To boot from fpga on socfpga gen5, we need to set CONFIG_SPL_TEXT_BASE to
0xC0000000 (hps2fpgaslaves base address).
Since converting CONFIG_SPL_TEXT_BASE to Kconfig hasn't been successful so
far, let's make this value overridable in socfpga_common.h, so that we can
have different board configs override this in socfpga_common.h.
Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Neil Armstrong [Wed, 20 Feb 2019 10:36:12 +0000 (11:36 +0100)]
fastboot: common: fix default fastboot_boot on 64-bit
When booting on a 64-bit system, the boot_addr_start buffer is not
large enough to contain a 64-bit number, thus leading to a crash
even if fastboot_buf_addr is valid, only the high part of the address
will be printed to boot_addr_start :
fastboot with fastboot_buf_addr = 0x0000000006000000:
downloading of 92239872 bytes finished
Booting kernel at 0x00000000...
usb: udc-uclass: Fixed problem when no alias is defined in DT
commit 801f1fa442 "dm: usb: udc: Use SEQ_ALIAS to index the USB gadget
ports" changed the way the udevice if found. It uses the alias to find
a udevice for a given USB port number. In the commit log it was stated
that if no alias is provided, the bind order will be used instead. However
it doesn't work. Fixing this by adding a call to uclass_get_device() if
uclass_get_device_by_seq() fails.
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com> Tested-by: Vignesh R <vigneshr@ti.com>
Alison Wang [Wed, 6 Mar 2019 06:49:14 +0000 (14:49 +0800)]
armv7: ls102xa: Add workaround for DDR erratum A-008850
Barrier transactions from CCI400 need to be disabled till
the DDR is configured, otherwise it may lead to system hang.
The patch adds workaround to fix the erratum.
Signed-off-by: Shengzhou Liu <Shengzhou.Liu@nxp.com> Signed-off-by: Alison Wang <alison.wang@nxp.com> Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
If node /board_info/ports does not exist in the DPC file,
function mc_fixup_dpc() will skip not only MAC address fixup,
but also the cache flush at the end. This may cause the other
fixup changes (e.g. ICID related ones) to be ignored by MC.
Fixes: 1161dbcc0a36 ("drivers: net: fsl-mc: Include MAC addr fixup to DPL") Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com> Acked-by: Joe Hershberger <joe.hershberger@ni.com> Reviewed-by: Prabhakar Kushwaha <prabhakar.kushwaha@nxp.com>
Pankaj Bansal [Fri, 8 Feb 2019 08:59:24 +0000 (08:59 +0000)]
drivers: net: ldpaa_eth: check if the dpmac is enabled
some dpmacs in armv8a based freescale layerscape SOCs can be
configured via both serdes(sgmii, xfi, xlaui etc) bits and via
EC*_PMUX(rgmii) bits in RCW.
e.g. dpmac 17 and 18 in LX2160A can be configured as SGMII from
serdes bits and as RGMII via EC1_PMUX/EC2_PMUX bits
Now if a dpmac is enabled by serdes bits then it takes precedence
over EC*_PMUX bits. i.e. in LX2160A if we select serdes protocol
that configures dpmac17 as SGMII and set the EC1_PMUX as RGMII,
then the dpmac is SGMII and not RGMII.
Therefore, in fsl_rgmii_init function of SOC, we will check if the
dpmac is enabled or not? if it is (fsl_serdes_init has already enabled
the dpmac), then don't enable it.
MAINTAINERS: Remove unsupported statuses - Odd Fixes and Obsolete
The MAINTAINERS file was copied from Linux Kernel along with all its
statuses of maintainership. However tools/genboardscfg.py accepts only
Maintained, Supported and Orphan. Remove then the Odd Fixes and
Obsolete from MAINTAINERS file to avoid confusion.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Loading Environment from SPI Flash... SF: Detected s25fl128l with page size 256 Bytes, erase size 4 KiB, total 16 MiB
*** Warning - bad CRC, using default environment
Reason is the above commit, as it only reads until \0\0
is found, and assumes the rest of the Environment
space is filled with 0x00, which is not the case when
saving an Environment under linux with fw_setenv.
