Mirrors/u-boot
2
0
Fork 0
mirror of https://github.com/AsahiLinux/u-boot synced 2024-11-17 10:18:38 +00:00
Code Issues Projects Releases Packages Wiki Activity

Pull request for efi-2022-10-rc3

Browse source 
Documentation:
 
 * Add HTML documentation for patman
 * Improve binman documentation
 * Man-page for gpio
 
 UEFI:
 
 * move udevice pointer into struct efi_object
 * fix efi_convert_device_path_to_text()
 
 Other:
 
 * fs/erofs: silence messages from  erofs_probe()
 -----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCAAdFiEEbcT5xx8ppvoGt20zxIHbvCwFGsQFAmL3bF4ACgkQxIHbvCwF
 GsSeUQ/+MApJO7tESoo7ukZ150WGip3muq75SG2Xv0t67zx2n+shE2bs3xv+fGpR
 0Seqom8uo9s8jsSvhYWiE7GrNWcm7sIbAsdPL3678jJ34Fc+x8uy5jatfbCR+wep
 VwiUk8hPzIa+yjLLig/3obgNysQkfaMKxDjBf2J0Q3FdY0wC95lUf0n3x+ijzwSB
 mppxxnbJbVMtu0qMk7QIv7U/tA+uUB83JEVQA3X3MgpVL+NRb0TZsqwlvme/WGJ3
 ZMKv++tR8J7RkuEYq8yGDx+bMTUr2Ipmd0t2qMc4JwT7/3KODwstLlFPl1ODl26f
 hU64zCFR3NgIB3mxspUXUZVdc7OMKHfk+ZlzEh3uCKGG2fU6LzNnVkpfWuZIF7qN
 mtK+BQPyJ6xuOabW7bGrIitcMeub6KQqKPRNOq61Mwr1PfbpkVUQ8Rnk2fB16RDq
 0F87n8EzxYJV7caaCU2DedTPtZIKSeMWCNt4OA2NDczFpMErP9TjRfxtkTA1poFo
 0kyf8uMjSfkKHJ3VtK+VCr0NbvVRXqJZJ3lm5x4CwpMHFb25VidEr71WfRYhGpb0
 08Px8ueqp8tBaylltKFXwOym1NYl6dYQ7OQY71Tmb8V//JBY+P7Nw+h2Z6Xn6NPV
 Rb/f8UoAc3/C3a+GfAA7NdfInXWQ+dTfuQWDTk3CXpubIWTV2a0=
 =fm+u
 -----END PGP SIGNATURE-----

Merge tag 'efi-2022-10-rc3' of https://source.denx.de/u-boot/custodians/u-boot-efi

Pull request for efi-2022-10-rc3

Documentation:

* Add HTML documentation for patman
* Improve binman documentation
* Man-page for gpio

UEFI:

* move udevice pointer into struct efi_object
* fix efi_convert_device_path_to_text()

Other:

* fs/erofs: silence messages from  erofs_probe()
This commit is contained in:
Tom Rini 2022-08-13 07:37:48 -04:00
commit 20d4c6052f
24 changed files with 855 additions and 279 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
cmd/efidebug.c
View file

@ -77,7 +77,7 @@ static int do_efi_capsule_update(struct cmd_tbl *cmdtp, int flag,
ret = EFI_CALL(RT->update_capsule(&capsule, 1, 0));
if (ret) {
printf("Cannot handle a capsule at %p", capsule);
printf("Cannot handle a capsule at %p\n", capsule);
return CMD_RET_FAILURE;
}

1
doc/Makefile
View file

@ -56,6 +56,7 @@ quiet_cmd_sphinx = SPHINX $@ --> file://$(abspath $(BUILDDIR)/$3/$4)
PYTHONDONTWRITEBYTECODE=1 \
BUILDDIR=$(abspath $(BUILDDIR)) SPHINX_CONF=$(abspath $(srctree)/$(src)/$5/$(SPHINX_CONF)) \
$(SPHINXBUILD) \
-j$(shell nproc) \
-b $2 \
-c $(abspath $(srctree)/$(src)) \
-d $(abspath $(BUILDDIR)/.doctrees/$3) \

6
doc/develop/bootstd.rst
View file

@ -21,7 +21,7 @@ feature is typically called `distro boot` (see :doc:`distro`) because it is
a way for distributions to boot on any hardware.
Traditionally U-Boot has relied on scripts to implement this feature. See
disto_boodcmd_ for details. This is done because U-Boot has no native support
distro_bootcmd_ for details. This is done because U-Boot has no native support
for scanning devices. While the scripts work remarkably well, they can be hard
to understand and extend, and the feature does not include tests. They are also
making it difficult to move away from ad-hoc CONFIGs, since they are implemented
@ -52,7 +52,7 @@ BootLoaderSpec_ format. which looks something like this::
initrd /initramfs-5.3.7-301.fc31.armv7hl.img
As you can see it specifies a kernel, a ramdisk (initrd) and a directory from
which to load devicetree files. The details are described in disto_boodcmd_.
which to load devicetree files. The details are described in distro_bootcmd_.
The bootflow is provided by the distro. It is not part of U-Boot. U-Boot's job
is simply to interpret the file and carry out the instructions. This allows
@ -669,7 +669,7 @@ Other ideas:
not need to specify things like `pxefile_addr_r`
.. _disto_boodcmd: https://github.com/u-boot/u-boot/blob/master/include/config_distro_bootcmd.h
.. _distro_bootcmd: https://github.com/u-boot/u-boot/blob/master/include/config_distro_bootcmd.h
.. _BootLoaderSpec: http://www.freedesktop.org/wiki/Specifications/BootLoaderSpec/
.. _distro_boot: https://github.com/u-boot/u-boot/blob/master/boot/distro.c
.. _bootflow_h: https://github.com/u-boot/u-boot/blob/master/include/bootflow.h

4
doc/develop/codingstyle.rst
View file

@ -248,5 +248,5 @@ will be much faster (10-100x or more) if they can directly call run_command()
and ut_check_console_line() instead of using Python to send commands over a
pipe to U-Boot.
Tests run all supported CI systems (gitlab, travis, azure) using scripts in the
root of the U-Boot tree.
Tests run all supported CI systems (GitLab, Azure) using scripts in the root of
the U-Boot tree.

