-by: Yegor Yefremov
-
If board is booted with transitions happening on DCAN1 pins then
the following warning is seen in the kernel at boot when the
hwmod layer initializes."omap_hwmod: dcan1: _wait_target_disable failed"
This is because DCAN1 module's SWAKEUP mechanism is broken
and it fails to correctly turn OFF if it sees a transition on the
DCAN1 pins. Suggested workaround is to keep DCAN1 pins in safe mode
while enabling/disabling DCAN1 module.The hwmod layer enables and disables all modules at boot
and we have no opportunity to put the DCAN1 pins in safe mode
at that point.DCAN1 is not used by u-boot so it doesn't matter to it if these
pins are in safe mode. The kernel driver correctly configures
the right mode when DCAN1 is active.Signed-off-by: Roger Quadros
[trini: s/PULLUP/PULL_UP/ based on DRA7xx EVM version of this patch]
Signed-off-by: Tom Rini -
If board is booted with transitions happening on DCAN1 pins then
the following warning is seen in the kernel at boot when the
hwmod layer initializes."omap_hwmod: dcan1: _wait_target_disable failed"
This is because DCAN1 module's SWAKEUP mechanism is broken
and it fails to correctly turn OFF if it sees a transition on the
DCAN1 pins. Suggested workaround is to keep DCAN1 pins in safe mode
while enabling/disabling DCAN1 module.The hwmod layer enables and disables all modules at boot
and we have no opportunity to put the DCAN1 pins in safe mode
at that point.DCAN1 is not used by u-boot so it doesn't matter to it if these
pins are in safe mode. The kernel driver correctly configures
the right mode when DCAN1 is active.Signed-off-by: Roger Quadros
-
-fdelete-null-pointer-checks flag controls global dataflow analyses and
eliminate useless checks for null pointers; It assume that if a pointer is
checked after it has already been dereferenced, it cannot be null.
This flag is enabled by default.gcc v4.9 has more optimizations added to this option. Hence it is very
aggressive with GCC v4.9 series. Add -fno-delete-null-pointer-checks to
disable the optimizationSigned-off-by: Rohit Dharmakan
Signed-off-by: Prabhakar Kushwaha -
This also came from Linux - according to this thread it has a GPL v2
license like arch/arm/mach-omap2/mux.h:http://lists.denx.de/pipermail/u-boot/2015-June/217827.html
Signed-off-by: Simon Glass
Reported-by: Ingrid Viitanen
Reviewed-by: Tom Rini -
The word "partition" is doubled. Keep decent forms for the
following lines.Also, fix some other typos while we are here.
Signed-off-by: Masahiro Yamada
Acked-by: Stephen Warren -
bf533-stamp, bf538f-ezkit, and cm-bf548 are very space limited.
This was introduced by:
6e0d26c0502e (net: Handle ethaddr changes as an env callback)
by enabling CONFIG_REGEX, which is too big for these boards.This patch disables CONFIG_REGEX at the expense of working with more
than the first ethaddr.Signed-off-by: Joe Hershberger
