Since commit 2614a20847 ("common: command: tempory buffer should
have size of command line buf"), there have been consistent Travis CI
failures on my builds (interestingly not for Tom, even though building
the same commit id) due to a SEGV in building the byteswapped
omapimage:
arm: pcm051_rev3
make[2]: *** [MLO.byteswap] Error 139
^^^ error code for a SEGV
Turns out that the word-based byte-swapping loop in omapimage.c is to
blame. With the loop condition
while (swapped <= (sbuf->st_size / sizeof(uint32_t)))
there had been one-too-many iterations for all file sizes divisible by
the sizeof(uint32_t). I.e. we had 1 iteration for 0 bytes (and also 1
through 3 bytes) and 2 iterations at 4 bytes... clearly overshooting
on 0 and 4 bytes.
This commit fixes the calculation of an up-rounded word-count and
makes sure to keep the zero-based loop-counter below the number of
words to be processed.
References: 2614a20 ("common: command: tempory buffer should have size of command line buf")
Fixes: 79b9ebb ("omapimage: Add support for byteswapped SPI images")
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Martin Elshuber <martin.elshuber@theobroma-systems.com>
Now, lib/libfdt/fdt_wip.c is the same as scripts/dtc/libfdt/fdt_wip.c
Change the former to a wrapper of the latter.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Up to now we were able to read/write environment data from/to UBI
volumes only indirectly by gluebi driver. This driver creates NAND MTD
on top of UBI volumes, which is quite a workaroung for this use case.
Add support for direct read/write UBI volumes in order to not use
obsolete gluebi driver.
Forward-ported from this patch:
http://patchwork.ozlabs.org/patch/619305/
Original patch:
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Forward port:
Signed-off-by: S. Lockwood-Childs <sjl@vctlabs.com>
The chapter in which the table explaining the image format changed
chapter as the document evolved. This should help people track the
info down faster.
Signed-off-by: Jean-Francois Dagenais <jeff.dagenais@gmail.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
The patch is adding external pmufw "Platform Management Unit firmware"
to boot.bin image. Boot.bin is a Xilinx format which bootrom is capable
to read and boot the system. pmufw is copied to the header data section
follows by u-boot-spl.bin. pmufw is consumed by PMU unit (Microblaze)
and SPL runs on a53-0.
This is generated command line when PMUFW_INIT_FILE is setup.
./tools/mkimage -T zynqmpimage -R ./"" -n
./"board/xilinx/zynqmp/pmufw.bin" -d spl/u-boot-spl.bin spl/boot.bin
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
MRC (Memory Reference Code) is a binary blob used to set up the SDRAM
controller on some Intel boards. Add a test for this feature.
With this test coverage on binman is back up to 100%.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a test for this feature. It allows SPL to hold a pointer to the
microcode block. This is used for 64-bit U-Boot on x86.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a main program so that the tests can be executed directly, without
going through the main binman program.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a test that the 'entry' module works with or without importlib.
The tests are numbered so that they are executed in the correct order.
Signed-off-by: Simon Glass <sjg@chromium.org>
There is a little check at the top of entry.py which decides if importlib
is available. At present this has no test coverage. To add this we will
need to import the module twice, once with importlib and once without.
In preparation for allowing a test to control the importing of this
module, remove all global imports of the 'entry' module.
Signed-off-by: Simon Glass <sjg@chromium.org>
The tests were broken by two separate commits which adjusted the output
when boards are listed. Fix this by adding back a PowerPC board and
putting the name of each board in the test.
Fixes: b9f7d881 (powerpc, 5xx: remove some "5xx" remains)
Fixes: 8d7523c5 (buildman: Allow showing the list of boards with -n)
Signed-off-by: Simon Glass <sjg@chromium.org>
Accessing the network slows down the test and limits the environment in
which it can be run. Add an option to disable network tests.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present these tests use the same filename as patman. This adds
confusion when running all tests, since error messages look very similar.
In fact binman tries to run the wrong tests at present.
Rename the tests.
