On sun9i, the GTBUS manages transaction priority and bandwidth
for multiple read ports when accessing DRAM. The initialisation
mirrors the settings from Allwinner's boot0 for now, even though
this may not be optimal for all applications (e.g. headless
systems might want to give priority to IO modules).
Adding a common callout to gtbus_init() from the SPL clock init
with a weakly defined implementation in sunxi/clock.c to fallback
to for platforms that don't require this.
[wens@csie.org: Moved gtbus_sun9i.c to arch/arm/mach-sunxi/; style cleanup]
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Acked-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
This adds DRAM initialisation code for sun9i, which calculates the
appropriate timings based on timing information for the supplied
DDR3 bin and the clock speeds used.
With this DRAM setup, we have verified DDR3 clocks of up to 792MHz
(i.e. DDR3-1600) on the A80-Q7 using a dual-channel configuration.
[wens@csie.org: Moved dram_sun9i.c to arch/arm/mach-sunxi/; style cleanup]
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
[hdegoede@redhat.com: Drop some huge non-documenting #if 0 ... #endif blocks]
[hdegoede@redhat.com: Fix checkpatch warnings]
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
OHCI has a known limitation of allowing only 32-bit DMA buffer
addresses, so we have a lot of u32 variables around, which are assigned
to pointers and vice versa. This obviously creates issues with 64-bit
systems, so the compiler complains here and there.
To allow compilation for 64-bit boards which use only memory below 4GB
anyway (and to avoid more invasive fixes), adjust some casts and types
and assume that the EDs and TDs are all located in the lower 4GB.
This fixes compilation of the OHCI driver for the Pine64.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
ARMv7 Tegra boards aren't currently covered by any other travis-ci jobs.
Add a new job to build them.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Use buildman to compile any U-Boot binary tested by test/py. This
re-uses all the work done elsewhere to make buildman work within
Travis-CI, in particular related to toolchain downloading and buildman
config file creation.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Add the LD11 SoC data and adjuts the printf() format because this is
a 64-bit SoC. Otherwise, 16-digits pointer addresses would break
the log format.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Do not hard-code the number of DX blocks because it is a different
value for LD11 SoC.
Move the macro NR_DATX8_PER_DDRPHY to ddrphy-training.c since it
is the last user.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The DDR PHY register view of LD11 is slightly different from that
of LD4/Pro4/sLD8, but it will be possible to share the register
macros (and I want to re-use as much code as possible). Change
the code in the more flexible form.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The USB boot without the stand-by MPU is available on ES3 or later
of LD11 SoC, but the code in this if-conditional block must not be
run when booting from USB. Check if the boot device is USB, and
skip the code in the case.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
At the moment, the clk driver is not clever enough to automatically
enable parent clocks like Linux. Enable the STDMAC clock explicitly
if USB is enabled.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
This PHY might be used for other SoCs in the future.
Avoid including the SoC name in the header name.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The environment fdt_file is useful to remember the appropriate DTB
file name. Adjust it to the recent renaming in the upstream kernel.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Invoking exit prevents any subsequent build commands from running, and
future patches will add extra commands.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
This places build results into a board-specific directory rather than a
buildman-thread-specific directory. This is required so that we can
access the directory from test.py, and there's no risk of a particular
build's results being over-written by another build performed by the
same thread.
In theory, this can lead to slower builds when building many different
boards in a single buildman thread, since it removes the possibility of
incremental builds between boards. In practice however I didn't notice
longer build times when when enabling this option; if anything build
times decreased although I suspect that's simply due to general
variations in build performance across different machines within the
Travis CI infra-structure.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Any time an x86 toolchain is used, we need to edit ~/.buildman to
reference it. Move the editing logic into a central place so that it
doesn't have to be duplicated everywhere that uses the x86 toolchain;
future patches will add additional cases where it's used.
It would be nice if we could unconditionally write all of ~/.buildman at
once. Unfortunately, buildman fails if any toolchain mentioned in a
toolchain-prefix entry doesn't exist, even if it doesn't need to use it
for the current build.
