Simon Glass [Thu, 2 May 2019 16:52:20 +0000 (10:52 -0600)]
x86: Update device tree for Chromium OS verified boot
The standard image generated by U-Boot on x86 is u-boot.rom. Add a
separate image called image.bin for verified boot. This supports
verification in TPL of which SPL/U-Boot to start, then jumping to the
correct one, with SPL setting up the SDRAM and U-Boot proper providing
the user interface if needed.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 2 May 2019 16:52:19 +0000 (10:52 -0600)]
x86: Update device tree for TPL
Add TPL binaries to the device x86 binman desciption. When enabled, TPL
will start first, doing the 16-bit init, then jump to SPL and finally
U-Boot proper.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 2 May 2019 16:52:16 +0000 (10:52 -0600)]
x86: Add documentation on the samus flashmap
There are quite a few variables which control where things appear in the
final ROM image. Add a flashmap in the documentation to make this easier
to figure out.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
[bmeng: squashed "x86: Update the memory map a little" in]
[bmeng: fixed typo of 'documentation' in the commit title] Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:59:08 +0000 (21:59 -0600)]
x86: Support PCI VGA ROM when TPL is used
When TPL is in use, U-Boot proper should support initing the VGA ROM even
though the 32-bit init portion is in SPL. Update the condition to handle
this.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 2 May 2019 16:52:15 +0000 (10:52 -0600)]
x86: sysreset: Implement the get_last() method
Add a default implementation of this method which always indicates that
the last reset was a power-on reset. This is the most likely type of reset
and without a PCH-specific driver we cannot determine any other type.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Thu, 2 May 2019 16:52:13 +0000 (10:52 -0600)]
x86: sysreset: Separate out the EFI code
The EFI implementation of reset sits inside the driver and is called
directly from outside the driver, breaking the normal driver-model
conventions. Worse, it passed NULL as the device pointer, hoping that
the called function won't use it, which breaks as soon as code is added
to use it.
Separate out the implementation to improve the situation enough to allow
a future patch to add new sysreset features.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:57 +0000 (21:58 -0600)]
x86: Support saving MRC data from SPL
When SPL is used to set up the memory controller we want to save the MRC
data in SPL to avoid needing to pass it up to U-Boot proper to save. Add a
function to handle that.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:54 +0000 (21:58 -0600)]
x86: broadwell: Allow booting from SPL
At present broadwell only supports booting straight into U-Boot proper.
Add a separate init file to boot from SPL into U-Boot proper, and select
it when SPL is in use.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:51 +0000 (21:58 -0600)]
x86: broadwell: Split CPU init
Split the CPU init into two parts - the 'full' init which happens in the
first U-Boot phase, and the rest of the init that happens on subsequent
stages.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:50 +0000 (21:58 -0600)]
x86: broadwell: Move init of debug UART to cpu.c
At present the debug UART is set up in sdram.c which is not the best place
since it has nothing in particular to do with SDRAM. Since we want to
support initing this in SPL too, move it to a common file.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
[bmeng: added 'broadwell' tag in the commit title] Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:49 +0000 (21:58 -0600)]
x86: broadwell: Allow SDRAM init from SPL
At present, for broadwell, SDRAM is always set up in U-Boot proper since
the 64-bit mode (which uses SDRAM init in SPL) is not supported.
Update the code to allow SDRAM init in SPL instead so that U-Boot proper
can be loaded into SDRAM and run from there. This allows U-Boot to be
compressed to reduce space, since it is not necessary to run it directly
from flash. It could later allow us to support 64-bit U-Boot on broadwell.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:46 +0000 (21:58 -0600)]
x86: Support booting with TPL
Some boards want to use TPL as the first phase of U-Boot. This allows
selection of A or B SPL phases, thus allowing the memory init to be
upgraded in the field.
Add a new Kconfig option for this.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:42 +0000 (21:58 -0600)]
x86: Add a way to reinit the cpu
We cannot init the CPU fully both than once during a boot. Add a new
function which can be called to figure out the CPU identity, but which
does not change anything. For x86_64, this is empty for now.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Unfortunately this has a dramatic impact on the pre-relocation memory
used on x86 platforms (increasing it by 2KB) since it increases the
overhead for each PCI device from 220 bytes to 412 bytes.