Tom Rini [Wed, 13 Mar 2019 20:07:41 +0000 (16:07 -0400)]
Merge tag 'u-boot-imx-20190313' of git://git.denx.de/u-boot-imx
Small fixes in several i.MX boards
----------------------------------
- imx8: add pinctrl driveri (mx8m), fix documentation and
fix reported CPU frequency. Fabio is co-maintainer
- pico-imx6ul: switch to DM
- local fixes for ventana, mx6ul_14x14_evk, engicam,
imx6(q)_logic, liteboard
Adam Ford [Mon, 18 Feb 2019 23:58:17 +0000 (17:58 -0600)]
MTD: mxs_nand_spl: Redo the way nand_init initializes
Currently the spl system calls nand_init which does nothing.
It isn't until an attempt to load from NAND that it gets initialized.
Subsequent attempts to load just skip the initialization because
NAND is already initialized.
This moves the contents of mxs_nand_init to nand_init. In the event
of an error, it clears the number of nand chips found. Any
attempts to use nand will check if there are nand chips available
instead of actually doing the initialization at that time. If there
are none, it will return an error to the higher level calls.
Adam Ford [Wed, 20 Feb 2019 14:53:56 +0000 (08:53 -0600)]
arm: dts: imx6qdl-u-boot: Enable spba-bus@2000000 simple bus
spba-bus has a few nodes under it including the UART1 and
some ESPI buses. In order to use them in SPL, the
u-boot,dm-spl flag needs to be added to the spba-bus@2000000
container.
Adam Ford [Wed, 20 Feb 2019 04:07:22 +0000 (22:07 -0600)]
imx: serial_mxc: use CONFIG_IS_ENABLED instead of ifdef
Kconfig allows boards to configured with DM_SERIAL and still
have SPL_DM_SERIAL disabled. This patch changes the ifdef's
to CONFIG_IS_ENABLED to allow the modes to differ between
SPL and U-Boot.
Signed-off-by: Adam Ford <aford173@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Martyn Welch [Fri, 22 Feb 2019 19:05:48 +0000 (19:05 +0000)]
imx: mx6qsabrelite: Update the SabreLite README
The information in the SabreLite README is inaccurate and sparse. The
upstream U-Boot can boot the SabreLite from SPI-NOR. Additionally, the
freely available imx_loader tool can be easily used to boot a board with a
corrupted SPI, the official Freescale/NXP manufacturing tools are not
required.
Reformat the document, adding a description of how to boot from SPI-NOR
and adding a brief description of how to recover the board should the
SPI-NOR be corrupted using imx_loader.
Fabio Estevam [Thu, 14 Feb 2019 13:37:51 +0000 (11:37 -0200)]
warp7: Fix the write to the LDOGCTL PMIC register
The third parameter of the pmic_clrsetbits() function is the mask
to the register and the correct mask is 1 not 0.
Since the LDOGCTL only contains a single valid bit (bit 0),
we can use pmic_reg_write() and write 1 directly, which fixes
the problem in a simpler way and use the original pmic function
that was used prior to the DM PMIC conversion.
Marcin Niestroj [Sat, 19 Jan 2019 16:06:46 +0000 (17:06 +0100)]
ARM: liteboard: move towards driver model and device-tree boot
This patch mostly enables DM drivers in board defconfig and all their
dependencies. Additionally we remove USB code that is on longer
executed after enabling CONFIG_DM_USB. Enable CONFIG_PINCTRL, so we
can get rid of ethernet pin configuration.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Chris Spencer [Mon, 4 Feb 2019 10:05:34 +0000 (10:05 +0000)]
imx8mq_evk_defconfig: Enable pinctrl driver
The Ethernet controller is not able to initialise correctly without the
pinctrl driver.
This config setting was enabled in the initial version of this file,
but was removed by a savedefconfig resync because the parameter did not
actually exist at that point.
