imd: Use %zu for printing size_t ...

To avoid compiler warnings

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/stm32'

Merge branch 'for-next/misc'

Merge branch 'for-next/block'

state: treat state with all-invalid buckets as dirty ...

The state.dirty flag controls whether state_save will actually
persist state. It is cleared when we successfully load or save
state and set on writing a state parameter.

When the state however becomes corrupt during barebox runtime and
state.dirty == 0, reinitializing the state to defaults is quite
cumbersome:

1. We reset twice. After the first reset, the dirty flag is reset
   and before the second, state_save will reinitialize to defaults
2. We write any state variable and then run the state -s command

Both workarounds are quite obscure, improve the user experience
by having state -l set the dirty flag when it fails, so a subsequent
state -s may persist the default values to state.

Steps to reproduce:

  barebox$ state -l
  state: Using bucket 0@0x00000000
  barebox$ memcpy -s /dev/zero -d /dev/eeprom0.state 0 0 0x400
  barebox$ state -s
  barebox$ state -l
  ERROR: state: No meta data header found
  ERROR: state: No meta data header found
  ERROR: state: No meta data header found
  ERROR: state: Failed to find any valid state copy in any bucket
  ERROR: state: Failed to read state with format raw, -2
  state: No such file or directory

Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

state: backend_storage: deal gracefully with runtime bucket corruption ...

Corrupting an already selected bucket and then reading it again will
crash barebox when it attempts the refresh:

  barebox$ state -l
  barebox$ mw -d /dev/eeprom0.state 0 0x42
  barebox$ state -l
  ERROR: state: No meta data header found
  state: Using bucket 1@0x00000040
  unable to handle NULL pointer dereference at address 0x00000000
  pc : [<4fe4f1ea>]    lr : [<4fe0bcb1>]
  sp : 4ffefd5c  ip : 00000000  fp : 2ff68f04
  r10: 4ffefdc8  r9 : 4b434d63  r8 : 30155f50
  r7 : 00000024  r6 : 2ff68b60  r5 : 2ff68e90  r4 : 00000000
  r3 : 00000024  r2 : 00000024  r1 : 30155f50  r0 : 00000000
  Flags: Nzcv  IRQs off  FIQs off  Mode SVC_32
  WARNING: [<4fe4f1ea>] (memcmp+0x14/0x1a) from [<4fe0bcb1>] (bucket_refresh.isra.0+0x4d/0x78)
  WARNING: [<4fe0bcb1>] (bucket_refresh.isra.0+0x4d/0x78) from [<4fe0be1d>] (state_storage_read+0xd1/0x104)
  WARNING: [<4fe0be1d>] (state_storage_read+0xd1/0x104) from [<4fe0a5bd>] (state_do_load+0x1d/0x78)
  WARNING: [<4fe0a5bd>] (state_do_load+0x1d/0x78) from [<4fe04137>] (execute_command+0x23/0x4c)

The memcmp called here is an optimization to skip I/O if the used bucket
and the one to be refreshed compare equal. Unfortunately, if the now
corrupt bucket was previously the used one, bucket->len will hold the
old value and we'll run into a NULL pointer dereference.

While this is quite inconvenient, it appears it doesn't affect
correctness: after the reset, the corrupt bucket will be refreshed
as expected.

Improve upon this by setting the length to zero when we are NULLing the
buffer. The zero length of the corrupted bucket will then compare unequal
to used_bucket->len in bucket_refresh() and ensure we will always refresh
the buffer if it becomes corrupted without an intermittent reset.

Fixes: 238008b4bd8f ("state: Drop cache bucket")
Cc: Enrico Jörns <ejo@pengutronix.de>
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Revert "block: Adjust cache sizes" ...

On 12/13/19 2:12 PM, Sascha Hauer wrote:
> On Tue, Dec 10, 2019 at 03:44:52PM +0100, Hubert Feurstein wrote:
>> With v2015.06.0 the indicated progress of the copy command is very
>> smooth. Calling "cp -v /dev/zero /dev/mmc3.root" takes about 80
>> seconds for 256MB. But with v2019.12.0 the progress is very bumpy and
>> the copy takes about 280 seconds.
>>
>> I've tracked this down to this commit which destroys the performance:
>> "block: Adjust cache sizes" (b6fef20)
>
> We could just revert this patch. I can't find any workload that gets
> faster with b6fef20c1215. It's rather the other way round.

Do this by reverting commit b6fef20.

Reported-by: Hubert Feurstein <h.feurstein@gmail.com>
Suggested-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

reset_source: introduce reset_source_to_string ...

We have a few places where driver code prints reset_source_name() to
console after setting the reset source.

