Simon Glass [Sun, 21 Mar 2021 05:24:38 +0000 (18:24 +1300)]
dtoc: Support adding subnodes alongside existing ones
So far we have only needed to add subnodes to empty notds, so have not
had to deal with ordering. However this feature is needed for binman's
expanded nodes, since there may be another node in the same section.
While libfdt adds new properties after existing properties, it adds new
subnodes before existing subnodes. This means that we must reorder the
nodes in the cached version, so that the ordering remains consistent.
Update the sync implementation to sync existing subnodes first, then
add new ones, then tidy up the ordering in the cached version. Update the
test to cover this behaviour.
Also improve the comment about property syncing while we are here.
Simon Glass [Sun, 21 Mar 2021 05:24:37 +0000 (18:24 +1300)]
dtoc: Add a subnode test for multiple nodes
Add a new test that adds a subnode alongside an existing one, as well as
adding properties to a subnode. This will expand to adding multiple
subnodes in future patches. Put a node after the one we are adding to so
we can check that things sync correctly.
The testAddNode() test should be in the TestNode class since it is a node
test, so move it.
Simon Glass [Sun, 21 Mar 2021 05:24:36 +0000 (18:24 +1300)]
dtoc: Tweak ordering of fdt-offsets refreshing
Once the tree has been synced, thus potentially moving things around in the
fdt, we set _cached_offsets to False so that a refresh will happen next
time a property is accessed.
This 'lazy' refresh doesn't really save much time, since refresh is a very
fast operation, just a single walk of the tree. Also, having the refresh
happen in the bowels of property access it makes it harder to figure out
what is going on.
Simplify the code by always doing a refresh before and after a sync. Set
_cached_offsets to True immediately after this, in the Refresh() function,
since this makes more sense than doing it in the caller.
Simon Glass [Sun, 21 Mar 2021 05:24:35 +0000 (18:24 +1300)]
dtoc: Tidy up property-offset handling
If a property does not yet have an offset, then that means it exists in
the cache'd fdt but has not yet been synced back to the flat tree. Use
the dirty flag for this so we don't need to check the offset too. Improve
the comments for Prop and Node to make it clear what an offset of None
means.
Also clear the dirty flag after the property is synced.
Simon Glass [Sun, 21 Mar 2021 05:24:33 +0000 (18:24 +1300)]
binman: Support default alignment for sections
Sometimes it is useful to specify the default alignment for all entries
in a section, such as when word-alignment is necessary, for example. It
is tedious and error-prone to specify this individually for each section.
Simon Glass [Sun, 21 Mar 2021 05:24:32 +0000 (18:24 +1300)]
binman: Support obtaining section contents immediately
Generally the content of sections is not built until the final assembly
of the image. This is partly to avoid wasting time, since the entries
within sections may change multiple times as binman works through its
various stages. This works quite well since sections exist in a strict
hierarchy, so they can be processed in a depth-first manner.
However the 'collection' entry type does not have this luxury. If it
contains a section within its 'content' list, then it must produce the
section contents, if available. That section is typically a sibling
node, i.e. not part oc the collection's hierarchy.
Add a new 'required' argument to section.GetData() to support this. When
required is True, any referenced sections are immediately built. If this
is not possible (because one of the subentries does not have its data yet)
then an error is produced.
The test for this uses a 'collection' entry type, referencing a section as
its first member. This forces a call to _BuildSectionData() with required
set to False, at first, then True later, when the image is assembled.
Simon Glass [Sun, 21 Mar 2021 05:24:31 +0000 (18:24 +1300)]
binman: Add support for a collection of entries
The vblock entry type includes code to collect the data from a number of
other entries (not necessarily subentries) and concatenating it. This is
a useful feature for other entry types.
Make it a base class, so that vblock can use it, along with other entry
types.
Simon Glass [Sun, 21 Mar 2021 05:24:30 +0000 (18:24 +1300)]
binman: Allow disabling expanding an entry
At present there is a command-line flag to disable substitution of expanded
entries. Add an option to the entry node as well, so it can be controlled
at the node level.
Add a test to cover this. Fix up the comment to the checkSymbols() function
it uses, while we are here.
