Eric Benard [Tue, 11 Dec 2012 11:36:21 +0000 (11:36 +0000)]
m28evk/mx28evk: fix nand_update_full
- since commit 418396e212b59bf907dbccad997ff50f7eb61b16 nand write.raw
can take the number of page to be written as an argument. nand_update_full
is passing the size (in bytes) to nand write.raw. This value was previously
ignored but now breaks the write.
- this patch updates the default environment of these boards to provide a
pagecount instead of a size to nand write.raw.
- tested on a mx28evk with a 4k page NAND and on a custom board with a
2k page NAND.
Signed-off-by: Eric Bénard <eric@eukrea.com> Cc: Marek Vasut <marex@denx.de> Cc: Fabio Estevam <fabio.estevam@freescale.com> Acked-by: Marek Vasut <marex@denx.de>
Tom Rini [Thu, 13 Dec 2012 20:58:27 +0000 (13:58 -0700)]
omap3_evm: Let CONFIG_EFI_PARTITION be set for SPL
The #ifdef here is not enough to stop part_efi.c from being built, only
being unused. And with recent changes this now leads to warnings. The
easiest solution here is to just let the garbage collection at link time
do its job.
Joe Hershberger [Wed, 12 Dec 2012 04:16:32 +0000 (22:16 -0600)]
tools/env: Add environment variable flags support
Currently just validates variable types as decimal, hexidecimal,
boolean, ip address, and mac address. Call
env_acl_validate_setenv_params() from setenv() in fw_env.c.
If the entry is not found in the env .flags, then look in the static
one. This allows the env to override the static definitions, but prevents
the need to have every definition in the environment distracting you.
Need to build in _ctype for isdigit for Linux.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Joe Hershberger [Wed, 12 Dec 2012 04:16:31 +0000 (22:16 -0600)]
env: Add environment variable flags
Currently just validates variable types as decimal, hexidecimal,
boolean, ip address, and mac address.
If the entry is not found in the env ".flags", then look in the static
one. This allows the env to override the static definitions, but prevents
the need to have every definition in the environment distracting you.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Joe Hershberger [Wed, 12 Dec 2012 04:16:30 +0000 (22:16 -0600)]
env: Add a silent env handler
The silent variable now updates the global data flag anytime it is
changed as well as after the env relocation (in case its value is
different from the default env in such cases as NAND env)
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Joe Hershberger [Wed, 12 Dec 2012 04:16:23 +0000 (22:16 -0600)]
env: Hide '.' variables in env print by default
When printing all variables with env print, don't print variables that
begin with '.'. If env print is called with a '-a' switch, then
include variables that begin with '.' (just like the ls command).
Variables printed explicitly will be printed even without the -a.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Joe Hershberger [Wed, 12 Dec 2012 04:16:21 +0000 (22:16 -0600)]
env: Refactor apply into change_ok
Move the read of the old value to inside the check function. In some
cases it can be avoided all together and at the least the code is only
called from one place.
Also name the function and the callback to more clearly describe what
it does.
Pass the ENTRY instead of just the name for direct access to the whole
data structure.
Pass an enum to the callback that specifies the operation being approved.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Joe Hershberger [Wed, 12 Dec 2012 04:16:18 +0000 (22:16 -0600)]
Make linux kernel string funcs available to tools
isspace() and strim() are not in the typical user-mode string.h, so
put them in a separate compilation unit so that they can be built into
tools that need them independent of the other common string functions.
This allows code shared by u-boot and the linux user-mode tools to link.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Piotr Wilczek [Tue, 11 Dec 2012 10:09:47 +0000 (11:09 +0100)]
gpt: Support for new "gpt" command
New command - "gpt" is supported. It restores the GPT partition table.
It looks into the given environment variable for partitions definition.
It can be enabled at target configuration file with CONFIG_CMD_GPT.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com> Signed-off-by: Piotr Wilczek <p.wilczek@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Lukasz Majewski [Tue, 11 Dec 2012 10:09:43 +0000 (11:09 +0100)]
part:efi: Move part_efi.h file to ./include
This move is necessary to export gpt header and GPT partition entries to be
used with other commands or subsystems.
