There exists a tiny MMU, configurable only by the VC (running the
closed firmware), which maps from the ARM's physical addresses to bus
addresses. These bus addresses determine the caching behavior in the
VC's L1/L2 (note: separate from the ARM's L1/L2) according to the top
2 bits. The bits in the bus address mean:
From the VideoCore processor:
0x0... L1 and L2 cache allocating and coherent
0x4... L1 non-allocating, but coherent. L2 allocating and coherent
0x8... L1 non-allocating, but coherent. L2 non-allocating, but coherent
0xc... SDRAM alias. Cache is bypassed. Not L1 or L2 allocating or coherent
From the GPU peripherals (note: all peripherals bypass the L1
cache. The ARM will see this view once through the VC MMU):
0x0... Do not use
0x4... L1 non-allocating, and incoherent. L2 allocating and coherent.
0x8... L1 non-allocating, and incoherent. L2 non-allocating, but coherent
0xc... SDRAM alias. Cache is bypassed. Not L1 or L2 allocating or coherent
The 2835 firmware always configures the MMU to turn ARM physical
addresses with 0x0 top bits to 0x4, meaning present in L2 but
incoherent with L1. However, any bus addresses we were generating in
the kernel to be passed to a device had 0x0 bits. That would be a
reserved (possibly totally incoherent) value if sent to a GPU
peripheral like USB, or L1 allocating if sent to the VC (like a
firmware property request). By setting dma-ranges, all of the devices
below it get a dev->dma_pfn_offset, so that dma_alloc_coherent() and
friends return addresses with 0x4 bits and avoid cache incoherency.
This matches the behavior in the downstream 2708 kernel (see
BUS_OFFSET in arch/arm/mach-bcm2708/include/mach/memory.h).
Signed-off-by: Eric Anholt <eric@anholt.net>
Tested-by: Noralf Trønnes <noralf@tronnes.org>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
Acked-by: Lee Jones <lee.jones@linaro.org>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Device tree node names should contain the node's reg property address value.
The i2c0 node was apparently forgotten in commit 25b2f1bd0b (ARM: bcm2835:
node name unit address cleanup).
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
This patch converts all bcm2835 dts and dtsi files to use the pinctrl
header file.
Reviewed-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
DT nodes should be named according to the type of object that they
represent rather than the identity. DT nodes that contain a reg
property should include a unit address in their name. Fix these issues.
Add clock-output-names properties to the nodes so that the clocks get
named something meaningful. This works around the fact that the fixed
clock driver names clocks after the short node name, i.e. not including
the unit address.
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
DT nodes that contain a reg property should include a unit address in
their name. Add the missing unit addresses.
The unit address in a node name must match the value in the reg property.
Fix the cases where they don't match.
Don't fix the /clocks/* node names yet; that causes problems the clock
driver to attempt to register multiple clocks with the same name, which
fails.
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
This adds the definitions for the BCM2835 I2S driver
to the device tree. Some GPIO settings are needed for
the correct pin functions.
Signed-off-by: Florian Meier <florian.meier@koalo.de>
[swarren: fixed DT node sort order, simplified DT label name, removed
RPI .dts file changs, since use of I2S is a user-added option.]
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
This adds the definitions for the BCM2835 dmaengine driver
to the device tree. The dma-channel-mask is currently
fixed. Later it should be set via the firmware.
Signed-off-by: Florian Meier <florian.meier@koalo.de>
[swarren, fixed DT node sort order]
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
The following patch enables performance counter support on Raspberry-Pi.
We have this working on the 2708 based rasp-pi kernels by manually putting
the device registration in the platform files.
This change does things properly in a device tree. The boot messages look
proper, but my rasp-pi hangs somewhere in USB enabling when running a
stock 3.13-rc6 kernel so I have been unable to fully test this change.
I also understand that the rasp-pi 1176 pmu support is missing the
overflow interrupt. I'm not sure if that's true of all 2835
implementations. If not, then this patch will need to be changed a bit.
Signed-off-by: Vince Weaver <vincent.weaver@maine.edu>
[swarren, fixed DT node sort order]
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
The BCM2835 SoC contains a DWC2 USB controller. Add this to the DT.
Set up the pin controller to fully enable the USB controller on the
Raspberry Pi. The GPIO setup works because the default output value for
GPIO 6 (LAN_RUN/n_reset) just happens to be 1, which enables the
USB/LAN chip.
Note that you'll need a U-Boot which enables power to the USB controller;
search for U-Boot patch "ARM: rpi_b: power on SDHCI and USB HW modules".
