platform/kernel/u-boot.git
4 years agolog: uclass_get_name() depends on CONFIG_SPL_DM
Heinrich Schuchardt [Mon, 8 Jun 2020 16:04:22 +0000 (18:04 +0200)]
log: uclass_get_name() depends on CONFIG_SPL_DM

If CONFIG_SPL_DM=n and CONFIG_SPL_LOG=y a build error occurs:

ld.bfd: common/built-in.o: in function `log_get_cat_name':
common/log.c:48: undefined reference to `uclass_get_name'
make[1]: *** [scripts/Makefile.spl:422: spl/u-boot-spl] Error 1

Call uclass_get_name() only if DM is enabled.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agosandbox: make RAM size configurable
Heinrich Schuchardt [Sun, 7 Jun 2020 16:47:35 +0000 (18:47 +0200)]
sandbox: make RAM size configurable

Up to now the RAM size of the sandbox is hard coded as 128 MiB. This does
not allow testing the correct handling of addresses outside the 32bit
range. 128 MiB is also rather small when tracing functions where the trace
is written to RAM.

Provide configuration variable CONFIG_SANDBOX_RAM_SIZE_MB to set the RAM
size in MiB. It defaults to 128 MiB with a minimum of 64 MiB.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agopatman: Drop import of test_util in test_util
Simon Glass [Sun, 7 Jun 2020 12:45:50 +0000 (06:45 -0600)]
patman: Drop import of test_util in test_util

This module doesn't need to import itself. It causes problems on
very old Python 3 (e.g. 3.4.0). Drop it.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agopatman: Avoid importing gitutil in settings
Simon Glass [Sun, 7 Jun 2020 12:45:49 +0000 (06:45 -0600)]
patman: Avoid importing gitutil in settings

Pass this module in so that settings does not need to import it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Stefan Bosch <stefan_b@posteo.net>
4 years agopatman: Pass in maintainer dirs to avoid and import
Simon Glass [Sun, 7 Jun 2020 12:45:48 +0000 (06:45 -0600)]
patman: Pass in maintainer dirs to avoid and import

Adjust the get_maintainer module to accept a list of directories to search
for the script. This avoids needing to import gitutil.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Stefan Bosch <stefan_b@posteo.net>
4 years agopatman: Use a dict in gitutil to avoid importing series
Simon Glass [Sun, 7 Jun 2020 12:45:47 +0000 (06:45 -0600)]
patman: Use a dict in gitutil to avoid importing series

Only a few members of this class are used and only in a test. To avoid
importing the module, convert the test to use a dict.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Stefan Bosch <stefan_b@posteo.net>
4 years agopatman: Avoid circular dependency between command and tools
Simon Glass [Sun, 7 Jun 2020 12:45:46 +0000 (06:45 -0600)]
patman: Avoid circular dependency between command and tools

This seems to cause problems in some cases. Split the dependency by
copying the code to command.

Reported-by: Stefan Bosch <stefan_b@posteo.net>
Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agopatman: Drop unnecessary import in gitutil
Simon Glass [Sun, 7 Jun 2020 12:45:45 +0000 (06:45 -0600)]
patman: Drop unnecessary import in gitutil

The checkpatch module is not used, so drop it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Stefan Bosch <stefan_b@posteo.net>
4 years agosandbox: spi: sandbox_sf_state_name() is required
Heinrich Schuchardt [Sun, 7 Jun 2020 07:27:51 +0000 (09:27 +0200)]
sandbox: spi: sandbox_sf_state_name() is required

Compiling drivers/mtd/spi/sandbox.c fails when compiled with
CONFIG_LOG=n:

In file included from include/common.h:20,
                 from drivers/mtd/spi/sandbox.c:13:
drivers/mtd/spi/sandbox.c:295:15: error: format ‘%s’ expects argument of
type ‘char *’, but argument 7 has type ‘int’ [-Werror=format=]
  295 |   log_content(" cmd: transition to %s state\n",
      |               ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
include/linux/printk.h:37:21: note: in definition of macro ‘pr_fmt’
   37 | #define pr_fmt(fmt) fmt
      |                     ^~~
include/log.h:128:30: note: in expansion of macro ‘log_nop’
  128 | #define log_content(_fmt...) log_nop(LOG_CATEGORY, \
      |                              ^~~~~~~
drivers/mtd/spi/sandbox.c:295:3: note: in expansion of macro
‘log_content’
  295 |   log_content(" cmd: transition to %s state\n",
      |   ^~~~~~~~~~~
drivers/mtd/spi/sandbox.c:295:37: note: format string is defined here
  295 |   log_content(" cmd: transition to %s state\n",
      |                                    ~^
      |                                     |
      |                                     char *
      |                                    %d

Supply function sandbox_sf_state_name() independent of CONFIG_LOG.

Fixes: c3aed5db591e ("sandbox: spi: Add more logging")
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agosandbox: handling out of memory
Heinrich Schuchardt [Thu, 4 Jun 2020 17:28:22 +0000 (19:28 +0200)]
sandbox: handling out of memory

assert() only works in debug mode. So checking a successful memory
allocation should not use assert().

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agoMerge https://gitlab.denx.de/u-boot/custodians/u-boot-x86
Tom Rini [Thu, 9 Jul 2020 13:54:22 +0000 (09:54 -0400)]
Merge https://gitlab.denx.de/u-boot/custodians/u-boot-x86

- Add two- and three-argument versions of CONFIG_IS_ENABLED in
  linux/kconfig.h
- Adds a new feature which supports copying modified parts of
  the frame buffer to the uncached hardware buffer
- Enable the copy framebuffer on various x86 targets

4 years agoMerge tag 'for-v2020.10' of https://gitlab.denx.de/u-boot/custodians/u-boot-i2c
Tom Rini [Thu, 9 Jul 2020 12:22:44 +0000 (08:22 -0400)]
Merge tag 'for-v2020.10' of https://gitlab.denx.de/u-boot/custodians/u-boot-i2c

i2c changes for v2020.10
- Add support for I2C controllers found on Octeon II/III and Octeon TX
  TX2 SoC platforms.
- Add I2C controller support for Cortina Access CAxxxx SoCs
- new rtc methods, rtc command, and tests
- imx_lpi2c: Improve the codes to use private data
- stm32f7_i2c.c: Add new compatible "st,stm32mp15-i2c"
- stm32f7_i2c.c: Add Fast Mode Plus support
- pwm: Add PWM driver for SiFive SoC

4 years agoMerge https://gitlab.denx.de/u-boot/custodians/u-boot-marvell
Tom Rini [Thu, 9 Jul 2020 12:21:26 +0000 (08:21 -0400)]
Merge https://gitlab.denx.de/u-boot/custodians/u-boot-marvell

- Armada 38x DDR3 fixes, enhancements (Chris)
- Armada 38x UTMI PHY SerDes fix (Chris)
- Helios4 update - sync with clearfog (Dennis)
- LaCie Kirkwood board rework - enable DM (Simon)
- net/mvpp2 memory init fix (Sven)

4 years agonet: mvpp2: fix second cp110 initialization
Sven Auhagen [Wed, 1 Jul 2020 15:43:43 +0000 (17:43 +0200)]
net: mvpp2: fix second cp110 initialization

Since the mdio code got upstreamed it is not possible
to activate network ports on CP110 Master and Slave.

