Enabling timers and clocks in PRCMU and cleaning up mailbox.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: John Rigby <john.rigby@linaro.org>
This is to allow the prcmu functions to be used by multiple
u8500-based processors.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: John Rigby <john.rigby@linaro.org>
We can safely use the same reset code written in C for both Davinci and
C6X platforms. In addition the C version of the code is marginally
smaller on Davinci.
Tested-by: Matt Porter <mporter@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
This patch updates secure_emif_sdram_config with the
same value written to sdram_config during ddr3 initialization.
During suspend/resume, this value is copied into sdram_config.
With this, a write to sdram_config at the end of resume sequence
which triggers an init sequence can be avoided.
Without this register write in place, the DDR_RESET line goes
low for a few cycles during resume which is a violation of the
JEDEC spec.
Signed-off-by: Satyanarayana, Sandhya <sandhya.satyanarayana@ti.com>
Also enable the ohci port on hawkboard. These additions result in an
increased u-boot size -- adjust the same accordingly in the board's
config.
Move the usb header for da8xx platforms under arch-davinci.
Signed-off-by: Sughosh Ganu <urwithsughosh@gmail.com>
Make sure that when we setup the stack before calling s_init() we have
the stack have 8-byte alignment for ABI compliance.
Tested-by: Allen Martin <amartin@nvidia.com>
Signed-off-by: Tom Rini <trini@ti.com>
Make the lowlevel_init function that these platforms have which just
sets up the stack and calls a C function available to all armv7
platforms. As part of this we change some of the macros that are used
to be more clear. Previously (except for am335x evm) we had been
setting CONFIG_SYS_INIT_SP_ADDR to a series of new defines that are
equivalent to simply referencing NON_SECURE_SRAM_END. On am335x evm we
should have been doing this initially and do now.
Cc: Sricharan R <r.sricharan@ti.com>
Tested-by: Allen Martin <amartin@nvidia.com>
Signed-off-by: Tom Rini <trini@ti.com>
- Correct the MMC1 base offset
- Remove MMC2 (that area is reserved and not MMC2).
- Add the real BOOT_DEVICE_MMC2 value
Signed-off-by: Tom Rini <trini@ti.com>
This gets us rid of duplication of the same file.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Acked-by: Stefano Babic <sbabic@denx.de>
If VDDIO has a brownout, then the VDD5V_GT_VDDIO becomes unreliable
but this wasn't clear on code so a comment has been added to clarify
it.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
As the register accessing mode is the same for all i.MXS SoCs we ought
to use 'mxs' prefix intead of 'mx28'.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Most code can be shared between i.MX23 and i.MX28 as both are from
i.MXS family; this source directory structure makes easy to share code
among them.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Acked-by: Stefano Babic <sbabic@denx.de>
The mx28 prefix has been added to the initialization data and function
so it is clear by which SoC it is used as i.MX233 will have a specific
one. While on that, we also change it to static.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Acked-by: Marek Vasut <marex@denx.de>
The information now is gathered from HW_DIGCTL_CHIPID register and
includes the chip modem and revision on the output.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
In case an unidentified CPU type is detected it now returns
i.MX??, in a const char.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Cc: Marek Vasut <marex@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Markus Hubig <mhubig@imko.de>
Cc: Andreas Bießmann <andreas.devel@googlemail.com>
Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
Add at91sam9x5ek board support, this board support the following SoCs
AT91SAM9G15, AT91SAM9G25, AT91SAM9G35, AT91SAM9X25, AT91SAM9X35
Using at91sam9x5ek_nandflash to configure for the board
Now only supports NAND with software ECC boot up
Signed-off-by: Bo Shen <voice.shen@atmel.com>
[move MAINTAINERS entry to right place]
Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
For the DA8xx family of SoCs, the set_cpu_clk_info() function was not
initialising the DSP frequency, leading to 'bdinfo' command output such as:
[...snip...]
ARM frequency = 300 MHz
DSP frequency = -536870913 MHz
DDR frequency = 300 MHz
This commit provides a separate implementation of set_cpu_clk_info() for
the DA8xx SoCs that initialises the DSP frequency to zero (since
currently the DSP is not enabled by U-Boot on any DA8xx platform). The
separate implementation is justified because there is no common code
between DA8xx and the other SoC families. It is now much easier to
understand the flow of the two separate functions.
Signed-off-by: Laurence Withers <lwithers@guralp.com>
Cc: Tom Rini <trini@ti.com>
Cc: Hadli, Manjunath <manjunath.hadli@ti.com>
Cc: Heiko Schocher <hs@denx.de>
Replace a magic number for the DDR2/mDDR PHY clock ID with a proper
definition. In addition, don't request this clock ID on DA830 hardware,
which does not have a DDR2/mDDR PHY (or associated PLL controller).