Additionally the part_efi.h file has been cleaned-up to supress checkpatch's
warnings.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Joe Hershberger [Tue, 11 Dec 2012 17:52:50 +0000 (17:52 +0000)]
mips: serial: Fix busted manual relocation
serial_initialize() must be called after relocation to adjust the
pointers to putc(), getc(), etc. This is busted ever since the
serial driver-model-ification series.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com> Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
It could happen (1 out of 100 times) that NAND did not start up correctly after
warm rebooting, so we end up with various failures or DMA timed out due to a
stalled BCH. When resetting BCH together with GPMI, the issue could not be
observed anymore (after 10000+ reboots). We probably need the consistent state
already before sending commands to NAND. This behaviour was observed in barebox
and kernel, so I assume it affects U-Boot as well. I chose to keep the extra
reset for BCH when changing the flash layout to be on the safe side.
Signed-off-by: Wolfram Sang <w.sang@pengutronix.de> Acked-by: Marek Vasut <marex@denx.de>
Simon Glass [Wed, 5 Dec 2012 14:46:43 +0000 (14:46 +0000)]
console: Enable function to display console info
The CONFIG_SYS_CONSOLE_INFO_QUIET option should suppress the console
information, but allow boards to display it later if required. Adjust
the code to support this.
This is used to avoid printing the information while the LCD display
is not ready, since it only becomes ready when stdio init is complete.
Add a single command to read the EDID information over I2C.
For example:
SMDK5250 # i2c dev 7
Setting bus to 7
SMDK5250 # i2c edid 50
EDID version: 1.4
Product ID code: 305c
Manufacturer: AUO
Serial number: 00000000
Manufactured in week: 0 year: 2011
Video input definition: digital signal, voltage level 0, blank to black
Monitor is non-RGB
Maximum visible display size: 26 cm x 14 cm
Power management features: no active off, no suspend, no standby
Estabilished timings:
Standard timings:
1366x768 60 Hz (detailed)
1366x768 60 Hz (detailed)
Monitor ID: 2VD2K.B116XW
Signed-off-by: Tom Wai-Hong Tam <waihong@chromium.org> Signed-off-by: Sean Paul <seanpaul@chromium.org> Signed-off-by: Simon Glass <sjg@chromium.org>
This implements a library for accessing EDID data from an LCD panel.
This is used to obtain information about the panel such as its
resolution and type.
This is a tidied-up version of the original code pulled from
https://github.com/ynezz/u-boot-edid.
The changes we made are:
- removed bit fields in the struct;
- removed endianness cases in the struct;
- fixed some wrong definitions;
- fixed to fit 80 columns;
- fixed some code styles.
Signed-off-by: Tom Wai-Hong Tam <waihong@chromium.org> Signed-off-by: Simon Glass <sjg@chromium.org>
Simon Glass [Wed, 5 Dec 2012 14:46:36 +0000 (14:46 +0000)]
Add generic hash API
We have a SHA1 command and want to add a SHA256 command also. Instead of
duplicating the code, create a generic hash API which can process
commands for different algorithms.
Simon Glass [Wed, 5 Dec 2012 14:46:35 +0000 (14:46 +0000)]
Add strcasecmp() and strncasecmp()
strncasecmp() is present as strnicmp() but disabled. Make it available
and define strcasecmp() also. There is a only a small performance penalty
to having strcasecmp() call strncasecmp(), so do this instead of a
standalone function, to save code space.
Update the prototype in arch-specific headers as needed to avoid warnings.
Simon Glass [Wed, 5 Dec 2012 14:46:33 +0000 (14:46 +0000)]
sha1: Use const where possible, and unsigned for input len
In preparation for making the hash function common, we may as well use
const where we can. Also the input length cannot be negative, but may
be very large, so use unsigned.
designware_i2c: Fixed the setting of the i2c bus speed
There are three couple (hcnt/lcnt) of registers for each
speed (SS/FS/HS). The driver needs to set the proper couple
of regs according to what speed we are setting.