64
doc/develop/crash_dumps.rst
View file

@ -12,8 +12,8 @@ Creating a crash dump voluntarily
---------------------------------
For describing the analysis of a crash dump we need an example. U-Boot comes
with a command 'exception' that comes in handy here. The command is enabled
by::
with a command :doc:`exception <../usage/cmd/exception>` that comes in handy
here. The command is enabled by::
CONFIG_CMD_EXCEPTION=y
@ -122,3 +122,63 @@ If we want to dive deeper, we can disassemble the U-Boot binary::
This example is based on the ARMv8 architecture but the same procedures can be
used on other architectures as well.
Crashs in UEFI binaries
-----------------------
If UEFI images are loaded when a crash occurs, their load addresses are
displayed. If the process counter points to an address in a loaded UEFI
binary, the relative process counter position is indicated. Here is an
example executed on the U-Boot sandbox::
=> load host 0:1 $kernel_addr_r buggy.efi
5632 bytes read in 0 ms
=> bootefi $kernel_addr_r
Booting /buggy.efi
Buggy world!
Segmentation violation
pc = 0x19fc264c, pc_reloc = 0xffffaa4688b1664c
UEFI image [0x0000000019fc0000:0x0000000019fc6137] pc=0x264c '/buggy.efi'
The crash occured in UEFI binary buggy.efi at relative position 0x264c.
Disassembly may be used to find the actual source code location::
$ x86_64-linux-gnu-objdump -S -D buggy_efi.so
0000000000002640 <memset>:
2640: f3 0f 1e fa endbr64
2644: 48 89 f8 mov %rdi,%rax
2647: 48 89 f9 mov %rdi,%rcx
264a: eb 0b jmp 2657 <memset+0x17>
264c: 40 88 31 mov %sil,(%rcx)
Architecture specific details
-----------------------------
ARMv8
~~~~~
On the ARM 64-bit architecture CONFIG_ARMV8_SPL_EXCEPTION_VECTORS controls
if the exception vector tables are set up in the Secondary Program Loader (SPL).
Without initialization of the tables crash dumps cannot be shown. The feature is
disabled by default on most boards to reduce the size of the SPL.
RISC-V
~~~~~~
On the RISC-V architecture CONFIG_SHOW_REGS=y has to be specified to show
all registers in crash dumps.
Sandbox
~~~~~~~
The sandbox U-Boot binary must be invoked with parameter *-S* to display crash
dumps:
.. code-block:: bash
./u-boot -S -T
Only with CONFIG_SANDBOX_CRASH_RESET=y the sandbox reboots after a crash.

1
doc/develop/index.rst
View file

@ -14,6 +14,7 @@ General
process
release_cycle
system_configuration
sending_patches
Implementation
--------------

1
doc/develop/patman.rst Symbolic link
View file

@ -0,0 +1 @@
../../tools/patman/patman.rst

4
doc/develop/process.rst
View file

@ -139,6 +139,10 @@ like this:
as the Linux kernel. Following this and adding a ``Signed-off-by:`` line
that contains the developer's name and email address is required.
* Please note that when importing code from other projects you must say
where it comes from, and what revision you are importing. You must not
however copy ``Signed-off-by`` or other tags.
#. Everybody who can is invited to review and test the changes. Typically, we
follow the same guidelines as the Linux kernel for `Acked-by
<https://www.kernel.org/doc/html/latest/process/submitting-patches.html#when-to-use-acked-by-cc-and-co-developed-by>`_

16
doc/develop/sending_patches.rst Normal file
View file

@ -0,0 +1,16 @@
.. SPDX-License-Identifier: GPL-2.0+
Sending patches
===============
.. toctree::
:maxdepth: 2
patman
You can use a tool called patman to prepare, check and sent patches. It creates
change logs, cover letters and patch notes. It also simplified the process of
sending multiple versions of a series.
See more details at :doc:`patman`.

1
doc/git-mailrc
View file

@ -111,7 +111,6 @@ alias x86 uboot, sjg, bmeng
alias dm uboot, sjg
alias cfi uboot, stroese
alias dfu uboot, lukma
alias efi uboot, agraf
alias eth uboot, jhersh
alias kerneldoc uboot, marex
alias fdt uboot, sjg

90
doc/usage/cmd/gpio.rst Normal file
View file

@ -0,0 +1,90 @@
.. SPDX-License-Identifier: GPL-2.0+:
gpio command
============
Synopsis
--------
::
gpio <input|set|clear|toggle> <pin>
gpio read <name> <pin>
gpio status [-a] [<bank>|<pin>]
The gpio command is used to access General Purpose Inputs/Outputs.
gpio input
----------
Switch the GPIO *pin* to input mode.
gpio set
--------
Switch the GPIO *pin* to output mode and set the signal to 1.
gpio clear
----------
Switch the GPIO *pin* to output mode and set the signal to 0.
gpio toggle
-----------
Switch the GPIO *pin* to output mode and reverse the signal state.
gpio read
---------
Read the signal state of the GPIO *pin* and save it in environment variable
*name*.
gpio status
-----------
Display the status of one or multiple GPIOs. By default only claimed GPIOs
are displayed.
-a
Display GPIOs irrespective of being claimed.
bank
Name of a bank of GPIOs to be displayed.
pin
Name of a single GPIO to be displayed or manipulated.
Examples
--------
Switch the status of a GPIO::
=> gpio set a5
gpio: pin a5 (gpio 133) value is 1
=> gpio clear a5
gpio: pin a5 (gpio 133) value is 0
=> gpio toggle a5
gpio: pin a5 (gpio 133) value is 1
=> gpio read myvar a5
gpio: pin a5 (gpio 133) value is 1
=> echo $myvar
1
=> gpio toggle a5
gpio: pin a5 (gpio 133) value is 0
=> gpio read myvar a5
gpio: pin a5 (gpio 133) value is 0
=> echo $myvar
0
Configuration
-------------
The *gpio* command is only available if CONFIG_CMD_GPIO=y.
The *gpio read* command is only available if CONFIG_CMD_GPIO_READ=y.
Return value
------------
If the command succeds the return value $? is set to 0. If an error occurs, the
return value $? is set to 1.

1
doc/usage/index.rst
View file

@ -45,6 +45,7 @@ Shell commands
cmd/fatload
cmd/fdt
cmd/for
cmd/gpio
cmd/load
cmd/loadm
cmd/loady

4
fs/erofs/super.c
View file

@ -65,14 +65,14 @@ int erofs_read_superblock(void)
ret = erofs_blk_read(data, 0, 1);
if (ret < 0) {
erofs_err("cannot read erofs superblock: %d", ret);
erofs_dbg("cannot read erofs superblock: %d", ret);
return -EIO;
}
dsb = (struct erofs_super_block *)(data + EROFS_SUPER_OFFSET);
ret = -EINVAL;
if (le32_to_cpu(dsb->magic) != EROFS_SUPER_MAGIC_V1) {
erofs_err("cannot find valid erofs superblock");
erofs_dbg("cannot find valid erofs superblock");
return ret;
}

4
include/efi_loader.h
View file

@ -375,6 +375,7 @@ enum efi_object_type {
* @protocols: linked list with the protocol interfaces installed on this
* handle
* @type: image type if the handle relates to an image
* @dev: pointer to the DM device which is associated with this EFI handle
*
* UEFI offers a flexible and expandable object model. The objects in the UEFI
* API are devices, drivers, and loaded images. struct efi_object is our storage
@ -392,6 +393,7 @@ struct efi_object {
/* The list of protocols */
struct list_head protocols;
enum efi_object_type type;
struct udevice *dev;
};
enum efi_image_auth_status {
@ -690,6 +692,8 @@ struct efi_device_path *efi_get_dp_from_boot(const efi_guid_t guid);
const char *guid_to_sha_str(const efi_guid_t *guid);
int algo_to_len(const char *algo);
int efi_link_dev(efi_handle_t handle, struct udevice *dev);
/**
* efi_size_in_pages() - convert size in bytes to size in pages
*

3
lib/efi_driver/efi_block_device.c
View file

@ -158,8 +158,7 @@ static int efi_bl_bind(efi_handle_t handle, void *interface)
* FIXME: necessary because we won't do almost nothing in
* efi_disk_create() when called from device_probe().
*/
ret = dev_tag_set_ptr(bdev, DM_TAG_EFI, handle);
if (ret)
if (efi_link_dev(handle, bdev))
/* FIXME: cleanup for bdev */
return ret;

1
lib/efi_loader/efi_device_path_to_text.c
View file

@ -432,6 +432,7 @@ static uint16_t EFIAPI *efi_convert_device_path_to_text(
*(u8 **)&device_path += device_path->length;
}
*str = 0;
text = efi_str_to_u16(buffer);
out:

15
lib/efi_loader/efi_disk.c
View file

@ -46,7 +46,6 @@ struct efi_disk_obj {
struct efi_device_path *dp;
unsigned int part;
struct efi_simple_file_system_protocol *volume;
struct udevice *dev; /* TODO: move it to efi_object */
};
/**
@ -124,16 +123,16 @@ static efi_status_t efi_disk_rw_blocks(struct efi_block_io *this,
return EFI_BAD_BUFFER_SIZE;
if (CONFIG_IS_ENABLED(PARTITIONS) &&
device_get_uclass_id(diskobj->dev) == UCLASS_PARTITION) {
device_get_uclass_id(diskobj->header.dev) == UCLASS_PARTITION) {
if (direction == EFI_DISK_READ)
n = dev_read(diskobj->dev, lba, blocks, buffer);
n = dev_read(diskobj->header.dev, lba, blocks, buffer);
else
n = dev_write(diskobj->dev, lba, blocks, buffer);
n = dev_write(diskobj->header.dev, lba, blocks, buffer);
} else {
/* dev is a block device (UCLASS_BLK) */
struct blk_desc *desc;
desc = dev_get_uclass_plat(diskobj->dev);
desc = dev_get_uclass_plat(diskobj->header.dev);
if (direction == EFI_DISK_READ)
n = blk_dread(desc, lba, blocks, buffer);
else
@ -552,8 +551,7 @@ static int efi_disk_create_raw(struct udevice *dev)
return -1;
}
disk->dev = dev;
if (dev_tag_set_ptr(dev, DM_TAG_EFI, &disk->header)) {
if (efi_link_dev(&disk->header, dev)) {
efi_free_pool(disk->dp);
efi_delete_handle(&disk->header);
@ -609,8 +607,7 @@ static int efi_disk_create_part(struct udevice *dev)
log_err("Adding partition for %s failed\n", dev->name);
return -1;
}
disk->dev = dev;
if (dev_tag_set_ptr(dev, DM_TAG_EFI, &disk->header)) {
if (efi_link_dev(&disk->header, dev)) {
efi_free_pool(disk->dp);
efi_delete_handle(&disk->header);

13
lib/efi_loader/efi_helper.c
View file

@ -158,3 +158,16 @@ int algo_to_len(const char *algo)
return 0;
}
/** efi_link_dev - link the efi_handle_t and udevice
*
* @handle: efi handle to associate with udevice
* @dev: udevice to associate with efi handle
*
* Return: 0 on success, negative on failure
*/
int efi_link_dev(efi_handle_t handle, struct udevice *dev)
{