Signed-off-by: Laurence Withers <lwithers@guralp.com>
Cc: Tom Rini <trini@ti.com>
Cc: Prabhakar Lad <prabhakar.csengg@gmail.com>
On the DA830, UART2's clock is derived from PLL controller 0 output 2.
On the DA850, it is in the ASYNC3 group, and may be switched between PLL
controller 0 or 1. Fix the definition of the ID to match.
Signed-off-by: Laurence Withers <lwithers@guralp.com>
Cc: Tom Rini <trini@ti.com>
Cc: Prabhakar Lad <prabhakar.csengg@gmail.com>
Tidy up the clock IDs defined for the DA8xx SOCs. With this new structure in
place, it is clear how to define new clock IDs, and how these map to the
numbers presented in the technical reference manual.
Signed-off-by: Laurence Withers <lwithers@guralp.com>
Cc: Tom Rini <trini@ti.com>
Cc: Prabhakar Lad <prabhakar.csengg@gmail.com>
Signed-off-by: Tom Rini <trini@ti.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Albert Aribaud <albert.u.boot@aribaud.net>
Cc: U-Boot DM <u-boot-dm@lists.denx.de>
Cc: Tom Rini <trini@ti.com>
Acked-by: Tom Rini <trini@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
- Move definition of the EEPROM contents to <asm/arch/sys_proto.h>
- Make some defines a little less generic now.
- Pinmux must be done by done by SPL now.
- Create 3 pinmux functions, uart0, i2c0 and board.
- Add pinmux specific to Starter Kit EVM for MMC now.
Signed-off-by: Tom Rini <trini@ti.com>
- Board requires gpio0 #7 to be set to power DDR3.
- Board uses DDR3, add a way to determine which DDR type to call
config_ddr with.
- Both of the above require filling in the header structure early, move
it into the data section.
Signed-off-by: Tom Rini <trini@ti.com>
The intention has always been (and boards are to support) an i2c EEPROM
that will identify what hardware they are, allowing a single binary to
support multiple boards. As such, remove the 'evm.c' file as there is
nothing EVM centric in it currently, only SoC peripheral configuration.
Signed-off-by: Tom Rini <trini@ti.com>
In order to support DDR3 as well as DDR2, we need to perform the same
init sequence, but with different values. So change config_ddr() to
toggle setting pointers/etc for what DDR2 wants, and then calling.
Signed-off-by: Tom Rini <trini@ti.com>
The ddr_regs struct was incorrectly offset after the dt0wiratio0 entry.
Correct this by documenting a missing register that will be used at some
point in the future (when write leveling is supported). Further, the
cmdNcs{force,delay} fields are undocumented and we have been setting
them to zero, remove. Next, setting of the
'DATAn_REG_PHY_USE_RANK0_DELAYS field belongs with the rest of the
ddr_data entries, so program it there. Finally, comment on how we are
configuring the DATA1 registers that correspond to the DATA0 (dt0)
registers defined in the struct.
Signed-off-by: Tom Rini <trini@ti.com>
The various ratio1 fields are not documented in any of the documentation
I can find. Removing these and testing has yielded success, so remove
the code that sets them and move their locations into the reserved
fields.
Signed-off-by: Tom Rini <trini@ti.com>
This function sets a number of related registers to the same value (the
registers in question all have the same field descriptions and are
related in operation). Rather than defining a struct and setting the
value repeatedly, just pass in the value.
Signed-off-by: Tom Rini <trini@ti.com>
Rather than defining our own structs to note what to use when
programming the EMIF and related re-use the emif_regs struct.
Signed-off-by: Tom Rini <trini@ti.com>
A number of memory initalization functions were int and always returned
0. Further it's not feasible to be doing error checking here, so simply
turn them into void functions.
Signed-off-by: Tom Rini <trini@ti.com>
- Remove the call to set ddrctrl->ddrioctrl as it's all zeros.
- Comment what we're really setting in ddrctrl->ddrckectrl which is that
we're operating in the normal mode where EMIF/PHY clock is controlled
by the PHY.
Signed-off-by: Tom Rini <trini@ti.com>
EMIF parameters are calculated based on the AC timing
parameters from the SDRAM datasheet and the DDR frequency.
Current values for these paramters in AM335x U-Boot code,
though reliable, are not fully optimal. The most optimal
settings can be derived based on the guidelines published
at [1]. A pre-computed set of values with the most optimum
settings for AM335x EVM and BeagleBone can be found at [2].