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Olof Johansson <olof@lixom.net>
The I2C controller node needs #address-cells and #size-cells properties,
but these are currently missing. Add them. This allows child nodes to be
parsed correctly.
Cc: stable@vger.kernel.org
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Olof Johansson <olof@lixom.net>
Stephen Warren reported the recent commit 78506f2 (add support for
extended FIFO-size of PL011-r1p5) breaks the serial port on the
BCM2835 ARM SoC.
A UART compatible with the ARM PL011-r1p5 should have 32-deep FIFOs.
The BCM2835 UART just looks like an ARM PL011-r1p5, but has 16-deep
FIFOs just like PL011-r1p4 or earlier revisions. As a workaround for
this compatibility issue, this patch overrides the HW UART periphid
register values with the actually compatible UART periphid 0x00241011
(r1p3 or r1p4).
Reported-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Jongsung Kim <neidhard.kim@lge.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Olof Johansson <olof@lixom.net>
This branch adds two devices to the BCM2835 SoC device tree: the SPI
controller and the HW random number generator.
The SPI controller isn't actually instantiated in the Raspberry Pi
device tree, since there are no on-board SPI devices; it's up to the
end-user to modify their own device-tree to describe whatever they
have attached.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=4Sjv
-----END PGP SIGNATURE-----
Merge tag 'bcm2835-for-3.10-dt' of git://git.kernel.org/pub/scm/linux/kernel/git/swarren/linux-rpi into next/dt
From Stephen Warren <swarren@wwwdotorg.org>:
ARM: bcm2835: device tree updates
This branch adds two devices to the BCM2835 SoC device tree: the SPI
controller and the HW random number generator.
The SPI controller isn't actually instantiated in the Raspberry Pi
device tree, since there are no on-board SPI devices; it's up to the
end-user to modify their own device-tree to describe whatever they
have attached.
* tag 'bcm2835-for-3.10-dt' of git://git.kernel.org/pub/scm/linux/kernel/git/swarren/linux-rpi:
ARM: bcm2835: add Broadcom BCM2835 RNG to the device tree
ARM: bcm2835: add SPI device to DT
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
This adds a device tree binding for random number generator present on
Broadcom BCM2835 SoC, used in Raspberry Pi and Roku 2 devices.
Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
Tested-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
The BCM2835 has a single instance of the "SPI0"-type SPI master
controller. Instantiate it in the SoC .dtsi file, Don't enable it in
the Raspberry Pi board .dts file, since we have no idea what is actually
connected, and hence no idea what to set the bus clock rate to.
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
BCM2835-ARM-Peripherals.pdf states that the I2C module's input clock is
nominally 150MHz, and that value is currently reflected in bcm2835.dtsi.
However, practical measurements show that the rate is actually 250MHz,
and this agrees with various downstream kernels.
Switch the I2C clock's frequency to 250MHz so that the generated bus
clock rate is accurate.
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Olof Johansson <olof@lixom.net>
Both clock nodes in the current device tree are named "clock" and hence
end up being the same node. Rename the nodes to different names to avoid
this. In fact, fixed-clock uses the node name as the clock name, so name
the nodes after the clock they represent. Move the clocks into a
"clocks" sub-node to group them and avoid any possible naming conflicts
with other nodes also named after the device type.
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
The BCM2835 has 3 identical I2C controllers. Instantiate them all in the
SoC .dtsi file, and enable the relevant two in the Raspberry Pi board
.dts file.
Note that on the Raspberry Pi Model B revision 1, I2C0 is connected to
the general-purpose expansion header, and I2C1 is connected to the camera
connector. Revision 2 of the board swaps these assignments:-(
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Add the SDHCI device node to the SoC DT file. Add a dummy fixed-clock
to satisfy the SDHCI driver's clock lookup; eventually this should be
replaced by a real clock implementation. Add board specific properties
to the Raspberry Pi board file.
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Enable GPIO and pinctrl in Kconfig.
Add required <mach/gpio.h> for gpiolib.
Instantiate the BCM2835 GPIO module in bcm2835.dtsi.
Add a pinctrl definition to bcm2835-rpi-b.dts that sets up all of the
board's required pinmux configuration. GPIO aren't specified; that's
left to gpio_request().
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Implement the machine restart hook using the SoC's watchdog timer module.
To support this, define a DT binding for the watchdog module, and add it
to the device tree.
