Tom Rini [Tue, 9 May 2017 02:14:21 +0000 (22:14 -0400)]
ot1200: Fix unused variable warning
The variable pwm_pad is never referenced, drop.
Cc: Christian Gmeiner <christian.gmeiner@gmail.com> Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Tom Rini [Tue, 9 May 2017 02:14:19 +0000 (22:14 -0400)]
socrates: Fix a misleading indentation warning
With gcc-6 and later we see a warning about the fact that we have a
construct of "if (test);\n\tstatement". Upon reviewing the code, the
intention here is as the compiler suggests, we only want to execute the
indented statement if the test was true.
Cc: Sergei Poselenov <sposelenov@emcraft.com> Signed-off-by: Tom Rini <trini@konsulko.com>
Lokesh Vutla [Fri, 5 May 2017 07:29:10 +0000 (12:59 +0530)]
arm: am33xx: Add support for mulitiple PLL input frequencies
am335x supports various sysclk frequencies which can be determined
using sysboot pins. PLLs should be configures based on this
sysclk frequency. Add PLL configurations for all supported
frequencies.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com> Reviewed-by: Tom Rini <trini@konsulko.com>
Lokesh Vutla [Fri, 5 May 2017 07:29:08 +0000 (12:59 +0530)]
arm: am33xx: Fix MPU opp selection
Update MPU frequencies and voltages as per the latest
DM[1] dated: OCT 2011 Revised APRIL 2016, Section 5.4.
Below is the consolidated data:
MPU values for PG 2.0 and later(Package ZCZ and ZCE):
-------------------------------------------------------
| | ZCZ | ZCE |
|-------------------------------------------------------|
| | VDD[V] | ARM [MHz] | VDD[V] | ARM [MHz] |
|-------|----------|------------|----------|------------|
| NITRO | 1.325 | 1000 | NA | NA |
|-------|----------|------------|----------|------------|
| TURBO | 1.26 | 800 | NA | NA |
|-------|----------|------------|----------|------------|
|OPP120 | 1.20 | 720 | NA | NA |
|-------|----------|------------|----------|------------|
|OPP100 | 1.10 | 600 | 1.10 | 600 |
|-------|----------|------------|----------|------------|
| OPP50 | 0.95 | 300 | 0.95 | 300 |
-------------------------------------------------------
There is no eFuse blown on PG1.0 Silicons due to which there is
no way to detect the maximum frequencies supported. So default
to OPP100 for which both frequency and voltages are common on both
the packages.
[1] http://www.ti.com/lit/ds/symlink/am3356.pdf
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com> Reviewed-by: Tom Rini <trini@konsulko.com>
B, Ravi [Thu, 4 May 2017 10:15:29 +0000 (15:45 +0530)]
common: dfu: ignore reset for spl-dfu
The SPL-DFU feature enable to load and
execute u-boot from RAM over usb from
PC using dfu-util.
Hence dfu-reset should not be issued
when dfu-util -R switch is issued.
Vikas Manocha [Wed, 3 May 2017 23:38:55 +0000 (16:38 -0700)]
armv7m: add MPU configuration support
Cortex-M archs support option memory protection unit (MPU). MPU is used
to set the memory types, attributes, access permissions for different regions,
cache policies of the device.
e.g. using MPU it is possible to configure memory region as device memory
or strongly ordered, memory attributes like execute never, cache policies
like write-back or write-through.
dm: mmc: omap_hsmmc: Add pre-reloc flag to the driver
For platforms that don't use device tree in SPL the only
way to mark this driver as 'required by relocation' is
with the DM_FLAG_PRE_RELOC flag. Add this to ensure that
the driver is bound.
Reviewed-by: Tom Rini <trini@konsulko.com> Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Add debug UART functions to permit omap specific ns16550 to
provide an early debug UART. This is mostly in common with
DEBUG_UART_NS16550 except for Mode definition register which
is required for selecting UART mode(16x auto-baud or 13x mode).
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Tom Rini <trini@konsulko.com>
James Balean [Wed, 19 Apr 2017 02:06:35 +0000 (21:06 -0500)]
Add 16-bit single register pin controller support
Enables the pinctrl-single driver to support 16-bit registers. Only
32-bit registers were supported previously. Reduced width registers are
required for some platforms, such as OMAP.