Signed-off-by: Simon Glass <sjg@chromium.org>
These test programs are includedd as binary files in U-Boot to avoid
having to build them (and associated toolchain differences). Instructions
on building are in the files themselves, but it seems better to provide
a Makefile which can be manually run when desired.
Add a Makefile, separate from the normal build system, to handle this.
Signed-off-by: Simon Glass <sjg@chromium.org>
There is a debugging option in the Makefile to allow people to figure out
which u-boot.dtsi files are used in the build. But is it not easy to use
since it only shows files it finds, not those it is looking for. Update it
and update the mention of it to the docs.
Signed-off-by: Simon Glass <sjg@chromium.org>
With all targets converted to generate prepadded images, this removes
the spl_boot0 field from our config structure and removes the unused
code-path (for images that are not prepadded): i.e. spl_boot0 is now
implied as 'true' and the code is specialised by removing the other
case.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Enable the spl_boot0 in SPL and use the pre-padding TAG memory,
the mkimage do not need to pad it but only need to replace the value
with correct TAG value.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
[Updated:]
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Prior to this commit, tools/Makefile pulls all libfdt files from
lib/libfdt.
lib/libfdt/ and scripts/dtc/libfdt have the same copies for the
followings 6 files:
fdt.c fdt_addresses.c fdt_empty_tree.c fdt_overlay.c fdt_strerr.c
fdt_sw.c
This commit changes them to #include ones from scripts/dtc/libfdt.
Unfortunately, U-Boot locally modified the following 3 files:
fdt_ro.c fdt_wip.c fdt_rw.c
I did not touch them in order to avoid unpredictable impact.
The fdt_region.c is U-Boot own file. This is also borrowed from
lib/libfdt/.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The pylibfdt is used by dtoc (and, indirectly by binman), but there
is no reason why it must be generated in the tools/ directory.
Recently, U-Boot switched over to the bundled DTC, and the directory
structure under scripts/dtc/ now mirrors the upstream DTC project.
So, scripts/dtc/pylibfdt is the best location.
I also rewrote the Makefile in a cleaner Kbuild style.
The scripts from the upstream have been moved as follows:
lib/libfdt/pylibfdt/setup.py -> scripts/dtc/pylibfdt/setup.py
lib/libfdt/pylibfdt/libfdt.i -> scripts/dtc/pylibfdt/libfdt.i_shipped
The .i_shipped is coped to .i during building because the .i must be
located in the objtree when we build it out of tree.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
U-Boot bundles a patched copy of libfdt, so it's wrong to attempt to
include it <like/this>. This breaks the build for me when I have dtc
fully installed in my host -- as happened earlier tonight with
Buildroot, for example.
There are several other occurrences throughout the code where '<libfdt'
matches. I'm not modifying these because I have no clue why the
<systemwide> include style is being used -- IMHO wrongly.
Signed-off-by: Jan Kundrát <jan.kundrat@cesnet.cz>
This function is called when signing configuration nodes. Adjust
the error message.
I do not know why we do not need to show the error message in case of
ENOSPC. Remove the if-conditional that seems unnecessary.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
If -K option is missing when you sign image nodes, it fails with
an unclear error message:
tools/mkimage Can't add hashes to FIT blob: -1
It is hard to figure out the cause of the failure.
In contrast, when you sign configuration nodes, -K is optional because
fit_config_process_sig() returns successfully if keydest is unset.
Probably this is a preferred behavior when you want to update FIT with
the same key; you do not have to update the public key in this case.
So, this commit changes fit_image_process_sig() to continue signing
without keydest. If ->add_verify_data() fails, show a clearer error
message, which has been borrowed from fit_config_process_sig().
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The "algo_name" points to a property in a blob being edited. The
pointer becomes stale when fit_image_write_sig() inserts signatures.
Then crypto->add_verify_data() writes wrong data to the public key
destination.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The fw_env utility family has a default environment compiled in
which ties it quite strongly to the U-Boot source/config it has
been built with. Allow to display the U-Boot version it has been
built with using the -v/--version argument.
Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Reviewed-by: Lukasz Majewski <lukma@denx.de>
Both "conf_name" and "sig_name" point to the name of config node.