Fixes: 1bac199e8c87 ("configs: Resync with savedefconfig") Signed-off-by: Chris Spencer <christopher.spencer@sea.co.uk> Reviewed-by: Fabio Estevam <festevam@gmail.com>
Bin Meng [Tue, 26 Feb 2019 09:52:19 +0000 (01:52 -0800)]
x86: Make sure i8254 is setup correctly before generating beeps
The i8254 timer control IO port (0x43) should be setup correctly
by using PIT counter 2 to generate beeps, however in U-Boot other
codes like TSC driver utilizes PIT for TSC frequency calibration
and configures the counter 2 to a different mode that does not
beep. Fix this by always ensuring the PIT counter 2 is correctly
initialized so that the i8254 beeper driver works as expected.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
arm: exynos: odroid: Fix build if BOARD_TYPES are not set
CONFIG_BOARD_TYPES is necessary for Odroid X/X2/U3 boards to detect
proper revision. However building should succeed even without it.
While moving code around, document also the reference clock selection.
This fixes the build error without CONFIG_BOARD_TYPES:
board/samsung/odroid/odroid.c: In function 'board_usb_init':
board/samsung/odroid/odroid.c:473:8: error: 'gd_t' {aka 'volatile struct global_data'} has no member named 'board_type'
if (gd->board_type == ODROID_TYPE_U3)
^~
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Various places in the code set "boardname" env property. It was used
for booting from ITB images and choosing proper DTB file name. Instead
of duplicating it, use existing U-Boot wide - "board_name".
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Reviewed-by: Lukasz Majewski <lukma@denx.de> Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Last activity from Arndale (Exynos5250) board maintainer Chander Kashyap
was in January 2014 (Signed-off). Recently his samsung.com email
bounces with 550 (5.1.1 Recipient address rejected: User unknown).
Add Krzysztof Kozlowski as odd fixer for this board.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
arm: dts: exynos: Add ramp delay property to LDO regulators to Odroid XU3 family
Add startup time to LDO regulators of S2MPS11 PMIC on Odroid XU3/XU4/HC1
family of boards to be sure the voltage is proper before relying on the
regulator.
The datasheet for all the S2MPS1x family is inconsistent here and does
not specify unambiguously the value of ramp delay for LDO. It mentions
30 mV/us in one timing diagram but then omits it completely in LDO
regulator characteristics table (it is specified for bucks).
However the vendor kernels for Galaxy S5 and Odroid XU3 use values of 12
mV/us or 24 mV/us.
Without the ramp delay value the consumers do not wait for voltage
settle after changing it. Although the proper value of ramp delay for
LDOs is unknown, it seems safer to use at least some value from
reference kernel than to leave it unset.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Reviewed-by: Lukasz Majewski <lukma@denx.de> Tested-by: Anand Moon <linux.amoon@gmail.com> Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
arm: exynos: Wait till ADC stabilizes before checking Odroid HC1 revision
Fix detection of Odroid HC1 (Exynos5422) after reboot if kernel disabled
the LDO4/VDD_ADC regulator.
The LDO4 supplies both ADC block and the ADC input AIN9. Voltage on
AIN9 will rise slowly, so use delay of 5 milliseconds instead of
timers-based loop to wait for voltage stabilization.
First reads on Odroid HC1 return 305, 1207, 1297 and finally 1308
(reference value is 1309).
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Tested-by: Anand Moon <linux.amoon@gmail.com> Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
arm: exynos: Detect revision later, when all resources are ready
Detection of board revision is done early - before power setup. In case of
Odroid XU3/XU4/HC1 family, the detection is done using ADC which
is supplied by LDO4/VDD_ADC regulator. This regulator could be turned
off (e.g. by kernel before reboot). If ADC is used early, the
regulators are not yet available and the detection won't work.
Split the revision detection out of set_board_type() into separate
function called later - either when displaying board info (in late mode)
or during misc_init_r. The idea is that set_board_type() will be called
early so its method of detection are limited to flattened device tree
(exynos5-dt-types.c for Exynos5) or GPIO (odroid.c for Exynos4412). The
newly added set_board_revision() can be called only later, when
resources like regulator are available.
