Masahiro Yamada [Thu, 30 Oct 2014 06:30:58 +0000 (15:30 +0900)]
sparc: move CONFIG_SYS_TEXT_BASE to Kconfig
Defining CONFIG_SYS_TEXT_BASE in config.mk is very old style.
Create CONFIG_SYS_TEXT_BASE option in Kconfig, but let it
depend on CONFIG_SPARC because we do not want to disturb
the other architectures that still define CONFIG_SYS_TEXT_BASE
in their header files.
Hans de Goede [Wed, 24 Sep 2014 12:06:11 +0000 (14:06 +0200)]
usb: kbd: Add (optional) support for using an interrupt queue for polling
Waiting an interrupt packet to complete in usb_kbd_poll_for_event, causes
a 40 ms latency for each call to usb_kbd_testc, which is undesirable.
Using control messages leads to lower (but still not 0) latency, but some
devices do not work well with control messages (e.g. my kvm behaves funny
with them).
This commit adds support for using the int_queue mechanism which at least
the ehci-hcd driver supports. This allows polling with 0 latency, while
using interrupt packets.
Hans de Goede [Sat, 20 Sep 2014 15:03:52 +0000 (17:03 +0200)]
usb: Remove unnecessary portnr lookup from usb_new_device
If the device has a parent, it is instantiated from usb_hub_port_connect_change
and the portnr is right there in dev->portnr, so there is no need for this
whole dance to look it up.
net: sh-eth: Remove definition of RX_DESC_SIZE and TX_DESC_SIZE
RX_DESC_SIZE and TX_DESC_SIZE mean the size of the data descriptor for sh-eth.
We can acquire this in sizeof. It is not necessary to define these in define.
Initialization of L2CTLR[5] was set only as R8A7790 by commit 237faf095fb43abbed6e40266ef7efccc8b9308b.
However, initialization of cash needs to be performed continuously.
This changes into the processing which continues initialization of
L2CTLR[5] into L2CTLR cash and performs it.
Ye.Li [Thu, 30 Oct 2014 10:53:49 +0000 (18:53 +0800)]
imx: mx6 sabreauto: Add board support for USB EHCI
On mx6 sabreauto board, there are two USB ports:
0: OTG
1: HOST
The EHCI driver is enabled for this board, but the IOMUX and VBUS power
control is not implemented, which cause both USB port failed to work.
This patch fix the problem by adding the board support codes.
Since the power control uses the GPIO pin from port expander MAX7310,
the PCA953X driver is enabled for accessing the MAX7310.
The ID pin of OTG Port needs to configure the GPR1 bit 13 for selecting
its daisy chain. Add a new function "imx_iomux_set_gpr_register" to
handle GPR register setting.
Ye.Li [Thu, 30 Oct 2014 10:30:54 +0000 (18:30 +0800)]
imx: mx6slevk: Add support for USDHC1 and USDHC3 slots
There are three SD/MMC sockets on mx6slevk boards. Implements the
full support for them.
The default boot socket is USDHC2, so the MMC environment is set
to that device.
Ye.Li [Thu, 30 Oct 2014 10:20:55 +0000 (18:20 +0800)]
imx: gpt: Add High frequency clock source support for GPT
Introduce a new configuration "CONFIG_MXC_GPT_HCLK". When it is set,
the GPT will select a high frequency clock as clock source.
Otherwise, the GPT will stay to use 32Khz OSC as clock source.
In the implementation, since only the GPT on i.MX6 series provide the
clock source option for 24Mhz OSC. For others (only i.MX5 and i.MX6
compile the driver), if the configuration is set, the perclk will be
selected as clock source.
MX6Q/D Rev 1.0 and MX6SL are special in the implementation, because they
don't have the 24Mhz OSC clock source option, so also select the perclk
for them. For MX6SL, we will set the OSC 24Mhz to perclk in CCM, so
eventually the clock comes from OSC 24Mhz.
Ye.Li [Thu, 30 Oct 2014 09:54:08 +0000 (17:54 +0800)]
imximage: Fix the bootdata.size calculation
In system boot chapter of i.MX6 reference manual, the "Image Vector Table"
figure shows the bootdata.start points to the beginning of the destination
memory. It means the bootdata.size should contain the IVT offset part,
but the calculation in imximage tool does not have.
