When possible use DMA for reading from CFI flash, this provides upto 5x
improvement in read performance with high speed CFI compliant flashes
like HyperFlash.

Code will gracefully fallback to CPU copy when DMA is unavailable.
Signed-off-by: NVignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: NStefan Roese <sr@denx.de>

Caller would need gracefully handle failures of dma_get_device(),
therefore reduce pr_err() to pr_debug() when DMA device is not found.
Signed-off-by: NVignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: NStefan Roese <sr@denx.de>

bloblist enhancement for alignment
Update ofnode/dev_read phandle function
sandbox keyboard enhancements and fixes

When using SDL for input the SDL key codes are first converted to Linux key
codes and then to matrix entries of the cross wired keyboard.

We must not map any key code to two different places on the keyboard. So
comment out one backslash position.

Update the rest of the file from Linux 5.7.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Add missing SDL key scan codes, e.g.

* shift, ctrl, meta, alt
* brace/bracket
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Add unitary test for phandle access functions
- ofnode_count_phandle_with_args
- ofnode_parse_phandle_with_args
- dev_count_phandle_with_args
- dev_read_phandle_with_args
Signed-off-by: NPatrick Delaunay <patrick.delaunay@st.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

The result of fdt_node_offset_by_phandle is negative for error,
so this patch corrects the check of this result in
fdtdec_parse_phandle_with_args.

This patch allows to have the same behavior with or without OF_LIVE
for the function dev_read_phandle_with_args with cell_name = NULL and
with invalid phandle.
Signed-off-by: NPatrick Delaunay <patrick.delaunay@st.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

The cell_count argument is required when cells_name is NULL.

This patch adds this parameter in live tree API
- of_count_phandle_with_args
- ofnode_count_phandle_with_args
- dev_count_phandle_with_args

This parameter solves issue when these API is used to count
the number of element of a cell without cell name. This parameter
allow to force the size cell.

For example:
  count = dev_count_phandle_with_args(dev, "array", NULL, 3);
Signed-off-by: NPatrick Delaunay <patrick.delaunay@st.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Update the test for node.of_offset because an invalid offset is not
always set to -1 because the return value of the libfdt functions are:
+ an error with a value < 0
+ a valid offset with value >=0

For example, in ofnode_get_by_phandle() function, we have:
node.of_offset = fdt_node_offset_by_phandle(gd->fdt_blob, phandle);
and this function can return -FDT_ERR_BADPHANDLE (-6).

Without this patch, the added test dm_test_ofnode_get_by_phandle failed.
Signed-off-by: NPatrick Delaunay <patrick.delaunay@st.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Include missing log.h and change _ofnode_to_np() to ofnode_to_np() so
that compiling with OF_CHECKS enabled does not break.
Signed-off-by: NStefan Roese <sr@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Adjust a few comments to make the meaning clearer.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Some blobs need a larger alignment than the default. For example, ACPI
tables often start at a 4KB boundary. Add support for this.

Update the size of the test blob to allow these larger records.
Signed-off-by: NSimon Glass <sjg@chromium.org>

The intention which bloblists is that each blob's data is aligned in
memory. At present it is only the headers that are aligned.

Update the code to correct this and add a little more documentation.
Signed-off-by: NSimon Glass <sjg@chromium.org>

When running these tests on sandbox any failures result in very large or
long pointer values which are a pain to work with. Map them to an address
so it is easier to diagnose failures.
Signed-off-by: NSimon Glass <sjg@chromium.org>

It is helpful to be able to see basic statistics about the bloblist and
also to list its contents. Add a 'bloblist' command to handle this.

Put the display functions in the bloblist modules rather than in the
command code itself. That allows showing a list from SPL, where commands
are not available.

Also make bloblist_first/next_blob() static as they are not used outside
this file.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Reformat the command line options chapter so that the command line options
clearly stand out.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Add doc/arch/sandbox.rst to the scope of SANDBOX.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Pull request for UEFI sub-system for efi-2021-01-rc1

The following bugs in the UEFI system are resolved:

* illegal free in EFI_LOAD_FILE2_PROTOCOL implementation
* incorrect documentation of EFI_LOAD_FILE2_PROTOCOL implementation
* output of CRC32 as decimal instead hexadecimal in unit test
* use EfiReservedMemoryType for no-map reserved memory
* avoid unnecessary resets in UEFI unit tests
* call EFI bootmgr even without having /EFI/boot

- generate unique mac address from SoC serial on S400 board
- Add USB support for GXL and AXG SoCs
- Update Gadget code to use the new GXL and AXG USB glue driver
- Add a VIM3 board support to add dynamic PCIe enable in OS DT
- Fix AXG pinmux with requesting GPIOs
- Add missing GPIOA_18 for AXG pinctrl
- Add Amlogic PWM driver

The device tree may contain a /reserved-memory node. The no-map property
of the sub-nodes signals if the memory may be accessed by the UEFI payload
or not.

