Jagan Teki [Tue, 17 Sep 2019 06:10:38 +0000 (11:40 +0530)]
ram: rk3288: Initialize dram for TPL builds
Few of the rk3288 boards like tinker, vyasa are using
TPL, SPL bootchain so the dram initialization must needed
during TPL stage. So add proper ifconstruct to satisfy
both TPL, SPL and SPL-only bootchain boards.
This eventually fixing TPL to SPL handoff, otherwise missing
dram initilaztion at TPL stage would leads to SPL hang.
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Reviewed-by: Kever Yang<kever.yang@rock-chips.com>
Jagan Teki [Tue, 17 Sep 2019 06:10:37 +0000 (11:40 +0530)]
configs: vyasa-rk3288: Fix SPL_TEXT_BASE
The initial Vyasa-rk3288 TPL implementation is to reuse the
SPL_TEXT_BASE for TPL and SPL as 0x0 and 0xff704000 respectively.
But the below commit implements the reverse way of using TEXT_BASE's
like 0xff704000 for TPL and 0x0 for SPL and which indeed update
the SPL_TEXT_BASE for vyasa-rk3288 board.
"rockchip: Kconfig: enable TPL support for rk3328"
(sha1: 3f47db0275e0f1c1e89d8f13f8b4cdc86b1038ca)
So, fix by dropping the legacy SPL_TEXT_BASE and it will
reassign 0x0 by default based on Kconfig definition.
Reported-by: Jagan Teki <jagan@amarulasolutions.com> Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Reviewed-by: Kever Yang<kever.yang@rock-chips.com>
configs: rockpro64-rk3399: Enable CONFIG_MISC_INIT_R and ROCKCHIP_EFUSE
This enables reading the cpuid from e-fuse, and deriving a static
MAC address from it. Without this, the ethernet interface on the
rockpro64 can't be used to boot.
Signed-off-by: Hugh Cole-Baker <sigmaris@gmail.com> Cc: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
mtd: spi-nor: enable protection ops for SST26 flash series
Commit c4e8862308d4 (mtd: spi: Switch to new SPI NOR framework)
performs switch from previous 'spi_flash' infrastructure without
proper testing/investigations which results in a regressions for
SST26 flash series.
Enable protection ops for SST26 flash series which were
previously enabled by
Commit 3d4fed87a5fa (mtd: sf: Add support of sst26wf* flash ICs
protection ops)
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com> Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
mtd: spi-nor: add missing SST26* flash IC protection ops
Commit c4e8862308d4 (mtd: spi: Switch to new SPI NOR framework)
performs switch from previous 'spi_flash' infrastructure without
proper testing/investigations which results in a regressions for
SST26 flash series.
Add missing SST26* flash IC protection ops which were introduced
previously by
Commit 3d4fed87a5fa (mtd: sf: Add support of sst26wf* flash ICs
protection ops)
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com> Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Jagan Teki [Mon, 22 Jul 2019 11:52:56 +0000 (17:22 +0530)]
spi: Add spi_write_then_read
Add support for SPI synchronous write followed by read,
this is common interface call from spi-nor to spi drivers.
Reviewed-by: Simon Glass <sjg@chromium.org> Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Tested-by: Adam Ford <aford173@gmail.com> #da850-evm
spi: soft_spi: Fix data abort if slave is not probed
In case spi_get_bus_and_cs callback is used, spi bus is first probed
then slave devices are probed. To avoid a data abort in soft_spi probe
function, we need to check that (slave != NULL).
If slave is NULL, cs_flags and clk_flags will be initialized with
respectively GPIOD_ACTIVE_LOW and 0.
The prescaler value computation can yield wrong result if given 0x1f at
the beginning: the value is computed to be 0x20, but the maximum value
the register can hold 0x1f, so the actual stored value in this case is
0, which is obviously wrong.
Set the upper bound of the value to 0x1f with the min macro.
Signed-off-by: Marek Behún <marek.behun@nic.cz> Reviewed-by: Stefan Roese <sr@denx.de> Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Faiz Abbas [Tue, 27 Aug 2019 08:11:23 +0000 (13:41 +0530)]
ARM: dts: dra74x: Fix iodelay configuration for mmc3
According to the latest am572x[1] and dra74x[2] data manuals, mmc3
default, hs, sdr12 and sdr25 modes use iodelay values given in
MMC3_MANUAL1. Set the MODE_SELECT bit for these so that manual mode is
selected and correct iodelay values are configured.