Simon Glass [Sun, 21 Mar 2021 03:50:07 +0000 (16:50 +1300)]
x86: coral: Show memory config and SKU ID on startup
Provide the model information through sysinfo so that it shows up on
boot. For memconfig 4 pins are provided, for 16 combinations. For SKU
ID there are two options:
- two pins provided in a ternary arrangement, for 9 combinations.
- reading from the EC
Add a binding doc and drop the unused #defines as well.
Simon Glass [Sun, 21 Mar 2021 03:50:06 +0000 (16:50 +1300)]
sysinfo: Allow showing model info from sysinfo
Some boards may want to show the SKU ID or other information obtained at
runtime. Allow this to come from sysinfo. The board can then provide a
sysinfo driver to provide it.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Simon Glass [Mon, 15 Mar 2021 05:11:24 +0000 (18:11 +1300)]
sandbox: Correct uninit conflict
It is not possible to remove the state before driver model is uninited,
since the devices are allocated in the memory buffer. Also it is not
possible to uninit driver model afterwards, since the RAM has been
freed.
Drop the uninit altogether, since it is not actually necessary.
Simon Glass [Mon, 15 Mar 2021 05:11:21 +0000 (18:11 +1300)]
command: Fix operation of !CONFIG_CMDLINE
The U_BOOT_CMDREP_COMPLETE() macro produces a build error if CONFIG_CMDLINE
is not enabled. Fix this by updating the macro to provide the 'repeatable'
arugment in this case.
Simon Glass [Mon, 15 Mar 2021 05:11:17 +0000 (18:11 +1300)]
sf: Support querying write-protect
This feature was dropped from U-Boot some time ago:
f12f96cfaf5 (sf: Drop spl_flash_get_sw_write_prot")
However, we do need a way to see if a flash device is write-protected,
since if it is, it may not be possible to write to do (i.e. failing to
write is expected).
I am not sure of the correct layer to implement this, so this patch is a
stab at it. If spi-flash makes sense then I will add to the 'sf' also.
Re the points mentioned in the removal commit:
1) This kind of requirement can be achieved using existing
flash operations and flash locking API calls instead of
making a separate flash API.
Which uclass is this?
2) Technically there is no real hardware user for this API to
use in the source tree.
I do want coral (at least) to support this.
3) Having a flash operations API for simple register read bits
also make difficult to extend the flash operations.
This new patch only mentions write-protect being on or off, rather than
the actual mechanism.
4) Instead of touching generic code, it is possible to have
this functionality inside spinor operations in the form of
flash hooks or fixups for associated flash chips.
That sounds to me like what drivers are for. But we still need some sort
of API for it to be accessible.
Simon Glass [Mon, 15 Mar 2021 05:11:16 +0000 (18:11 +1300)]
bootstage: Warning if space is exhausted
At present bootstage silently ignores new records if it runs out of
space. It is sometimes obvious by looking at the report, but the IDs are
not contiguous, so it is easy to miss.
Simon Glass [Mon, 15 Mar 2021 05:11:12 +0000 (18:11 +1300)]
sandbox: image: Allow sandbox to load any image
Sandbox is special in that it is used for testing and it does not match
any particular target architecture. Allow it to load an image from any
architecture, so that 'bootm' can be used as needed.
Simon Glass [Mon, 15 Mar 2021 05:11:11 +0000 (18:11 +1300)]
sandbox: Provide a way to bind fixed/removeable devices
At present when a file is bound to a host device it is always marked as
removeable. Arguably the device is removeable, since it can be unbound at
will. However while it is bound, it is not considered removable by the
user. Also it is useful to be able to model both fixed and removeable
devices for code that distinguishes them.
Add a -r flag to the 'host bind' command and plumb it through to provide
this feature.
Simon Glass [Mon, 15 Mar 2021 05:11:09 +0000 (18:11 +1300)]
sandbox: Disintangle declarations in do_host_bind()
This function has a strange mix of declarations and argument parsing
which is a bit hard to follow and harder to modify. Separate out the
declarations at the start of the function and adjust the ordering of
the code slightly.
Simon Glass [Mon, 15 Mar 2021 05:11:07 +0000 (18:11 +1300)]
sandbox: Only delete the executable if requested
At present sandbox removes its executable after failing to run it,
since there is no other way that it would get cleaned up.