The latter should be the name of the signature node.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Last user of this option went away in commit:
fdc7718999 ("board: usb_a9263: Update to support DT and DM")
Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
All BMP files were being treated as though they had a 40 byte header.
There are several BMP header formats consisting of additional data.
This was causing some of the header to be read as color information,
skewing the color palette.
Signed-off-by: Jon Smith <jtsmith@pdiarm.com>
fit_handle_file function does not quote input and output files while preparing
command to run DTC to convert .its to .itb. This results in a failure if input
or output files contain spaces in their names. Quote input and output files in
DTC command to avoid this failure.
Signed-off-by: Mirza, Taimoor <Taimoor_Mirza@mentor.com>
Move the % arch outside the double quote so that the missing toolchain
message is displayed correctly.
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The recent changes to these files did not completely fix the previous
issues, or introduced different (minor) issues. In cmd/gpt.c we need to
dereference str_disk_guid to be sure that malloc worked. In
cmd/nvedit.c we need to be careful that we can also fit in that leading
space when adding to the string. And in tools/fit_image.c we need to
re-work the error handling slightly in fit_import_data() so that we only
call munmap() once. We have two error paths here, one where we have an
fd to close and one where we do not. Adjust labels to match this.
Reported-by: Coverity (CID: 167366, 167367, 167370)
Signed-off-by: Tom Rini <trini@konsulko.com>
Coverity has found some problems with the return paths in parts of this
code. We have a case where we were going to the wrong part of the
unwind (open() failed so we cannot close the fd), a case where we were
only free()ing our buf on the error path and finally a case where we did
not munmap in the failure path.
Reported-by: Coverity (CID: 138492, 138495, 143064)
Signed-off-by: Tom Rini <trini@konsulko.com>
Add support for rk3128 package header in mkimage tool.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Corresponds to 375506d (File writing nit) from upstream
(https://github.com/ulfalizer/Kconfiglib).
Adds proper 'imply' support and fixes a few minor issues, one of which
previously triggered the following weird warning:
configs/taurus_defconfig: /tmp/tmpisI45S:6: warning: assignment to SPL_LDSCRIPT changes mode of containing choice from "arch/$(ARCH)/cpu/u-boot-spl.lds" to "y"
The change in 8639f69 (genconfig.py: Print defconfig next to warnings)
was reapplied.
tools/moveconfig.py previously depended on a hack that merged 'select's
with 'imply's. It was modified to look at the union of
Symbol.get_selected_symbols() and Symbol.get_implied_symbols(), which
should give the same behavior.
tools/genboardscfg.py was verified to produce identical board.cfg's
before and after the change.
Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
In some cases, such as FreeBSD, the path to an alternative dtc needs to
be used. Rather than override the one given in the Makefile on the
command line, make this part of the build configuration.
Signed-off-by: Tom Rini <trini@konsulko.com>
Add a header that indicates that the files generated by dtoc should not be
modified.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
At present dtoc has a very simplistic view of phandles. It assumes that
a property has only a single phandle with a single argument (i.e. two
cells per property).
This is not true in many cases. Enhance the implementation to scan all
phandles in a property and to use the correct number of arguments (which
can be 0, 1, 2 or more) when generating the C code. For the struct
definitions, use a struct which can hold the maximum number of arguments
used by the property.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
We want to support more than one phandle argument. It makes sense to use
an array for this rather than discrete struct members. Adjust the code to
support this. Rename the member to 'arg' instead of 'id'.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
When writing values from properties which contain phandles, dtoc currently
writes 8 phandles per line. Change this to write one phandle per line.
This helps reduce line length, since phandles are generally longer and may
have arguments.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
Rather than naming the phandle struct according to the number of cells it
uses (e.g. struct phandle_2_cell) name it according to the number of
arguments it has (e.g. struct phandle_1_arg). This is a more intuitive
naming.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
Update this function to return more detail about a property that contains
phandles. This will allow (in a future commit) more accurate handling of
these properties.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
This function will need to have access to class members once we enhance it
to support multiple phandle values. In preparation for that, move it into
the class.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
Now that the Fdt class can map phandles to the associated nodes, use that
instead of a separate implementation.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
Add a map from phandles to nodes. This can be used by clients of the the
class instead of maintaining this themselves.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
At present dtoc assumes that all 'reg' properties have both an address and
a size. For I2C devices we do not have this. Adjust dtoc to cope.
