A recent change adjusted the symbol calculation to work on x86 but broke
it for Tegra. In fact this is because they have different needs.

On x86 devices the code is linked to a ROM address and the end-at-4gb
property is used for the image. In this case there is no need to add the
base address of the image, since the base address is already built into
the offset and image-pos properties.

On other devices we must add the base address since the offsets start at
zero.

In addition the base address is currently added to the 'offset' and 'size'
values. It should in fact only be added to 'image-pos', since 'offset' is
relative to its parent and 'size' is not actually an address. This code
should have been adjusted when support for 'image-pos' and 'size' was
added, but it was not.

To correct these problems:
- move the code that handles adding the base address to section.py, which
  can check the end-at-4gb property and which property
  (offset/size/image-pos) is being read
- add the base address only when needed (only for image-pos and not if the
  image uses end-at-4gb)
- add a note to the documentation
- add a separate test to cover x86 behaviour

Fixes: 15c981cc (binman: Correct symbol calculation with non-zero image base)
Signed-off-by: NSimon Glass <sjg@chromium.org>
Tested-by: NStephen Warren <swarren@nvidia.com>

Update this tool to use Python 3 to meet the 2020 deadline.

Unfortunately this introduces a test failure due to a problem in pylibfdt
on Python 3. I will investigate.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Some tests have crept in with Python 2 strings and constructs. Convert
then.
Signed-off-by: NSimon Glass <sjg@chromium.org>

When preparing to possible expand or contract an entry we reset the size
to the original value from the binman device-tree definition, which is
often None.

This causes binman to forget the original size of the entry. Remember this
so that it can be used when needed.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Build this swig module with Python 3.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present all the 'command' methods return bytes. Most of the time we
actually want strings, so change this. We still need to keep the internal
representation as bytes since otherwise unicode strings might break over
a read() boundary (e.g. 4KB), causing errors. But we can convert the end
result to strings.

Add a 'binary' parameter to cover the few cases where bytes are needed.
Signed-off-by: NSimon Glass <sjg@chromium.org>

This comment references the wrong FSP component. Fix it.
Signed-off-by: NSimon Glass <sjg@chromium.org>
Reviewed-by: NBin Meng <bmeng.cn@gmail.com>

This entry is used to hold an Intel FSP-T (Firmware Support Package
Temp-RAM init) binary. Add support for this in binman.
Signed-off-by: NSimon Glass <sjg@chromium.org>
Reviewed-by: NBin Meng <bmeng.cn@gmail.com>

This entry is used to hold an Intel FSP-S (Firmware Support Package
Silicon init) binary. Add support for this in binman.
Signed-off-by: NSimon Glass <sjg@chromium.org>
Reviewed-by: NBin Meng <bmeng.cn@gmail.com>

At present binman adds the image base address to the symbol value before
it writes it to the binary. This is not correct since the symbol value
itself (e.g. image position) has no relationship to the image base.

Fix this and update the tests to cover this case.
Signed-off-by: NSimon Glass <sjg@chromium.org>
Reviewed-by: NBin Meng <bmeng.cn@gmail.com>

This code is not needed so drop it.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Sometimes binman takes multiple passes to complete packing an image. Add
logging to indicate this.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present the symbol information is written to binaries just before
binman exits. This is fine for entries within sections since the section
contents is calculated when it is needed, so the updated symbol values are
included in the image that is written.

However some binaries are inside entries which have already generated
their contents and do not notice that the entries have changed (e.g. Intel
IFWI).

Move the symbol writing earlier to cope with this.
Signed-off-by: NSimon Glass <sjg@chromium.org>

The Intel IFWI (Integrated Firmware Image) is effectively a section with
other entries inside it. Support writing symbol information into entries
within it.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Add support for the ProcessContents() method in this entry so that it is
possible to support entries which change after initial creation.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present this class reads its entries in the constructor. This is not
how things should be done now. Update it.
Signed-off-by: NSimon Glass <sjg@chromium.org>

This comment references another entry type. Fix it.
Signed-off-by: NSimon Glass <sjg@chromium.org>

The Intel FSP supports initialising memory early during boot using a binary
blob called 'fspm'. Add support for this.
Signed-off-by: NSimon Glass <sjg@chromium.org>

It is useful to be able to access the size of an image in SPL, with
something like:

binman_sym_declare(unsigned long, u_boot_any, size);

...
   ulong u_boot_size = binman_sym(ulong, u_boot_any, size);

Add support for this and update the tests.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present these are large enough to hold 20 bytes of symbol data. Add
four more bytes so we can add another test.