This means that all PCI devices have the controller struct associated
with them. The solution is to move the regions[] member out of the array,
makes its size dynamic, or split UCLASS_PCI into controllers and
non-controllers, as the comment suggests.
For now, revert the commit to get things running again.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:38 +0000 (21:58 -0600)]
spl: Allow sandbox to build a device-tree file
At present only OF_SEPARATE is considered valid for building a device-tree
file in SPL. However sandbox uses OF_HOSTFILE instead. Update the logic to
handle this and make it easier to understand.
Note that the new logic is not quite the same as the old logic. It was
previously assumed that checking for:
producing 'yy' meant that the first two were 'y' and the last was empty.
Strictly speaking it would be possible for any two of the three to be 'y'
and still yield the same result. However, that was not the intention of
the new logic, since OF_PLATDATA always ensures that no device-tree file
is included. So in effect the new logic is the same, with the addition of
OF_HOSTFILE as an option for OP_SEPARATE.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:39 +0000 (21:58 -0600)]
binman: Allow sections to have an offset
At present sections are always placed automatically. Even if an 'offset'
property is provided it is ignored. Update the logic to support an offset
for sections.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:34 +0000 (21:58 -0600)]
binman: Don't show image-skip message by default
This message is not very important since it is simply indicating that the
user's instructions are being followed. Only show it when the verbosity
level is above the default.
Also drop the unnecessary extra newline on this message, which causes two
line breaks.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Fri, 26 Apr 2019 03:58:33 +0000 (21:58 -0600)]
binman: Don't generate an error in 'text' entry constructor
It is not good practice to raise an exception in a constructor. In this
case the 'text' entry may not actually be used, if -i is used to filter
out the images that get built.
Move the exception to where the data is actually used.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Chris Brandt [Wed, 23 Aug 2017 19:53:59 +0000 (14:53 -0500)]
ARM: dts: renesas: Add RZ/A1 GR-Peach board
Add board code and DTs for Renesas RZ/A1 SoC-based GR-Peach,
which is a cheap development platform with RZ/A1H SoC. The
DTs are imported from Linux 5.0.11, commit d5a2675b207d .
Currently supported are UART, ethernet and RPC SPI. The board
can be booted from RPC SPI by writing the u-boot.bin binary
to the beginning of the SPI NOR, e.g. using the "sf" command.
The board can also be booted via JTAG by setting text base to
0x20020000, loading u-boot.bin there via JTAG and executing it
from that address.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Chris Brandt <chris.brandt@renesas.com> Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Ang, Chee Hong [Fri, 3 May 2019 08:18:27 +0000 (01:18 -0700)]
ARM: socfpga: stratix10: Probe FPGA status before bridge enable
Send CONFIG_STATUS and RECONFIG_STATUS mailbox commands to Secure
Device Manager (SDM) to get the status of FPGA and make sure the
FPGA is in user mode before enable the bridge.
Signed-off-by: Ang, Chee Hong <chee.hong.ang@intel.com>
Ang, Chee Hong [Fri, 3 May 2019 08:19:08 +0000 (01:19 -0700)]
ARM: socfpga: stratix10: Disable FPGA2SOC reset
Software must never reset FPGA2SOC bridge. This bridge must only be
reset by POR/COLD/WARM reset. Asserting the FPGA2SOC reset by software
can cause the SoC to lock-up if there are traffics being drived into
FPGA2SOC bridge.
Signed-off-by: Ang, Chee Hong <chee.hong.ang@intel.com>
Marek Behún [Thu, 2 May 2019 13:29:12 +0000 (15:29 +0200)]
cmd: pxe: add board specific PXE default path
The list of PXE default paths contains ARCH and SOC specific paths, but
one PXE server can serve different board with the same ARCH and SOC.
This is the case for Turris Omnia and Turris Mox, where ARCH=arm and
SOC=mvebu.