handle->dev = dev;
return dev_tag_set_ptr(dev, DM_TAG_EFI, handle);
}

200
tools/binman/binman.rst
View file

@ -42,7 +42,7 @@ the devicetree description of the image.
Binman is designed primarily for use with U-Boot and associated binaries such
as ARM Trusted Firmware, but it is suitable for use with other projects, such
as Zephyr. Binman also provides facilities useful in Chromium OS, such as CBFS,
vblocks and and the like.
vblocks and the like.
Binman provides a way to process binaries before they are included, by adding a
Python plug-in.
@ -118,6 +118,10 @@ flash.
For U-Boot, binman should not be used to create ad-hoc images in place of
FIT.
Note that binman can itself create a FIT. This helps to move mkimage
invocations out of the Makefile and into binman image descriptions. It also
helps by removing the need for ad-hoc tools like `make_fit_atf.py`.
Relationship to mkimage
-----------------------
@ -140,6 +144,9 @@ seems better to use the mkimage tool to generate binaries and avoid blurring
the boundaries between building input files (mkimage) and packaging then
into a final image (binman).
Note that binman can itself invoke mkimage. This helps to move mkimage
invocations out of the Makefile and into binman image descriptions.
Using binman
============
@ -170,6 +177,164 @@ This simplifies the U-Boot Makefile somewhat, since various pieces of logic
can be replaced by a call to binman.
Invoking binman within U-Boot
-----------------------------
Within U-Boot, binman is invoked by the build system, i.e. when you type 'make'
or use buildman to build U-Boot. There is no need to run binman independently
during development. Everything happens automatically and is set up for your
SoC or board so that binman produced the right things.
The general policy is that the Makefile builds all the binaries in INPUTS-y
(the 'inputs' rule), then binman is run to produce the final images (the 'all'
rule).
There should be only one invocation of binman in Makefile, the very last step
that pulls everything together. At present there are some arch-specific
invocations as well, but these should be dropped when those architectures are
converted to use binman properly.
As above, the term 'binary' is used for something in INPUTS-y and 'image' is
used for the things that binman creates. So the binaries are inputs to the
image(s) and it is the image that is actually loaded on the board.
Again, at present, there are a number of things created in Makefile which should
be done by binman (when we get around to it), like `u-boot-ivt.img`,
`lpc32xx-spl.img`, `u-boot-with-nand-spl.imx`, `u-boot-spl-padx4.sfp` and
`u-boot-mtk.bin`, just to pick on a few. When completed this will remove about
400 lines from `Makefile`.
Since binman is invoked only once, it must of course create all the images that
are needed, in that one invocation. It does this by working through the image
descriptions one by one, collecting the input binaries, processing them as
needed and producing the final images.
The same binaries may be used by multiple images. For example binman may be used
to produce an SD-card image and a SPI-flash image. In this case the binaries
going into the process are the same, but binman produces slightly different
images in each case.
For some SoCs, U-Boot is not the only project that produces the necessary
binaries. For example, ARM Trusted Firmware (ATF) is a project that produces
binaries which must be incorporate, such as `bl31.elf` or `bl31.bin`. For this
to work you must have built ATF before you build U-Boot and you must tell U-Boot
where to find the bl31 image, using the BL31 environment variable.
How do you know how to incorporate ATF? It is handled by the atf-bl31 entry type
(etype). An etype is an implementation of reading a binary into binman, in this
case the `bl31.bin` file. When you build U-Boot but do not set the BL31
environment variable, binman provides a help message, which comes from
`missing-blob-help`::
See the documentation for your board. You may need to build ARM Trusted
Firmware and build with BL31=/path/to/bl31.bin
The mechanism by which binman is advised of this is also in the Makefile. See
the `-a atf-bl31-path=${BL31}` piece in `cmd_binman`. This tells binman to
set the EntryArg `atf-bl31-path` to the value of the `BL31` environment
variable. Within binman, this EntryArg is picked up by the `Entry_atf_bl31`
etype. An EntryArg is simply an argument to the entry. The `atf-bl31-path`
name is documented in :ref:`etype_atf_bl31`.
Invoking binman outside U-Boot
------------------------------
While binman is invoked from within the U-Boot build system, it is also possible
to invoke it separately. This is typically used in a production build system,
where signing is completed (with real keys) and any missing binaries are
provided.
For example, for build testing there is no need to provide a real signature,
nor is there any need to provide a real ATF BL31 binary (for example). These can
be added later by invoking binman again, providing all the required inputs
from the first time, plus any that were missing or placeholders.
So in practice binman is often used twice:
- once within the U-Boot build system, for development and testing
- again outside U-Boot to assembly and final production images
While the same input binaries are used in each case, you will of course you will
need to create your own binman command line, similar to that in `cmd_binman` in
the Makefile. You may find the -I and --toolpath options useful. The
device tree file is provided to binman in binary form, so there is no need to
have access to the original `.dts` sources.
Assembling the image description
--------------------------------
Since binman uses the device tree for its image description, you can use the
same files that describe your board's hardware to describe how the image is
assembled. Typically the images description is in a common file used by all
boards with a particular SoC (e.g. `imx8mp-u-boot.dtsi`).
Where a particular boards needs to make changes, it can override properties in
the SoC file, just as it would for any other device tree property. It can also
add a image that is specific to the board.
Another way to control the image description to make use of CONFIG options in
the description. For example, if the start offset of a particular entry varies
by board, you can add a Kconfig for that and reference it in the description::
u-boot-spl {
};
fit {
offset = <CONFIG_SPL_PAD_TO>;
...
};
The SoC can provide a default value but boards can override that as needed and
binman will take care of it.
It is even possible to control which entries appear in the image, by using the
C preprocessor::
#ifdef CONFIG_HAVE_MRC
intel-mrc {
offset = <CONFIG_X86_MRC_ADDR>;
};
#endif
Only boards which enable `HAVE_MRC` will include this entry.
Obviously a similar approach can be used to control which images are produced,
with a Kconfig option to enable a SPI image, for example. However there is
generally no harm in producing an image that is not used. If a board uses MMC
but not SPI, but the SoC supports booting from both, then both images can be
produced, with only on or other being used by particular boards. This can help
reduce the need for having multiple defconfig targets for a board where the
only difference is the boot media, enabling / disabling secure boot, etc.
Of course you can use the device tree itself to pass any board-specific
information that is needed by U-Boot at runtime (see binman_syms_ for how to
make binman insert these values directly into executables like SPL).
There is one more way this can be done: with individual .dtsi files for each
image supported by the SoC. Then the board `.dts` file can include the ones it
wants. This is not recommended, since it is likely to be difficult to maintain
and harder to understand the relationship between the different boards.
Producing images for multiple boards
------------------------------------
When invoked within U-Boot, binman only builds a single set of images, for
the chosen board. This is set by the `CONFIG_DEFAULT_DEVICE_TREE` option.
However, U-Boot generally builds all the device tree files associated with an
SoC. These are written to the (e.g. for ARM) `arch/arm/dts` directory. Each of
these contains the full binman description for that board. Often the best
approach is to build a single image that includes all these device tree binaries
and allow SPL to select the correct one on boot.
However, it is also possible to build separate images for each board, simply by
invoking binman multiple times, once for each device tree file, using a
different output directory. This will produce one set of images for each board.
Example use of binman for x86
-----------------------------
@ -188,19 +353,25 @@ the configuration of the Intel-format descriptor.
Installing binman
-----------------
First install prerequisites, e.g::
First install prerequisites, e.g:
.. code-block:: bash
sudo apt-get install python-pyelftools python3-pyelftools lzma-alone \
liblz4-tool
You can run binman directly if you put it on your PATH. But if you want to
install into your `~/.local` Python directory, use::
install into your `~/.local` Python directory, use:
.. code-block:: bash
pip install tools/patman tools/dtoc tools/binman
Note that binman makes use of libraries from patman and dtoc, which is why these
need to be installed. Also you need `libfdt` and `pylibfdt` which can be
installed like this::
installed like this:
.. code-block:: bash
git clone git://git.kernel.org/pub/scm/utils/dtc/dtc.git
cd dtc
@ -209,7 +380,9 @@ installed like this::