The downstream rpi-split branch contains a full watchdog timer driver
implementation, which also implements the restart hook. However, the
restart function is largely separate from the watchdog driver, so for
simplicity, the restart hook is implemented here directly in the main
machine source file.
Overall structure (separate setup/restart) functions derived from the
picoxcell ARM support.
Watchdog register IO sequence taken from code by Simon Arlott. Note that
the watchdog module is not documented in BCM2835-ARM-Peripherals.pdf.
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split as of 2012/09/08, and modified as follows:
* s/bcm2708/bcm2835/.
* Modified device tree vendor prefix.
* Modified UART DT node to use a unit-address to create unique UART node
names, rather than using non-type names "uart0" and "uart1".
Note that UART 1 (the Broadcom "mini UART") is not yet present, but
I'm naming the DT node in anticipation that it will be added.
Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
The System Timer peripheral provides four 32-bit timer channels and a
single 64-bit free running counter. Each channel has an output compare
register, which is compared against the 32 least significant bits of the
free running counter values, and generates an interrupt.
Timer 3 is used as the Linux timer.
The BCM2835 also contains an SP804-based timer module. However, it
apparently has significant differences from the standard SP804 IP block,
and Broadcom's documentation recommends using the system timer instead.
This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split as of 2012/09/08, and modified as follows:
* s/bcm2708/bcm2835/.
* Modified device tree vendor prefix.
* Moved to drivers/clocksource/. This looks like the desired location for
such code now.
* Added DT binding docs.
* Moved struct sys_timer bcm2835_timer into time.c to encapsulate it more.
* Simplified bcm2835_time_init() to find one matching node and operate on
it, rather than looping over all matching nodes. This seems more
consistent with other clocksource code.
* Simplified bcm2835_time_init() using of_iomap().
* Renamed struct bcm2835_timer.index to match_mask to better represent its
purpose.
* s/printk(PR_INFO/pr_info(/
Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
The BCM2835 contains a custom interrupt controller, which supports 72
interrupt sources using a 2-level register scheme. The interrupt
controller, or the HW block containing it, is referred to occasionally
as "armctrl" in the SoC documentation, hence the symbol naming in the
code.
This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split as of 2012/09/08, and modified as follows:
* s/bcm2708/bcm2835/.
* Modified device tree vendor prefix.
* Moved implementation to drivers/irchip/.
* Added devicetree documentation, and hence removed list of IRQs from
bcm2835.dtsi.
* Changed shift in MAKE_HWIRQ() and HWIRQ_BANK() from 8 to 5 to reduce
the size of the hwirq space, and pass the total size of the hwirq space
to irq_domain_add_linear(), rather than just the number of valid hwirqs;
the two are different due to the hwirq space being sparse.
* Added the interrupt controller DT node to the top-level of the DT,
rather than nesting it inside a /axi node. Hence, changed the reg value
since /axi had a ranges property. This seems simpler to me, but I'm not
sure if everyone will like this change or not.
* Don't set struct irq_domain_ops.map = irq_domain_simple_map, hence
removing the need to patch include/linux/irqdomain.h or
kernel/irq/irqdomain.c.
* Simplified armctrl_of_init() using of_iomap().
* Removed unused IS_VALID_BANK()/IS_VALID_IRQ() macros.
* Renamed armctrl_handle_irq() to prevent possible symbol clashes.
* Made armctrl_of_init() static.
* Removed comment "Each bank is registered as a separate interrupt
controller" since this is no longer true.
* Removed FSF address from license header.
* Added my name to copyright header.
Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
The BCM2835 is an ARM SoC from Broadcom. This patch adds very basic
support for this SoC.
http://www.broadcom.com/products/BCM2835http://www.raspberrypi.org/wp-content/uploads/2012/02/BCM2835-ARM-Peripherals.pdf
Note that the documentation in the latter .pdf assumes the MMU setup
that's used on the "VideoCore" companion processor, and does not document
physical peripheral addresses. Subtract 0x5e000000 to obtain the physical
addresses. This is accounted for by the ranges property in the /soc node
in the device tree.
The BCM2835 SoC is used in the Raspberry Pi. This patch also adds a
minimal device tree for this board; enough to see some very early kernel
boot messages through earlyprintk. However, this patch does not yet
provide a useful booting system.
http://www.raspberrypi.org/.
This patch was extracted from git://github.com/lp0/linux.git branch
rpi-split from 3-4 months ago, and significantly stripped down and
modified since.
Signed-off-by: Chris Boot <bootc@bootc.net>
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Dom Cobley <dc4@broadcom.com>
Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Eric Anholt