[1] http://processors.wiki.ti.com/index.php/AM335x_EMIF_Configuration_tips
[2] http://processors.wiki.ti.com/index.php/OMAP_and_Sitara_CCS_support#AM335x
Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
Signed-off-by: Tom Rini <trini@ti.com>
Depending on if we have DDR2 or DDR3 on the board we will need to call
ddr_pll_config with a different value. This call can be delayed
slightly to the point where we know which type of memory we have.
Signed-off-by: Tom Rini <trini@ti.com>
We need to pass in the type of memory that is connected to the board.
The only reliable way to do this is to know what type of board we are
running on (which later will be knowable in s_init()). For now, pass in
the value of DDR2.
Signed-off-by: Tom Rini <trini@ti.com>
Rework the EMIF4/DDR code slightly to setup the structs that
config_cmd_ctrl and config_ddr_data take to be setup at compile time and
mark them as const. This lets us simplify the calling path slightly as
well as making it easier to deal with DDR3.
Signed-off-by: Tom Rini <trini@ti.com>
With the previous bugfix we now don't need to set two different REF_CTRL
values and instead set the final value.
Signed-off-by: Tom Rini <trini@ti.com>
When we change SDRAM_CONFIG this triggers a refresh based on all of the
parameters that we have programmed so we must do this last.
Signed-off-by: Tom Rini <trini@ti.com>
We do not need to check for EMIF_GCLK and L3_GCLK being active. This
was a hold-over from bringup and no longer required.
Signed-off-by: Tom Rini <trini@ti.com>
This patch adds pin mux settings for CPSW switch found on
TI AM335X based boards (MII and RGMII modes).
Signed-off-by: Chandan Nath <chandan.nath@ti.com>
[Ilya: split pinmux into separate patch]
Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com>
This patch adds platform-specific initialization for CPSW
switch on TI AM33XX SoCs.
Signed-off-by: Chandan Nath <chandan.nath@ti.com>
[Ilya: split init out of original patch]
Signed-off-by: Ilya Yanok <ilya.yanok@cogentembedded.com>
follow section 15.2.13.1 (Initializing Following Device Power Up or Reset) of
OMAP-L138 DSP+ARM Processor Technical Reference Manual
Signed-off-by: Mikhail Kshevetskiy <mikhail.kshevetskiy@gmail.com>
Acked-by: Christian Riesch <christian.riesch@omicron.at>
Tested-by: Christian Riesch <christian.riesch@omicron.at>
This reverts commit 5347560f5427bcdd48a563b62180481606ac8044,
which was applied only to get release 2012.07 functional on as
many ARM targets as possible despite mis-aligned accesses.
DA850/OMAP-L138 does not support strict MMC/SD boot mode. SPL will
be in SPI flash and U-Boot image will be in MMC/SD card. SPL will
do the low level initialization and then loads the u-boot image
from MMC/SD card.
Define CONFIG_SPL_MMC_LOAD macro in the DA850/OMAP-L138
configuration file to enable this feature.
Tested-by: Christian Riesch <christian.riesch@omicron.at>
Signed-off-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Rajashekhara, Sudhakar <sudhakar.raj@ti.com>
Signed-off-by: Hadli, Manjunath <manjunath.hadli@ti.com>
initialize baudrate, flags, data and serial initialization,
only when CONFIG_SPL_LIBCOMMON_SUPPORT is defined.
Tested-by: Christian Riesch <christian.riesch@omicron.at>
Signed-off-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Rajashekhara, Sudhakar <sudhakar.raj@ti.com>
Signed-off-by: Hadli, Manjunath <manjunath.hadli@ti.com>
Acked-by: Christian Riesch <christian.riesch@omicron.at>
This patch adds support for MMC/SD on DA850/OMAP-L138.
Tested-by: Christian Riesch <christian.riesch@omicron.at>
Signed-off-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Rajashekhara, Sudhakar <sudhakar.raj@ti.com>
Signed-off-by: Hadli, Manjunath <manjunath.hadli@ti.com>
This patch uses the code in omap-common to support gpio modules 1-3
on am33xx based boards.