This installs the `libfdt.so` library into `~/lib` so you can use
`LD_LIBRARY_PATH=~/lib` when running binman. If you want to install it in the
system-library directory, replace the last line with::
system-library directory, replace the last line with:
.. code-block:: bash
make NO_PYTHON=1 PREFIX=/ install
@ -218,14 +391,20 @@ Running binman
Type::
.. code-block: bash
make NO_PYTHON=1 PREFIX=/ install
binman build -b <board_name>
to build an image for a board. The board name is the same name used when
configuring U-Boot (e.g. for sandbox_defconfig the board name is 'sandbox').
Binman assumes that the input files for the build are in ../b/<board_name>.
Or you can specify this explicitly::
Or you can specify this explicitly:
.. code-block:: bash
make NO_PYTHON=1 PREFIX=/ install
binman build -I <build_path>
where <build_path> is the build directory containing the output of the U-Boot
@ -254,6 +433,7 @@ file, typically <soc>-u-boot.dtsi, where <soc> is your CONFIG_SYS_SOC value.
You can use other, more specific CONFIG options - see 'Automatic .dtsi
inclusion' below.
.. _binman_syms:
Access to binman entry offsets at run time (symbols)
----------------------------------------------------
@ -264,13 +444,17 @@ is useful to be able to find the location of U-Boot so that it can be executed
when SPL is finished.
Binman allows you to declare symbols in the SPL image which are filled in
with their correct values during the build. For example::
with their correct values during the build. For example:
.. code-block:: c
binman_sym_declare(ulong, u_boot_any, image_pos);
declares a ulong value which will be assigned to the image-pos of any U-Boot
image (u-boot.bin, u-boot.img, u-boot-nodtb.bin) that is present in the image.
You can access this value with something like::
You can access this value with something like:
.. code-block:: c
ulong u_boot_offset = binman_sym(ulong, u_boot_any, image_pos);

146
tools/binman/entries.rst
View file

@ -11,6 +11,8 @@ features to produce new behaviours.
.. _etype_atf_bl31:
Entry: atf-bl31: ARM Trusted Firmware (ATF) BL31 blob
-----------------------------------------------------
@ -25,6 +27,8 @@ about ATF.
.. _etype_atf_fip:
Entry: atf-fip: ARM Trusted Firmware's Firmware Image Package (FIP)
-------------------------------------------------------------------
@ -179,6 +183,8 @@ FIPs so that binman and other tools can access the entire image correctly.
.. _etype_blob:
Entry: blob: Arbitrary binary blob
----------------------------------
@ -201,6 +207,8 @@ data.
.. _etype_blob_dtb:
Entry: blob-dtb: A blob that holds a device tree
------------------------------------------------
@ -210,6 +218,8 @@ obtained from the list of available device-tree files, managed by the
.. _etype_blob_ext:
Entry: blob-ext: Externally built binary blob
---------------------------------------------
@ -223,6 +233,8 @@ See 'blob' for Properties / Entry arguments.
.. _etype_blob_ext_list:
Entry: blob-ext-list: List of externally built binary blobs
-----------------------------------------------------------
@ -237,6 +249,8 @@ Args:
.. _etype_blob_named_by_arg:
Entry: blob-named-by-arg: A blob entry which gets its filename property from its subclass
-----------------------------------------------------------------------------------------
@ -255,6 +269,8 @@ See cros_ec_rw for an example of this.
.. _etype_blob_phase:
Entry: blob-phase: Section that holds a phase binary
----------------------------------------------------
@ -264,6 +280,8 @@ entry; similarly for SPL.
.. _etype_cbfs:
Entry: cbfs: Coreboot Filesystem (CBFS)
---------------------------------------
@ -416,6 +434,8 @@ both of size 1MB.
.. _etype_collection:
Entry: collection: An entry which contains a collection of other entries
------------------------------------------------------------------------
@ -429,6 +449,8 @@ the image, not necessarily child entries.
.. _etype_cros_ec_rw:
Entry: cros-ec-rw: A blob entry which contains a Chromium OS read-write EC image
--------------------------------------------------------------------------------
@ -440,6 +462,8 @@ updating the EC on startup via software sync.
.. _etype_fdtmap:
Entry: fdtmap: An entry which contains an FDT map
-------------------------------------------------
@ -488,6 +512,8 @@ without the fdtmap header, so it can be viewed with `fdtdump`.
.. _etype_files:
Entry: files: A set of files arranged in a section
--------------------------------------------------
@ -504,6 +530,8 @@ at run-time so you can obtain the file positions.
.. _etype_fill:
Entry: fill: An entry which is filled to a particular byte value
----------------------------------------------------------------
@ -520,6 +548,8 @@ byte value of a region.
.. _etype_fit:
Entry: fit: Flat Image Tree (FIT)
---------------------------------
@ -803,6 +833,8 @@ U-Boot SPL can then load the firmware (U-Boot proper) and all the loadables
.. _etype_fmap:
Entry: fmap: An entry which contains an Fmap section
----------------------------------------------------
@ -828,6 +860,8 @@ CBFS entries appear as a single entry, i.e. the sub-entries are ignored.
.. _etype_gbb:
Entry: gbb: An entry which contains a Chromium OS Google Binary Block
---------------------------------------------------------------------
@ -847,6 +881,8 @@ README.chromium for how to obtain the required keys and tools.
.. _etype_image_header:
Entry: image-header: An entry which contains a pointer to the FDT map
---------------------------------------------------------------------
@ -866,6 +902,8 @@ first/last in the entry list.
.. _etype_intel_cmc:
Entry: intel-cmc: Intel Chipset Micro Code (CMC) file
-----------------------------------------------------
@ -879,6 +917,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_descriptor:
Entry: intel-descriptor: Intel flash descriptor block (4KB)
-----------------------------------------------------------
@ -900,6 +940,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_fit:
Entry: intel-fit: Intel Firmware Image Table (FIT)
--------------------------------------------------
@ -911,6 +953,8 @@ At present binman only supports a basic FIT with no microcode.
.. _etype_intel_fit_ptr:
Entry: intel-fit-ptr: Intel Firmware Image Table (FIT) pointer
--------------------------------------------------------------
@ -919,6 +963,8 @@ This entry contains a pointer to the FIT. It is required to be at address
.. _etype_intel_fsp:
Entry: intel-fsp: Intel Firmware Support Package (FSP) file
-----------------------------------------------------------
@ -936,6 +982,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_fsp_m:
Entry: intel-fsp-m: Intel Firmware Support Package (FSP) memory init
--------------------------------------------------------------------
@ -953,6 +1001,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_fsp_s:
Entry: intel-fsp-s: Intel Firmware Support Package (FSP) silicon init
---------------------------------------------------------------------
@ -970,6 +1020,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_fsp_t:
Entry: intel-fsp-t: Intel Firmware Support Package (FSP) temp ram init
----------------------------------------------------------------------
@ -986,6 +1038,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_ifwi:
Entry: intel-ifwi: Intel Integrated Firmware Image (IFWI) file
--------------------------------------------------------------
@ -1020,6 +1074,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_me:
Entry: intel-me: Intel Management Engine (ME) file
--------------------------------------------------
@ -1040,6 +1096,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_mrc:
Entry: intel-mrc: Intel Memory Reference Code (MRC) file
--------------------------------------------------------
@ -1054,6 +1112,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_refcode:
Entry: intel-refcode: Intel Reference Code file
-----------------------------------------------
@ -1068,6 +1128,8 @@ See README.x86 for information about x86 binary blobs.
.. _etype_intel_vbt:
Entry: intel-vbt: Intel Video BIOS Table (VBT) file
---------------------------------------------------
@ -1081,6 +1143,8 @@ See README.x86 for information about Intel binary blobs.
.. _etype_intel_vga:
Entry: intel-vga: Intel Video Graphics Adaptor (VGA) file
---------------------------------------------------------
@ -1096,6 +1160,8 @@ See README.x86 for information about Intel binary blobs.
.. _etype_mkimage:
Entry: mkimage: Binary produced by mkimage
------------------------------------------
@ -1130,6 +1196,8 @@ this example which also produces four arguments::
.. _etype_opensbi:
Entry: opensbi: RISC-V OpenSBI fw_dynamic blob
----------------------------------------------
@ -1143,6 +1211,8 @@ https://github.com/riscv/opensbi for more information about OpenSBI.
.. _etype_powerpc_mpc85xx_bootpg_resetvec:
Entry: powerpc-mpc85xx-bootpg-resetvec: PowerPC mpc85xx bootpg + resetvec code for U-Boot
-----------------------------------------------------------------------------------------
@ -1155,11 +1225,13 @@ placed at offset 'RESET_VECTOR_ADDRESS - 0xffc'.
.. _etype_pre_load:
Entry: pre-load: Pre load image header
--------------------------------------
Properties / Entry arguments:
- key-path: Path of the directory that store key (provided by the environment variable KEY_PATH)
- pre-load-key-path: Path of the directory that store key (provided by the environment variable PRE_LOAD_KEY_PATH)
- content: List of phandles to entries to sign
- algo-name: Hash and signature algo to use for the signature
- padding-name: Name of the padding (pkcs-1.5 or pss)
