According to fsync specification [1] some special files (e.g., a pipe, FIFO,
or socket) don't support synchronization and return either EROFS or EINVAL.
On the linux side the sys_fsync -> do_fsync() checks if the requested file
has f_op->fsync defined. If not it returns EINVAL [2].
This commit prevents writing error messages for files (devices), which
do not support fsync().
[1] - http://man7.org/linux/man-pages/man2/fsync.2.html
[2] - http://elixir.free-electrons.com/linux/v4.13-rc6/source/fs/sync.c#L183
Signed-off-by: Lukasz Majewski <lukma@denx.de>
Acked-by: Michael Heimpold <mhei@heimpold.de>
We are now using an env_ prefix for environment functions. Rename setenv()
for consistency. Also add function comments in common.h.
Suggested-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Simon Glass <sjg@chromium.org>
Closing a file descriptor does not guarantee that the data has been
successfully saved to disk, as the kernel might defer the write.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
If realloc fails we should release the old buffer.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
fw_env_open allocates buffers to store the environment, but these
buffers are never freed. This becomes quite nasty using the fw_ tools as
library, because each access to the environment (even just reading a
variable) generates a memory leak equal to the size of the environment.
Fix this renaming fw_env_close() as fw_env_flush(), because the function
really flushes the environment from RAM to storage, and add a
fw_env_close function to free the allocated resources.
Signed-off-by: Stefano Babic <sbabic@denx.de>
Changes in the environment library are difficult to tracked by programs
using the library. Add simply an API version number that must be
increased each time when the API is changed.
This can be detected and a program can work with different versions of
the library.
Signed-off-by: Stefano Babic <sbabic@denx.de>
Move U-Boot private data into a separate file. This
lets export fw_env.h to be used by external programs
that want to change the environment using the library
built in tools/env.
Signed-off-by: Stefano Babic <sbabic@denx.de>
aes.h is a too generic name if this file can
be exported and used by a program.
Rename it to avoid any conflicts with
other files (for example, from openSSL).
Signed-off-by: Stefano Babic <sbabic@denx.de>
Richard reported U-Boot tools issues in OpenEmbedded/Yocto project.
OE needs to be able to change the default compiler. If we pass in
HOSTCC through the make command, it overwrites all HOSTCC instances,
including ones in tools/Makefile and tools/env/Makefile, which breaks
"make cross_tools" and "make env", respectively.
Add "override" directives to avoid overriding HOSTCC instances that
really need to point to the cross-compiler.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reported-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
commit 183923d3e4 enforces that the
environment must start at an erase block boundary.
For block devices the sample fw_env.config does not mandate a erase block size
for block devices. A missing setting defaults to the full env size.
Depending on the environment location the alignment check now errors out for
perfectly legal settings.
Fix this by defaulting to the standard blocksize of 0x200 for environments
stored in a block device.
That keeps the fw_env.config files for block devices working even with that
new check.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com>
The default lockname is set to /var/lock. This limits the
usage of this application where OS uses different lockfile
location parameter.
For example, In case of android, the default lock
path location is /data.
Hence by providing the command line option to input lockfile
path will be useful to reuse the tool across multiple
operating system.
usage: ./fw_printenv -l <lockfile path>
Signed-off-by: Ravi Babu <ravibabu@ti.com>
This allows to take advantage of the environment being block aligned.
This is not a new constraint. Writes always start at the begin of the
environment, since the header with CRC/length as there.
Every environment modification requires updating the header
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
flash_write_buf already looks up size/offset/#sector from struct
envdev_s. It can look up mtd_type as well. Same applies to
flash_read_buf. Makes the interface simpler
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
the offset is not modified by linux ioctl call
see mtd_ioctl{drivers/mtd/mtdchar.c}
Makes the interface less ambiguous, since the caller can
now exclude a modification of blockstart
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
instead of adhoc computation of the environment end,
use a function with a proper name
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
addon 183923d3e
MMC/SATA have no erase blocks, only blocks. Hence the warning
about erase block alignment might be confusing in such environment.
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
For double buffering to work, the target buffer must always be big
enough to hold all data. This can only be ensured if buffers are of
equal size, otherwise one must be smaller and we risk data loss
when copying from the bigger to the smaller buffer.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
56086921 added support for unaligned environments access.
U-boot itself does not support this:
- env_nand.c fails when using an unaligned offset. It produces an
error in nand_erase_opts{drivers/mtd/nand/nand_util.c}
- in env_sf/env_flash the unused space at the end is preserved, but
not in the beginning. block alignment is assumed
- env_sata/env_mmc aligns offset/length to the block size of the
underlying device. data is silently redirected to the beginning of
a block
There is seems no use case for unaligned environment. If there is
some useful data at the beginning of the the block (e.g. end of u-boot)
that would be very unsafe. If the redundant environments are hosted by
the same erase block then that invalidates the idea of double buffering.