If CONFIG_SYS_BOARD is defined, also try "default-$ARCH-$SOC-$BOARD"
path.
Adam Ford [Tue, 30 Apr 2019 10:21:42 +0000 (05:21 -0500)]
ARM: da850evm: Enable da850-ohci USB host controller
The DA850 EVM has one USB 1.1 OHCI Host controller. With the
host controller now support DM_USB, this patch enables
the respective functions for the da850evm.
Adam Ford [Tue, 30 Apr 2019 10:21:41 +0000 (05:21 -0500)]
usb: ohci: ohci-da8xx: Enable da850-ohci driver with DM support
This patch reuses some former code for the hawkboard, combines it
with some some similar DM_USB compatible code for the OHCI driver,
and enables the use of the da850's OHCI controller with DM_USB
compatibility.
Marek Behún [Mon, 29 Apr 2019 20:40:44 +0000 (22:40 +0200)]
lib: add Zstandard decompression support
Add the zstd library from Linux kernel (only decompression support).
There are minimal changes to build with U-Boot, otherwise the files are
identical to Linux commit dc35da16 from March 2018, the files had not
been touched since in kernel. Also SPDX lincese tags were added.
Marek Behún [Mon, 29 Apr 2019 20:40:43 +0000 (22:40 +0200)]
lib: Add xxhash support
This adds the xxhash support from Linux. Files are almost identical to
those added to Linux in commit 5d240522 ("lib: Add xxhash module") (they
haven't been touched since in Linux). The only difference is to add some
includes to be compatible with U-Boot. Also SPDX lincese tags were
added.
Remove the RDLVL_MASK, RDLVLGATE_MASK, WRLVL_MASK & enable
PHY_INVERT_CLKOUT to enable Hardware leveling for am437x
as recommended by EMIF Tools app note:
display_options_get_banner_priv() overwrites bytes before the start of the
buffer if the buffer size is less then 3. This case occurs in the Sandbox
when executing the `ut_print` command.
Correctly handle small buffer sizes. Adjust the print unit test to catch
when bytes before the buffer are overwritten.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Simon Glass <sjg@chromium.org>
spl: fit: Always enable tracking of os-type if SPL_OS_BOOT is enabled
FIT_IMAGE_TINY is used to reduce the size of the SPL by removing os-type
tracking and recording the loadables into the loaded FDT. When this option
is enabled, it is assumed that the next stage firmware is u-boot.
However this does not play well with the SPL_OS_BOOT option that enables
loading different type of next stage firmware, like the OS itself.
When SPL_OS_BOOT is used, do not disable os-tracking. The added footprint
is about 300 Bytes.
Add a PL310 cache controller driver that is usually found on
ARMv7(32-bit) devices. The driver configures the cache settings that can
be found in the device tree files.
This initial revision only configures basic settings(data & instruction
prefetch, shared-override, data & tag latency). I believe these are the
settings that affect performance the most. Comprehensive settings can be
done by the OS.
Reviewed-by: Simon Glass <sjg@chromium.org> Signed-off-by: Dinh Nguyen <dinguyen@kernel.org>
Linux commit 8ecd7f5970c5 ("ARM: 8483/1: Documentation: l2c: Rename
l2cc to l2c2x0")
Linux docs:
Documentation/devicetree/bindings/arm/l2c2x0.txt
Copied from Linux kernel v5.0.
"The documentation in the l2cc.txt is specific to the L2 cache
controllers L2C210/L2C220/L2C310 (also known as PL210/PL220/PL310
and variants) and not generic as the file name implies. It's not
valid for integrated L2 controllers as found in e.g.
Cortex-A15/A7/A57/A53."
Reviewed-by: Simon Glass <sjg@chromium.org> Signed-off-by: Dinh Nguyen <dinguyen@kernel.org>
arch: armv8: Provide a way to disable cache maintenance ops
On AM654 SoC(arm64) which is IO coherent and has L3 Cache, cache
maintenance operations being done to support non-coherent platforms
causes issues.
