kbuild: clean up *.lex.c and *.tab.[ch] patterns from top-level Makefile ...

Linux commit 9a8dfb3

Files suffixed by .lex.c, .tab.[ch] are generated lexers, parsers,
respectively.  Clean them up globally from the top Makefile.

Some of the final host programs those lexer/parser are linked into
are necessary for building external modules, but the intermediates
are unneeded.  They can be cleaned away by 'make clean' instead of
'make mrproper'.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

.gitignore: move *.lex.c *.tab.[ch] patterns to the top-level .gitignore ...

Linux commit 5988930

These patterns are common to host programs that require lexer and parser.
Move them to the top .gitignore.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

kbuild: add %.lex.c and %.tab.[ch] to 'targets' automatically ...

Linux commit b23d1a2
(adjust commit log for barebox)

This will be needed for Barebox to switch to the compile-time
generation of lexers and parsers.

Files generated by if_changed* must be added to 'targets' to include
*.cmd files.  Otherwise, they would be regenerated every time.

The build system automatically adds objects to 'targets' where
appropriate, such as obj-y, extra-y, etc. but does nothing for
intermediate files.  So, each Makefile needs to add them by itself.

There are some common cases where objects are generated by chained
rules.  Lexers and parsers are compiled like follows:

   %.lex.o <- %.lex.c <- %.l
   %.tab.o <- %.tab.c <- %.y

They are common patterns, so it is reasonable to take care of them
in the core Makefile instead of requiring each Makefile to do so.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

kbuild: add .SECONDARY special target ...

Based on the following Linux commits:

 - 54a702f70589 ("kbuild: mark $(targets) as .SECONDARY and remove
   .PRECIOUS markers")

 - 8e9b61b293d9 ("kbuild: move .SECONDARY special target to
   Kbuild.include")

This consolidates those two commits and adjusts commit log for barebox.

GNU Make automatically deletes intermediate files that are updated
in a chain of pattern rules.

Example 1) %.dtb.o <- %.dtb.S <- %.dtb <- %.dts
Example 2) %.o <- %.c <- %.c_shipped

arch/arm/crypto/Makefile marks such targets as .PRECIOUS to prevent Make
from deleting them, but the correct way is to use .SECONDARY.

  .SECONDARY
    Prerequisites of this special target are treated as intermediate
    files but are never automatically deleted.

  .PRECIOUS
    When make is interrupted during execution, it may delete the target
    file it is updating if the file was modified since make started.
    If you mark the file as precious, make will never delete the file
    if interrupted.

Both can avoid deletion of intermediate files, but the difference is
the behavior when Make is interrupted; .SECONDARY deletes the target,
but .PRECIOUS does not.

The use of .PRECIOUS is relatively rare since we do not want to keep
partially constructed (possibly corrupted) targets.

.SECONDARY with no prerequisites causes all targets to be treated as
secondary. This agrees the policy of Kbuild.

scripts/Kbuild.include seems a suitable place to add it because it is
included from almost all sub-makes.

I deleted 'SECONDARY' (no dot) from images/Makefile, which I guess
a typo.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

kbuild: add .DELETE_ON_ERROR special target ...

Linux commit 9c2af1c

If Make gets a fatal signal while a shell is executing, it may delete
the target file that the recipe was supposed to update.  This is needed
to make sure that it is remade from scratch when Make is next run; if
Make is interrupted after the recipe has begun to write the target file,
it results in an incomplete file whose time stamp is newer than that
of the prerequisites files.  Make automatically deletes the incomplete
file on interrupt unless the target is marked .PRECIOUS.

The situation is just the same as when the shell fails for some reasons.
Usually when a recipe line fails, if it has changed the target file at
all, the file is corrupted, or at least it is not completely updated.
Yet the file’s time stamp says that it is now up to date, so the next
time Make runs, it will not try to update that file.

However, Make does not cater to delete the incomplete target file in
this case.  We need to add .DELETE_ON_ERROR somewhere in the Makefile
to request it.

scripts/Kbuild.include seems a suitable place to add it because it is
included from almost all sub-makes.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

kbuild: filter-out PHONY targets from "targets" ...

Linux commit 591f668

The variable "targets" contains object paths for which existing .*.cmd
files should be included.

scripts/Makefile.build automatically adds $(MAKECMDGOALS) to "targets"
as follows:

  targets += $(extra-y) $(MAKECMDGOALS) $(always)

The $(MAKECMDGOALS) is a PHONY target in several places.  PHONY targets
never create .*.cmd files.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

scripts/mkcompile_h: Remove trailing spaces from compiler version ...