API get_ti_sci_handle() is relying on the device-tree node name
to be "dmsc" for probing the ti_sci device. But with the introduction
of debug messages for dmsc, the node name changed to dmsc@44083000.
Because of this ti_sci is never probed cause a boot failure. Instead
of relying on device-tree node name, use the first available firmware
node for probing ti_sci.
Adam Ford [Mon, 2 Sep 2019 11:52:01 +0000 (06:52 -0500)]
ARM: omapl138_lcdk: Remove dead code
The header it littered with #ifdefs and #defines and that appear
to be legacy associations to the older da850-evm and in some cases
obsolete with either Kconfig or DM migrations. This patch removes
these legacy references.
Enable driver model for USB, MMC, SPI and SPI_FLASH. Also enable BLK.
This will remove the following compile warnings:
===================== WARNING ======================
This board does not use CONFIG_DM_MMC. Please update
the board to use CONFIG_DM_MMC before the v2019.04 release.
====================================================
===================== WARNING ======================
This board does not use CONFIG_DM_USB. Please update
the board to use CONFIG_DM_USB before the v2019.07 release.
====================================================
===================== WARNING ======================
This board does not use CONFIG_DM_SPI_FLASH. Please update
the board to use CONFIG_SPI_FLASH before the v2019.07 release.
====================================================
Target was compile tested, build was clean.
Add device tree from Linux for driver model conversion
and enable OF_CONTROL. This will remove the following compile
warning:
==================================================
Device Tree Source is not correctly specified.
Please define 'CONFIG_DEFAULT_DEVICE_TREE'
or build with 'DEVICE_TREE=<device_tree>' argument
===================================================
Target was compile tested, build was clean.
In case dma_ring_reset_quirk is not set the k3_ringacc_ring_reset_dma will
just exit without ring reset. Fix it, by adding ring reset call in case
dma_ring_reset_quirk is not.
The code that generates the compatible property concatenates the
ethernet phy id and clause-compatible information without
separating them with a comma, resulting into no ethernet phy driver
getting loaded by Linux kernel.
Suffix phy_id_compatible_str with comma to fix this
Pankaj Bansal [Sat, 17 Aug 2019 01:07:32 +0000 (01:07 +0000)]
board: fsl: lx2160a: implement board_fix_fdt
lx2160a rev1 and rev2 SoC has different pcie controller.
The pcie controller device tree node fields "compatible"
and registers names needs to be updated accordingly
This change in device tree is handled as part of
fdt fixups. These changes would only be applied
if the soc revision is not rev1.
Chuanhua Han [Thu, 8 Aug 2019 09:04:58 +0000 (17:04 +0800)]
armv8: fsl-layerscape: Update I2C clock divider
By default, i2c input clock is programmed at
platform clk / 2 in u-boot, but this is not
correct for all the platforms,
Update I2C clock divider's default values as per
SoC (LS1012A, LS1028A, LX2160A and LS1088A).
Signed-off-by: Chuanhua Han <chuanhua.han@nxp.com> Reviewed-by: Priyanka Jain <priyanka.jain@nxp.com>
Thomas Schaefer [Thu, 8 Aug 2019 08:00:30 +0000 (16:00 +0800)]
armv8: ls1028a: configure PMU's PCTBENR to enable WDT
The SP805-WDT module on LS1028A requires configuration of PMU's
PCTBENR register to enable watchdog counter decrement and reset
signal generation. The watchdog clock needs to be enabled first.
board/lx2160a: Fix MC firmware loading for SD boot
During boot, u-boot reads MC, DPL, DPC firmware from SD card
and copies to DDR. Update DDR addresses to which these firmwares
are copied as per memory map of these firmwares on SD-card
so that isolation between the regions of various firmwares
is maintained to avoid geting overwritten.
NXP fsl_esdhc controller supports two reference clocks:
platform clock and peripheral clock
Peripheral clock can provide higher clock frequency
which is required to be used for tuning of SD UHS mode
and eMMC HS200/HS400 modes.