This working is probe order dependent, because reset_source_name()
prints the highest priority reset_source so far, which doesn't
necessarily have to be the one that was just computed.

Implement reset_source_to_string, so drivers can be migrated to use it.

Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/parameter'

Merge branch 'for-next/optee'

Merge branch 'for-next/misc'

block: Implement discard_range ...

This implements the discard_range hook. When a range of data is
discarded then we do not have to read it from the device and can
pass a zeroed buffer instead.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

commands: bbu: add support for imd checksum ...

Add support for verifying an image by use of the imd checksum.
If the checksum that is saved in the image and the one that is
calculated over the image differ, barebox_update aborts.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

imd: add support for checksum generation/verification ...

Add a new imd type "checksum". This type consists of the CRC32 checksum
of the whole barebox image minus the checksum itself.
The checksum can be written to the imd field with the bareboximd host-tool.
It can be verified with said tool or with "imd" on the target.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ARM: add optee early loading function ...

Add a OP-TEE early loading function which expects a pointer to a valid
tee binary and the device tree. OP-TEE will then be started and barebox
will continue to run in normal mode.

The function start_optee_early should be used in a boards lowlevel.c
file. Ensure that barebox has been relocated and a proper c environment
has been setup beforehand. Depending on the OP-TEE configuration, the
fdt will be modified. If the internal barebox device tree is passed,
OP-TEE will overwrite barebox PBL memory during this modification. Copy
the fdt to a save memory location beforehand to avoid a corruption of
barebox PBL memory.

This also moves the OP-TEE Kconfig symbols into a separate menu.

Signed-off-by: Rouven Czerwinski <r.czerwinski@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

optee: move optee_verify_header() to common ...

Subsequent patches will use this to verify the header in the PBL, move
it to common to make it potentially available for both.

Signed-off-by: Rouven Czerwinski <r.czerwinski@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

imd: replace magicvalue with sizeof(struct) ...

Instead of using "8" as the size of an imd_header, use the sizeof operator.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

common: make CONFIG_BAREBOX_UPDATE visible ...

The barebox update infrastructure is also useful without
CONFIG_CMD_BAREBOX_UPDATE (e.g. from a own barebox_main() with
CONFIG_SHELL_NONE).

Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

globalvar: Allow NULL pointers to be returned by dev_add_param() ...

The following commit will change the return value of dev_add_param()
from -ENOSYS to NULL if CONFIG_PARAMETER is disabled. After that,
building without CONFIG_PARAMETER will return a NULL pointer to
__nvvar_add() instead of an error.

Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

state: remove param member from state_uint32, state_enum32, state_mac ...

After adding the parameter, the pointer isn't required anymore.

Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

state: remove param member from struct state_string ...

After adding the parameter, the pointer isn't required anymore.

Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/zynq'

filetype: add Zynq image type ...

Use the 2 invariant words (width detection and image identification)
from the Zynq image header to detect the filetype.

Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Allow usage of default environment without environment file storage ...

Currently, the default environment is only used when the
barebox environment on the persistent store is not valid
or when ENVFS_FLAGS_FORCE_BUILT_IN is set in the super block.

However, ENVFS_FLAGS_FORCE_BUILT_IN can be cleared and the
environmnet variables in the persistent store will be
used again. This may not be desirable.

This patch allows building CONFIG_DEFAULT_ENVIRONMENT
independent of CONFIG_ENV_HANDLING. This can be useful
if you never want to load or write values from the
persistent store and you only need to read environment variables
from your default environment.

If CONFIG_ENV_HANDLING is not set, a message will be printed to the
user indicating that changes to non-volatile variables won't be
persisted.

Move envfs functions that are needed when CONFIG_DEFAULT_ENVIRONMENT
and/or CONFIG_ENV_HANDLING is set to a new file common/envfs-core.c.

Signed-off-by: Albert Schwarzkopf <a.schwarzkopf@phytec.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/zynq'

Merge branch 'for-next/randfixes'

efi: fix off-by-one in mem_malloc_init(..., end) ...

The second (end) parameter of mem_malloc_init() denotes the last address
in the malloc region, so we need to subtract one from the current value
to arrive at the correct end. So far this went not noticed, because iomem
doesn't yet display barebox malloc memory region.

Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ARM: PBL break is not available on ARMv4 ...

ARMv4 does not support the bkpt mnemonic. Rather than trying to work
around this, just disable this feature for ARMv4. On Tegra the PBL
runs on the ARM720T co-processor, so we also need to disable the option
for this SoC architecture.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

common: provide stubs for ctrlc_* functions ...

Currently the stubs are only provided for CONFIG_CONSOLE_SIMPLE, but
they are also needed for CONSOLE_NONE. Move them to the header.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ratp: add more build dependencies ...

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>