In the newer versions of designware i2c IP there is the possibility
of configuring it with IC_EMPTYFIFO_HOLD_MASTER_EN=1, which basically
requires the s/w to generate the stop bit condition directly, as
the h/w will not automatically generate it when TX_FIFO is empty.
To avoid generation of an extra 0x0 byte sent as data, the
IC_STOP command must be sent along with the last IC_CMD.
This patch always writes bit[9] of ic_data_cmd even in the
older versions, assuming that it is a noop there.
Marek Vasut [Fri, 30 Nov 2012 18:17:07 +0000 (18:17 +0000)]
mxs: i2c: Implement algorithm to set up arbitrary i2c speed
This algorithm computes the values of TIMING{0,1,2} registers for the
MX28 I2C block. This algorithm was derived by using a scope, but the
result seems correct.
The resulting values programmed into the registers do not correlate
with the contents in datasheet. When using the values from the datasheet,
the I2C clock were completely wrong.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Stefano Babic <sbabic@denx.de> Cc: Fabio Estevam <fabio.estevam@freescale.com> Cc: Wolfgang Denk <wd@denx.de>
Marek Vasut [Fri, 30 Nov 2012 18:17:06 +0000 (18:17 +0000)]
mxs: i2c: Restore speed setting after block reset
The I2C block reset configures the I2C bus speed to strange value.
Read the I2C speed from the block before reseting the block and
restore it afterwards, so the I2C operates correctly. This issue
can be replicated by doing unsuccessful I2C transfer, after such
transfer finishes, the I2C block clock speed is misconfigured.
Marek Vasut [Mon, 12 Nov 2012 14:34:31 +0000 (14:34 +0000)]
i2c: mxs: Fix TIMING2 register value
According to FSL, the value in the TIMING2 register shall be 0x00300030
instead of what's written in the datasheet. This new value correlates
with older STMP36xx datasheet. Issues were detected in Linux when this
register was misconfigured, so write this correct value.
Marek Vasut [Mon, 12 Nov 2012 14:34:28 +0000 (14:34 +0000)]
i2c: mxs: Abstract out the MXS I2C speed setup
This patch pulls out the I2C speed setup from the i2c_init() call
and implements the bus configuration lookup table with register
values that needs to be programmed into the I2C IP to run at
particular speed.
This patch is a first step towards implementing run-time I2C bus
speed configuration for the MXS I2C IP.
Marek Vasut [Mon, 12 Nov 2012 14:34:25 +0000 (14:34 +0000)]
i2c: Use __weak instead of __attribute__((weak, alias))
Use __weak from linux/compiler.h instead of __attribute__((weak, alias))
to define overridable function. This patch is intended as a cleanup patch
to bring some consistency into the code.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Heiko Schocher <hs@denx.de>
In file included from ubifs.h:2137:0,
from ubifs.c:26:
misc.h: In function 'ubifs_zn_dirty':
misc.h:38:2: warning: passing argument 2 of 'test_bit' discards 'const' qualifier from pointer target type [enabled by default]
../include/asm/bitops.h:569:23: note: expected 'volatile void *' but argument is of type 'const long unsigned int *'
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Zhi-zhou Zhang [Sat, 24 Nov 2012 05:07:12 +0000 (05:07 +0000)]
MIPS: fix a latent bug on initialize $gp
If bal is 8 bytes aligned, the _gp will not be 8 bytes aligned.
then the following ld insntrustion generates a Adel exception.
So here make _gp be always aligned in 8 bytes.
Signed-off-by: Zhi-zhou Zhang <zhizhou.zh@gmail.com> Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
The filename buffer is allocated dynamically. It must be cache aligned.
Moreover, it is necessary to erase its content before we use it for
file name operations.
This prevents from corruption of written file names.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Łukasz Majewski [Wed, 5 Dec 2012 08:06:38 +0000 (08:06 +0000)]
fs:ext4:write: Store block device descriptor in file system structure
The device block descriptor (block_dev_desc_t) )shall be stored at
ext4 early code (at ext4fs_set_blk_dev in this case) to be available
for latter use (like put_ext4()).