Signed-off-by: James Balean <james@balean.com.au> Cc: Felix Brack <fb@ltec.ch> Cc: Simon Glass <sjg@chromium.org> Reviewed-by: Felix Brack <fb@ltec.ch> Tested-by: Felix Brack <fb@ltec.ch> Reviewed-by: Simon Glass <sjg@chromium.org>
B, Ravi [Tue, 18 Apr 2017 11:57:27 +0000 (17:27 +0530)]
boot: fdt: Perform arch_fixup_fdt() on the given device tree for falcon boot
In single stage bootmode or falcon boot mode, the SPL shall update the
device tree that we load with the normal fixups done via
arch_fixup_fdt(), when possible (ie we have enough information in this
restricted environment to be able to do that still). This will include
for example updating them memory nodes.
Signed-off-by: Ravi Babu <ravibabu@ti.com>
[trini: Reword commit message]
Philipp Tomsich [Mon, 17 Apr 2017 15:45:13 +0000 (17:45 +0200)]
spl: Makefile: include /config in the (reduced) FDT used by the SPL stage
When OF control is enabled for the SPL stage, nodes are removed from
the DTB to reduce its size. While /chosen is kept, /config is removed.
There's no reason why /chosen should be kept over /config (and as we
would like to put properties into /config that control the SPL stage),
we add '/config' to the list of nodes to be retained for the SPL stage.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
This adds documentation on the u-boot,spl-payload-offset property
(which overrides CONFIG_SYS_SPI_U_BOOT_OFFS during the SPI loading in
the SPL stage, if present).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Mon, 17 Apr 2017 15:45:11 +0000 (17:45 +0200)]
spl: spi: override CONFIG_SYS_SPI_U_BOOT_OFFS via /config-property
For the RK3399-Q7, we need some flexibility (depending on the feature
set we include in the SPL stage and how large our SPI flash is) in
positioning the SPL payload (i.e. the FIT image containing U-Boot, ATF
and the M0 payload) in our SPI flash.
To avoid having to deal with this through different U-Boot images, we
introduce a the '/config/u-boot,spl-payload-offset' property node
allow it to override the default setting.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Simon Glass [Wed, 5 Apr 2017 23:53:18 +0000 (17:53 -0600)]
arm: Support cache invalidate
At present there is not operation to invalidate a cache range. This seems
to be needed to fill out the cache operations. Add an implementation based
on the flush operation.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: York Sun <york.sun@nxp.com>
Alex Deymo [Sun, 2 Apr 2017 08:49:50 +0000 (01:49 -0700)]
disk: Return the partition number in part_get_info_by_name()
Similar to what blk_get_device_part_str() does, this patch makes
part_get_info_by_name() return the partition number in case of a match.
This is useful when the partition number is needed and not just the
descriptor.
Signed-off-by: Alex Deymo <deymo@google.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Alex Deymo [Sun, 2 Apr 2017 08:49:47 +0000 (01:49 -0700)]
image: Update include/android_image.h
Update the Android image header format to the latest version published
in AOSP. The original code moved to a new repository, so this patch also
updates the reference to that path.
Signed-off-by: Alex Deymo <deymo@google.com> Reviewed-by: Simon Glass <sjg@chromium.org>
usb: gadget: avoid variable name clipping in cb_getvar
Hi,
A kind reminder to look at this patch (already reviewed by Marek and acked by Lukasz), and if possible to put it in the next pull list, or the one after is timing is too short.
Thanks in advance for your time
Best Regards
Nicolas
-----Original Message-----
From: Nicolas LE BAYON
Sent: mardi 25 avril 2017 10:18
To: Nicolas LE BAYON <nicolas.le.bayon@st.com>; u-boot@lists.denx.de; lukma@denx.de; marex@denx.de Cc: nlebayon@gmail.com; Patrice CHOTARD <patrice.chotard@st.com>; Jean-philippe ROMAIN <jean-philippe.romain@st.com>
Subject: [U-Boot][PATCH v7] usb: gadget: avoid variable name clipping in cb_getvar
From: Nicolas Le Bayon <nicolas.le.bayon@st.com>
Instead of using a fixed-size array to store variable name, preferring a dynamic allocation treats correctly all variable name lengths.
Variable names are growing through releases and features. By this way, name clipping is prevented.
Signed-off-by: Nicolas Le Bayon <nicolas.le.bayon@st.com> Reviewed-by: Marek Vasut <marex@denx.de> Acked-by: Lukasz Majewski <lukma@denx.de>
Tom Rini [Wed, 10 May 2017 21:40:11 +0000 (17:40 -0400)]
Merge git://git.denx.de/u-boot-rockchip
This adds a new firefly-rk3399 board, MIPI support for rk3399 and
rk3288, rk818 pmic support, mkimage improvements for rockchip and a few
other things.