@ -1193,6 +1265,8 @@ For example, this creates an image with a pre-load header and a binary::
.. _etype_scp:
Entry: scp: System Control Processor (SCP) firmware blob
--------------------------------------------------------
@ -1203,6 +1277,8 @@ This entry holds firmware for an external platform-specific coprocessor.
.. _etype_section:
Entry: section: Entry that contains other entries
-------------------------------------------------
@ -1338,6 +1414,8 @@ available. This is set by the `SetAllowMissing()` method, if
.. _etype_tee_os:
Entry: tee-os: Entry containing an OP-TEE Trusted OS (TEE) blob
---------------------------------------------------------------
@ -1351,6 +1429,8 @@ https://github.com/OP-TEE/optee_os for more information about OP-TEE.
.. _etype_text:
Entry: text: An entry which contains text
-----------------------------------------
@ -1399,6 +1479,8 @@ by setting the size of the entry to something larger than the text.
.. _etype_u_boot:
Entry: u-boot: U-Boot flat binary
---------------------------------
@ -1420,6 +1502,8 @@ Note that this entry is automatically replaced with u-boot-expanded unless
.. _etype_u_boot_dtb:
Entry: u-boot-dtb: U-Boot device tree
-------------------------------------
@ -1435,6 +1519,8 @@ binman to know which entries contain a device tree.
.. _etype_u_boot_dtb_with_ucode:
Entry: u-boot-dtb-with-ucode: A U-Boot device tree file, with the microcode removed
-----------------------------------------------------------------------------------
@ -1451,6 +1537,8 @@ it available to u-boot-ucode.
.. _etype_u_boot_elf:
Entry: u-boot-elf: U-Boot ELF image
-----------------------------------
@ -1462,6 +1550,8 @@ relocated to any address for execution.
.. _etype_u_boot_env:
Entry: u-boot-env: An entry which contains a U-Boot environment
---------------------------------------------------------------
@ -1471,6 +1561,8 @@ Properties / Entry arguments:
.. _etype_u_boot_expanded:
Entry: u-boot-expanded: U-Boot flat binary broken out into its component parts
------------------------------------------------------------------------------
@ -1486,6 +1578,8 @@ image, so that the entries positions are provided to the running U-Boot.
.. _etype_u_boot_img:
Entry: u-boot-img: U-Boot legacy image
--------------------------------------
@ -1500,6 +1594,8 @@ applications.
.. _etype_u_boot_nodtb:
Entry: u-boot-nodtb: U-Boot flat binary without device tree appended
--------------------------------------------------------------------
@ -1514,6 +1610,8 @@ section containing u-boot and u-boot-dtb
.. _etype_u_boot_spl:
Entry: u-boot-spl: U-Boot SPL binary
------------------------------------
@ -1541,6 +1639,8 @@ unless --no-expanded is used or the node has a 'no-expanded' property.
.. _etype_u_boot_spl_bss_pad:
Entry: u-boot-spl-bss-pad: U-Boot SPL binary padded with a BSS region
---------------------------------------------------------------------
@ -1563,6 +1663,8 @@ binman uses that to look up the BSS address.
.. _etype_u_boot_spl_dtb:
Entry: u-boot-spl-dtb: U-Boot SPL device tree
---------------------------------------------
@ -1575,6 +1677,8 @@ to activate.
.. _etype_u_boot_spl_elf:
Entry: u-boot-spl-elf: U-Boot SPL ELF image
-------------------------------------------
@ -1586,6 +1690,8 @@ be relocated to any address for execution.
.. _etype_u_boot_spl_expanded:
Entry: u-boot-spl-expanded: U-Boot SPL flat binary broken out into its component parts
--------------------------------------------------------------------------------------
@ -1609,6 +1715,8 @@ this is non-empty (and not 'n' or '0') then this expanded entry is selected.
.. _etype_u_boot_spl_nodtb:
Entry: u-boot-spl-nodtb: SPL binary without device tree appended
----------------------------------------------------------------
@ -1633,6 +1741,8 @@ binman uses that to look up symbols to write into the SPL binary.
.. _etype_u_boot_spl_with_ucode_ptr:
Entry: u-boot-spl-with-ucode-ptr: U-Boot SPL with embedded microcode pointer
----------------------------------------------------------------------------
@ -1643,6 +1753,8 @@ process.
.. _etype_u_boot_tpl:
Entry: u-boot-tpl: U-Boot TPL binary
------------------------------------
@ -1670,6 +1782,8 @@ unless --no-expanded is used or the node has a 'no-expanded' property.
.. _etype_u_boot_tpl_bss_pad:
Entry: u-boot-tpl-bss-pad: U-Boot TPL binary padded with a BSS region
---------------------------------------------------------------------
@ -1692,6 +1806,8 @@ binman uses that to look up the BSS address.
.. _etype_u_boot_tpl_dtb:
Entry: u-boot-tpl-dtb: U-Boot TPL device tree
---------------------------------------------
@ -1704,6 +1820,8 @@ to activate.
.. _etype_u_boot_tpl_dtb_with_ucode:
Entry: u-boot-tpl-dtb-with-ucode: U-Boot TPL with embedded microcode pointer
----------------------------------------------------------------------------
@ -1714,6 +1832,8 @@ process.
.. _etype_u_boot_tpl_elf:
Entry: u-boot-tpl-elf: U-Boot TPL ELF image
-------------------------------------------
@ -1725,6 +1845,8 @@ be relocated to any address for execution.
.. _etype_u_boot_tpl_expanded:
Entry: u-boot-tpl-expanded: U-Boot TPL flat binary broken out into its component parts
--------------------------------------------------------------------------------------
@ -1748,6 +1870,8 @@ this is non-empty (and not 'n' or '0') then this expanded entry is selected.
.. _etype_u_boot_tpl_nodtb:
Entry: u-boot-tpl-nodtb: TPL binary without device tree appended
----------------------------------------------------------------
@ -1772,6 +1896,8 @@ binman uses that to look up symbols to write into the TPL binary.
.. _etype_u_boot_tpl_with_ucode_ptr:
Entry: u-boot-tpl-with-ucode-ptr: U-Boot TPL with embedded microcode pointer
----------------------------------------------------------------------------
@ -1780,6 +1906,8 @@ process.
.. _etype_u_boot_ucode:
Entry: u-boot-ucode: U-Boot microcode block
-------------------------------------------
@ -1830,6 +1958,8 @@ Entry types that have a part to play in handling microcode:
.. _etype_u_boot_with_ucode_ptr:
Entry: u-boot-with-ucode-ptr: U-Boot with embedded microcode pointer
--------------------------------------------------------------------
@ -1846,6 +1976,8 @@ complicated. Otherwise it is the same as the u-boot entry.
.. _etype_vblock:
Entry: vblock: An entry which contains a Chromium OS verified boot block
------------------------------------------------------------------------
@ -1869,6 +2001,8 @@ and kernel are genuine.
.. _etype_x86_reset16:
Entry: x86-reset16: x86 16-bit reset code for U-Boot
----------------------------------------------------
@ -1885,6 +2019,8 @@ For 64-bit U-Boot, the 'x86_reset16_spl' entry type is used instead.
.. _etype_x86_reset16_spl:
Entry: x86-reset16-spl: x86 16-bit reset code for U-Boot
--------------------------------------------------------
@ -1901,6 +2037,8 @@ For 32-bit U-Boot, the 'x86_reset_spl' entry type is used instead.
.. _etype_x86_reset16_tpl:
Entry: x86-reset16-tpl: x86 16-bit reset code for U-Boot
--------------------------------------------------------
@ -1917,6 +2055,8 @@ For 32-bit U-Boot, the 'x86_reset_tpl' entry type is used instead.
.. _etype_x86_start16:
Entry: x86-start16: x86 16-bit start-up code for U-Boot
-------------------------------------------------------
@ -1935,6 +2075,8 @@ For 64-bit U-Boot, the 'x86_start16_spl' entry type is used instead.
.. _etype_x86_start16_spl:
Entry: x86-start16-spl: x86 16-bit start-up code for SPL
--------------------------------------------------------
@ -1953,6 +2095,8 @@ For 32-bit U-Boot, the 'x86-start16' entry type is used instead.
.. _etype_x86_start16_tpl:
Entry: x86-start16-tpl: x86 16-bit start-up code for TPL
--------------------------------------------------------

5
tools/binman/entry.py
View file

@ -750,6 +750,11 @@ features to produce new behaviours.
first_line = lines[0]
rest = [line[4:] for line in lines[1:]]
hdr = 'Entry: %s: %s' % (name.replace('_', '-'), first_line)
# Create a reference for use by rST docs
ref_name = f'etype_{module.__name__[6:]}'.lower()
print('.. _%s:' % ref_name)
print()
print(hdr)
print('-' * len(hdr))
print('\n'.join(rest))

1
tools/patman/README.rst Symbolic link
View file

@ -0,0 +1 @@
patman.rst

3
tools/patman/main.py
View file

@ -164,7 +164,8 @@ elif args.cmd == 'send':
elif args.full_help:
tools.print_full_help(
os.path.join(os.path.dirname(os.path.realpath(sys.argv[0])), 'README')
os.path.join(os.path.dirname(os.path.realpath(sys.argv[0])),
'README.rst')
)
else:

548
tools/patman/README → tools/patman/patman.rst
View file

@ -1,19 +1,31 @@
# SPDX-License-Identifier: GPL-2.0+
# Copyright (c) 2011 The Chromium OS Authors.
.. SPDX-License-Identifier: GPL-2.0+
.. Copyright (c) 2011 The Chromium OS Authors
.. Simon Glass <sjg@chromium.org>
.. v1, v2, 19-Oct-11
.. revised v3 24-Nov-11
.. revised v4 04-Jul-2020, with Patchwork integration
What is this?
=============
Patman patch manager
====================
This tool is a Python script which:
- Creates patch directly from your branch
- Cleans them up by removing unwanted tags
- Inserts a cover letter with change lists
- Runs the patches through checkpatch.pl and its own checks
- Optionally emails them out to selected people
It also has some Patchwork features:
- shows review tags from Patchwork so you can update your local patches
- pulls these down into a new branch on request
- lists comments received on a series
It is intended to automate patch creation and make it a less
@ -24,9 +36,9 @@ It is configured almost entirely by tags it finds in your commits.
This means that you can work on a number of different branches at