It adds base address and register definitions, enables clocks to the
modules, and enables building the common gpio code for CONFIG_AM33XX
as well as CONFIG_OMAP
Signed-off-by: Steve Sakoman <steve@sakoman.com>
This patch moves all bootcount implementations into a common
directory: drivers/bootcount. The generic bootcount driver
is now usable not only by powerpc platforms, but others as well.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Heiko Schocher <hs@denx.de>
Cc: Valentin Longchamp <valentin.longchamp@keymile.com>
Cc: Christian Riesch <christian.riesch@omicron.at>
Cc: Manfred Rudigier <manfred.rudigier@omicron.at>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Rob Herring <rob.herring@calxeda.com>
Cc: Reinhard Meyer <reinhard.meyer@emk-elektronik.de>
Tested-by: Valentin Longchamp <valentin.longchamp@keymile.com>
Tested-by: Christian Riesch <christian.riesch@omicron.at>
Acked-by: Rob Herring <rob.herring@calxeda.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
All the global flag defines are the same across all arches. So unify them
in one place, and add a simple way for arches to extend for their needs.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
struct s3c24x0_i2c is being moved to common local header file so that
the same can be used by s3c series and exynos series SoCs.
Signed-off-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Doug Anderson <dianders@chromium.org>
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
This patch defined EXYNOS5_I2C_SPACING used to calculate I2C channel
base address.
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
This patch adds the base address for I2C.
Signed-off-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
This adds i2c clock information for EXYNOS5.
Signed-off-by: Alim Akhtar <alim.akhtar@samsung.com>
Signed-off-by: Doug Anderson <dianders@chromium.org>
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
bad_cache_range actually returned true if the range was OK, but it was used
according to its name, which resulted in all valid dcache range invalidate/flush
operations being dropped. Hence, most DMA transfers resulted in garbage data.
This patch renames this function according to what it does, and it fixes the
interpretation of its return value by other functions. The chosen naming is the
same as for ARM926EJ-S in order to be consistent.
Signed-off-by: Benoît Thébaudeau <benoit.thebaudeau@advansee.com>
Cc: Albert Aribaud <albert.u.boot@aribaud.net>
Acked-by: Stefano Babic <sbabic@denx.de>
Use the same name as defined in the datasheet.
DSP_CLK -> DISP_CLK
Signed-off-by: Ashok Kumar Reddy Kourla <ashokkourla2000@gmail.com>
Acked-by: Marek Vasut <marex@denx.de>
Create a weak-aliased arch_cpu_init, so that we can get rid of CONFIG_ARCH_CPU_INIT
and always call arch_cpu_init.
This way we do not need to define CONFIG_ARCH_CPU_INIT in every board file, since
arch_cpu_init() is supposed to handle common CPU level code.
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Acked-by: Marek Vasut <marex@denx.de>
Acked-by: Stefano Babic <sbabic@denx.de>
Recent compiler generates unaligned memory access in armv7 default.
But current U-Boot does not allow unaligned memory access, so it causes
data abort exception.
This patch add compile option "-mno-unaligned-access" if it is available.
Signed-off-by: Tetsuyuki Kobayashi <koba@kmckk.co.jp>
Tested-by: Gary Thomas <gary@mlbassoc.com>
In currently, when __bss_start is equal to __bss_end__,
The bss loop will clear all the things in memory space.
But just only when __bss_end__ greater than __bss_start__,
we do the clear bss section operation.
Signed-off-by: Zhong Hongbo <bocui107@gmail.com>
Acked-by: Andreas Bießmann <andreas.devel@googlemail.com>
* 'next' of git://git.denx.de/u-boot-video:
ipu_common: Add ldb_clk for use in parenting the pixel clock
ipu_common: Do not hardcode the ipu_clk frequency
ipu_common: Rename MXC_CCM_BASE
ipu_common: Let clk_ipu_enable/disable only run on MX51 and MX53
ipu_common: Only apply the erratum to MX51
video: Rename CONFIG_VIDEO_MX5
mx6: Allow mx6 to access the IPUv3 registers
common lcd: minor coding style changes
Signed-off-by: Wolfgang Denk <wd@denx.de>
The following platforms had their config files changed
flea3, imx31_phycore, mx35pdk, mx53ard, mx53evk, mx53smd
and mx53loco.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Acked-by: Stefano Babic <sbabic@denx.de>
* 'master' of git://git.denx.de/u-boot-arm:
tegra: define fdt_load/fdt_high variables
tegra: enable bootz command
tegra: usb: Fix device enumeration problem of USB1
tegra: trimslice: set up serial flash pinmux
tegra: add pin_mux_spi() board initialization function
tegra: add GMC/GMD funcmux entry for SFLASH
tegra: bootcmd: start USB only when needed
tegra: bootcmd enhancements
tegra: add enterrcm command
tegra: enable CONFIG_ENV_VARS_UBOOT_CONFIG
Add env vars describing U-Boot target board
tegra: usb: fix wrong error check
tegra: add ULPI on USB2 funcmux entry
tegra: fix leftover CONFIG_TEGRA2_MMC & _SPI build switches
tegra: Add Tamonten Evaluation Carrier support
tegra: Use SD write-protect GPIO on Tamonten
tegra: Implement gpio_early_init() on Tamonten
tegra: Allow boards to perform early GPIO setup
tegra: plutux: Add device tree support
tegra: medcom: Add device tree support
tegra: Rework Tamonten support
beagle: add eeprom expansion board info for bct brettl4
Signed-off-by: Wolfgang Denk <wd@denx.de>
* 'master' of git://git.denx.de/u-boot-usb:
CONFIG: EXYNOS5: USB: Enable USB 2.0 on smdk5250
EXYNOS5: USB: Fix incorrect USB base addresses
EXYNOS: Add power Enable/Disable for USB-EHCI
USB: EXYNOS: Set USB 2.0 HOST Link mode
EXYNOS5: Add structure for PMU registers
EXYNOS5: Fix system register structure
USB: EXYNOS: Incorporate EHCI review comments
Signed-off-by: Wolfgang Denk <wd@denx.de>
A known hardware issue of USB1 port where bit 1 (connect status
change) of PORTSC register will be set after issuing Port Reset
(like "usb reset" in u-boot command line).