This is necessary to fix the detection of Odroid HC1 after reboot, if
kernel turned off the LDO4 regulator.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Reviewed-by: Lukasz Majewski <lukma@denx.de> Tested-by: Anand Moon <linux.amoon@gmail.com> Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Alexander Graf [Wed, 30 Jan 2019 10:46:34 +0000 (11:46 +0100)]
x86: Add efi runtime reset
Our selftest will soon test the actual runtime reset function rather than
the boot time one. For this, we need to ensure that the runtime version
actually succeeds on x86 to keep our travis tests work.
So this patch implements an x86 runtime reset function. It is missing
shutdown functionality today, but OSs usually implement that via ACPI
and this function does more than the stub from before, so it's at least
an improvement.
Eventually we will want to have full DM functionality in runtime services.
But this fixes a travis failure and doesn't clutter the code too heavily, so
we should pull it in without the amazing new RTS DM framework.
Function term_read_reply tries to read from the serial console until
the end_char was read. This can hang forever if we are, for some reason,
not able to read the full response (e.g. serial buffer too small,
frame error). This patch moves the timeout detection into
term_read_reply() to assure we will make progress.
Fixes: 6bb591f704 ("efi_loader: query serial console size reliably") Signed-off-by: Matthias Brugger <mbrugger@suse.com>
Throw missing error when an incomplete reply for the cursor position is
received.
Change type of argument of term_get_char() *s32. This renders the function
reusable in efi_cin_read_key().
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Andy Shevchenko [Thu, 28 Feb 2019 08:10:07 +0000 (10:10 +0200)]
x86: edison: Add the rest of UARTs present on board
Intel Edison has three UART ports, i.e.
port 0 - Bluetooth
port 1 - auxiliary, available for general purpose use
port 2 - debugging, usually console output is here
Enable all of them for future use.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Andy Shevchenko [Thu, 28 Feb 2019 15:19:54 +0000 (17:19 +0200)]
x86: acpi: Not every platform has serial console a first device
We may not do an assumption that current console device is always a first
of UCLASS_SERIAL one.
For example, on properly described Intel Edison board the console UART
is a third one.
Use current serial device as described in global data.
Fixes: a61cbad78e67 ("dm: serial: Adjust serial_getinfo() to use proper API") Cc: Simon Glass <sjg@chromium.org> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Marek Vasut [Sat, 9 Mar 2019 20:58:09 +0000 (21:58 +0100)]
ddr: socfpga: Clean up ddr_setup()
Replace the current rather convoluted code using ad-hoc polling
mechanism with a more straightforward code. Use wait_for_bit_le32()
to poll the DDRCALSTAT register instead of local reimplementation.
It makes no sense to pull for 5 seconds before giving up and trying
to restart the EMIF, so instead wait 500 mSec for the calibration to
complete and if this fails, restart the EMIF and try again. Perform
this 32 times instead of 3 times as the original code did.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Chin Liang See <chin.liang.see@intel.com> Cc: Dinh Nguyen <dinguyen@kernel.org> Cc: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> Cc: Tien Fong Chee <tien.fong.chee@intel.com>
Marek Vasut [Sat, 9 Mar 2019 20:57:58 +0000 (21:57 +0100)]
ddr: socfpga: Clean up EMIF reset
The EMIF reset code can well use wait_for_bit_le32() instead of all that
convoluted polling code. Reduce the timeout from 100 seconds to 1 second,
since if the EMIF fails to reset itself in 1 second, it's unlikely longer
wait would help. Make sure to clear the EMIF reset request even if the
SEQ2CORE_INT_RESP_BIT isn't asserted.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Chin Liang See <chin.liang.see@intel.com> Cc: Dinh Nguyen <dinguyen@kernel.org> Cc: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> Cc: Tien Fong Chee <tien.fong.chee@intel.com>
Marek Vasut [Fri, 8 Mar 2019 18:11:55 +0000 (19:11 +0100)]
ddr: socfpga: Fix EMIF clear timeout
The current EMIF clear timeout handling code was applying bitwise
operations to signed data types and as it was, was extremely hard
to read. Replace it with simple wait_for_bit(). Expand the error
handling to make it more readable too.