Reported-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
When using 32-bit addresses dtoc works correctly. For 64-bit addresses it
does not since it ignores the #address-cells and #size-cells properties.
Update the tool to use fdt64_t as the element type for reg properties when
either the address or size is larger than one cell. Use the correct value
so that C code can obtain the information from the device tree easily.
Alos create a new type, fdt_val_t, which is defined to either fdt32_t or
fdt64_t depending on the word size of the machine. This type corresponds
to fdt_addr_t and fdt_size_t. Unfortunately we cannot just use those types
since they are defined to phys_addr_t and phys_size_t which use
'unsigned long' in the 32-bit case, rather than 'unsigned int'.
Add tests for the four combinations of address and size values (32/32,
64/64, 32/64, 64/32). Also update existing uses for rk3399 and rk3368
which now need to use the new fdt_val_t type.
Signed-off-by: Simon Glass <sjg@chromium.org>
Suggested-by: Heiko Stuebner <heiko@sntech.de>
Reported-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
Large arrays can result in lines with hundreds or thousands of characters
which is not very editor-friendly. To avoid this, addjust the tool to
group values 8 per line.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
When dealing with multi-cell values we need a type that can hold this
value. Add this and a function to process it from a list of cell values.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
We need to be able to search back up the tree for #address-cells and
#size-cells. Record the parent of each node to make this easier.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Kever Yang <kever.yang@rock-chips.com>
In order to display the company's logo via the API of DM_VIDEO,
and add the logo files of both Atmel and Microchip.
Signed-off-by: Wenyou Yang <wenyou.yang@microchip.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
This section of the settings file may be missing. Handle that gracefully
rather than emitting an error.
Also update patman to write this section when a new settings file is
created.
Fixes: e11aa602 (patman: add support for omitting bouncing addresses)
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Chris Packham <judge.pckham@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Add support for reading a list of bouncing addresses from a in-tree file
(doc/bounces) and from the ~/.patman config file. These addresses are
stripped from the Cc list.
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com <mailto:philipp.tomsich@theobroma-systems.com>>
Adjust this code so that it can work with Python 2 and 3.
Fixes: d73fcb1 (moveconfig: Support building a simple config database)
Reported-by: Chris Packham <judge.packham@gmail.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
At present we sometimes see warnings of the form:
/tmp/tmpMA89kB:36: warning: overriding the value of CMD_SPL.
Old value: "y", new value: "y".
This is not very useful as it does not show whch defconfig file it relates
to. Update the tool to show this.
Signed-off-by: Simon Glass <sjg@chromium.org>
Some files are generated during libfdt build. Ignore them.
This was wrongly put in the .gitignore in the root directory before.
Now let's remove entries there and put them in the right place.
Fixes: 34e2c285 ("gitignore: add intermediates from libfdt build")
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
This was broken by the recent environment refactoring. Specifically:
$ make environ
scripts/Makefile.build:59: tools/environ/Makefile: No such file or directory
make[1]: *** No rule to make target 'tools/environ/Makefile'. Stop.
make: *** [Makefile:1469: environ] Error 2
Fix this by updating the Makefile and adjusting the #include filesnames in
two C files.
Fixes: ec74f5f (Makefile: Rename 'env' target to 'environ')
Reported-by: Måns Rullgård <mans@mansr.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
With the move of environment code from common/ to env/ a number of
changes needed to be made to various make targets. We missed updating
some of the files required for out of tree builds of the tools. Correct
the 'environ' target to know that we need to work under tools/env/ still
(not tools/environ/) and then update the wrappers in env_attr.c and
env_flags.c to point to the new correct file.
Reported-by: Peter Robinson <pbrobinson@gmail.com>
Tested-by: Peter Robinson <pbrobinson@gmail.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
This should not be printed by default. Prefix it with $(Q).