We found this issue when booting from QuadSPI NOR on i.MX6SX. The u-boot
runs into abnormal (crash or stop) after booting. After checked the destination
memory where the image is loaded to, there are hundreds of bytes at
the image end are not loaded into memory. Since there is a 4096 bytes
round in the calculation, for the booting devices using smaller IVT offset,
such as SD and SPI booting, they are not easy to reproduce.
MIPS: always keep all sections in u-boot ELF binary.
Always keep all sections in u-boot ELF binary. Move all unneeded
sections after _end to avoid allocating space in the final binary.
Also remove .deadcode section which is now obsolete.
removed section .deadcode because the original symptoms were not
visible anymore. Unfortuneatly the binutils bug still exists.
The size of .rel.dyn section is often bigger than needed for all
entries. But objcopy only allocates space as much as required for all
reloc entries. Thus there is a gap between the last entry and
__rel_dyn_end in u-boot.bin. If u-boot is booted from RAM (e.g. in
SPL scenarios) that area could contain garbage data which could lead
to CPU exceptions during relocation.
MIPS: kconfig: add options for vct board variant select
The vct board can be built with various variants which are
configured via CONFIG_SYS_EXTRA_OPTIONS. This is deprecated. Thus
add new Kconfig options for those board variants.
MIPS: kconfig: add options for dbau1x00 board variant select
The dbau1x00 board can be built with various variants which are
configured via CONFIG_SYS_EXTRA_OPTIONS. This is deprecated. Thus
add new Kconfig options for those board variants.
Add new Kconfig option to let the user select the targets
CPU type. Each target have to select SUPPORTS_CPU_MIPS[32,64]_R[1,2]
to indicate which CPU types are supported.
Add new Kconfig option to let the user select the targets
endianess. Each target have to select SUPPORTS_BIG_ENDIAN
and/or SUPPORTS_LITTLE_ENDIAN to indicate which endianess
is supported.
Masahiro Yamada [Thu, 30 Oct 2014 06:50:15 +0000 (15:50 +0900)]
Remove the CREDITS file
This file is not maintained these days.
We use MAINTAINERS for the maintainership of the supported boards.
For dead boards, we have some clues in doc/README.scrapyard and
also imperishable history in git-log.
Masahiro Yamada [Thu, 30 Oct 2014 09:28:01 +0000 (18:28 +0900)]
Remove unused files
[1] arch/arm/include/asm/arch-at91/at91_shdwn.h
The top9000 was the last board to use this header file.
It was removed by commit d58a9451e733 (ppc/arm: zap EMK boards).
[2] board/matrix_vision/common/*
Some Matrix Vision boards were dropped by commit e7a565638a7a
(powerpc: mpc83xx: remove board support for MERGERBOX and MVBLM7)
and commit af55e35d3389
(powerpc: mpc5xxx: remove board support for MVBC_P and MVSMR).
Since then these files have been unused.
[3] include/usb/omap1510_udc.h
The omap5912osk was the last board to use this header file.
It was removed by commit 62d636aa2ac2
(omap: remove omap5912osk board support).
This enables the VMMC2 LDO, which powers the MMC2 device.
When the device starts from MMC2, this has already been enabled by the BootROM,
but when starting from peripheral boot (USB, UART), it is not the case.
Stefan Roese [Tue, 28 Oct 2014 10:32:24 +0000 (11:32 +0100)]
tools/kwbimage: Fix compilation warning
This patch fixes a compilation warning of kwbimage.c:
tools/kwbimage.c: In function ‘kwbimage_set_header’:
tools/kwbimage.c:784:8: warning: ‘headersz’ may be used uninitialized in this function [-Wmaybe-uninitialized]
memcpy(ptr, image, headersz);
^
Instead of using multiple if statements, use a switch statement with
a default entry. And return with error if an unsupported version
is configured in the cfg file.
Andreas Bießmann [Fri, 24 Oct 2014 21:39:11 +0000 (23:39 +0200)]
tools/kwbimage.c: fix build on darwin
kwbimage uses get_current_dir_name(3) which is a gnu extension and not
available on darwin host. Fix this by converting to portable getcwd(3)
function.
This patch fixes the following error:
---8<---
HOSTCC tools/kwbimage.o
tools/kwbimage.c:399:16: warning: implicit declaration of function 'get_current_dir_name' is invalid in C99 [-Wimplicit-function-declaration]
char *cwd = get_current_dir_name();
^
tools/kwbimage.c:399:10: warning: incompatible integer to pointer conversion initializing 'char *' with an expression of type 'int' [-Wint-conversion]
char *cwd = get_current_dir_name();
^ ~~~~~~~~~~~~~~~~~~~~~~
2 warnings generated.