Peripheral clock is enabled by default by defining config
option FSL_ESDHC_USE_PERIPHERAL_CLK if eMMC HS200/HS400 modes
are supported.
Tom Rini [Thu, 12 Sep 2019 03:08:34 +0000 (23:08 -0400)]
Merge tag 'u-boot-rockchip-20190912' of https://gitlab.denx.de/u-boot/custodians/u-boot-rockchip
- add idbloader.img target for rockchip tpl+spl;
- usb ehci/ohci: go on process if clock driver don't have clk_enable();
- remove clk_enable() for rockchip clock drivers;
- add boot order for rockpro64
Kever Yang [Wed, 28 Aug 2019 08:23:47 +0000 (16:23 +0800)]
usb: ohci-generic: don't probe fail if there is no clk_enable() ops
Some clock driver do not have a clk_enable() call back, and we should not
treat this as fail in ehci probe like other modules, eg. clk_enabl_bulk()
do not return fail if ret value is '-ENOSYS'
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Reviewed-by: Patrice Chotard <patrice.chotard@st.com>
Kever Yang [Wed, 28 Aug 2019 08:23:46 +0000 (16:23 +0800)]
usb: ehci-generic: don't probe fail if there is no clk_enable() ops
Some clock driver do not have a clk_enable() call back, and we should not
treat this as fail in ehci probe like other modules, eg. clk_enabl_bulk()
do not return fail if ret value is '-ENOSYS'
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Reviewed-by: Patrice Chotard <patrice.chotard@st.com>
Michal Suchanek [Sun, 18 Aug 2019 08:55:28 +0000 (10:55 +0200)]
dwc2: use the nonblock argument in submit_int_msg
An USB 1.1 keyboard connected to dwc2 through a high-speed hub does not
report status until it changes. With this patch you can enable keyboard
by pressing a key while USB devices are probed. Without a keypress no
state is reported and the probe times out. We don't want to wait for a
keypress or timeout while polling for keypresses so implement an int_msg
nonblock variant that exits early on error.
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
- Tangier ACPI table fixes
- Support getting high memory size on QEMU x86
- Show UEFI images involved in crash for x86
- EFI loader conventional memory map fix
Tom Rini [Tue, 10 Sep 2019 12:51:17 +0000 (08:51 -0400)]
Merge tag 'u-boot-amlogic-20190910' of https://gitlab.denx.de/u-boot/custodians/u-boot-amlogic
- Add support for dis_u2_susphy_quirk in the xhci-dwc3 driver to fix boot when
a device is plugged only in the OTG capable port for libretech-ac and libretech-cc
Andy Shevchenko [Thu, 29 Aug 2019 14:04:20 +0000 (17:04 +0300)]
x86: tangier: Fix off-by-one error when preparing CSRT
Intel iDMA 32-bit controller has 17 bits for the maximum block size value.
Due to nature of the binary number representation the maximum value is
2^17 - 1. The original code misses the latter part in equation.
Fixes: 5e99fde34a77 ("x86: tangier: Populate CSRT for shared DMA controller") Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Andy Shevchenko [Thu, 29 Aug 2019 14:04:19 +0000 (17:04 +0300)]
x86: tangier: Reserve PCI ECAM in motherboard resources
Per PCI firmware specification the ACPI has to reserve the memory
which is defined as PCI ECAM.