Fixes ee95d10b: ("fdt: Build the new python libfdt module")
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Since commit 3809e302 "Makefile: honor PYTHON configuration properly",
the build commands of libfdt are printed while previously were not.
This adds the missing '--quiet' back.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
According to fsync specification [1] some special files (e.g., a pipe, FIFO,
or socket) don't support synchronization and return either EROFS or EINVAL.
On the linux side the sys_fsync -> do_fsync() checks if the requested file
has f_op->fsync defined. If not it returns EINVAL [2].
This commit prevents writing error messages for files (devices), which
do not support fsync().
[1] - http://man7.org/linux/man-pages/man2/fsync.2.html
[2] - http://elixir.free-electrons.com/linux/v4.13-rc6/source/fs/sync.c#L183
Signed-off-by: Lukasz Majewski <lukma@denx.de>
Acked-by: Michael Heimpold <mhei@heimpold.de>
Move the imximage.h header file to a common location so we can make
use of it from U-Boot too.
Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Reviewed-by: Łukasz Majewski <lukma@denx.de>
On some systems `python` is `python3` (for instance, Archlinux). The
`PYTHON` variable can be used to point to `python2` to have a successful
build.
The use of `PYTHON` is currently limited in the Makefile and needs to be
extended in other places:
First, pylibfdt is required to be a Python 2 binding (binman imports
pylibfdt and is only compatible Python 2), so its setup.py needs to be
called accordingly. An alternative would be to change the libfdt
setup.py shebang to python2, but the binding is actually portable. Also,
it would break on system where there is no such thing as `python2`.
Secondly, the libfdt import checks need to be done against Python 2 as
well since the Python 2 compiled modules (in this case _libdft.so) can
not be imported from Python 3.
Note on the libfdt imports: "@if ! PYTHONPATH=tools $(PYTHON) -c 'import
libfdt'; then..." is probably simpler than the currently sub-optimal
pipe.
Reviewed-by: Jonathan Gray <jsg@jsg.id.au>
We are now using an env_ prefix for environment functions. Rename setenv()
for consistency. Also add function comments in common.h.
Suggested-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Simon Glass <sjg@chromium.org>
About a quarter of the files in common/ relate to the environment. It
seems better to put these into their own subdirectory and remove the
prefix.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
This commit adds support for RK3368 SoC in mkimage.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
It seems that gcc 6.3 at least is smart enough to warn about the _val
variable being unassigned in the default case in the set_hdr_field()
macro, but not smart enough to figure out that the default case is never
taken. This results in warnings such as the following:
pfx##hdr32[idx].field = _val; \
^
../tools/mips-relocs.c:51:11: note: _val was declared here
uint64_t _val; \
^
../tools/mips-relocs.c:88:2: note: in expansion of macro set_hdr_field
set_hdr_field(p, idx, field, val)
^~~~~~~~~~~~~
../tools/mips-relocs.c:408:3: note: in expansion of macro set_phdr_field
set_phdr_field(i, p_filesz, load_sz);
^~~~~~~~~~~~~~
../tools/mips-relocs.c: In function main:
../tools/mips-relocs.c:77:25: warning: _val may be used uninitialized
in this function [-Wmaybe-uninitialized]
Avoid this by assigning _val = 0 in the default case, and asserting that
we didn't actually hit it for good measure.
For reference gcc 7.1.1 seems to be smart enough to not hit the above
warning without this patch.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Fixes: 011dd93ca97a ("MIPS: Stop building position independent code")
Cc: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Cc: Tom Rini <trini@konsulko.com>
Cc: u-boot@lists.denx.de
U-Boot has up until now built with -fpic for the MIPS architecture,
producing position independent code which uses indirection through a
global offset table, making relocation fairly straightforward as it
simply involves patching up GOT entries.
Using -fpic does however have some downsides. The biggest of these is
that generated code is bloated in various ways. For example, function
calls are indirected through the GOT & the t9 register:
8f998064 lw t9,-32668(gp)
0320f809 jalr t9
Without -fpic the call is simply:
0f803f01 jal be00fc04 <puts>
This is more compact & faster (due to the lack of the load & the
dependency the jump has on its result). It is also easier to read &
debug because the disassembly shows what function is being called,
rather than just an offset from gp which would then have to be looked up
in the ELF to discover the target function.