Eric Gao [Tue, 2 May 2017 10:23:56 +0000 (18:23 +0800)]
rockchip: dts: Add mipi dsi support for rk3399
Add dts config for mipi display, include vop, mipi controller, panel, backlight
. And Enable rk808 for lcd_3v3 in another patch.
Signed-off-by: Eric Gao <eric.gao@rock-chips.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Eric Gao [Tue, 2 May 2017 10:23:55 +0000 (18:23 +0800)]
rockchip: board: evb_rk3399: initialize pwm0 for dispaly backlight
Enable pwm0 for display of rk3399 evb board. The PWM do not have decicated
interrupt number in dts and can not get periph_id by pinctrl framework. So
init them here.
Signed-off-by: Eric Gao <eric.gao@rock-chips.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Eric Gao [Tue, 2 May 2017 10:23:54 +0000 (18:23 +0800)]
rockchip: video: vop: Reserve enough space for mipi dispaly
plat->size here is used to reserve frame buffer space befor relocation.
our mipi panel use 24 bitwidth, and vop require 32bit align. So the frame
buffer size should be at least 1920*1200*32/8.
Signed-off-by: Eric Gao <eric.gao@rock-chips.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Eric Gao [Tue, 2 May 2017 10:23:49 +0000 (18:23 +0800)]
rockchip: include: grf: Add GRF register declaration for mipi dsi
Add GRF register declaration for mipi dsi.
Signed-off-by: Eric Gao <eric.gao@rock-chips.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Jacob Chen [Tue, 2 May 2017 06:54:51 +0000 (14:54 +0800)]
power: regulator: rk808: add rk818 support
Add support for the rk818 regulator. The regulator module consists
of 4 DCDCs, 9 LDOs, 1 switch and 1 BOOST converter which is used to
power OTG and HDMI5V.
Signed-off-by: Jacob Chen <jacob-chen@iotwrt.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Fri, 28 Apr 2017 16:33:58 +0000 (18:33 +0200)]
rockchip: pinctrl: rk3399: add support for the HDMI I2C pins
To add HDMI support for the RK3399, this commit provides the needed
pinctrl functionality to configure the HDMI I2C pins (used for reading
the screen's EDID).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Fri, 28 Apr 2017 16:33:57 +0000 (18:33 +0200)]
rockchip: clk: rk3399: allow requests for HDMI clocks
This allows requests (via the DTS) for PCLK_HDMI_CTRL/PCLK_VIO_GRF,
which are clock gates in the HDMI output path for the RK3399.
As these are enabled by default (i.e. after reset), we don't implement
any logic to actively open/close these clock gates and simply assume
that their reset-default has not been changed.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Fri, 28 Apr 2017 15:31:44 +0000 (17:31 +0200)]
rockchip: ARM64: puma-rk3399: get DRAM size from DMC init
With the RK3399 DRAM controller (DMC) driver providing all the
infrastructure, retrieve the DRAM size from the DMC init in the
board-specific code (instead of hard-coding) for the RK3399-Q7 (Puma).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Fri, 28 Apr 2017 15:11:55 +0000 (17:11 +0200)]
rockchip: clk: rk3399: allow requests for PCLK_EFUSE1024NS
The (non-secure) efuse node in the DTS requests PCLK_EFUSE1024NS.
To allow us to add a efuse-driver (and more importantly, to allow
probes of such a driver to succeed), we need need to accept requests
for PCLK_EFUSE1024NS and return a non-error result.
As PCLK_EFUSE1024NS is enabled by default (i.e. after reset), we don't
implement any logic to manage this clock gate and simply assume that
the reset-default has not been changed.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Tested-by: Klaus Goger <klaus.goger@theobroma-systems.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Kever Yang [Mon, 24 Apr 2017 02:27:52 +0000 (10:27 +0800)]
dm: sandbox: pwm: add test for pwm_set_invert()
Add test case for new interface set_invert().
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Simon Glass <sjg@chromium.org>
Fix typo in subject and build error in sandbox_pwm_set_invert(): Signed-off-by: Simon Glass <sjg@chromium.org>
Kever Yang [Mon, 24 Apr 2017 02:27:49 +0000 (10:27 +0800)]
power: regulator: pwm: support pwm polarity setting
The latest kernel PWM drivers enable the polarity settings. When system
run from U-Boot to kerenl, if there are differences in polarity set or
duty cycle, the PMW will re-init:
close -> set polarity and duty cycle -> enable the PWM.
The power supply controled by pwm regulator may have voltage shaking,
which lead to the system not stable.