Fixes: 39665beed6f7 ("x86: tangier: Enable ACPI support for Intel Tangier") Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Andy Shevchenko [Thu, 29 Aug 2019 14:04:18 +0000 (17:04 +0300)]
x86: acpi: Annotate struct acpi_table_header with __packed
GCC 9.2 starts complaining about possible pointer misalignment of
pointers to the unpacked (alignment=4) structures in the packed
(alignment=1) ones:
CC arch/x86/cpu/tangier/acpi.o
arch/x86/cpu/tangier/acpi.c: In function ‘acpi_create_fadt’:
arch/x86/cpu/tangier/acpi.c:22:37: warning: taking address of packed
member of ‘struct acpi_fadt’ may result in an unaligned pointer value
[-Waddress-of-packed-member]
22 | struct acpi_table_header *header = &(fadt->header);
CC arch/x86/lib/acpi_table.o
arch/x86/lib/acpi_table.c: In function ‘acpi_create_spcr’:
arch/x86/lib/acpi_table.c:366:37: warning: taking address of packed
member of ‘struct acpi_spcr’ may result in an unaligned pointer value
[-Waddress-of-packed-member]
366 | struct acpi_table_header *header = &(spcr->header);
Fix the potential issues by annotating embedded structures with
__packed even though they are packed naturally.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
[bmeng: add GCC version number in the commit message] Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Neil Armstrong [Mon, 9 Sep 2019 18:52:39 +0000 (18:52 +0000)]
usb: xhci-dwc3: Add support for dis_u2_susphy_quirk
This quirk is necessary for the Amlogic GXL SoCs otherwise the
Port 2 PHY doesn't get out of suspend and U-Boot resets the board after:
XHCI timeout on event type 33... cannot recover.
BUG: failure at drivers/usb/host/xhci-ring.c:474/xhci_wait_for_event()!
BUG!
This quirk is also handled in the dwc3 core code, but until the
xhci-dwc3 driver uses the dwc3 core, the quirk must be handled here
to fix USB support on the Amlogic libretech-cc and libretech-ac board
when a device is only plugged in the OTG port.
Cc: Yuri Frolov <crashing.kernel@gmail.com> Cc: Bin Meng <bmeng.cn@gmail.com> Fixes: dc9cdf859e ("usb: dwc3: Add DWC3 controller driver support") Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Park, Aiden [Tue, 3 Sep 2019 17:43:46 +0000 (17:43 +0000)]
x86: efi_loader: Use efi_add_conventional_memory_map()
Use efi_add_conventional_memory_map() to configure EFI conventional memory
properly with ram_top value. This will give 32-bit mode U-Boot proper
conventional memory regions even if e820 has an entry which is greater than
32-bit address space.
Signed-off-by: Aiden Park <aiden.park@intel.com> Tested-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
[bmeng: fixed some typos in the commit message] Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Bin Meng [Thu, 29 Aug 2019 09:53:03 +0000 (02:53 -0700)]
x86: Drop weak version board_get_usable_ram_top()
Every x86 platform provides board_get_usable_ram_top(), hence there
is no need to provide a weak version board_get_usable_ram_top(), not
to mention there is another weak version board_get_usable_ram_top()
in common/board_f.c.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Aiden Park <aiden.park@intel.com>
Andy Shevchenko [Mon, 26 Aug 2019 01:48:15 +0000 (09:48 +0800)]
x86: acpi: Slightly reduce binary size of ACPI tables for Tangier
Using ACPI predefined macros, such as Zero or One, will reduce a binary
size of resulting ACPI tables.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
[bmeng: manually fixed the conflicts when applying] Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Tom Rini [Mon, 9 Sep 2019 14:55:45 +0000 (10:55 -0400)]
arm: ti: Add missing "=" from previous fix
While the original patch to fix a regression in distro boot for mmc on
these platforms had the correct syntax, I broke the change while
applying. Add back in the missing "=" here so that the syntax is
correct.
Reported-by: Andre Heider <a.heider@gmail.com> Fixes: 27e0f3bcf075 ("arm: ti: Fix regression in distro boot for mmc") Signed-off-by: Tom Rini <trini@konsulko.com>
efi_loader: volume name in EFI_FILE_PROTOCOL.GetInfo()
We cannot determine the volume name in U-Boot. Instead of providing a dummy
volume name in case of EFI_FILE_SYSTEM_INFO and EFI_UNSUPPORTED in case of
EFI_FILE_SYSTEM_VOLUME_LABEL consistently return an empty string.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
EFI_FILE_PROTOCOL.Read() is used both to read files and directories.
When reaching the end of a directory we always have to return buffer size
zero irrespective of the incoming buffer size. (The described scenario for
a Shim quirk cannot arise because every directory has at least '.' and '..'
as entries.)
Even when the buffer_size is too small multiple times we have to keep a
reference to our last read directory entry.
When we return to the start of the directory via SetPosition() we must
remove the reference to a previously kept directory entry.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
projects / u-boot.git / log