The problem is in mvpp2_base_probe which is called for each
CP110 and it initializes the buffer area for descs and rx_buffers.

This should only happen once though and the bd space is actually
set to 0 after the first run of the function.

This leads to an error when the second CP110 tries the initialization
again and disables all network ports on this CP110.

This patch adds a static variable to check if the buffer area is
initialized only once globally.

Signed-off-by: Sven Auhagen <sven.auhagen@voleatech.de>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agoarm: kirkwood: enable DM_ETH for LaCie board
Simon Guinot [Sun, 28 Jun 2020 17:00:31 +0000 (19:00 +0200)]
arm: kirkwood: enable DM_ETH for LaCie board

This patch enables DM_ETH for the following Kirkwood-based LaCie boards:

- d2 Network v2
- Internet Space v2
- 2Big Network v2
- Network Space v2
- Network Space Lite v2
- Network Space Max v2
- Network Space Mini v2

Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agoarm: kirkwood: enable DM_USB for LaCie board
Simon Guinot [Sun, 28 Jun 2020 17:00:30 +0000 (19:00 +0200)]
arm: kirkwood: enable DM_USB for LaCie board

This patch enables DM_USB and USB_STORAGE for the following
Kirkwood-based LaCie boards:

- d2 Network v2
- Internet Space v2
- 2Big Network v2
- Network Space v2
- Network Space Lite v2
- Network Space Max v2

Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agoarm: kirkwood: switch LaCie boards to sata_mv driver
Simon Guinot [Sun, 28 Jun 2020 17:00:29 +0000 (19:00 +0200)]
arm: kirkwood: switch LaCie boards to sata_mv driver

This patch switches the SATA driver from mvsata_ide to sata_mv for the
following Kirkwood-based LaCie boards:

- d2 Network v2
- Internet Space v2
- 2Big Network v2
- Network Space v2
- Network Space Lite v2
- Network Space Max v2
- Network Space Mini v2

Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agoarm: kirkwood: convert LaCie boards to DM_SPI_FLASH
Simon Guinot [Sun, 28 Jun 2020 17:00:28 +0000 (19:00 +0200)]
arm: kirkwood: convert LaCie boards to DM_SPI_FLASH

This patch converts the following Kirkwood-based LaCie boards to DM,
DM_SPI and DM_SPI_FLASH:

- d2 Network v2
- Internet Space v2
- 2Big Network v2
- Network Space v2
- Network Space Lite v2
- Network Space Max v2
- Network Space Mini v2

Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>
Reviewed-by: Stefan Roese <sr@denx.de>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
4 years agoarm: kirkwood: add DT spi0 alias to LaCie boards
Simon Guinot [Sun, 28 Jun 2020 17:00:27 +0000 (19:00 +0200)]
arm: kirkwood: add DT spi0 alias to LaCie boards

The spi0 alias is needed by the environment code to retrieve the SPI
flash. This patch adds some -u-boot.dtsi files, providing the spi0
aliases, for all the following Kirkwood-based LaCie boards:

- d2 Network v2
- Internet Space v2
- 2Big Network v2
- Network Space v2
- Network Space Lite v2
- Network Space Max v2
- Network Space Mini v2

Note that this -u-boot.dtsi files will be removed as soon as the spi0
aliases will be available in the upstream Linux dtsi files.

Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Stefan Roese <sr@denx.de>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
4 years agoarm: mvebu: helios4: sync helios4 config to clearfog and dts to kernel
Dennis Gilmore [Sat, 27 Jun 2020 20:00:16 +0000 (15:00 -0500)]
arm: mvebu: helios4: sync helios4 config to clearfog and dts to kernel

The helios4 is built on the same microsom as the clearfog, by syncing the config
we enable the same featureset that exists in the som on the helios4. The current
config does not boot as some of the clearfog changes needed to be made on the
helios4 also, generally speaking most changes for the clearfog should also be
made on the helios4.

Signed-off-by: Dennis Gilmore <dennis@ausil.us>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agoarm: mvebu: a38x: Adjust UTMI PHY parameters
Chris Packham [Thu, 25 Jun 2020 00:48:51 +0000 (12:48 +1200)]
arm: mvebu: a38x: Adjust UTMI PHY parameters

When running USB compliance tests on our Armada-385 hardware platforms
we have seen some eye mask violations. Marvell's internal documentation
says: Based on silicon test results, it is recommended to change the
impedance calibration threshold setting to 0x6 prior to calibration.

Port changes from Marvell's u-boot fork[1] to address this.

[1] - https://github.com/MarvellEmbeddedProcessors/u-boot-marvell/commit/a6221551

Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agoarm: mvebu: a38x: Fix typo
Chris Packham [Thu, 25 Jun 2020 00:48:50 +0000 (12:48 +1200)]
arm: mvebu: a38x: Fix typo

Fix spelling of Alignment.

Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agomv_ddr: ddr3: Update {min,max}_read_sample calculation
Chris Packham [Wed, 27 May 2020 01:31:30 +0000 (13:31 +1200)]
mv_ddr: ddr3: Update {min,max}_read_sample calculation

Measurements on actual hardware shown that the read ODT is early by 3
clocks. Adjust the calculation to avoid this.

Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
[upstream https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell/pull/22]
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Tested-by: Baruch Siach <baruch@tkos.co.il>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agomv_ddr: ddr3: Use correct bitmask for read sample delay
Chris Packham [Wed, 27 May 2020 01:31:29 +0000 (13:31 +1200)]
mv_ddr: ddr3: Use correct bitmask for read sample delay

In the Armada 385 functional spec (MV-S109094-00 Rev. C) the read sample
delay fields are 5 bits wide. Use the correct bitmask of 0x1f when
extracting the value.

Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
[upstream https://github.com/MarvellEmbeddedProcessors/mv-ddr-marvell/pull/22]
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agox86: apl: Re-enable loading of SPL
Simon Glass [Wed, 8 Jul 2020 15:02:15 +0000 (09:02 -0600)]
x86: apl: Re-enable loading of SPL

At present the SPL loader is not included in the TPL image so SPL cannot
be loaded. Fix it by including this file for both SPL and TPL.

Signed-off-by: Simon Glass <sjg@chromium.org>
Fixes: c87f9ce2273 ("x86: Don't build some unused objects in TPL")
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agospi: Remove unnecessary #ifdefs in header file
Simon Glass [Wed, 8 Jul 2020 15:02:14 +0000 (09:02 -0600)]
spi: Remove unnecessary #ifdefs in header file

These prevent use of compile-time checks such as:

    if (CONFIG_IS_ENABLED(DM_SPI))

since, for example, if CONFIG_SPL_DM_SPI is not enabled then the
definitions are not included by spi.h and the C code will not build.

The #ifdefs are unnecessary since there are no conflicts with the pre-DM
code. In any case we have almost switched over to driver model for SPI.

Drop these #ifdefs from spi.h to fix a build warning on chromebook_coral
in the following patch.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agobootstage: Fix 'stacked' typo
Simon Glass [Fri, 3 Jul 2020 16:37:07 +0000 (10:37 -0600)]
bootstage: Fix 'stacked' typo

This should be 'stashed'. Fix it.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agolinux/kconfig.h: create two- and three-argument versions of CONFIG_IS_ENABLED
Rasmus Villemoes [Fri, 3 Jul 2020 16:37:06 +0000 (10:37 -0600)]
linux/kconfig.h: create two- and three-argument versions of CONFIG_IS_ENABLED

This adds a bunch of preprocessor magic to extend the capabilities of
CONFIG_IS_ENABLED. The existing semantics of

  CONFIG_IS_ENABLED(FOO)

expanding to a 1 or 0 (depending on build context and the defined-ness
or not of the appropriate CONFIG_FOO/CONFIG_SPL_FOO/CONFIG_TPL_FOO)
are of course preserved. With this, one is also allowed a two-argument
form

  CONFIG_IS_ENABLED(FOO, (something))

which expands to something precisely when CONFIG_IS_ENABLED(FOO) would
expand to 1, and expands to nothing otherwise. It is, in other words,
completely equivalent to the three lines

  #if CONFIG_IS_ENABLED(FOO)
  something
  #endif

The second argument must be parenthesized in order to allow any
tokens, including a trailing comma, to appear - one use case for this
is precisely to make it a bit more ergonomic to build an array and
only include certain items depending on .config. That should increase
both readability and not least "git grep"ability.

A third variant is also introduced,

  CONFIG_IS_ENABLED(FOO, (xxx), (yyy))

which corresponds to

  #if CONFIG_IS_ENABLED(FOO)
  xxx
  #else
  yyy
  #endif

Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agolinux/kconfig.h: remove unused helper macros
Rasmus Villemoes [Fri, 3 Jul 2020 16:37:05 +0000 (10:37 -0600)]
linux/kconfig.h: remove unused helper macros

U-Boot does not have loadable modules, and nothing currently uses any
of the (CONFIG_)?IS_(BUILTIN|MODULE) macros - only
the (CONFIG_)?IS_ENABLED variants are ever used.

While I understand the desire to keep this somewhat synchronized with
linux, we've already departed by the introduction of the
CONFIG_IS_ENABLED extra logic, and deleting these makes the next patch
much simpler, since I won't have to duplicate a lot of logic for no
real gain (as there are no users).

Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agolinux/kconfig.h: simplify logic for choosing CONFIG_{SPL_, TPL_, }*
Rasmus Villemoes [Fri, 3 Jul 2020 16:37:04 +0000 (10:37 -0600)]
linux/kconfig.h: simplify logic for choosing CONFIG_{SPL_, TPL_, }*

Instead of using the arg1_or_junk trick to pick between two choices,
with a bit of duplication between the branches (and most of the
CONFIG_TPL_BUILD case being redundant, as _IS_TPL is known to be
defined to 1 in that case), simply define a prefix that we inject
between CONFIG_ and the given config symbol.

This only requires one level of indirection (to get the
_CONFIG_PREFIX macro expanded before the token concatenation takes
place), and makes it easy to, say, introduce a CONFIG_HOSTTOOL_
prefix. [I would expect most HOSTTOOL_ symbols to just be def_bool y,
but it would allow us to clean up some of the ifdef HOSTCC mess in the
sources shared between U-Boot and host tools.]

Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agoarm: cmd_stm32prog: Fix the CONFIG_IS_ENABLED() usage
Bin Meng [Wed, 8 Jul 2020 07:19:22 +0000 (00:19 -0700)]
arm: cmd_stm32prog: Fix the CONFIG_IS_ENABLED() usage

Add parentheses around CONFIG_IS_ENABLED() in the if statement, to
fix potential build failures.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Patrick Delaunay <patrick.delaunay@st.com>
4 years agox86: minnowmax: Drop screen resolution to 1024x768
Simon Glass [Fri, 3 Jul 2020 03:12:37 +0000 (21:12 -0600)]
x86: minnowmax: Drop screen resolution to 1024x768

This seems like a more reasonable resolution for this board, since it is
quite slow. It also allows it to work with a 5" LCD display in my lab.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agominnowmax: Enable the copy framebuffer
Simon Glass [Fri, 3 Jul 2020 03:12:36 +0000 (21:12 -0600)]
minnowmax: Enable the copy framebuffer

Update the video driver to support this feature and enable it on
minnowmax to speed up the display.

With this change, the time taken to print the environment to the display
without CONFIG_CONSOLE_SCROLL_LINES is reduced from over 13 seconds to
300ms, at 1280x1024.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agochromebook_link: Enable the copy framebuffer
Simon Glass [Fri, 3 Jul 2020 03:12:35 +0000 (21:12 -0600)]
chromebook_link: Enable the copy framebuffer

Update the video driver to support this feature and enable it on link.
Also remove the multi-line scrolling since normal scrolling is fast enough
now.