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com> Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Acked-by: Simon Glass <sjg@chromium.org>
Simon Glass [Sun, 23 Apr 2017 00:42:22 +0000 (18:42 -0600)]
dtoc: Handle nodes with phandles that depend on the same
At present dtoc assumes that nodes which are phandles do not themselves
reference other phandle nodes. Unfortunately this is not necessarilly
true. As a result we can currently output C code which does not compile
because a node declaration can be referenced before it is declared.
Adjust the code to explicitly output all phandle nodes needed by node
before the node itself is output.
This fixes building with the latest rk3399-firefly.dts from Linux, which
has reordered the nodes.
Signed-off-by: Simon Glass <sjg@chromium.org> Tested-by: Kever Yang <kever.yang@rock-chips.com>
Philipp Tomsich [Tue, 25 Apr 2017 07:52:07 +0000 (09:52 +0200)]
rockchip: mmc: handle deprecation of 'clock-freq-min-max'
The 'clock-freq-min-max' property was deprecated in the upstream
(i.e. Linux) DTS bindings in favor of the 'max-frequency' property.
With the latest RK3399 DTSI does no longer include the deprecated
property and the rockchip_dw_mmc driver requiring it to be present,
the driver doesn't bind to the node in the RK3399 DTSI any longer
(thus breaking access to the SD card on the RK3399-Q7 board).
To fix this, we implement a similar logic as in the Linux driver: if
the deprecated property is present, we issue a warning (if DEBUG is
enabled); if it is missing, we require 'max-frequency' to be set and
use it to create a min/max value-pair.
Philipp Tomsich [Tue, 25 Apr 2017 07:52:06 +0000 (09:52 +0200)]
rockchip: clk: rk3399: adapt MMC clk configuration to the updated RK3399 DTS
The clocking of the designware MMC controller in the upstream
(i.e. Linux) RK3399 has changed/does not match what the current DTS in
U-Boot uses: the first clock entry now is HCLK_SDMMC instead of
SCLK_SDMMC.
With the simple clock driver used for the RK3399, this needs a change
in the selector understood by the various case statements in the driver
to ensure that the driver still loads successfully.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Kever Yang [Thu, 20 Apr 2017 09:03:46 +0000 (17:03 +0800)]
rockchip: reserve memory for rk3399 ATF data
There are 3 regions used by rk3399 ATF:
- bl31 code, located at 0x10000;
- cortex-m0 code and data, located at 0xff8c0000;
- bl31 data, located at 0xff8c1000 ~ 0xff8c4000;
SPL_TEXT_BASE starts from 0xff8c2000, we need to reserve memory
for ATF data, or else there will be memory corrupt after SPL
loads the ATF image.
Kever Yang [Thu, 20 Apr 2017 08:15:34 +0000 (16:15 +0800)]
rockchip: pinctrl: rk3399: add gmac io strength support
GMAC controller need to init the tx io driver strength to 13mA,
just like the description in dts pinctrl node, or else the controller
may only work in 100MHz Mode, and fail to work at 1000MHz mode.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com <mailto:philipp.tomsich@theobroma-systems.com>> Reviewed-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Mon, 17 Apr 2017 15:50:38 +0000 (17:50 +0200)]
rockchip: dts: rk3399-puma: Add DDR3-1600 timings and use for Puma
With the validation done for DDR3-1600 (i.e. 800 MHz bus clock), we
add the timings (rk3399-sdram-ddr3-1600.dtsi) and change rk3399-puma.dts
to use these by default.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Drop blank line at end of file: Signed-off-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Mon, 17 Apr 2017 15:48:06 +0000 (17:48 +0200)]
rockchip: mkimage: remove placeholder functions from rkimage
The imagetool framework checks whether function pointer for the verify,
print and extract actions are available and will will handle their
absence appropriately.
This change removes the unnecessary functions and uses the driver
structure to convey available functionality to imagetool. This is in
fact better than having verify just return 0 (which previously broke
dumpimage, as dumpimage assumed that we had handled the image and did
not continue to probe further).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Mon, 17 Apr 2017 15:48:05 +0000 (17:48 +0200)]
rockchip: mkimage: play nice with dumpimage
Dumpimage (it invoked with "-T rkspi" or "-T rksd") would not work due
to check_params failing. These changes ensure that we can both be called
with an empty imagename.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Mon, 17 Apr 2017 15:48:02 +0000 (17:48 +0200)]
rockchip: mkimage: Update comments for header size
The calculation of the variable header size in rkcommon_vrec_header
had been update twice in the earlier series (introducing boot0-style
images to deal with the alignment of the first instruction in 64bit
binaries). Unfortunately, I didn't update the comment twice (so it
remained out-of-date).