once, and keep the settings with each branch rather than having to
git format-patch, git send-email, etc. with the correct parameters
each time. So for example if you put:
each time. So for example if you put::
Series-to: fred.blogs@napier.co.nz
Series-to: fred.blogs@napier.co.nz
in one of your commits, the series will be sent there.
@ -35,30 +47,33 @@ patches automatically (unless you use -m to disable this).
How to use this tool
====================
--------------------
This tool requires a certain way of working:
- Maintain a number of branches, one for each patch series you are
working on
working on
- Add tags into the commits within each branch to indicate where the
series should be sent, cover letter, version, etc. Most of these are
normally in the top commit so it is easy to change them with 'git
commit --amend'
series should be sent, cover letter, version, etc. Most of these are
normally in the top commit so it is easy to change them with 'git
commit --amend'
- Each branch tracks the upstream branch, so that this script can
automatically determine the number of commits in it (optional)
automatically determine the number of commits in it (optional)
- Check out a branch, and run this script to create and send out your
patches. Weeks later, change the patches and repeat, knowing that you
will get a consistent result each time.
patches. Weeks later, change the patches and repeat, knowing that you
will get a consistent result each time.
How to configure it
===================
-------------------
For most cases of using patman for U-Boot development, patman can use the
file 'doc/git-mailrc' in your U-Boot directory to supply the email aliases
you need. To make this work, tell git where to find the file by typing
this once:
this once::
git config sendemail.aliasesfile doc/git-mailrc
@ -68,19 +83,16 @@ out where to send patches pretty well.
During the first run patman creates a config file for you by taking the default
user name and email address from the global .gitconfig file.
To add your own, create a file ~/.patman like this:
To add your own, create a file ~/.patman like this::
>>>>
# patman alias file
# patman alias file
[alias]
me: Simon Glass <sjg@chromium.org>
[alias]
me: Simon Glass <sjg@chromium.org>
u-boot: U-Boot Mailing List <u-boot@lists.denx.de>
wolfgang: Wolfgang Denk <wd@denx.de>
others: Mike Frysinger <vapier@gentoo.org>, Fred Bloggs <f.bloggs@napier.net>
<<<<
u-boot: U-Boot Mailing List <u-boot@lists.denx.de>
wolfgang: Wolfgang Denk <wd@denx.de>
others: Mike Frysinger <vapier@gentoo.org>, Fred Bloggs <f.bloggs@napier.net>
Aliases are recursive.
@ -90,263 +102,281 @@ used. Failing that you can put it into your path or ~/bin/checkpatch.pl
If you want to avoid sending patches to email addresses that are picked up
by patman but are known to bounce you can add a [bounces] section to your
.patman file. Unlike the [alias] section these are simple key: value pairs
that are not recursive.
that are not recursive::
>>>
[bounces]
gonefishing: Fred Bloggs <f.bloggs@napier.net>
<<<
[bounces]
gonefishing: Fred Bloggs <f.bloggs@napier.net>
If you want to change the defaults for patman's command-line arguments,
you can add a [settings] section to your .patman file. This can be used
for any command line option by referring to the "dest" for the option in
patman.py. For reference, the useful ones (at the moment) shown below
(all with the non-default setting):
>>>
[settings]
ignore_errors: True
process_tags: False
verbose: True
smtp_server: /path/to/sendmail
patchwork_server: https://patchwork.ozlabs.org
<<<
(all with the non-default setting)::
[settings]
ignore_errors: True
process_tags: False
verbose: True
smtp_server: /path/to/sendmail
patchwork_server: https://patchwork.ozlabs.org
If you want to adjust settings (or aliases) that affect just a single
project you can add a section that looks like [project_settings] or
[project_alias]. If you want to use tags for your linux work, you could
do:
[project_alias]. If you want to use tags for your linux work, you could do::
>>>
[linux_settings]
process_tags: True
<<<
[linux_settings]
process_tags: True
How to run it
=============
-------------
First do a dry run:
$ ./tools/patman/patman send -n
.. code-block:: bash
./tools/patman/patman send -n
If it can't detect the upstream branch, try telling it how many patches
there are in your series:
there are in your series
$ ./tools/patman/patman -c5 send -n
.. code-block:: bash
./tools/patman/patman -c5 send -n
This will create patch files in your current directory and tell you who
it is thinking of sending them to. Take a look at the patch files.
it is thinking of sending them to. Take a look at the patch files:
$ ./tools/patman/patman -c5 -s1 send -n
.. code-block:: bash
./tools/patman/patman -c5 -s1 send -n
Similar to the above, but skip the first commit and take the next 5. This
is useful if your top commit is for setting up testing.
How to install it
=================
-----------------
The most up to date version of patman can be found in the U-Boot sources.
However to use it on other projects it may be more convenient to install it as
a standalone application. A distutils installer is included, this can be used
to install patman:
$ cd tools/patman && python setup.py install
.. code-block:: bash
cd tools/patman && python setup.py install
How to add tags
===============
---------------
To make this script useful you must add tags like the following into any
commit. Most can only appear once in the whole series.
Series-to: email / alias
Email address / alias to send patch series to (you can add this
multiple times)
Email address / alias to send patch series to (you can add this
multiple times)
Series-cc: email / alias, ...
Email address / alias to Cc patch series to (you can add this
multiple times)
Email address / alias to Cc patch series to (you can add this
multiple times)
Series-version: n
Sets the version number of this patch series
Sets the version number of this patch series
Series-prefix: prefix
Sets the subject prefix. Normally empty but it can be RFC for
RFC patches, or RESEND if you are being ignored. The patch subject
is like [RFC PATCH] or [RESEND PATCH].
In the meantime, git format.subjectprefix option will be added as
well. If your format.subjectprefix is set to InternalProject, then
the patch shows like: [InternalProject][RFC/RESEND PATCH]
Sets the subject prefix. Normally empty but it can be RFC for
RFC patches, or RESEND if you are being ignored. The patch subject
is like [RFC PATCH] or [RESEND PATCH].
In the meantime, git format.subjectprefix option will be added as
well. If your format.subjectprefix is set to InternalProject, then
the patch shows like: [InternalProject][RFC/RESEND PATCH]
Series-postfix: postfix
Sets the subject "postfix". Normally empty, but can be the name of a
tree such as net or net-next if that needs to be specified. The patch
subject is like [PATCH net] or [PATCH net-next].
Sets the subject "postfix". Normally empty, but can be the name of a
tree such as net or net-next if that needs to be specified. The patch
subject is like [PATCH net] or [PATCH net-next].
Series-name: name
Sets the name of the series. You don't need to have a name, and
patman does not yet use it, but it is convenient to put the branch
name here to help you keep track of multiple upstreaming efforts.
Sets the name of the series. You don't need to have a name, and
patman does not yet use it, but it is convenient to put the branch
name here to help you keep track of multiple upstreaming efforts.
Series-links: [id | version:id]...
Set the ID of the series in patchwork. You can set this after you send
out the series and look in patchwork for the resulting series. The
URL you want is the one for the series itself, not any particular patch.
E.g. for http://patchwork.ozlabs.org/project/uboot/list/?series=187331
the series ID is 187331. This property can have a list of series IDs,
one for each version of the series, e.g.
Set the ID of the series in patchwork. You can set this after you send
out the series and look in patchwork for the resulting series. The
URL you want is the one for the series itself, not any particular patch.
E.g. for http://patchwork.ozlabs.org/project/uboot/list/?series=187331
the series ID is 187331. This property can have a list of series IDs,
one for each version of the series, e.g.
Series-links: 1:187331 2:188434 189372
::
Patman always uses the one without a version, since it assumes this is
the latest one. When this tag is provided, patman can compare your local
branch against patchwork to see what new reviews your series has
collected ('patman status').
Series-links: 1:187331 2:188434 189372
Patman always uses the one without a version, since it assumes this is
the latest one. When this tag is provided, patman can compare your local
branch against patchwork to see what new reviews your series has
collected ('patman status').
Series-patchwork-url: url
This allows specifying the Patchwork URL for a branch. This overrides
both the setting files and the command-line argument. The URL should
include the protocol and web site, with no trailing slash, for example
'https://patchwork.ozlabs.org/project'
This allows specifying the Patchwork URL for a branch. This overrides
both the setting files and the command-line argument. The URL should
include the protocol and web site, with no trailing slash, for example
'https://patchwork.ozlabs.org/project'
Cover-letter:
This is the patch set title
blah blah
more blah blah
END
Sets the cover letter contents for the series. The first line