It might be that unaligned access was allowed in the past, and that
people with legacy u-boot are trapped. But at the time of 56086921
it wasn't supported and due to reasons above I guess it was never
introduced.
I prefer to remove that (unused) feature in favor of simplicity
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
Acked-by: Stefan Agner <stefan.agner@toradex.com>
Format warnings (-Wformat) were shown in printf() calls after defining
DEBUG macro.
Update format string and explicitly cast variables to suppress all
warnings.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Try to avoid adhoc iteration of the environment. Reuse fw_getenv
to find the variables that should be printed. Only use open-coded
iteration when printing all variables.
For backwards compatibility, keep emitting a newline when
printing with value_only.
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
forward declaration not needed when re-ordered
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
there are two groups of functions:
- application ready tools: fw_setenv/fw_getenv/fw_parse_script
these are used, when creating a single binary containing multiple
tools (busybox like)
- file access like: open/read/write/close
above functions are implemented on top of these. applications
can use those to modify several variables without creating a
temporary batch script file
tested with "./scripts/kernel-doc -html -v tools/env/fw_env.h"
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
A negative value for the offset is treated as a backwards offset for
from the end of the device/partition for block devices. This aligns
the behavior of the config file with the syntax of CONFIG_ENV_OFFSET
where the functionality has been introduced with
commit 5c088ee841 ("env_mmc: allow negative CONFIG_ENV_OFFSET").
Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Currently flash_read completes a crucial part of the environment
device configuration, the device type (mtd_type). This is rather
confusing as flash_io calls flash_read conditionally, and one might
think flash_write, which also makes use of mtd_type, gets called
before flash_read. But since flash_io is always called with O_RDONLY
first, this is not actually the case in reality.
However, it is much cleaner to complete and verify the config early
in parse_config. This also prepares the code for further extension.
Signed-off-by: Stefan Agner <stefan.agner@toradex.com>
Reviewed-by: Andreas Fenkart
If users of the library are happy with the default, e.g. config file
name. They can pass NULL as the opts pointer. This simplifies the
transition of existing library users.
FIXES a compile error. since common_args has been removed by
a previous patch
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
checkpatch complains about in succeding patch. Prefer to fix all
declarations in a dedicated patch.
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
env library is broken as the config file pointer is only initialized
in main(). When running in the env library parse_config() fails:
Cannot parse config file '(null)': Bad address
Ensure that config file pointer is always initialized.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
fw_senten/fw_printenv can be compiled as a tools library,
excluding the fw_env_main object.
Reported-by: Stefano Babic <sbabic@denx.de>
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
disabled original parsing, but not yet removed since the
argument indexing needs to be fixed
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
goal is to use getopt for all argument parsing instead of adhoc
parsing in fw_getenv/fw_setenv functions
Signed-off-by: Andreas Fenkart <andreas.fenkart@digitalstrom.com>
With gcc 5.2 and later we get a bunch of "error: unknown type name" for
'uint8_t', 'uint32_t' and friends.
Signed-off-by: Peter Robinson <pbrobinson@gmail.com>
Sometimes it can be useful to link the fw_ tools instead
of having the fw_setenv/fw_printenv installed.
Patch exports the tool as library and allowes to link it
with own programs.
Signed-off-by: Stefano Babic <sbabic@denx.de>
CC: Tom Rini <trini@konsulko.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
When for example generating/manipulating SD card/eMMC images which
contain U-Boot and its environment(s), it is handy to use a given
configuration file instead of the compiled-in default one.
And since the default configuration file is expected under /etc
it's hard for an usual linux user account without special permissions
to use fw_printenv/fw_setenv for this purpose.
So allow to pass an optional filename via a new '-c' command
line argument.
Example:
$ ln -s fw_printenv tools/env/fw_setenv
$ cat fw_env.config
test.img 0x20000 0x20000
test.img 0x40000 0x20000
$ tools/env/fw_printenv -c ./fw_env.config fdt_file
fdt_file=imx28-duckbill.dtb
$ tools/env/fw_setenv -c ./fw_env.config fdt_file imx28-duckbill-spi.dtb
$ tools/env/fw_printenv -c ./fw_env.config fdt_file
fdt_file=imx28-duckbill-spi.dtb
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
When building tools-only (or env) we need to be sure that we do use
<linux/kconfig.h> and do not use <generated/autoconf.h>. This will fix
problems such as running 'make defconfig' or 'make sandbox_config' and
then 'make tools-only'.
Based on the responses below to the thread add linux/kconfig.h higher in
the includes and drop the now unneeded autoconf.h lower down to ensure
the default environment is included correctly
http://lists.denx.de/pipermail/u-boot/2015-June/216849.html
Signed-off-by: Peter Robinson <pbrobinson@gmail.com>
This patch fixes cross-compiling U-Boot tools with the musl C library:
* including <sys/types.h> is needed for ulong
* defining _GNU_SOURCE is needed for loff_t
Tested for target at91sam9261ek_dataflash_cs3.
Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Cc: Tom Rini <trini@konsulko.com>