This change brings the comment back in-sync with what the code is
doing.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Mon, 17 Apr 2017 15:48:01 +0000 (17:48 +0200)]
rockchip: mkimage: rewrite padding calculation for SD/MMC and SPI images
In (first) breaking and (then) fixing the rkspi tool, I realised that
the calculation of the required padding (for the header-size and the
2K-in-every-4K SPI layout) was not as self-explainatory as it could
have been. This change rewrites the code (using new, common functions
in rkcommon.c) and adds verbose in-line comments to ensure that we
won't fall into the same pit in the future...
Tested on the RK3399 (with has a boot0-style payload) with SD/MMC and SPI.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Mon, 17 Apr 2017 15:48:00 +0000 (17:48 +0200)]
rockchip: mkimage: rkspi: include the header sector in the SPI size calculation
Our earlier change broke the generation of SPI images, by excluding the
2K used for header0 from the size-calculation.
This commit makes sure that these are included before calculating the
required total size (including the padding from the 2K-from-every-4K
conversion).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Thu, 20 Apr 2017 20:05:55 +0000 (22:05 +0200)]
rockchip: spl: rk3399: spi: enable SPL_SPI_LOAD if SPI is enabled for SPL
To include the ability to load from an SPI flash in SPL, it's not
sufficient to define SPL_SPI_SUPPORT and SPL_SPI_FLASH_SUPPORT via
Kconfig... so we conditionally define SPL_SPI_LOAD if SPI support
is already enabled for SPL via Kconfig.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
rockchip: spi: enable support for the rk_spi driver for the RK3399
The existing Rockchip SPI (rk_spi.c) driver also matches the hardware
block found in the RK3399. This has been confirmed both with SPI NOR
flashes and general SPI transfers on the RK3399-Q7 for SPI1 and SPI5.
This change adds the 'rockchip,rk3399-spi' string to its compatible
list to allow reuse of the existing driver.
X-AffectedPlatforms: RK3399-Q7 Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Tested-by: Jakob Unterwurzacher <jakob.unterwurzacher@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Thu, 20 Apr 2017 20:05:53 +0000 (22:05 +0200)]
rockchip: pinctrl: rk3399: add support for the SPI5 controller
This commit adds support for the pin-configuration of the SPI5
controller of the RK3399 through the following changes:
* grf_rk3399.h: adds definition for configuring the SPI5 pins
in the GPIO2C group
* periph.h: defines PERIPH_ID_SPI3 through PERIPH_ID_SPI5
* pinctrl_rk3399.c: adds the reverse-mapping from the IRQ# to
PERIPH_ID_SPI5; dispatches PERIPH_ID_SPI3
through SPI5 to the appropriate pin-config
function; implements the pin-configuration
for PERIPH_ID_SPI5 using the GPIO2C group
X-AffectedPlatforms: RK3399-Q7 Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Tested-by: Jakob Unterwurzacher <jakob.unterwurzacher@theobroma-systems.com> Acked-by: Simon Glass <sjg@chromium.org>
Philipp Tomsich [Thu, 20 Apr 2017 20:05:52 +0000 (22:05 +0200)]
rockchip: spi: rewrite rkspi_set_clk for a more conservative baudrate setting
The baudrate in rkspi was calculated by using an integer division
(which implicitly discarded any fractional result), then rounding to
an even number and finally clamping to 0xfffe using a bitwise AND
operator. This introduced two issues:
1) for very small baudrates (overflowing the 0xfffe range), the
bitwise-AND generates rather random-looking (wildly varying)
actual output bitrates
2) for higher baudrates, the calculation tends to 'err towards a
higher baudrate' with the actual error increasing as the dividers
become very small. E.g., with a 99MHz input clock, a request
for a 20MBit baudrate (99/20 = 4.95), a 24.75 MBit would be use
(which amounts to a 23.75% error)... for a 34 MBit request this
would be an actual outbout of 49.5 Mbit (i.e. a 45% error).
This change rewrites the divider selection (i.e. baudrate calculation)
by making sure that
a) for the normal case: the largest representable baudrate below the
requested rate will be chosen;
b) for the denormal case (i.e. when the divider can no longer be
represented), the lowest representable baudrate is chosen.
Even though the denormal case (b) may be of little concern in real
world applications (even with a 198MHz input clock, this will only
happen at below approx. 3kHz/3kBit), our board-verification team kept
complaining.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Tested-by: Klaus Goger <klaus.goger@theobroma-systems.com>
projects / u-boot.git / log