Another disadvantage of -fpic is that each function begins with a
sequence to calculate the value of the gp register, for example:
3c1c0004 lui gp,0x4
279c3384 addiu gp,gp,13188
0399e021 addu gp,gp,t9
Without using -fpic this sequence no longer appears at the start of each
function, reducing code size considerably.
This patch switches U-Boot from building with -fpic to building with
-fno-pic, in order to gain the benefits described above. The cost of
this is an extra step during the build process to extract relocation
data from the ELF & write it into a new .rel section in a compact
format, plus the added complexity of dealing with multiple types of
relocation rather than the single type that applied to the GOT. The
benefit is smaller, cleaner, more debuggable code. The relocate_code()
function is reimplemented in C to handle the new relocation scheme,
which also makes it easier to read & debug.
Taking maltael_defconfig as an example the size of u-boot.bin built
using the Codescape MIPS 2016.05-06 toolchain (gcc 4.9.2, binutils
2.24.90) shrinks from 254KiB to 224KiB.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Cc: u-boot@lists.denx.de
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Tested-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
It is annoying to have to set up and maintain two sets of toolchains, one
for buildman and one for moveconfig.
Adjust moveconfig to make use to buildman's toolchains. This should make
things easier.
One missing feature is the ability to specify the toolchain on the command
line with a special environment variable, e.g. CROSS_COMPILE_ARM. I'm not
sure if that is useful, but if it is it could be implemented in buildman.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Tested-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Add an option to specify 'all' to enable all flags. Also print an error
if an unrecognised flag is used. At present it just prints usage
information which is not very helpful.
Signed-off-by: Simon Glass <sjg@chromium.org>
FreeBSD recently switch to it's BSDL dtc. While it support most of the
features of the GPL one it still lacks the incbin directive.
Add the possibility to specify which dtc we want to use for compiling dts
and generating fit image.
Signed-off-by: Emmanuel Vadot <manu@bidouilliste.com>
Acked-by: Simon Glass <sjg@chromium.org>
Closing a file descriptor does not guarantee that the data has been
successfully saved to disk, as the kernel might defer the write.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
Change is consistent with other SOCs and it is in preparation
for adding SOMs. SOC's related files are moved from cpu/ to
mach-imx/<SOC>.
This change is also coherent with the structure in kernel.
Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Fabio Estevam <fabio.estevam@nxp.com>
CC: Akshay Bhat <akshaybhat@timesys.com>
CC: Ken Lin <Ken.Lin@advantech.com.tw>
CC: Marek Vasut <marek.vasut@gmail.com>
CC: Heiko Schocher <hs@denx.de>
CC: "Sébastien Szymanski" <sebastien.szymanski@armadeus.com>
CC: Christian Gmeiner <christian.gmeiner@gmail.com>
CC: Stefan Roese <sr@denx.de>
CC: Patrick Bruenn <p.bruenn@beckhoff.com>
CC: Troy Kisky <troy.kisky@boundarydevices.com>
CC: Nikita Kiryanov <nikita@compulab.co.il>
CC: Otavio Salvador <otavio@ossystems.com.br>
CC: "Eric Bénard" <eric@eukrea.com>
CC: Jagan Teki <jagan@amarulasolutions.com>
CC: Ye Li <ye.li@nxp.com>
CC: Peng Fan <peng.fan@nxp.com>
CC: Adrian Alonso <adrian.alonso@nxp.com>
CC: Alison Wang <b18965@freescale.com>
CC: Tim Harvey <tharvey@gateworks.com>
CC: Martin Donnelly <martin.donnelly@ge.com>
CC: Marcin Niestroj <m.niestroj@grinn-global.com>