With this change, the time taken to print the environment to the display
without CONFIG_CONSOLE_SCROLL_LINES is reduced from about 930ms to 29ms.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agochromebook_samus: Enable the copy framebuffer
Simon Glass [Fri, 3 Jul 2020 03:12:34 +0000 (21:12 -0600)]
chromebook_samus: Enable the copy framebuffer

Update the video driver to support this feature and enable it on samus.
Also remove the multi-line scrolling since normal scrolling is fast enough
now.

With this change, the time taken to print the environment to the display
without CONFIG_CONSOLE_SCROLL_LINES is reduced from about 430ms to 12ms.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agox86: video: Support copy framebuffer with probed devices
Simon Glass [Fri, 3 Jul 2020 03:12:33 +0000 (21:12 -0600)]
x86: video: Support copy framebuffer with probed devices

For PCI video devices that are not mentioned in the devicetree, U-Boot
does not bind a driver before relocation, since PCI is not fully probed
at that point. Furthermore it is possible for the video device to be on
a secondary bus which is not even scanned.

This is fine if the framebuffer is allocated in fixed memory, as it
normally is on x86. But when using this as a copy framebuffer, we also
need U-Boot to allocate its own cached framebuffer for working in. Since
the video driver is never bound before relocation, the framebuffer size
is never set and U-Boot does no allocation.

Add a new CONFIG option to reserve 16MB of memory for this eventuality.
This allows vesa devices to use the copy framebuffer.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Correctly handle multiple framebuffers
Simon Glass [Fri, 3 Jul 2020 03:12:32 +0000 (21:12 -0600)]
video: Correctly handle multiple framebuffers

At present video_bottom is set to the bottom of each framebuffer when it
is allocated. This is not correct, since it should hold the bottom of the
entire area available for framebuffers.

Fix this by adding a private address in the uclass which keeps track of
the next available spot for a framebuffer.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agox86: fsp: video: Allocate a frame buffer when needed
Simon Glass [Fri, 3 Jul 2020 03:12:31 +0000 (21:12 -0600)]
x86: fsp: video: Allocate a frame buffer when needed

When the copy framebuffer is in use, we must also have the standard U-Boot
framebuffer available. Update the FSP driver to support this.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: pci: Set up the copy framebuffer
Simon Glass [Fri, 3 Jul 2020 03:12:30 +0000 (21:12 -0600)]
video: pci: Set up the copy framebuffer

When using a copy framebuffer we need to tell the video subsystem its
address. U-Boot's normally allocated framebuffer is used as the working
buffer, but nothing is displayed until it is copied to the copy
framebuffer.

For this to work the video driver must request that a framebuffer be
allocated separately from the hardware framebuffer, so add a check for
that.

Also add a log category so that logging appears correctly.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: sandbox: Add support for the copy framebuffer
Simon Glass [Fri, 3 Jul 2020 03:12:29 +0000 (21:12 -0600)]
video: sandbox: Add support for the copy framebuffer

Enable this feature on sandbox by updating the SDL driver to have two
framebuffers.

Update the video tests to check that the copy framebuffer is kept in sync.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
4 years agovideo: Add comments to struct sandbox_sdl_plat
Simon Glass [Fri, 3 Jul 2020 03:12:28 +0000 (21:12 -0600)]
video: Add comments to struct sandbox_sdl_plat

This struct is not commented but needs it. Also fix the comment in
check_vidconsole_output() about the encoding for the rotation value.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
4 years agovideo: Update the copy framebuffer when writing bitmaps
Simon Glass [Fri, 3 Jul 2020 03:12:27 +0000 (21:12 -0600)]
video: Update the copy framebuffer when writing bitmaps

Adjust the bitmap code to sync to the copy framebuffer when done.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Update rotated console to support copy buffer
Simon Glass [Fri, 3 Jul 2020 03:12:26 +0000 (21:12 -0600)]
video: Update rotated console to support copy buffer

Update the implementation to keep a track of what it changes in the frame
buffer and then tell the copy buffer about it. Use the special
vidconsole_memmove() helper so that memmove() operations are also
reflected in the copy buffer.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Update truetype console to support copy buffer
Simon Glass [Fri, 3 Jul 2020 03:12:25 +0000 (21:12 -0600)]
video: Update truetype console to support copy buffer

Update the implementation to keep a track of what it changes in the frame
buffer and then tell the copy buffer about it. Use the special
vidconsole_memmove() helper so that memmove() operations are also
reflected in the copy buffer.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Update normal console to support copy buffer
Simon Glass [Fri, 3 Jul 2020 03:12:24 +0000 (21:12 -0600)]
video: Update normal console to support copy buffer

Update the implementation to keep a track of what it changes in the frame
buffer and then tell the copy buffer about it. Use the special
vidconsole_memmove() helper so that memmove() operations are also
reflected in the copy buffer.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Add helpers for vidconsole for the copy framebuffer
Simon Glass [Fri, 3 Jul 2020 03:12:23 +0000 (21:12 -0600)]
video: Add helpers for vidconsole for the copy framebuffer

Add a convenience function to call video_sync_copy() for a vidconsole.
Also add a memmove() helper, which does the memmove() as well as the sync.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Clear the copy framebuffer when clearing the screen
Simon Glass [Fri, 3 Jul 2020 03:12:22 +0000 (21:12 -0600)]
video: Clear the copy framebuffer when clearing the screen

Update video_clear() to also sync to the copy framebuffer.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Set up the copy framebuffer when enabled
Simon Glass [Fri, 3 Jul 2020 03:12:21 +0000 (21:12 -0600)]
video: Set up the copy framebuffer when enabled

This framebuffer is separately mapped. Update the video post-probe
function to set this up.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Add support for copying to a hardware framebuffer
Simon Glass [Fri, 3 Jul 2020 03:12:20 +0000 (21:12 -0600)]
video: Add support for copying to a hardware framebuffer

Some architectures use a cached framebuffer and flush the cache as needed
so that changes are visible. This is supported by U-Boot.

However x86 uses an uncached framebuffer with a 'write-combining' feature
to speed up writes.  Reads are permitted but they are extremely expensive.

Unfortunately, reading from the frame buffer is quite common, e.g. to
scroll it. This makes scrolling very slow.

Add a new feature which supports copying modified parts of the frame
buffer to the uncached hardware buffer. This speeds up scrolling by at
least 10x on x86 so the extra complexity cost seems worth it.