The sandbox/x86 build definition currently does nothing more than edit
~/.buildman; no builds are run. Fix this by not defining a custom script
for this build, and hence preventing that stanza from replacing the
default script.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
The phrase "if [ $? -ne 0 ]; then exit $?; fi" doesn't work correctly;
by the time the "exit" statement runs, $? has already been over-written
by the result of the [ command. Fix this by explicitly storing $? and
then using that stored value in both the test and the error-case exit
statement.
This change also converts from textual comparison to integer comparison,
since the exit code is an integer and there's no need to convert it to
a string for comparison.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Travis CI seems to be confused when there's a colon in an echo command,
and this is currently worked around using a variable that contains the
text we want to echo. Use = syntax instead so that we can remove the
work-around; it's rather confusing until you find out what it's for.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
There were two sub-jobs to build arm1136. Remove the duplicate.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Travis CI names sub-jobs after the first environment variable that is set
for a script. This doesn't produce meaningful results for any of the non-
buildman jobs. Add a dummy variable to give the jobs meaningful names.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Add driver model support while retaining the existing legacy code.
This allows the driver to support boards that have converted to
driver model as well as those that have not.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Stephen Warren <swarren@nvidia.com>
Enable an early debug UART to debug problems when an ICE or other
debug mechanism is not available.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Andreas Bießmann <andreas.devel@googlemail.com>
If 'ethaddr' is not set, we will get the ethernet address from AT24MAC,
and set it to 'ethaddr' variable.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Signed-off-by: Songjun Wu <songjun.wu@microchip.com>
Reviewed-by: Andreas Bießmann <biessmann@corscience.de>
Since the introduction of pinctrl and clk driver, and the dts file,
remove unneeded the pin configurations and the clock enabling code.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Andreas Bießmann <andreas.devel@googlemail.com>
Move the config options from the include/configs/sama5d2_xplained.h
to configs/sama5d2_xplained_*_defconfig.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Andreas Bießmann <andreas.devel@googlemail.com>
Add support to enable an early debug UART for debugging.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Andreas Bießmann <andreas@biessmann.org>
Add ATMEL_USART option to support to enable the Atmel usart driver
from Kconfig.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Andreas Bießmann <andreas@biessmann.org>
Due to the peripheral and generated clock driver improvement,
remove the unnecessary clock calling.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
Due to the peripheral clock driver improvement, remove the
unnecessary clock calling.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Andreas Bießmann <andreas@biessmann.org>
Change the error return value -ENODEV from to -EINVAL for more
reasonable.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Due to the peripheral clock driver improvement, remove the
unnecessary clock calling.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Due to the peripheral clock driver improvement, remove the
unnecessary clock calling.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
For the peripheral clock, provide the clock ops for the clock
provider, such as spi0_clk. The .of_xlate is to get the clk->id,
the .enable is to enable the spi0 peripheral clock, the .get_rate
is to get the clock frequency.
The driver for periph32ck node is responsible for recursively
binding its children as clk devices, not provide the clock ops.
So do the generated clock and system clock.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
In order to make clk->dev available in ops->of_xlate() to get the
clock ID from the 'reg' property of the clock node, assign the
clk->dev before calling ops->of_xlate().
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Simon Glass <sjg@chromium.org>
Fix the warning from dtc like,
---8<----
Warning (unit_address_vs_reg): Node /ahb/apb/pmc@f0014000/periph64ck/sdmmc0_hclk has a reg or ranges property, but no unit name
--->8----
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
The at91-pmc and at91-sckc aren't the clock providers, change their
class ID from UCLASS_CLK to UCLASS_SIMPLE_BUS, they also don't
need to bind the child nodes explicitly, the .post_bind callback
of simple_bus uclass will do it for them.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
add at least icache support for at91 based boards.
This speeds up NOR flash access on an at91sam9g15
based board from 15.2 seconds reading 8 MiB from
a SPI NOR flash to 5.7 seconds.
Signed-off-by: Heiko Schocher <hs@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Andreas Bießmann <andreas@biessmann.org>
on at91sam9x5 PRES offset is 4 in the PMC master
clock register.
Signed-off-by: Heiko Schocher <hs@denx.de>
Acked-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Andreas Bießmann <andreas@biessmann.org>