For example, here is how U-Boot prepares/handles a buffer to receive
data from a device (DMA Write). This may vary slightly depending on the
driver framework:
Start DMA to write to destination buffer
Wait for DMA to be done (dma_receive()/dma_memcpy())
Invalidate destination buffer (invalidate_dcache_range())
Read from destination buffer
The invalidate after the DMA is needed in order to read latest data from
memory that’s updated by DMA write. Also, in case random prefetch has
pulled in buffer data during the “wait for DMA” before the DMA has
written to it. This works well for non-coherent architectures.
In case of coherent architecture with L3 cache, DMA write would directly
update L3 cache contents (assuming cacheline is present in L3) without
updating the DDR memory. So invalidate after “wait for DMA” in above
sequence would discard latest data and read will cause stale data to be
fetched from DDR. Therefore invalidate after “wait for DMA” is not
always correct on coherent architecture.
Therefore, provide a Kconfig option to disable cache maintenance ops on
coherent architectures. This has added benefit of improving the
performance of DMA transfers as we no longer need to invalidate/flush
individual cache lines(especially for buffer thats several KBs in size).
In order to facilitate use of same Kconfig across different
architecture, I have added the symbol to top level arch/Kconfig file.
Patch currently disables cache maintenance ops for arm64 only.
flush_dcache_all() and invalidate_dcache_all() are exclusively used
during enabling/disabling dcache and hence are not disabled.
Alex Kiernan [Thu, 18 Apr 2019 20:34:55 +0000 (20:34 +0000)]
Refactor IMAGE_ENABLE_VERIFY to handle builds without SPL verification
If building with SPL_LOAD_FIT_FULL and FIT_SIGNATURE, but without
SPL_FIT_SIGNATURE then the build fails with:
common/built-in.o: In function `fit_image_verify_with_data':
common/image-fit.c:1220: undefined reference to `fit_image_verify_required_sigs'
common/image-fit.c:1244: undefined reference to `fit_image_check_sig'
common/built-in.o: In function `fit_image_load':
common/image-fit.c:1857: undefined reference to `fit_config_verify'
Refactor so that host builds still depend on FIT_SIGNATURE, but target
builds check FIT_SIGNATURE/SPL_FIT_SIGNATURE dependent on the build.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Sun, 5 May 2019 00:02:31 +0000 (20:02 -0400)]
Merge tag 'mips-pull-2019-05-03' of git://git.denx.de/u-boot-mips
- mscc: small fixes, enhance network support for Serval, Luton and Ocelot
- mt7620: rename arch to more generic name mtmips
- mips: pass initrd addresses via DT as physical addresses
Marek Vasut [Mon, 18 Mar 2019 03:49:21 +0000 (04:49 +0100)]
cmd: mmc: Make Mode: printout consistent
The "Mode :" line is the only one in "mmc info" output that has a
space in front of the colon. Drop the space to make it consistent
with the rest of the output, e.g.:
=> mmc dev 1 ; mmc info
switch to partitions #0, OK
mmc1 is current device
Device: sd@ee160000
Manufacturer ID: 3
OEM: 5344
Name: SL08G
Bus Speed: 50000000
Mode : SD High Speed (50MHz)
^------------------------------ Remove this space
Rd Block Len: 512
SD version 3.0
High Capacity: Yes
Capacity: 7.4 GiB
Bus Width: 1-bit
Erase Group Size: 512 Bytes
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Jaehoon Chung <jh80.chung@samsung.com> CC: Tom Rini <trini@konsulko.com>
Marek Vasut [Sat, 13 Apr 2019 09:42:34 +0000 (11:42 +0200)]
net: ravb: Avoid unsupported internal delay mode for R-Car E3/D3
According to the R-Car Gen3 Hardware Manual Rev 1.50 of Nov 30, 2018, the
TX clock internal delay mode isn't supported on R-Car E3 (r8a77990) or D3
(r8a77995).
Avoid setting the APSR:TDM bit on these SoCs. Moreover, only set APSR:TDM
when the DT explicitly specifies RGMII ID or TXID mode instead of setting
it unconditionally when the PHY link speed is 1000 Mbit/s.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org> Cc: Joe Hershberger <joe.hershberger@ni.com>