will become the subject of the cover letter
Sets the cover letter contents for the series. The first line
will become the subject of the cover letter::
Cover-letter:
This is the patch set title
blah blah
more blah blah
END
Cover-letter-cc: email / alias
Additional email addresses / aliases to send cover letter to (you
can add this multiple times)
Additional email addresses / aliases to send cover letter to (you
can add this multiple times)
Series-notes:
blah blah
blah blah
more blah blah
END
Sets some notes for the patch series, which you don't want in
the commit messages, but do want to send, The notes are joined
together and put after the cover letter. Can appear multiple
times.
Sets some notes for the patch series, which you don't want in
the commit messages, but do want to send, The notes are joined
together and put after the cover letter. Can appear multiple
times::
Series-notes:
blah blah
blah blah
more blah blah
END
Commit-notes:
blah blah
blah blah
more blah blah
END
Similar, but for a single commit (patch). These notes will appear
immediately below the --- cut in the patch file.
Similar, but for a single commit (patch). These notes will appear
immediately below the --- cut in the patch file::
Signed-off-by: Their Name <email>
A sign-off is added automatically to your patches (this is
probably a bug). If you put this tag in your patches, it will
override the default signoff that patman automatically adds.
Multiple duplicate signoffs will be removed.
Commit-notes:
blah blah
blah blah
more blah blah
Tested-by: Their Name <email>
Reviewed-by: Their Name <email>
Acked-by: Their Name <email>
These indicate that someone has tested/reviewed/acked your patch.
When you get this reply on the mailing list, you can add this
tag to the relevant commit and the script will include it when
you send out the next version. If 'Tested-by:' is set to
yourself, it will be removed. No one will believe you.
Signed-off-by: Their Name <email>
A sign-off is added automatically to your patches (this is
probably a bug). If you put this tag in your patches, it will
override the default signoff that patman automatically adds.
Multiple duplicate signoffs will be removed.
Tested-by / Reviewed-by / Acked-by
These indicate that someone has tested/reviewed/acked your patch.
When you get this reply on the mailing list, you can add this
tag to the relevant commit and the script will include it when
you send out the next version. If 'Tested-by:' is set to
yourself, it will be removed. No one will believe you.
Example::
Tested-by: Their Name <fred@bloggs.com>
Reviewed-by: Their Name <email>
Acked-by: Their Name <email>
Series-changes: n
- Guinea pig moved into its cage
- Other changes ending with a blank line
<blank line>
This can appear in any commit. It lists the changes for a
particular version n of that commit. The change list is
created based on this information. Each commit gets its own
change list and also the whole thing is repeated in the cover
letter (where duplicate change lines are merged).
This can appear in any commit. It lists the changes for a
particular version n of that commit. The change list is
created based on this information. Each commit gets its own
change list and also the whole thing is repeated in the cover
letter (where duplicate change lines are merged).
By adding your change lists into your commits it is easier to
keep track of what happened. When you amend a commit, remember
to update the log there and then, knowing that the script will
do the rest.
By adding your change lists into your commits it is easier to
keep track of what happened. When you amend a commit, remember
to update the log there and then, knowing that the script will
do the rest.
Example::
Series-changes: n
- Guinea pig moved into its cage
- Other changes ending with a blank line
<blank line>
Commit-changes: n
- This line will not appear in the cover-letter changelog
<blank line>
This tag is like Series-changes, except changes in this changelog will
only appear in the changelog of the commit this tag is in. This is
useful when you want to add notes which may not make sense in the cover
letter. For example, you can have short changes such as "New" or
"Lint".
This tag is like Series-changes, except changes in this changelog will
only appear in the changelog of the commit this tag is in. This is
useful when you want to add notes which may not make sense in the cover
letter. For example, you can have short changes such as "New" or
"Lint".
Example::
Commit-changes: n
- This line will not appear in the cover-letter changelog
<blank line>
Cover-changes: n
- This line will only appear in the cover letter
<blank line>
This tag is like Series-changes, except changes in this changelog will
only appear in the cover-letter changelog. This is useful to summarize
changes made with Commit-changes, or to add additional context to
changes.
This tag is like Series-changes, except changes in this changelog will
only appear in the cover-letter changelog. This is useful to summarize
changes made with Commit-changes, or to add additional context to
changes.
Example::
Cover-changes: n
- This line will only appear in the cover letter
<blank line>
Patch-cc: Their Name <email>
This copies a single patch to another email address. Note that the
Cc: used by git send-email is ignored by patman, but will be
interpreted by git send-email if you use it.
This copies a single patch to another email address. Note that the
Cc: used by git send-email is ignored by patman, but will be
interpreted by git send-email if you use it.
Series-process-log: sort, uniq
This tells patman to sort and/or uniq the change logs. Changes may be
multiple lines long, as long as each subsequent line of a change begins
with a whitespace character. For example,
This tells patman to sort and/or uniq the change logs. Changes may be
multiple lines long, as long as each subsequent line of a change begins
with a whitespace character. For example,
- This change
continues onto the next line
- But this change is separate
Example::
Use 'sort' to sort the entries, and 'uniq' to include only
unique entries. If omitted, no change log processing is done.
Separate each tag with a comma.
- This change
continues onto the next line
- But this change is separate
Use 'sort' to sort the entries, and 'uniq' to include only
unique entries. If omitted, no change log processing is done.
Separate each tag with a comma.
Change-Id:
This tag is stripped out but is used to generate the Message-Id
of the emails that will be sent. When you keep the Change-Id the
same you are asserting that this is a slightly different version
(but logically the same patch) as other patches that have been
sent out with the same Change-Id.
This tag is stripped out but is used to generate the Message-Id
of the emails that will be sent. When you keep the Change-Id the
same you are asserting that this is a slightly different version
(but logically the same patch) as other patches that have been
sent out with the same Change-Id.
Various other tags are silently removed, like these Chrome OS and
Gerrit tags:
Gerrit tags::
BUG=...
TEST=...
Review URL:
Reviewed-on:
Commit-xxxx: (except Commit-notes)
BUG=...
TEST=...
Review URL:
Reviewed-on:
Commit-xxxx: (except Commit-notes)
Exercise for the reader: Try adding some tags to one of your current
patch series and see how the patches turn out.
Where Patches Are Sent
======================
----------------------
Once the patches are created, patman sends them using git send-email. The
whole series is sent to the recipients in Series-to: and Series-cc.
You can Cc individual patches to other people with the Patch-cc: tag. Tags
in the subject are also picked up to Cc patches. For example, a commit like
this:
this::
>>>>
commit 10212537b85ff9b6e09c82045127522c0f0db981
Author: Mike Frysinger <vapier@gentoo.org>
Date: Mon Nov 7 23:18:44 2011 -0500
commit 10212537b85ff9b6e09c82045127522c0f0db981
Author: Mike Frysinger <vapier@gentoo.org>
Date: Mon Nov 7 23:18:44 2011 -0500
x86: arm: add a git mailrc file for maintainers
@ -354,46 +384,47 @@ Date: Mon Nov 7 23:18:44 2011 -0500
Patch-cc: sandbox, mikef, ag
Patch-cc: afleming
<<<<
will create a patch which is copied to x86, arm, sandbox, mikef, ag and
afleming.
If you have a cover letter it will get sent to the union of the Patch-cc
lists of all of the other patches. If you want to sent it to additional
people you can add a tag:
people you can add a tag::
Cover-letter-cc: <list of addresses>
Cover-letter-cc: <list of addresses>
These people will get the cover letter even if they are not on the To/Cc
list for any of the patches.
Patchwork Integration
=====================
---------------------
Patman has a very basic integration with Patchwork. If you point patman to
your series on patchwork it can show you what new reviews have appears since
your series on patchwork it can show you what new reviews have appeared since
you sent your series.
To set this up, add a Series-link tag to one of the commits in your series
(see above).
Then you can type
Then you can type:
.. code-block:: bash
patman status
and patman will show you each patch and what review tags have been collected,
for example:
for example::
...
21 x86: mtrr: Update the command to use the new mtrr
Reviewed-by: Wolfgang Wallner <wolfgang.wallner@br-automation.com>
+ Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
22 x86: mtrr: Restructure so command execution is in
Reviewed-by: Wolfgang Wallner <wolfgang.wallner@br-automation.com>
+ Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
...
...