This will be treated as an error and stops later device enumeration.
Therefore we clear that bit after Port Reset in order to proceed
later device enumeration.
Signed-off-by: Jim Lin <jilin@nvidia.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>
Tegra's boot ROM supports a mode whereby code may be downloaded and flash
programmed over a USB connection. On dev boards, this is typically entered
by holding down a "force recovery" button and resetting the CPU. However,
not all boards have such a button (one example is the Compulab Trimslice),
so a method to enter RCM from software is useful.
This change implements the command "enterrcm" to do this, and enables it
for all Tegra boards by default. Even on boards other than Trimslice,
controlling this over a UART may be useful, e.g. to allow simple remote
control without the need for mechanical button actuators, or hooking up
relays/... to the button.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
loop_count runs down from 10000, so the correct condition to error out
is ==0.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
CC: Stephen Warren <swarren@wwwdotorg.org>
CC: Tom Warren <twarren.nvidia@gmail.com>
Tested-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>
This is needed as a prerequisite for Tegra USB ULPI support
within U-Boot.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
CC: Stephen Warren <swarren@wwwdotorg.org>
CC: Tom Warren <twarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
This patch corrects the base addresses for USB_PHY and USB_OTG.
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
Acked-by: Chander Kashyap <chander.kashyap@linaro.org>
Acked-by: Minkyu Kang <mk7.kang@samsung.com>
This patch adds functions to enable/disable the power of USB
host controller for EXYNOS5.
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Signed-off-by: Che-Liang Chiou <clchiou@chromium.org>
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
This patch corrects the SYSREG structure.
We have removed the sysreg.h added in the previous patchset
version as the sysreg structure is already defined in system.h.
Signed-off-by: Rajeshwari Shinde <rajeshwari.s@samsung.com>
Acked-by: Minkyu Kang <mk7.kang@samsung.com>
This reverts commit fa042186b9.
It causes build warnings like this:
cpu.c:48:1: warning: -fstack-usage not supported for this target
[enabled by default]
Signed-off-by: Wolfgang Denk <wd@denx.de>
The following is the cleanup sequence in arch/arm/cpu/armv7/cpu.c
int cleanup_before_linux(void)
{
...
...
dcache_disable();
v7_outer_cache_disable();
invalidate_dcache_all();
}
1) invalidate_dcache_all call expects that all the caches has been
flushed, invalidated and there are no dirty entries prior to its
execution. In the above sequence dcache_disable() flushes, invalidates
the caches and turns off the mmu. But after it cleanups the cache
and before the mmu is disabled there is a cp_delay() function which
has STR instruction. On certain cores like the cortex-a15, cache hit
and a write can happen to a cache line even when the dcache is
disabled. So the above mentioned STR instruction creates a dirty entry
after cleaning. The mmu gets disabled after this.
2) invalidate_dcache_all invalidates the cache lines. Again on
cores like cortex-a15, invalidate instruction flushes the dirty
line as well. So some times the dirty line from sequence 1
can corrupt the memory resulting in a crash.
Fixing this by moving the get_cr() and cp_delay() calls before
cleaning up the cache, thus avoiding the dirty entry.
Signed-off-by: R Sricharan <r.sricharan@ti.com>
Introduce arch_early_init_r() function, which can be useful for doing
early initialization after relocation has happened.
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