CC: Lukasz Majewski <lukma@denx.de>
CC: Adam Ford <aford173@gmail.com>
CC: "Albert ARIBAUD (3ADEV)" <albert.aribaud@3adev.fr>
CC: Boris Brezillon <boris.brezillon@free-electrons.com>
CC: Soeren Moch <smoch@web.de>
CC: Richard Hu <richard.hu@technexion.com>
CC: Wig Cheng <wig.cheng@technexion.com>
CC: Vanessa Maegima <vanessa.maegima@nxp.com>
CC: Max Krummenacher <max.krummenacher@toradex.com>
CC: Stefan Agner <stefan.agner@toradex.com>
CC: Markus Niebel <Markus.Niebel@tq-group.com>
CC: Breno Lima <breno.lima@nxp.com>
CC: Francesco Montefoschi <francesco.montefoschi@udoo.org>
CC: Jaehoon Chung <jh80.chung@samsung.com>
CC: Scott Wood <oss@buserror.net>
CC: Joe Hershberger <joe.hershberger@ni.com>
CC: Anatolij Gustschin <agust@denx.de>
CC: Simon Glass <sjg@chromium.org>
CC: "Andrew F. Davis" <afd@ti.com>
CC: "Łukasz Majewski" <l.majewski@samsung.com>
CC: Patrice Chotard <patrice.chotard@st.com>
CC: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
CC: Hans de Goede <hdegoede@redhat.com>
CC: Masahiro Yamada <yamada.masahiro@socionext.com>
CC: Stephen Warren <swarren@nvidia.com>
CC: Andre Przywara <andre.przywara@arm.com>
CC: "Álvaro Fernández Rojas" <noltari@gmail.com>
CC: York Sun <york.sun@nxp.com>
CC: Xiaoliang Yang <xiaoliang.yang@nxp.com>
CC: Chen-Yu Tsai <wens@csie.org>
CC: George McCollister <george.mccollister@gmail.com>
CC: Sven Ebenfeld <sven.ebenfeld@gmail.com>
CC: Filip Brozovic <fbrozovic@gmail.com>
CC: Petr Kulhavy <brain@jikos.cz>
CC: Eric Nelson <eric@nelint.com>
CC: Bai Ping <ping.bai@nxp.com>
CC: Anson Huang <Anson.Huang@nxp.com>
CC: Sanchayan Maity <maitysanchayan@gmail.com>
CC: Lokesh Vutla <lokeshvutla@ti.com>
CC: Patrick Delaunay <patrick.delaunay@st.com>
CC: Gary Bisson <gary.bisson@boundarydevices.com>
CC: Alexander Graf <agraf@suse.de>
CC: u-boot@lists.denx.de
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
The verify_header callback in kwbimage.c only verifies v0 headers checksum.
Running 'mkimage -l' on a v1 image gives the following misleading output:
GP Header: Size ae000000 LoadAddr 34160600
Implement support for v1 headers. For that, factor out the header checksum code
to a separate main_hdr_checksum_ok() routine. This routine relies on the fact
that the checksum field offset is the same in both v0 and v1 headers. With this
patch applied 'mkimage -l' correctly identifies the image:
Image Type: MVEBU Boot from sdio Image
Image version:1
Data Size: 398904 Bytes = 389.55 KiB = 0.38 MiB
Load Address: 007fffc0
Entry Point: 00800000
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Stefan Roese <sr@denx.de>
The offset marking in kwbimage.h is inconsistent. main_hdr_v0 uses decimals,
main_hdr_v1 uses hex without '0x' prefix, secure_hdr_v1 uses hex with '0x'
prefix. Make all offset marks hex with '0x' prefix.
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Stefan Roese <sr@denx.de>
If a system module is named the same as one of those used by binman we
currently pick the system module. Adjust the ordering so that our modules
are chosen instead.
The module conflict reported was 'tools' from jira-python. I cannot access
that package to test it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Kevin Hilman <khilman@baylibre.com>
Acked-by: Kevin Hilman <khilman@baylibre.com>
This function uses several separate string replaces where a regular
expression might seem more reasonable. Add a comment justifying the way it
is currently done.
Signed-off-by: Simon Glass <sjg@chromium.org>
Collect the main logic of dtoc into a function and put it into
dtb_platdata. This will allow tests to use this function instead of
duplicating the code themselves.
Signed-off-by: Simon Glass <sjg@chromium.org>