In the EBBR specification (https://github.com/arm-software/ebbr) the
modeling of the reserved memory has been clarified.

If a reserved memory node in the device tree has the no-map property map,
create a EfiReservedMemoryType memory map entry else use
EfiBootServicesData.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

CONFIG_EFI_GRUB_ARM32 is only needed for architectures with caches that are
not managed via CP15 (or for some outdated buggy versions of GRUB). It
makes more sense to disable the setting per architecture than per defconfig.

Move QEMU's CONFIG_EFI_GRUB_ARM32_WORKAROUND=n from defconfig to Kconfig.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

There is an extra space.
Signed-off-by: NSean Anderson <seanga2@gmail.com>
Reviewed-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

Currently, the EFI bootmgr is only called if there is a EFI binary
inside the path for removable media is found, i.e. /EFI/boot/. This
doesn't make sense. It is the duty of the bootmgr to find out the
path and name of the EFI binary to boot. It should be called even
if there is no /EFI/boot directory.

Thus, call the bootmgr before we try to boot the EFI binary inside
the removable media path.

Also remove the ${fdtcontroladdr} parameter because the fallback is
handled in cmd/bootefi.c and that already takes care of correct settings
if the board has ACPI and thus no device tree at all.
Signed-off-by: NMichael Walle <michael@walle.cc>
Reviewed-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

When we do not execute a test requiring ExitBootServices do not reset the
system after testing.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

Print the CRC32 loaded via the EFI_LOAD_FILE2_PROTOCOL as a hexadecimal
number.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: NIlias Apalodimas <ilias.apalodimas@linaro.org>

Add code to use %x in efi_st_print().
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

strsep() changes the address that its first argument points to.
We cannot use the changed address as argument of free().
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: NIlias Apalodimas <ilias.apalodimas@linaro.org>

U-Boot offers a EFI_LOAD_FILE2_PROTOCOL which the Linux EFI stub can use to
load an initial RAM disk. Update the function comments of the
implementation.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: NIlias Apalodimas <ilias.apalodimas@linaro.org>

Rsync all defconfig files using moveconfig.py
Signed-off-by: NTom Rini <trini@konsulko.com>

Bring in the assorted changes that have been staged in the 'next' branch
prior to release.
Signed-off-by: NTom Rini <trini@konsulko.com>

This adds the driver for the PWM controller found in the Amlogic SoCs.

This PWM is only a set of Gates, Dividers and Counters:
PWM output is achieved by calculating a clock that permits calculating
two periods (low and high). The counter then has to be set to switch after
N cycles for the first half period.
The hardware has no "polarity" setting. This driver reverses the period
cycles (the low length is inverted with the high length) for
PWM_POLARITY_INVERSED.

Disabling the PWM stops the output immediately (without waiting for the
current period to complete first).
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>

Add the missing GPIOA_18 from the missing EE gpio list.
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>

The AXG pmx driver gpio request offset needs the pin base to have the
correct pin number.
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>
Reviewed-by: NMark Kettenis <kettenis@openbsd.org>

The VIM3 on-board  MCU can mux the PCIe/USB3.0 shared differential
lines using a FUSB340TMX USB 3.1 SuperSpeed Data Switch between
an USB3.0 Type A connector and a M.2 Key M slot.
The PHY driving these differential lines is shared between
the USB3.0 controller and the PCIe Controller, thus only
a single controller can use it.

This adds this dynamic switching right before booting Linux
and the configuration steps in the boards documentation.
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>
Reviewed-by: NKevin Hilman <khilman@baylibre.com>
Tested-by: NKevin Hilman <khilman@baylibre.com>
[narmstrong: fixed warning by replacing min() by min_t()]

Use the newly added VIM3 board support instead of the generic W400.
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>
Reviewed-by: NKevin Hilman <khilman@baylibre.com>
Tested-by: NKevin Hilman <khilman@baylibre.com>
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>

The VIM3 will need a specific code to enable PCIe if enabled in the MCU,
thus add a specific board support for VIM3 & VIM3L.
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>
Reviewed-by: NKevin Hilman <khilman@baylibre.com>
Tested-by: NKevin Hilman <khilman@baylibre.com>
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>

This imports the G12A & SM1 SoC and boards DT changes from the Linux
commit 9123e3a74ec7 ("Linux 5.9-rc1").
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>
Reviewed-by: NKevin Hilman <khilman@baylibre.com>
Tested-by: NKevin Hilman <khilman@baylibre.com>
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>

Enable USB Host & Gadget on the Amlogic S400 board.
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>

Add the correcly architectured USB Glue node for Meson AXG and the
S400 board in -u-boot.dtsi until support in upstream Linux then
backported.
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>

Add the board_usb_init()/cleanup() for USB gadget for AXG based
on the code for the G12A architecture.
Signed-off-by: NNeil Armstrong <narmstrong@baylibre.com>