Sets up MSELECT to let main CPUs talk to peripheral devices and starts
high performance A9 CPU cluster.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Allows to talk to external PMIC devices to bring up CPU rail.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
For proper startup we need to give clocks and IO signals some time to
stabilize. Tegra2 got away without them, but Tegra3 seems to be a bit
pickier.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Running at 1GHz, rather than 13MHz certainly makes things a bit faster.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
We run the system bus from the OSC clock during init, to avoid crashing
the system while reconfiguring the PLLs.
Switch to a more reasonable clock when we are done with this.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
To keep things clean I removed all support for the old way to build
images. There is now a single tegra_v7 defconfig which builds both
supported Tegra boards as images.
The new image generation also paves the way for integration of the
tegra-cbootimage tool to produce directly flashable images.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
It seems GCC 4.8 tries to be clever by not inlining some of those
functions. This causes havok, as it's absolutely required to inline
the early startup function, otherwise we may end up calling ARMv7 code
on the ARMv4 AVP.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
gpiolib user have nothing to define in their machine
specific gpio.h, so do not include it.
The only thing they could define would be ARCH_NR_GPIOS,
but currently no architecture defines it. Should an architecure
feel the need to do it this would be a good opportunity to
get rid of this limitation.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Only UART clocks are included for now, but the code should cover
other peripherals needs, too.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
This removes the existing Tegra CAR driver and replaces it with code
ported from the Linux clock framework.
In the current state only the relevant PLLs are supported, but this is
no functional regression from the existing code.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
In order to properly bring up the system PLLs we need a reliable
clocksource. To break the circular dependency between the clocksource
and the CAR driver, get the OSC frequency with a lowlevel function.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
See also:
commit eb84709192
Author: Sascha Hauer <s.hauer@pengutronix.de>
Date: Mon Mar 25 15:18:38 2013 +0100
clk: remove unused __clk_[get|put]
This is some unused code resulting from copying stuff from
the kernel. Remove it.
Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Taken from the Linux kernel, simplified and reworked to match barebox.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Tested-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
ODMdata tells us which UART to use for debugging purposes. This is
agreed upon in both the upstream Linux kernel and U-Boot, so do it the
same way in barebox.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Tested-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
All Tegra20 boards have a common startup sequence. Also there is an
agreement on how to find out about the installed amount of RAM and other
information needed by early startup. So as there is really no need to do
any lowlevel stuff per board, we can just do it at the ARCH level.
This also enables the first stage loading of barebox by detecting the
currently running CPU and booting the main CPU cluster if neccesary.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Tested-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Currently only implements system wide reset functionality.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Tested-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Only a basic set of clocks is supported. This is a temporary solution
and will go away as soon as the port of the Tegra common clock code from
the Linux kernel is ready to go.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Tested-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
putc already is a regular barebox function. To avoid conflicts and
confusions just let architectures define PUTC_LL directly instead
of going through this addiotional redirection.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
The FSF address has changed in the past. Instead of updating it
each time the address changes, just drop it completely treewide.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Lucas Stach
Sascha Hauer
Antony Pavlov