However, this is actually only wanted if the image was created within
sandbox. For the case where the image was generated by the build system,
such as u-boot-spl, we don't want to delete it.
Simon Glass [Mon, 15 Mar 2021 05:00:33 +0000 (18:00 +1300)]
x86: fsp: Don't notify if booted from coreboot
When booting from coreboot there is no need to notify the FSP of anything,
since coreboot has already done it. Nor it is possible, since the FSP
details are not provided by coreboot.
Simon Glass [Mon, 15 Mar 2021 05:00:32 +0000 (18:00 +1300)]
x86: coral: Fall back to coreboot video when FSP missing
When booting from coreboot the FSP video information is no-longer
available. Enable the coreboot driver so that we can get some sort of
display in this case.
Simon Glass [Mon, 15 Mar 2021 05:00:31 +0000 (18:00 +1300)]
x86: coral: Allow init of debug UART in U-Boot proper
At present the debug UART is only set up in SPL, on the assumption that
the boot flow will always pass through there. When booting from coreboot,
SPL is not used, so the debug UART is not available.
Move the code into a common place so that it can be used in U-Boot proper
also. Add the required init to start_from_spl.S as well.
Simon Glass [Mon, 15 Mar 2021 05:00:28 +0000 (18:00 +1300)]
x86: fsp: Don't enable FSP graphics if booted from coreboot
This driver cannot work when booted from coreboot, since the FSP
information is not available. Disable it in that case, so that the
coreboot video driver can be used instead.
Simon Glass [Mon, 15 Mar 2021 05:00:26 +0000 (18:00 +1300)]
video: Fix video on coreboot with the copy buffer
The copy buffer, if enabled, prevents booting from coreboot correctly,
since no memory is allocated for it. Allow it to fall back to disabled
in this situation. This ensures that a console is displayed, even if
it is slow.
Simon Glass [Mon, 15 Mar 2021 05:00:23 +0000 (18:00 +1300)]
x86: Allow installing an e820 when booting from coreboot
Move this code into a generic location so that it can be used by other x86
boards which want to boot from coreboot. Also ensure that this is called
if booting from coreboot.
Simon Glass [Mon, 15 Mar 2021 05:00:21 +0000 (18:00 +1300)]
x86: Move coreboot sysinfo parsing into generic x86 code
It is useful to be able to parse coreboot tables on any x86 build which is
booted from coreboot. Add a new Kconfig option to enable this feature and
move the code so it can be used on any board, if enabled.
Simon Glass [Mon, 15 Mar 2021 05:00:20 +0000 (18:00 +1300)]
x86: coreboot: Sync up timestamp codes
Add new timestamp codes that are present in coreboot, so that we can decode
these in U-Boot.
At present TS_U_BOOT_START_KERNEL is used twice. It should only be used
just before jumping to Linux, so update the other call site to use
TS_START_KERNEL.
Simon Glass [Mon, 15 Mar 2021 05:00:19 +0000 (18:00 +1300)]
x86: Move coreboot timestamp info into coreboot_tables.h
This all relates to the sysinfo structure provided by coreboot. Put the
timestamp definitions into the same file as the others. Tidy up a few
comments at the same time.
Simon Glass [Mon, 15 Mar 2021 05:00:18 +0000 (18:00 +1300)]
x86: Make coreboot sysinfo available to any x86 board
It is possible to boot U-Boot for chromebook_coral either 'bare metal' or
from coreboot. In the latter case we want to provide access to the coreboot
sysinfo tables. Move the definitions into a file available to any x86
board.
Simon Glass [Mon, 15 Mar 2021 05:00:16 +0000 (18:00 +1300)]
cbfs: Support reading compression information
CBFS now supports compressed filed. Add support for reading this
information so that the correct decompression can be applied. The
decompression itself is not implemented in CBFS.
Simon Glass [Mon, 15 Mar 2021 05:00:13 +0000 (18:00 +1300)]
cbfs: Allow file traversal with any CBFS
The file traversal functions currently use a single global CBFS. In some
cases we need to access multiple CBFSs to obtain different files. Add new
functions to support this.