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Łukasz Majewski [Wed, 5 Dec 2012 08:06:37 +0000 (08:06 +0000)]
fs:ext4:write: Add lldiv and do_div to perform 64-32 bits division
The ext4write code has been using direct calls to 64-32 division
(/ and %).
Officially supported u-boot toolchains (eldk-5.[12].x) generate calls
to __aeabi_uldivmod(), which is niether defined in the toolchain libs
nor u-boot source tree.
Due to that, when the ext4write command has been executed, "undefined
instruction" execption was generated (since the __aeabi_uldivmod()
is not provided).
To fix this error, lldiv() for division and do_div() for modulo have
been used.
Those two functions are recommended for performing 64-32 bit number
division in u-boot.
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Joshua Housh [Sun, 2 Dec 2012 17:09:26 +0000 (17:09 +0000)]
serial_pl011: Set RTS during initialization
If the pl011 is connected to another device which has hardware
flow-control on, characters are never received by the pl011.
Asserting RTS when flow-control is off will have no effect.
This is in line with how Linux behaves.
Signed-off-by: Joshua Housh <joshua.housh@calxeda.com> Tested-by: Marek Vasut <marex@denx.de>
Stefan Reinauer [Sat, 3 Nov 2012 11:41:39 +0000 (11:41 +0000)]
x86: drop unused code in coreboot.c
The function setup_pcat_compatibility() is weak and implemented as empty
function in board.c hence we don't have to override that with another
empty function.
monitor_flash_len is unused, drop it.
Signed-off-by: Stefan Reinauer <reinauer@chromium.org> Signed-off-by: Simon Glass <sjg@chromium.org>
Vadim Bendebury [Sat, 3 Nov 2012 11:41:37 +0000 (11:41 +0000)]
x86: Provide a way to throttle port80 accesses
Some systems (like Google Link device) provide the ability to keep a
history of the target CPU port80 accesses, which is extremely handy
for debugging. The problem is that the EC handling port 80 access is
orders of magnitude slower than the AP. This causes random loss of
trace data.
This change allows to throttle port 80 accesses such that in case the
AP is trying to post faster than the EC can handle, a delay is
introduced to make sure that the post rate is throttled. Experiments
have shown that on Link the delay should be at least 350,000 of tsc
clocks.
Throttling is not being enabled by default: to enable it one would
have to set MIN_PORT80_KCLOCKS_DELAY to something like 400 and rebuild
the u-boot image. With upcoming EC code optimizations this number
could be decreased (new new value should be established
experimentally).
Signed-off-by: Vadim Bendebury <vbendeb@chromium.org> Signed-off-by: Simon Glass <sjg@chromium.org>
Vadim Bendebury [Mon, 3 Dec 2012 13:59:20 +0000 (13:59 +0000)]
x86: Provide tick counter and frequency reference for Intel core architecture
Some u-boot modules rely on availability of get_ticks() and
get_tbclk() functions, reporting a free running clock and its
frequency respectively. Traditionally these functions return number
and frequency of timer interrupts.
Intel's core architecture processors however are known to run the
rdtsc instruction at a constant rate of the so called 'Max Non Turbo
ratio' times the external clock frequency which is 100MHz. This is
just as good for the timer tick functions in question.
Signed-off-by: Vadim Bendebury <vbendeb@chromium.org> Signed-off-by: Simon Glass <sjg@chromium.org>
Duncan Laurie [Mon, 3 Dec 2012 13:59:00 +0000 (13:59 +0000)]
x86: Fix MTRR clear to detect which MTRR to use
Coreboot was always using MTRR 7 for the write-protect
cache entry that covers the ROM and U-boot was removing it.
However with 4GB configs we need more MTRRs for the BIOS
and so the WP MTRR needs to move. Instead coreboot will
always use the last available MTRR that is normally set
aside for OS use and U-boot can clear it before the OS.
Signed-off-by: Duncan Laurie <dlaurie@chromium.org> Signed-off-by: Simon Glass <sjg@chromium.org>
projects / u-boot.git / log