Unfortunately at present this involves changing a few test files to make
room. We could adjust the test files to not specify sizes for entries.
Then we could make the tests check the actual sizes. But for now, leave it
as it is, since the effort is minor.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Entries which include a section and need to obtain its contents call
GetData(), as with any other entry. But the current implementation of this
method in entry_Section requires the size of the section to be known. If
it is unknown, an error is produced, since size is None:

   TypeError: can't multiply sequence by non-int of type 'NoneType'

There is no need to know the size in advance since the code can be
adjusted to build up the section piece by piece, instead of patching each
entry into an existing bytearray.

Update the code to handle this and add a test.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Two of the test files somehow were not converted to three digits. Fix
them, using the next available numbers.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present a small number of test files use hyphens instead of
underscores. Rename them for consistency.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present we only support symbols inside binaries which are at the top
level of an image. This restrictions seems unreasonable since more complex
images may want to group binaries within different sections.

Relax the restriction, adding a new _SetupTplElf() helper function.

Also fix a typo in the comment for testTpl().
Signed-off-by: NSimon Glass <sjg@chromium.org>

We use the Makefile for all ELF test files now, so drop all the code that
checks whether to get the test file from the Makefile or from the git
repo.

Also add a comment to the Makefile indicating that it is run from binman.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Remove this file from git and instead build it using the Makefile.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Remove this file from git and instead build it using the Makefile.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Remove this file from git and instead build it using the Makefile.

With this change a few things need to be adjusted:

1. The 'notes' section no-longer appears at the start of the ELF file
(before the code), so update testSymbols to adjust the offsets.

2. The dynamic linker is disabled to avoid errors like:

    "Not enough room for program headers, try linking with -N"

3. The interpreter note is moved to the end of the image, so that the
binman symbols appear first.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Remove this file from git and instead build it using the Makefile.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Remove this file from git and instead build it using the Makefile.

Update tools.GetInputFilename() to support reading files from an absolute
path, so that we can read the Elf test files easily. Also make sure that
the temp directory is report in ELF tests as this was commented out.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present the ELF test files are checked into the U-Boot tree. This is
covenient since the files never change and can be used on non-x86
platforms. However it is not good practice to check in binaries and in
this case it does not seem essential.

Update the binman test-file Makefile to support having source in a
different directory. Adjust binman to run it to build bss_data, as a
start. We can add other files as needed.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present this command silently fails if something goes wrong. Use the
tools.Run() function instead, since it reports errors.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present this entry does not work correctly when a FIT image is used as
the input. It updates the FIT instead of the output image. The test passed
because the FIT image happened to have the right data already.

Fix it.
Signed-off-by: NSimon Glass <sjg@chromium.org>

A Firmware Image Table (FIT) is a data structure defined by Intel which
contains information about various things needed by the SoC, such as
microcode.

Add support for this entry as well as the pointer to it. The contents of
FIT are fixed at present. Future work is needed to support adding
microcode, etc.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present these two sections of code are linked together into a single
2KB chunk in a single file. Some Intel SoCs like to have a FIT (Firmware
Interface Table) in the ROM and the pointer for this needs to go at
0xffffffc0 which is in the middle of these two sections.

Make use of the new 'reset' entry and change the existing 16-bit entry to
include just the 16-bit data.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present binman has a single entry type for the 16-bit code code needed
to start up an x86 processor. This entry is intended to include both the
reset vector itself as well as the code to move to 32-bit mode.

However this is not very flexible since in some cases other data needs to
be included at the top of the SPI flash, in between these two pieces. For
example Intel requires that a FIT (Firmware Image Table) pointer be placed
0x40 bytes before the end of the ROM.

To deal with this, add a new reset entry for just the reset vector. A
subsequent change will adjust the existing 'start16' entry.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present the Intel IFWI entry uses 'replace' without the 'ifwi-' prefix.
This is a fairly generic name which might conflict with the main Entry
base class at some point, if more features are added. Add a prefix.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Some versions of binutils generate hidden symbols which are currently not
parsed by binman. Correct this.
Signed-off-by: NSimon Glass <sjg@chromium.org>

Recent versions of binutils add a '.note.gnu.property' into the ELF file.
This is not required and interferes with the expected output. Drop it.

Also fix testMakeElf() to use a different file for input and output.
Signed-off-by: NSimon Glass <sjg@chromium.org>

At present if libfdt is not available binman can't do anything much.
Improve the situation a little.

Ideally there should be a test to cover this, but I'm not quite sure how
to fake this.
Signed-off-by: NSimon Glass <sjg@chromium.org>
(fixed up missing ReadChildData() enty test)