As a starting point, add the Kconfig, update the video structures to keep
track of the buffer and add a function to do the copy.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Add a comment for struct video_uc_platdata
Simon Glass [Fri, 3 Jul 2020 03:12:19 +0000 (21:12 -0600)]
video: Add a comment for struct video_uc_platdata

Add a few notes to explain the purpose of each member of this struct.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Drop unnecessary #ifdef around vid_console_color()
Simon Glass [Fri, 3 Jul 2020 03:12:18 +0000 (21:12 -0600)]
video: Drop unnecessary #ifdef around vid_console_color()

All of the functions in this file only apply if DM_VIDEO is enabled. Drop
the #ifdef as it just clutters things up. Add the needed forward
declaration.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Adjust rotated console to start at right edge
Simon Glass [Fri, 3 Jul 2020 03:12:17 +0000 (21:12 -0600)]
video: Adjust rotated console to start at right edge

At present when the console is rotated 180 degrees it starts almost a
whole character to the left of the right edge (typically 7 pixels with
an 8-pixel-wide font). On a display which aligns with the font width,
this just wastes space. On a display that does not this can result in
x_frac going negative for the final character (the one on the left
side) and the overflow -EAGAIN check at the start of the function
failing.

Change the function to start at the rightmost pixel to fix these
problems.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agovideo: Split out expression parts into variables
Simon Glass [Fri, 3 Jul 2020 03:12:16 +0000 (21:12 -0600)]
video: Split out expression parts into variables

The functions in this file do similar things but not always in the same
way. To make the code easier to read and compare, use a separate 'linenum'
variable in every function. This is then multiplied by the line length to
get the offset within the frame buffer to modify. Also use an 'x' variable
to hold the pixel position within that line. This is multipled by the
pixel size and added to the offset.

Also move the pbytes declaration up a little with the other long lines.

A side effect of splitting out these variables is that they are promoted
to int, i.e. a signed type, from the unsigned short used in the
vidconsole_priv struct. This would be necessary should any of the
variables go negative. At present this can actually happen in
console_putc_xy_2(), if the display width is not a multiple of the
character size (see next patch).

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
4 years agosandbox: video: Allow selection of rotated console
Simon Glass [Fri, 3 Jul 2020 03:12:15 +0000 (21:12 -0600)]
sandbox: video: Allow selection of rotated console

Add a devicetree property to select a rotated console. This uses the same
encoding as vidconsole itself: 0=normal; 1=90 degrees clockwise, 2=upside
down, 3=90 degrees anticlockwise.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
4 years agovideo: Show an error when a vidconsole function fails
Simon Glass [Fri, 3 Jul 2020 03:12:14 +0000 (21:12 -0600)]
video: Show an error when a vidconsole function fails

At present these functions fail silently even when debugging, which is not
very helpful. Add a way to print a message to the serial output when an
error is detected.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agoconsole: Add a way to output to serial only
Simon Glass [Fri, 3 Jul 2020 03:12:13 +0000 (21:12 -0600)]
console: Add a way to output to serial only

In the video drivers it is useful to print errors while debugging but
doing so risks an infinite loop as the debugging info itself may go
through the video drivers.

Add a new console function that prints information only to the serial
device, thus making it safe for use in debugging.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agox86: fsp: Reinit the FPU after FSP meminit
Simon Glass [Fri, 3 Jul 2020 03:12:12 +0000 (21:12 -0600)]
x86: fsp: Reinit the FPU after FSP meminit

The APL FSP appears to leave the FPU in a bad state in that it has
registers in use. This causes an error when the next FPU operation is
performed.

Work around this by re-resetting the FPU after calling FSP-M. This allows
the freetype console to work correctly.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agopwm: Add PWM driver for SiFive SoC
Yash Shah [Thu, 23 Apr 2020 11:27:16 +0000 (16:57 +0530)]
pwm: Add PWM driver for SiFive SoC

Adds a PWM driver for PWM chip present in SiFive's HiFive Unleashed SoC
This driver is simple port of Linux pwm sifive driver from Linux v5.6

commit: 9e37a53eb051 ("pwm: sifive: Add a driver for SiFive SoC PWM")

Signed-off-by: Yash Shah <yash.shah@sifive.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
4 years agopwm: Add DT documentation for SiFive PWM Controller
Yash Shah [Thu, 23 Apr 2020 11:27:15 +0000 (16:57 +0530)]
pwm: Add DT documentation for SiFive PWM Controller

DT documentation for PWM controller added from Linux v5.6

commit: daa78cc3408e
("pwm: sifive: Add DT documentation for SiFive PWM Controller")

Signed-off-by: Yash Shah <yash.shah@sifive.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
4 years agoi2c: stm32f7: SYSCFG Fast Mode Plus support for I2C STM32F7
Patrick Delaunay [Mon, 6 Jul 2020 11:31:35 +0000 (13:31 +0200)]
i2c: stm32f7: SYSCFG Fast Mode Plus support for I2C STM32F7

Read SYSCFG bindings to set Fast Mode Plus bits if Fast Mode Plus
speed is selected.

Handle the stm32mp15 specific compatible to handle FastMode+
registers handling which is different on the stm32mp15 compared
to the stm32f7 or stm32h7.
Indeed, on the stm32mp15, the FastMode+ set and clear registers
are separated while on the other platforms (F7 or H7) the control
is done in a unique register.

Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
4 years agoi2c: stm32f7: add stm32mp15 compatible
Patrick Delaunay [Mon, 6 Jul 2020 11:26:52 +0000 (13:26 +0200)]
i2c: stm32f7: add stm32mp15 compatible

Add a new compatible "st,stm32mp15-i2c" introduced in Linux kernel v5.8

Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
4 years agotest: dm: rtc: add tests of rtc shell command
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:20 +0000 (22:01 +0200)]
test: dm: rtc: add tests of rtc shell command

Add tests of the "list", "read" and "write" subcommands of the rtc
shell command.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agosandbox: add rtc command to defconfigs
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:19 +0000 (22:01 +0200)]
sandbox: add rtc command to defconfigs

In order to allow adding unit tests of the rtc command, add it to the
various sandbox defconfigs.

Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agotest: dm: rtc: add test of dm_rtc_read, dm_rtc_write
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:18 +0000 (22:01 +0200)]
test: dm: rtc: add test of dm_rtc_read, dm_rtc_write

Define a few aux registers and check that they can be read/written
individually. Also check that one can access the time-keeping
registers directly and get the expected results.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agortc: i2c_rtc_emul: catch any write to the "reset" register
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:17 +0000 (22:01 +0200)]
rtc: i2c_rtc_emul: catch any write to the "reset" register

It's more natural that any write that happens to touch the reset
register should cause a reset, rather than just a write that starts at
that offset.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agortc: sandbox-rtc: fix set method
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:16 +0000 (22:01 +0200)]
rtc: sandbox-rtc: fix set method

The current set method is broken; a simple test case is to first set
the date to something in April, then change the date to 31st May:

=> date 040412122020.34
Date: 2020-04-04 (Saturday)    Time: 12:12:34
=> date 053112122020.34
Date: 2020-05-01 (Friday)    Time: 12:12:34

or via the amending of the existing rtc_set_get test case similarly:

$ ./u-boot -T -v
=> ut dm rtc_set_get
Test: dm_test_rtc_set_get: rtc.c
expected: 31/08/2004 18:18:00
actual: 01/08/2004 18:18:00

The problem is that after each register write,
sandbox_i2c_rtc_complete_write() gets called and sets the internal
time from the current set of registers. However, when we get to
writing 31 to mday, the registers are in an inconsistent state (mon is
still 4), so the mktime machinery ends up translating April 31st to
May 1st. Upon the next register write, the registers are populated by
sandbox_i2c_rtc_prepare_read(), so the 31 we just wrote to mday gets
overwritten by a 1.

Fix it by writing all registers at once, and for consistency, update
the get method to retrieve them all with one "i2c transfer".

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agortc: add rtc command
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:15 +0000 (22:01 +0200)]
rtc: add rtc command

Mostly as an aid for debugging RTC drivers, provide a command that can
be used to read/write arbitrary registers (assuming the driver
provides the read/write methods or their single-register-at-a-time
variants).

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agortc: pcf2127: provide ->write method
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:14 +0000 (22:01 +0200)]
rtc: pcf2127: provide ->write method

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agortc: pcf2127: provide ->read method
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:13 +0000 (22:01 +0200)]
rtc: pcf2127: provide ->read method

This simply consists of renaming the existing pcf2127_read_reg()
helper to follow the naming of the other
methods (i.e. pcf2127_rtc_<method name>) and changing the type of its
"len" parameter.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agortc: fall back to ->{read, write} if ->{read, write}8 are not provided
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:12 +0000 (22:01 +0200)]
rtc: fall back to ->{read, write} if ->{read, write}8 are not provided

Similar to how the dm_rtc_{read,write} functions fall back to using
the {read,write}8 methods, do the opposite in the rtc_{read,write}8
functions.

This way, each driver only needs to provide either ->read8 or ->read
to make both rtc_read8() and dm_rtc_read() work - without this, a
driver that provides ->read() would most likely just duplicate the
logic here for implementing a ->read8() method in term of its ->read()
method. The same remarks of course apply to the write case.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agortc: add dm_rtc_write() helper
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:11 +0000 (22:01 +0200)]
rtc: add dm_rtc_write() helper

Similar to dm_rtc_read(), introduce a helper that allows the caller to
write multiple consecutive 8-bit registers with one call. If the
driver provides the ->write method, use that, otherwise loop using
->write8.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agortc: add dm_rtc_read helper and ->read method
Rasmus Villemoes [Mon, 6 Jul 2020 20:01:10 +0000 (22:01 +0200)]
rtc: add dm_rtc_read helper and ->read method

Some users may want to read multiple consecutive 8-bit
registers. Instead of each caller having to implement the loop,
provide a dm_rtc_read() helper. Also, allow a driver to provide a
->read method, which can be more efficient than reading one register
at a time.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
4 years agoi2c: imx_lpi2c: Improve the codes to use private data
Ye Li [Wed, 10 Jun 2020 03:29:50 +0000 (20:29 -0700)]
i2c: imx_lpi2c: Improve the codes to use private data

Current driver calls the devfdt_get_addr to get the base address
of lpi2c controller in each sub-functions. Since the devfdt_get_addr
accesses the DTB and translate the address, it introduces much
overhead.
Improve the codes to use private variable which has recorded the
base address from probe.

Signed-off-by: Ye Li <ye.li@nxp.com>
Reviewed-by: Peng Fan <peng.fan@nxp.com>
4 years agoboard: presidio-asic: Add I2C support
Alex Nemirovsky [Mon, 1 Jun 2020 19:56:32 +0000 (12:56 -0700)]
board: presidio-asic: Add I2C support

Add I2C board support for Cortina Access Presidio Engineering Board

Signed-off-by: Alex Nemirovsky <alex.nemirovsky@cortina-access.com>
CC: Heiko Schocher <hs@denx.de>
Reviewed-by: Heiko Schocher <hs@denx.de>
4 years agoi2c: i2c-cortina: added CAxxxx I2C support
Arthur Li [Mon, 1 Jun 2020 19:56:31 +0000 (12:56 -0700)]
i2c: i2c-cortina: added CAxxxx I2C support

Add I2C controller support for Cortina Access CAxxxx SoCs

Signed-off-by: Arthur Li <arthur.li@cortina-access.com>
Signed-off-by: Alex Nemirovsky <alex.nemirovsky@cortina-access.com>
CC: Heiko Schocher <hs@denx.de>
Reviewed-by: Heiko Schocher <hs@denx.de>
hs: fixed build error, add include log.h

4 years agoi2c: octeon_i2c: Add I2C controller driver for Octeon
Suneel Garapati [Tue, 26 May 2020 12:13:07 +0000 (14:13 +0200)]
i2c: octeon_i2c: Add I2C controller driver for Octeon

Add support for I2C controllers found on Octeon II/III and Octeon TX
TX2 SoC platforms.

Signed-off-by: Aaron Williams <awilliams@marvell.com>
Signed-off-by: Suneel Garapati <sgarapati@marvell.com>
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Heiko Schocher <hs@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Cc: Aaron Williams <awilliams@marvell.com>
Cc: Chandrakala Chavva <cchavva@marvell.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Heiko Schocher <hs@denx.de>
Reviewed-by: Rayagonda Kokatanur <rayagonda.kokatanur@broadcom.com>
4 years agoMerge branch '2020-07-08-misc-features-and-fixes'
Tom Rini [Thu, 9 Jul 2020 00:20:24 +0000 (20:20 -0400)]
Merge branch '2020-07-08-misc-features-and-fixes'

- mem cmd improvements
- TPM fixes
- SPL/NAND/FIT fixes
- RSA improvements

4 years agolib: rsa: function to verify a signature against a hash
Heiko Stuebner [Fri, 22 May 2020 14:20:33 +0000 (16:20 +0200)]
lib: rsa: function to verify a signature against a hash

rsa_verify() expects a memory region and wants to do the hashing itself,
but there may be cases where the hashing is done via other means,
like hashing a squashfs rootfs.