This patch also changes the timeout for emif_clear() from 14 hours
to 1 second.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Chin Liang See <chin.liang.see@intel.com> Cc: Dinh Nguyen <dinguyen@kernel.org> Cc: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> Cc: Tien Fong Chee <tien.fong.chee@intel.com>
Marek Vasut [Wed, 6 Mar 2019 19:07:55 +0000 (20:07 +0100)]
ARM: socfpga: Fix A10 SoCDK Kconfig
The Kconfig checked for SoCFPGA Arria10 as a platform, instead of
checking for specific board configuration, which works with one
single platform in tree, but not with multiple. Fix it.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Chin Liang See <chin.liang.see@intel.com> Cc: Dinh Nguyen <dinguyen@kernel.org> Cc: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com> Cc: Tien Fong Chee <tien.fong.chee@intel.com>
Dinh Nguyen [Sun, 3 Mar 2019 17:02:10 +0000 (11:02 -0600)]
ARM: socfpga: fix data and tag latency values for pl310 cache controller
The values for the data and tag latency settings on the PL310 caches
controller is an (n-1). For example, the "arm,tag-latency" is specified
as <1 1 1>, so the values that should be written to register should be
0x000. And for the "arm,data-latency" specified as <2 1 1>, the register
value should be 0x010.
Eugeniu Rosca [Sat, 9 Mar 2019 13:04:57 +0000 (14:04 +0100)]
ARM: dts: rmobile: Zap redundant USB/SDHI nodes on M3N
v2019.01 commit cbff9f80cedd ("ARM: dts: rmobile: Sync Gen3 DTs with
Linux 4.19.6") made the sdhi/usb nodes available in r8a77965.dtsi.
Hence, remove the SDHI/USB nodes from r8a77965-u-boot.dtsi. This is
equivalent to partially reverting below v2019.01 commits:
- f529bc551b6d ("ARM: dts: rmobile: Extract USB nodes on M3N")
- 830b94f76867 ("ARM: dts: rmobile: Extract SDHI nodes on M3N")
Duplicating the nodes from <soc>.dtsi to <soc>-u-boot.dtsi is obviously:
- not needed if no U-boot-specific changes are needed in those nodes.
- potentially dangerous/error-prone, since the duplicated properties
override the properties originally defined in <soc>.dtsi. One
possible consequence is that <soc>.dtsi is getting an update from
Linux, while <soc>-u-boot.dtsi stays unchanged. In this situation,
the obsolete property values from <soc>-u-boot.dtsi will take
precedence masking some of the <soc>.dtsi updates, potentially
leading to all kind of obscure issues.
Below is the dtdiff of r8a77965-salvator-x-u-boot.dtb (the only "user"
of r8a77965-u-boot.dtsi) before and after the patch (slightly
reformatted to avoid 'git am/apply' issues and to reduce the width).
What below output means is there is already a mismatch in some of
SDHI/USB nodes between r8a77965.dtsi and r8a77965-u-boot.dtsi. Since no
U-Boot customization is needed in SDHI/USB DT nodes, get rid of them in
r8a77965-u-boot.dtsi.
The update didn't fully work on R-Car3 due to the
CONFIG_SYS_BOOTMAPSZ=8MiB definition left by v2016.09 commit e525d34b476e ("ARM: rmobile: Add support salvator-x board"),
which is explained in below README excerpt:
----8<----
CONFIG_SYS_BOOTMAPSZ:
[..] If CONFIG_SYS_BOOTMAPSZ is undefined,
then the value in "bootm_size" will be used instead.
----8<----
Allow the original commit to accomplish its purpose on R-Car3 targets
by removing the CONFIG_SYS_BOOTMAPSZ definition.
projects / u-boot.git / log