...
Undefined symbols for architecture x86_64:
"_get_current_dir_name", referenced from:
_image_headersz_v1 in kwbimage.o
ld: symbol(s) not found for architecture x86_64
--->8---
Nikolay Dimitrov [Sun, 12 Oct 2014 10:47:51 +0000 (13:47 +0300)]
novena: Fix ethernet PHY reset sequence
This patch fixes conflict between PHY pins becoming outputs after reset and
imx6 still driving the pins. It also fixes the reset timing as recommended by
the PHY datasheet.
The older 'mr' variant and the generic variant of the
OT1200 differ in some places. As the name suggests the
generic variant supports more boot devices.
In order to be compatible with the 'mr' variant we define
some 'feature' GPIOs. On the 'mr' variant this pads are
not connected so we define their state with the help
of the internal pullups.
On the generic variant this GPIOs are connected and
represent the state of the hardware.
Markus Niebel [Thu, 23 Oct 2014 13:47:05 +0000 (15:47 +0200)]
tqma6: fix sf detection
Commit 155fa9af95ac5be857a7327e7a968a296e60d4c8 changed the way
to define a GPIO line, which can be used to force CS high
across multiple transactions. In order to fix sf detection
change board code to make use of board_spi_cs_gpio(..).
Marek Vasut [Fri, 24 Oct 2014 21:39:07 +0000 (23:39 +0200)]
ARM: mx6: Add support for Kosagi Novena
Add support for the Kosagi Novena board. Currently supported are:
- I2C busses
- FEC Ethernet
- MMC0, MMC1, Booting from MMC
- SATA
- USB ports
- USB Ethernet
Masahiro Yamada [Mon, 20 Oct 2014 11:45:22 +0000 (20:45 +0900)]
ARM: UniPhier: make pinmon command optional
Add CONFIG_CMD_PINMON to UniPhier-specific Kconfig and make the
"pinmon" command user-configurable. This command can be disabled
via the configuration if users do not need it.
Before this commit, the stack addresses for IRQ and FIQ modes,
IRQ_STACK_START and FIQ_STACK_START, were computed in interrupt_init but
they were not used.
This commit sets the stack pointers for IRQ and FIQ modes.
This commit relocates the exception vectors.
As ARM1176 and ARMv7 have the security extensions, it uses VBAR. For
the other ARM processors, it copies the relocated exception vectors to
the correct address: 0x00000000 or 0xFFFF0000.
A regression was introduced in commit 41623c91. The consequence of that
is the non-relocation of the section .vectors symbols :
_undefined_instruction, _software_interrupt, _prefetch_abort,
_data_abort, _not_used, _irq and _fiq.
Before commit 41623c91, the exception vectors were in a .text section.
The .text section has the attributes allocatable and executable [1].
In commit 41623c91, a specific section is created, called .vectors, with
the attribute executable only.
What have changed between commit 41623c91^ and 41623c91 is the attribute
of the section which contains the exception vectors.
An allocatable section is "a section [that] occupies memory during
process execution" [1] which is the case of the section .vectors.
Adding the lacking attribute (SHF_ALLOC or "a") for the definition of
the section .vectors fixed the issue.
To summarize, the fix has to mark .vectors as allocatable because the
exception vectors reside in "memory during execution" and they need to
be relocated.
This commit introduces a Kconfig symbol for each ARM CPU:
CPU_ARM720T, CPU_ARM920T, CPU_ARM926EJS, CPU_ARM946ES, CPU_ARM1136,
CPU_ARM1176, CPU_V7, CPU_PXA, CPU_SA1100.
Also, it adds the CPU feature Kconfig symbol HAS_VBAR which is selected
for CPU_ARM1176 and CPU_V7.
For each target, the corresponding CPU is selected and the definition of
SYS_CPU in the corresponding Kconfig file is removed.
Also, it removes redundant "string" type in some Kconfig files.
Marek Vasut [Sun, 19 Oct 2014 22:23:42 +0000 (00:23 +0200)]
i2c: mxs: Pass the i2c_adapter around
This patch permeates the struct i2c_adapter throughout the driver,
so that it can be used to determine which adapter should be used.
Note that the driver still supports only one adapter. Note that
the patch does shuffle mxs_i2c_set_bus_speed() to the top of the
driver, but the function remains unchanged.
projects / u-boot.git / log