So add rsa_verify_hash() to allow verifiying a signature against
an existing hash. As this entails the same verification routines
we can just move the relevant code over from rsa_verify() and also
call rsa_verify_hash() from there.

Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
4 years agocmd: Add a memory-search command
Simon Glass [Wed, 3 Jun 2020 01:26:49 +0000 (19:26 -0600)]
cmd: Add a memory-search command

It is useful to be able to find hex values and strings in a memory range.
Add a command to support this.

cmd: Fix 'md' and add a memory-search command
At present 'md.q' is broken. This series provides a fix for this. It also
implements a new memory-search command called 'ms'. It allows searching
memory for hex and string data.
END

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agocommand: Drop #ifdef for MEM_SUPPORT_64BIT_DATA
Simon Glass [Wed, 3 Jun 2020 01:26:48 +0000 (19:26 -0600)]
command: Drop #ifdef for MEM_SUPPORT_64BIT_DATA

This is defined only when __lp64__ is defined. That means that ulong is
64 bits long. Therefore we don't need to use a separate u64 type on those
architectures.

Fix up the code to take advantage of that, removing the preprocessor
conditions.

Also include the header file that defines MEM_SUPPORT_64BIT_DATA. It is
included by env.h in this file, but that might not last forever.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agodisplay_options: Drop #ifdef for MEM_SUPPORT_64BIT_DATA
Simon Glass [Wed, 3 Jun 2020 01:26:47 +0000 (19:26 -0600)]
display_options: Drop #ifdef for MEM_SUPPORT_64BIT_DATA

This is defined only when __lp64__ is defined. That means that ulong is
64 bits long. Therefore we don't need to use a separate u64 type on those
architectures.

Fix up the code to take advantage of that, removing the preprocessor
conditions.

Also include the missing header file that defines MEM_SUPPORT_64BIT_DATA

Fixes: 09140113108 ("command: Remove the cmd_tbl_t typedef")
Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agocmd: mem: Drop #ifdef for MEM_SUPPORT_64BIT_DATA
Simon Glass [Wed, 3 Jun 2020 01:26:46 +0000 (19:26 -0600)]
cmd: mem: Drop #ifdef for MEM_SUPPORT_64BIT_DATA

This is defined only when __lp64__ is defined. That means that ulong is
64 bits long. Therefore we don't need to use a separate u64 type on those
architectures.

Fix up the code to take advantage of that, removing the preprocessor
conditions.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agocmd: mem: Use a macro to avoid #ifdef in help
Simon Glass [Wed, 3 Jun 2020 01:26:45 +0000 (19:26 -0600)]
cmd: mem: Use a macro to avoid #ifdef in help

It is a bit painful to have #ifdefs in the middle of the help for each
command. Add a macro to avoid this.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agoUpdate MEM_SUPPORT_64BIT_DATA to be always defined
Simon Glass [Wed, 3 Jun 2020 01:26:44 +0000 (19:26 -0600)]
Update MEM_SUPPORT_64BIT_DATA to be always defined

Define this macro always so we don't need the preprocessor to check it.
Convert the users to #if instead of #ifdef.

Note that '#if MEM_SUPPORT_64BIT_DATA' does not give an error if the
macro is not define. It just assumes zero.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agospl: fit: improve spl_nand_fit_read(...) readability
Dario Binacchi [Wed, 27 May 2020 11:56:21 +0000 (13:56 +0200)]
spl: fit: improve spl_nand_fit_read(...) readability

Replacing the ret variable with err and handling first the error
condition about the value returned by the spl_nand_fit_read routine,
improves the code readability.
Furthermore, the 'else' int the 'else return ret' instruction was
useless.

cc: Michael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: Dario Binacchi <dariobin@libero.it>
4 years agospl: fit: nand: fix fit loading in case of bad blocks
Dario Binacchi [Wed, 27 May 2020 11:56:20 +0000 (13:56 +0200)]
spl: fit: nand: fix fit loading in case of bad blocks

The offset at which the image to be loaded from NAND is located is
retrieved from the itb header. The presence of bad blocks in the area
of the NAND where the itb image is located could invalidate the offset
which must therefore be adjusted taking into account the state of the
sectors concerned.

cc: Michael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Reviewed-by: Michael Trimarchi <michael@amarulasolutions.com>
4 years agospl: fit: fail fit loading in case of FDT appending error
Dario Binacchi [Wed, 27 May 2020 11:56:19 +0000 (13:56 +0200)]
spl: fit: fail fit loading in case of FDT appending error

If uboot does not embed its device tree and the FIT loading function
returns error in case of failure in the FDT append, the redundant itb
image could be loaded.

cc: Michael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Reviewed-by: Michael Trimarchi <michael@amarulasolutions.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agospl: fix format of function documentation
Dario Binacchi [Wed, 27 May 2020 11:56:18 +0000 (13:56 +0200)]
spl: fix format of function documentation

U-Boot adopted the kernel-doc annotation style.

cc: Michael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agospl: fit: select SPL_CRYPTO_SUPPORT for SPL_FIT_SIGNATURE
Heiko Stuebner [Thu, 18 Jun 2020 14:23:28 +0000 (16:23 +0200)]
spl: fit: select SPL_CRYPTO_SUPPORT for SPL_FIT_SIGNATURE

Verifying FIT images obviously needs the rsa parts of crypto
support and while main uboot always compiles crypto support,
it's optional for SPL and we should thus select the necessary
option to not end up in compile errors like:

    u-boot/lib/rsa/rsa-verify.c:328: undefined reference to `rsa_mod_exp'

So select SPL_CRYPTO_SUPPORT in SPL_FIT_SIGNATURE.

Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agospl: fit: select SPL_HASH_SUPPORT for SPL_FIT_SIGNATURE
Heiko Stuebner [Thu, 18 Jun 2020 14:23:27 +0000 (16:23 +0200)]
spl: fit: select SPL_HASH_SUPPORT for SPL_FIT_SIGNATURE

rsa-checsum needs support for hash functions or else will run into
compile errors like:
u-boot/lib/rsa/rsa-checksum.c:28: undefined reference to `hash_progressive_lookup_algo'

So similar to the main FIT_SIGNATURE entry selects HASH,
select SPL_HASH_SUPPORT for SPL_FIT_SIGNATURE.

Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agolib: rsa: add documentation to padding_pss_verify to document limitations
Heiko Stuebner [Thu, 18 Jun 2020 14:23:26 +0000 (16:23 +0200)]
lib: rsa: add documentation to padding_pss_verify to document limitations

padding_pss_verify only works with the default pss salt setting of -2
(length to be automatically determined based on the PSS block structure)
not -1 (salt length set to the maximum permissible value), which makes
verifications of signatures with that saltlen fail.

Until this gets implemented at least document this behaviour.

Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agolib: rsa: free local arrays after use in rsa_gen_key_prop()
Heiko Stuebner [Thu, 18 Jun 2020 14:23:25 +0000 (16:23 +0200)]
lib: rsa: free local arrays after use in rsa_gen_key_prop()

n, rr and rrtmp are used for internal calculations, but in the end
the results are copied into separately allocated elements of the
actual key_prop, so the n, rr and rrtmp elements are not used anymore
when returning from the function and should of course be freed.

Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agolib: rsa: fix allocated size for rr and rrtmp in rsa_gen_key_prop()
Heiko Stuebner [Tue, 7 Jul 2020 20:57:26 +0000 (22:57 +0200)]
lib: rsa: fix allocated size for rr and rrtmp in rsa_gen_key_prop()

When calculating rrtmp/rr rsa_gen_key_prop() tries to make
(((rlen + 31) >> 5) + 1) steps in the rr uint32_t array and
(((rlen + 7) >> 3) + 1) / 4 steps in uint32_t rrtmp[]
with rlen being num_bits * 2

On a 4096bit key this comes down to to 257 uint32_t elements
in rr and 256 elements in rrtmp but with the current allocation
rr and rrtmp only have 129 uint32_t elements.

On 2048bit keys this works by chance as the defined max_rsa_size=4096
allocates a suitable number of elements, but with an actual 4096bit key
this results in other memory parts getting overwritten.

So as suggested by Heinrich Schuchardt just use the actual bit-size
of the key as base for the size calculation, in turn making the code
compatible to any future keysizes.

Suggested-by: Heinrich Schuchardt <xypron.debian@gmx.de>
Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
rrtmp needs 2 + (((*prop)->num_bits * 2) >> 5) array elements.

Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
4 years agolib: rsa: bring exp_len in line when generating a key_prop
Heiko Stuebner [Thu, 18 Jun 2020 14:23:23 +0000 (16:23 +0200)]
lib: rsa: bring exp_len in line when generating a key_prop

The exponent field of struct key_prop gets allocated an uint64_t,
and the contents are positioned from the back, so an exponent of
"0x01 0x00 0x01" becomes 0x0 0x0 0x0 0x0 0x0 0x1 0x0 0x1"

Right now rsa_gen_key_prop() allocates a uint64_t but sets exp_len
to the size returned from the parser, while on the other hand the
when getting the key from the devicetree exp_len always gets set to
sizeof(uint64_t).

So bring that in line with the established code.

Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agolib: rsa: take spl/non-spl into account when building rsa_verify_with_pkey()
Heiko Stuebner [Thu, 18 Jun 2020 14:23:22 +0000 (16:23 +0200)]
lib: rsa: take spl/non-spl into account when building rsa_verify_with_pkey()

Right now in multiple places there are only checks for the full
CONFIG_RSA_VERIFY_WITH_PKEY option, not split into main,spl,tpl variants.

This breaks when the rsa functions get enabled for SPL, for example to
verify u-boot proper from spl.

So fix this by using the existing helpers to distinguis between
build-steps.

Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agolib: rsa: distinguish between tpl and spl for CONFIG_RSA_VERIFY
Heiko Stuebner [Thu, 18 Jun 2020 14:23:21 +0000 (16:23 +0200)]
lib: rsa: distinguish between tpl and spl for CONFIG_RSA_VERIFY

While the SPL may want to do signature checking this won't be
the case for TPL in all cases, as TPL is mostly used when the
amount of initial memory is not enough for a full SPL.

So on a system where SPL uses DM but TPL does not we currently
end up with a TPL compile error of:

    lib/rsa/rsa-verify.c:48:25: error: dereferencing pointer to incomplete type ‘struct checksum_algo’

To prevent that change the $(SPL_) to $(SPL_TPL_) to distinguish
between both. If someone really needs FIT signature checking in
TPL as well, a new TPL_RSA_VERIFY config symbol needs to be added.

Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agocmd: add a panic command
Heiko Stuebner [Mon, 29 Jun 2020 23:05:45 +0000 (01:05 +0200)]
cmd: add a panic command

Even in boot scripts it may be needed to "panic" when all options
are exhausted and the device specification specifies hanging
instead of resetting the board.

So add a new panic command that just wraps around the core panic
call in U-Boot and can take an optional message.

Signed-off-by: Heiko Stuebner <heiko.stuebner@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
4 years agodebug_uart: Add CR before and after announce string
Stefan Roese [Fri, 15 May 2020 05:09:03 +0000 (07:09 +0200)]
debug_uart: Add CR before and after announce string

Add linefeeds before and after the announce string. This makes the
output easier to read, especially if some text follows the announce
message without a specific additional CR.

Signed-off-by: Stefan Roese <sr@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
4 years agotpm: add #ifndef to fix redeclaration build errors
Johannes Holland [Mon, 11 May 2020 13:22:25 +0000 (15:22 +0200)]
tpm: add #ifndef to fix redeclaration build errors

tpm_tis_spi.c directly includes tpm_tis.h and tpm-v2.h which both
define the same enums (see e.g. TPM_ACCESS_VALID). Add an #ifndef to
prevent redeclaration errors.

Signed-off-by: Johannes Holland <johannes.holland@infineon.com>
4 years agotpm: add TPM2_GetRandom command support
Dhananjay Phadke [Thu, 4 Jun 2020 23:43:59 +0000 (16:43 -0700)]
tpm: add TPM2_GetRandom command support

Add support for TPM2 GetRandom command

Signed-off-by: Dhananjay Phadke <dphadke@linux.microsoft.com>
Reviewed-by: Simon Glass <sjg@chromium.org>