This is an adoption of the linux kernel commit

| commit adcc3f7
| Author: Jonathan Liu <net147@gmail.com>
| Date:   Mon Jun 12 18:23:17 2017 +1000
|
|    scripts/mkcompile_h: Remove trailing spaces from compiler version
|
|    Improves the output of "cat /proc/version" by getting rid of the
|    trailing space at the end of the compiler version when the kernel
|    is compiled using GCC.
|
|    Signed-off-by: Jonathan Liu <net147@gmail.com>
|    Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>

Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/imx'

scripts: imx-image: fix build with OpenSSL < 1.1.0 ...

When adding support for OpenSSL 1.1.x it was missed that EVP_PKEY_get0_RSA
is only available with that version. Earlier versions don't provide a fully
equivalent function, so add it to the compatibility helpers.

Fixes: 542a50d4eb (scripts: imx-image: fix build with OpenSSL 1.1.x)
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

checkpatch.pl: check for #if 0/#if 1 ...

The #if 0 or #if 1 is used to toggle features. Warn if #if 0 or #if 1
is present and suggest that they can be removed.

Based on 60f890105547f7a4 ("checkpatch: check for #if 0/#if 1")
Linux kernel commit.

Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

checkpatch.pl: use SPDX tag ...

Also add Joe's copyright.

Based on Linux kernel commit 882ea1d64eb3956
("scripts: use SPDX tag in get_maintainer and checkpatch").

Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

checkpatch.pl: check for common memset parameter issues against statements ...

Based on these Linux kernel commits

> 309c00c73f053a90 checkpatch: warn about memset with swapped arguments
> 554e165cf32610ec checkpatch: check for common memset parameter issues against statments

Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

checkpatch.pl: add ability to ignore various messages ...

Based on commit 000d1cc1829f938c ("checkpatch.pl: add
ability to ignore various messages") from Linux kernel.

This simplifies current linux kernel checkpatch.pl commit
adoption for barebox due to conversion of
1-argument ERROR($1), WARN($1) and CHK($1) subs into
2-argument ERROR($1, $2), WARN($1, $2) and CHK($1, $2) subs.

Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

scripts: imx-image: Add support for max_load_size option ...

When an image is loaded to SRAM we can normally only load a part of the
full image, limited by the available space in SRAM. We currently load
the pblb image which is the executable part of the PBL without the
compressed barebox payload. We are going to link the compressed barebox
image into the pbl image though, so we can't see the desired load size
from outside the image anymore.

This adds a max_load_size option to set this size explicitly.

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

scripts: imx-image: Fix typos in comments ...

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

Merge branch 'for-next/imx'

scripts: imx-image: fix first instruction in i.MX8MQ header ...

The instruction being patched in was a branch to 0x24000, instead of
the intended 0x9000. This doesn't matter when starting the image from
a regular boot source, but breaks starting the image via bootm, as we
rely on the jump in the first header instruction taking us to the
actual code.

As the code offset changes depending on the presence of the signed
HDMI firmware, compute the offet dynamically while building the image.

Fixes: 6d2132a9ded7 (scripts: imx-image: Add i.MX8MQ support)
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

kconfig: Avoid format overflow warning from GCC 8.1 ...

Linux commit 2ae89c7 adapted for
barebox:

In file included from scripts/kconfig/zconf.tab.c:2485:
scripts/kconfig/confdata.c: In function ‘conf_write’:
scripts/kconfig/confdata.c:773:22: warning: ‘%s’ directive writing likely 7 or more bytes into a region of size between 1 and 4097 [-Wformat-overflow=]
  sprintf(newname, "%s%s", dirname, basename);
                      ^~
scripts/kconfig/confdata.c:773:19: note: assuming directive output of 7 bytes
  sprintf(newname, "%s%s", dirname, basename);
                   ^~~~~~
scripts/kconfig/confdata.c:773:2: note: ‘sprintf’ output 1 or more bytes (assuming 4104) into a destination of size 4097
  sprintf(newname, "%s%s", dirname, basename);
  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
scripts/kconfig/confdata.c:776:23: warning: ‘.tmpconfig.’ directive writing 11 bytes into a region of size between 1 and 4097 [-Wformat-overflow=]
   sprintf(tmpname, "%s.tmpconfig.%d", dirname, (int)getpid());
                       ^~~~~~~~~~~
scripts/kconfig/confdata.c:776:3: note: ‘sprintf’ output between 13 and 4119 bytes into a destination of size 4097
   sprintf(tmpname, "%s.tmpconfig.%d", dirname, (int)getpid());
   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Increase the size of tmpname and newname to make GCC happy.

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

scripts: imx-usb-loader: list_imx_device_types(): move variable declaration out of for() loop ...