Simon Glass [Mon, 15 Mar 2021 05:00:12 +0000 (18:00 +1300)]
cbfs: Allow access to CBFS without a header
In some cases CBFS does not start with a header but is just a collection
of files. It is possible to support this so long as the size of the CBFS
is provided.
Update the cbfs_init_mem() function to support this.
Simon Glass [Mon, 15 Mar 2021 05:00:11 +0000 (18:00 +1300)]
smbios: Allow writing to the coreboot version string
When U-Boot is booted from coreboot the SMBIOS tables are written by
coreboot, not U-Boot. The existing method of updating the BIOS version
string does not work in that case, since gd->smbios_version is only set
when U-Boot writes the tables.
Add a new function which allows the version to be updated by parsing the
tables and writing the string in the correct place. Since coreboot
provides a pointer to the SMBIOS tables in its sysinfo structure, this
makes it easy to do the update.
Simon Glass [Mon, 15 Mar 2021 05:00:06 +0000 (18:00 +1300)]
x86: coral: Update the SD card-detect GPIO
Since the recent bug fix, it doesn't matter which GPIO phandle is used so
long as the GPIO number is right. Still, we may as well use the correct
one to avoid confusion.
Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
Simon Glass [Mon, 15 Mar 2021 05:00:04 +0000 (18:00 +1300)]
x86: Probe device if needed in intel_gpio_xlate()
The Intel GPIO binding allows GPIOs to be globally numbered, so that it
does not matter which GPIO bank is specified in the device tree. This is
convenient and avoid confusion since the banks do not have the same number
of GPIOs and the numbering is not sequential.
The GPIO uclass ensures that the device mentioned in the devicetree
binding is probed. It is fine for the driver to update gpio_desc to point
to a different driver, but this may not have been probed. If it has not
been, then it cannot be claimed since there is no uclass data.
We could handle this in the GPIO uclass but so far it is an unusual
situation so it is probably not worth the extra code. Handle this case in
the GPIO driver by probing the selected device if necessary.
Simon Glass [Mon, 15 Mar 2021 05:00:02 +0000 (18:00 +1300)]
x86: coral: Add information about building / running
Add detailed information on how to build the coral image, since it needs
binary blobs. Provide a way to avoid the memory-training delay. Also show
the console output from a sample run.
Tom Rini [Fri, 26 Mar 2021 16:15:26 +0000 (12:15 -0400)]
Merge tag 'dm-pull-26mar21-take2' of git://git.denx.de/u-boot-dm into next
dtoc support for of-platdata-inst
driver model support for of-platdata-inst
support of-platdata-inst on x86 / coral
binman support for exapanded entries
binman convert docs to reST
ti-sysc fix for duplicate uclass driver
patman minor improvements
pylibfdt build only if needed
correct obscure CI error with OF_PLATDATA_INST
Simon Glass [Thu, 18 Mar 2021 07:25:16 +0000 (20:25 +1300)]
binman: Drop repetitive heading for each entry
Many entries start 'Entry containing a'. This looks fine in the source
code but is annoying when viewed in the htmldocs table of contents. Drop
these unnecessary words.
Simon Glass [Thu, 18 Mar 2021 07:25:13 +0000 (20:25 +1300)]
binman: doc: Add documentation to htmldocs
Add a link to binman's documentation and adjust the files so that it is
accessible. Use the name README.rst so it is easy to discover when binman
is installed without U-Boot.
Simon Glass [Thu, 18 Mar 2021 07:25:09 +0000 (20:25 +1300)]
x86: Make use of binman expanded entries
We don't need to spell out the separate pieces of U-Boot phase binaries
anymore. Revert to using the simple entry and let binman do the expansion
itself as needed.
Simon Glass [Thu, 18 Mar 2021 07:25:08 +0000 (20:25 +1300)]
Makefile: Pass new entry args to binman
To support the use of 'expanded' entries, binman needs to be told whether
SPL and TPL have a devicetree and whether they need BSS padding. Add these
to the Makefile.
Simon Glass [Thu, 18 Mar 2021 07:25:06 +0000 (20:25 +1300)]
binman: Plumb expanded entries through fully
Add support for this feature in the control, image and section modules, so
that expanded entries will be selected by default. So far there are no
expanded entry types, so this is a nop.