21 x86: mtrr: Update the command to use the new mtrr
Reviewed-by: Wolfgang Wallner <wolfgang.wallner@br-automation.com>
+ Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
22 x86: mtrr: Restructure so command execution is in
Reviewed-by: Wolfgang Wallner <wolfgang.wallner@br-automation.com>
+ Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
...
This shows that patch 21 and 22 were sent out with one review but have since
attracted another review each. If the series needs changes, you can update
@ -402,6 +433,8 @@ series.
To automatically pull into these tags into a new branch, use the -d option:
.. code-block:: bash
patman status -d mtrr4
This will create a new 'mtrr4' branch which is the same as your current branch
@ -409,6 +442,8 @@ but has the new review tags in it. The tags are added in alphabetic order and
are placed immediately after any existing ack/review/test/fixes tags, or at the
end. You can check that this worked with:
.. code-block:: bash
patman -b mtrr4 status
which should show that there are no new responses compared to this new branch.
@ -417,7 +452,7 @@ There is also a -C option to list the comments received for each patch.
Example Work Flow
=================
-----------------
The basic workflow is to create your commits, add some tags to the top
commit, and type 'patman' to check and send them.
@ -425,7 +460,7 @@ commit, and type 'patman' to check and send them.
Here is an example workflow for a series of 4 patches. Let's say you have
these rather contrived patches in the following order in branch us-cmd in
your tree where 'us' means your upstreaming activity (newest to oldest as
output by git log --oneline):
output by git log --oneline)::
7c7909c wip
89234f5 Don't include standard parser if hush is used
@ -438,36 +473,46 @@ but that you don't want to submit because there is an existing patch for it
on the list. So you can tell patman to create and check some patches
(skipping the first patch) with:
.. code-block:: bash
patman -s1 send -n
If you want to do all of them including the work-in-progress one, then
(if you are tracking an upstream branch):
.. code-block:: bash
patman send -n
Let's say that patman reports an error in the second patch. Then:
.. code-block:: bash
git rebase -i HEAD~6
<change 'pick' to 'edit' in 89234f5>
<use editor to make code changes>
# change 'pick' to 'edit' in 89234f5
# use editor to make code changes
git add -u
git rebase --continue
Now you have an updated patch series. To check it:
.. code-block:: bash
patman -s1 send -n
Let's say it is now clean and you want to send it. Now you need to set up
the destination. So amend the top commit with:
.. code-block:: bash
git commit --amend
Use your editor to add some tags, so that the whole commit message is:
Use your editor to add some tags, so that the whole commit message is::
The current run_command() is really only one of the options, with
hush providing the other. It really shouldn't be called directly
in case the hush parser is bring used, so rename this function to
better explain its purpose.
better explain its purpose::
Series-to: u-boot
Series-cc: bfin, marex
@ -490,6 +535,8 @@ mmc and sparc, and the last one to sandbox.
Now to send the patches, take off the -n flag:
.. code-block:: bash
patman -s1 send
The patches will be created, shown in your editor, and then sent along with
@ -502,36 +549,42 @@ Also, the patch on the list that you were waiting for has been merged,
so you can drop your wip commit.
Take a look on patchwork and find out the URL of the series. This will be
something like http://patchwork.ozlabs.org/project/uboot/list/?series=187331
Add this to a tag in your top commit:
something like `http://patchwork.ozlabs.org/project/uboot/list/?series=187331`
Add this to a tag in your top commit::
Series-link: http://patchwork.ozlabs.org/project/uboot/list/?series=187331
Series-links: 187331
You can use then patman to collect the Acked-by tag to the correct commit,
creating a new 'version 2' branch for us-cmd:
.. code-block:: bash
patman status -d us-cmd2
git checkout us-cmd2
You can look at the comments in Patchwork or with:
.. code-block:: bash
patman status -C
Then you can resync with upstream:
git fetch origin (or whatever upstream is called)
.. code-block:: bash
git fetch origin # or whatever upstream is called
git rebase origin/master
and use git rebase -i to edit the commits, dropping the wip one.
Then update the Series-cc: in the top commit to add the person who reviewed
the v1 series:
Then update the `Series-cc:` in the top commit to add the person who reviewed
the v1 series::
Series-cc: bfin, marex, Heiko Schocher <hs@denx.de>
and remove the Series-prefix: tag since it it isn't an RFC any more. The
series is now version two, so the series info in the top commit looks like
this:
this::
Series-to: u-boot
Series-cc: bfin, marex, Heiko Schocher <hs@denx.de>
@ -541,7 +594,7 @@ this:
Finally, you need to add a change log to the two commits you changed. You
add change logs to each individual commit where the changes happened, like
this:
this::
Series-changes: 2
- Updated the command decoder to reduce code size
@ -551,7 +604,7 @@ this:
When you run patman it will collect all the change logs from the different
commits and combine them into the cover letter, if you have one. So finally
you have a new series of commits:
you have a new series of commits::
faeb973 Don't include standard parser if hush is used
1b2f2fe mmc: sparc: Stop using builtin_run_command()
@ -560,59 +613,63 @@ you have a new series of commits:
so to send them:
.. code-block:: bash
patman
and it will create and send the version 2 series.
General points
==============
--------------
1. When you change back to the us-cmd branch days or weeks later all your
information is still there, safely stored in the commits. You don't need
to remember what version you are up to, who you sent the last lot of patches
to, or anything about the change logs.
information is still there, safely stored in the commits. You don't need
to remember what version you are up to, who you sent the last lot of patches
to, or anything about the change logs.
2. If you put tags in the subject, patman will Cc the maintainers
automatically in many cases.
automatically in many cases.
3. If you want to keep the commits from each series you sent so that you can
compare change and see what you did, you can either create a new branch for
each version, or just tag the branch before you start changing it:
compare change and see what you did, you can either create a new branch for
each version, or just tag the branch before you start changing it:
git tag sent/us-cmd-rfc
...later...
git tag sent/us-cmd-v2
.. code-block:: bash
git tag sent/us-cmd-rfc
# ...later...
git tag sent/us-cmd-v2
4. If you want to modify the patches a little before sending, you can do
this in your editor, but be careful!
this in your editor, but be careful!
5. If you want to run git send-email yourself, use the -n flag which will
print out the command line patman would have used.
print out the command line patman would have used.
6. It is a good idea to add the change log info as you change the commit,
not later when you can't remember which patch you changed. You can always
go back and change or remove logs from commits.
not later when you can't remember which patch you changed. You can always
go back and change or remove logs from commits.
7. Some mailing lists have size limits and when we add binary contents to
our patches it's easy to exceed the size limits. Use "--no-binary" to
generate patches without any binary contents. You are supposed to include
a link to a git repository in your "Commit-notes", "Series-notes" or
"Cover-letter" for maintainers to fetch the original commit.
our patches it's easy to exceed the size limits. Use "--no-binary" to
generate patches without any binary contents. You are supposed to include
a link to a git repository in your "Commit-notes", "Series-notes" or
"Cover-letter" for maintainers to fetch the original commit.
8. Patches will have no changelog entries for revisions where they did not
change. For clarity, if there are no changes for this patch in the most
recent revision of the series, a note will be added. For example, a patch
with the following tags in the commit
change. For clarity, if there are no changes for this patch in the most
recent revision of the series, a note will be added. For example, a patch
with the following tags in the commit::
Series-version: 5
Series-changes: 2
- Some change
Series-version: 5
Series-changes: 2
- Some change
Series-changes: 4
- Another change
Series-changes: 4
- Another change
would have a changelog of
would have a changelog of:::
(no changes since v4)
@ -622,8 +679,9 @@ would have a changelog of
Changes in v2:
- Some change
Other thoughts
==============
--------------
This script has been split into sensible files but still needs work.
Most of these are indicated by a TODO in the code.
@ -633,6 +691,8 @@ It would be nice if this could handle the In-reply-to side of things.
The tests are incomplete, as is customary. Use the 'test' subcommand to run
them:
.. code-block:: bash
$ tools/patman/patman test
Error handling doesn't always produce friendly error messages - e.g.
@ -641,9 +701,3 @@ putting an incorrect tag in a commit may provide a confusing message.
There might be a few other features not mentioned in this README. They
might be bugs. In particular, tags are case sensitive which is probably
a bad thing.
Simon Glass <sjg@chromium.org>
v1, v2, 19-Oct-11
revised v3 24-Nov-11
revised v4 Independence Day 2020, with Patchwork integration