On older compilers this causes the following error:
| scripts/imx/imx-usb-loader.c: In function 'list_imx_device_types':
| scripts/imx/imx-usb-loader.c:252:2: error: 'for' loop initial declarations are only allowed in C99 mode
|   for (int i = 0; i < ARRAY_SIZE(imx_ids); i++) {
|   ^
| scripts/imx/imx-usb-loader.c:252:2: note: use option -std=c99 or -std=gnu99 to compile your code

Fixes: bcc2df673cdb ("scripts: imx-usb-loader: allow use of unknown USB IDs")
Reported-by: Niklas Reisser <Niklas.Reisser@de.bosch.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/ratp'

checkpatch: don't check TODO file presence ...

The commit b3baf2d86149 ("TODO: remove unused file") drops
the TODO file. Alas scripts/checkpatch.pl checks TODO file
presence in top_of_kernel_tree(). Without TODO checkpatch.pl
exits with 'Must be run from the top-level dir. of a kernel tree'.

Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

bbremote: implement support for GPIO operations ...

Signed-off-by: Aleksander Morgado <aleksander@aleksander.es>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

bbremote: implement i2c read/write support ...

Extend the bbremote script with operations to perform binary i2c
read/write operations.

E.g.:

    barebox:/ i2c_read -b 0 -a 0x68 -r 0x06A0 -c 4 -w
    0x8f 0x30 0x00 0x00

    barebox:/ i2c_write -b 0 -a 0x68 -r 0x06A0 -w 0x87 0x30 0x00 0x00

    barebox:/ i2c_read -b 0 -a 0x68 -r 0x06A0 -c 4 -w
    0x87 0x30 0x00 0x00

    barebox:/ i2c_write -b 0 -a 0x68 -r 0x06A0 -w 0x8f 0x30 0x00 0x00

    barebox:/ i2c_read -b 0 -a 0x68 -r 0x06A0 -c 4 -w
    0x8f 0x30 0x00 0x00

    ===================================================================

    $ ./scripts/bbremote --port /dev/ttyUSB2 i2c-read 0x00 0x68 0x06A0 0x01 4
    8f300000

    $ ./scripts/bbremote --port /dev/ttyUSB2 i2c-write 0x00 0x68 0x06A0 0x01 "87300000"
    4 bytes written

    $ ./scripts/bbremote --port /dev/ttyUSB2 i2c-read 0x00 0x68 0x06A0 0x01 4
    87300000

    $ ./scripts/bbremote --port /dev/ttyUSB2 i2c-write 0x00 0x68 0x06A0 0x01 "8f300000"
    4 bytes written

    $ ./scripts/bbremote --port /dev/ttyUSB2 i2c-read 0x00 0x68 0x06A0 0x01 4
    8f300000

Signed-off-by: Aleksander Morgado <aleksander@aleksander.es>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/mxs'

Merge branch 'for-next/misc'

Merge branch 'for-next/imx-hab'

ARM: i.MX: make HAB certificate pathes explicitly overwritable ...

When building HAB images for i.MX we have to specify some pathes to
the certificates. This can be done with Kconfig variables. For better
build system integration we also want to be able to specify the pathes
in environment variables. This currently doesn't work as we specify
the variables from the environment with the -D option to cpp, but also
include generated/autoconf.h which overwrites the variables with the
values from Kconfig.

To overcome this introduce a Kconfig switch that explcitly selects
whether we want to have the variables from Kconfig or the environment.
Also, only pass the variables from the environment when explicitly
wanted.

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

scripts: imx-usb-loader: make i.MX6SoloX better to type ...

Now that users can select the device type on the command line, having a
name without whitespace makes input for that device type easier.

Signed-off-by: Roland Hieber <r.hieber@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

scripts: imx-usb-loader: allow use of unknown USB IDs ...

Some vendors fuse their devices so that the IMX USB ROM loader
identifies itself with a different Vendor and Product ID on USB
enumeration. Currently, imx-usb-loader will refuse to detect and work
with such devices, so let's teach it.

Because we cannot easily detect the device type from the USB ID in this
case, introduce the new command line parameter -d <type> to specify the
device type to use on the device path specified with -p <path>, even if
the VID/PID pair of that device is unknown. The device name is sourced
from the "name" field of the imx_ids array of known devices at the top
of the file. Using "-d list" will print a list of known device types.
Using -d without -p will not do anything useful, except generate a
warning.

Signed-off-by: Roland Hieber <r.hieber@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

scripts: imx-usb-loader: fail early if the USB device path does not match ...

find_imx_dev() loops through all USB devices, tries to open them, and
then compares the chosen device path (given with -p on the command line)
to the path of the currently opened device. The device path can be
checked earlier, opening the device is not neccessary.

We fail early here because in the next commit we want to enable the user
to force using a device by specifying its path. Opening every single
device available on the system then leads to unnecessary error messages
for all devices that do not match the provided path.

Signed-off-by: Roland Hieber <r.hieber@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>