Simon Glass [Thu, 18 Mar 2021 07:25:05 +0000 (20:25 +1300)]
binman: Allow a way to select expanded entries
Add a new command-line option to disable expanded entries. This is needed
for most tests, since it is much easier to 'factor out' this function into
a separate test and keep the existing packing tests simple.
Add the option and select it by default from tests.
Simon Glass [Thu, 18 Mar 2021 07:25:04 +0000 (20:25 +1300)]
binman: Allow using an an 'expanded' entry type
As the first step in supporting expanded entries, add a way for binman to
automatically select an 'expanded' version of an entry type, if requested.
This is controlled by a class method.
Simon Glass [Thu, 18 Mar 2021 07:25:03 +0000 (20:25 +1300)]
binman: Use standard filenames for SPL/TPL devicetree
At present, before any entry expansion is done (such as a 'files' entry
expanding out to individual entries for each file it contains), we check
the binman definition (i.e. '/binman' node) to find out what devicetree
files are used in the images.
This is a pain, since the definition may change during expansion. For
example if there is no u-boot-spl-dtb entry in the definition at the start,
we assume that the SPL devicetree is not used. But if an entry later
expands to include this, then we don't notice.
In fact the flexibility provided by the current approach of checking the
definition is not really useful. We know that we can have SPL and TPL
devicetrees. We know the pathname to each, so we can simply check if the
files are present. If they are present, we can prepare them and update
them regardless of whether they are actually used. If they are not present,
we cannot prepare/update them anyway, i.e. an error will be generated.
Simplify state.Prepare() so it uses a hard-coded list of devicetree files.
Note that state.PrepareFromLoadedData() is left untouched, since in that
case we have a complete definition from the loaded file, but cannot of
course rely on the devicetree files that created it still being present.
So in that case we still check the image defitions.
Simon Glass [Thu, 18 Mar 2021 07:25:01 +0000 (20:25 +1300)]
binman: Drop unnecessary field in output_fdt_info
At present we store an entry as the third field in output_fdt_info[].
This is only used to get the type of the entry. Of course multiple entries
may have this same type. Also the entry type is the key to this dict, so
we can use that instead.
Drop the field and update GetUpdateNodes() to suit. Improve the comment for
output_fdt_info a little while here.
Simon Glass [Thu, 18 Mar 2021 07:25:00 +0000 (20:25 +1300)]
binman: Use the fake SPL/TPL only if requested
At present we always use the main devicetree for SPL/TPL as well when
setting up the state. But this it not needed if there is a real devicetree
for SPL or TPL. In fact it confuses things since we cannot distinguish
between one being provided and using the fake one.
Update the code to create the fakes only when requested. Put the mapping
in a constant so we can use it elsewhere.
Rename 'other_fname' to 'fname' while we are here since there is nothing
'other' about it.
Simon Glass [Thu, 18 Mar 2021 07:24:58 +0000 (20:24 +1300)]
binman: Add support for u-boot-tpl-bss-bad
This entry holds the padding between the end of of TPL binary and the
end of BSS. This region must be left empty so that the devicetree can be
appended correctly and remain accessible without interfering with BSS.
Simon Glass [Thu, 18 Mar 2021 07:24:54 +0000 (20:24 +1300)]
binman: Tidy up underscores in entry documentation
Several entries currently use an underscore in the entry-type name, but in
fact a hyphen is used. Update the docs to fix this as it might be
confusing.
Also simplify the 'filename' comment and fix the 'operation' typo.
Simon Glass [Thu, 18 Mar 2021 07:24:51 +0000 (20:24 +1300)]
binman: Allow extracting to current directory
Extracting files to the current directory is not normally a very friendly
thing to do, but it can be warranted, e.g. in a new temporary dir. At
present binman reports an error when such an attempt is made. Fix it.
Simon Glass [Mon, 15 Mar 2021 04:25:51 +0000 (17:25 +1300)]
x86: coral: Drop TPM and ACPI interrupts from TPL
These devices are not actually built in TPL but are currently active in
the TPL devicetree. For of-platdata-inst this means that we will try to
generate devices for them, which fails.
Update them to be active only in U-Boot proper.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Simon Glass <sjg@chromium.org>