New linux version will check the return code of parser on subpartitions.
The only valid case for skipping a parser with an error is -ENOENT.
Change the relevant entry to -ENOENT.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
This reverts commit ade045084b.
Skipping parser for subpartition will be handled now with -ENOENT.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
New linux version will check the return code of parser on subpartitions.
The only valid case for skipping a parser with an error is -ENOENT.
Change the relevant entry to -ENOENT.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
This reverts commit a51359aafd.
Skipping parser for subpartition will be handled now with -ENOENT.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
New linux version will check the return code of parser on subpartitions.
The only valid case for skipping a parser with an error is -ENOENT.
Change the relevant entry to -ENOENT.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
This reverts commit 6fa1775348.
Skipping parser for subpartition will be handled now with -ENOENT.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
New linux version will check the return code of parser on subpartitions.
The only valid case for skipping a parser with an error is -ENOENT.
Change the relevant entry to -ENOENT.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
This reverts commit 5aab294726.
Skipping parser for subpartition will be handled now with -ENOENT.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Rework each affected mtdsplit driver to make use of -ENOENT error
instead of -ENODEV to handle new kernel that checks error from parser on
subpartitions.
The only acceptable error is -ENOENT that skip the parser. This follow
pattern used upstream and also by an mtdsplit parser, mtdsplit_bcm_wfi,
and also by a workaround currently implemented for mtdsplit_mstc_boot.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
This reverts commit 627a0e7da0.
A better solution has been accepted upstream that doesn't hide all
unexpected error.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Return 0 if the current mtd is inactive or no valid header/rootfs found,
instead of -ENODEV.
Linux Kernel 6.7 and later versions handle all errors returned by mtd
parsers, including -ENODEV as error. So '0' needs to be returned if no
child partitions were not parsed.
Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/20697
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Currently the detection of external-data FIT images works by checking
if the FIT structure is more than 4 kiB. However, for boards with lots
of different DT-overlays and configurations the FIT structure can
exceed 4 kiB which results in the FIT splitter to fail detecting the
rootfs.
Increase the threshold for external-data FIT to 512 kiB as there aren't
any kernel images smaller than that, and FIT structure less than 512 kiB
will always be an external-data FIT.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Since Linux 6.7, introduced with commit 5c2f7727d437 ("mtd: mtdpart:
check for subpartitions parsing result"), errors during subpartition
parsing cause all MTD partitions to be torn down.
Since the current mtdsplit driver for devices using H3C VFS returns
-EINVAL if it does not find a file system containing an OpenWrt image,
this makes initial installation of OpenWrt impossible.
Work around this by returning 0 when the file system contains unexpected
data. Also print a message in this case to show what is going on.
Signed-off-by: Jan Hoffmann <jan@3e8.eu>
Link: https://github.com/openwrt/openwrt/pull/19475
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Add new mtdsplit parser "mstc-boot" for the devices manufactured by MSTC
(Mitra Star Technology Corp.). This is necessary to handle dual-boot on
those devices.
This parser splits kernel+rootfs or only rootfs(or UBI) based on the
image in the firmware partition or pre-defined partitions in dts, and
"bootnum" value in the "persist" (or "working") partition.
Note: "bootnum" is used for switching active firmware partitions on the
devices manufactured by MSTC and '1' or '2' are used on most
devices. But some devices use '0' or '1'. (example: I-O DATA
WN-DEAX1800GR)
Sequence:
1. obtain "bootnum" value
2. child nodes exsist (regardless of bootnum)
-> fixed partitions
(active parts : without bootnum (ex.: "kernel", "rootfs")
inactive parts: with bootnum (ex.: "kernel2", "rootfs2"))
3. current partition is active (dt bootnum == mtd bootnum)
-> image-based partitions
Device Tree:
- common
- mstc,bootnum : "bootnum" value for the mtd partition (0/1/2)
- mstc,persist : phandle of "persist" partition containing "bootnum"
value
- fixed partitions
- #address-cells: indicate cell count of address of child nodes (1)
- #size-cells : indicate cell count of size of child nodes (1)
- (child nodes) : define the child partitions
- reg : define the offset and size
- label-base : define the base name of the partition
- (example) : base:"kernel"->"kernel"(active)/"kernel2"(inactive)
example:
partition@3c0000 {
compatible = "mstc,boot";
reg = <0x3c0000 0x3240000>;
label = "firmware1";
mstc,bootnum = <1>;
mstc,persist = <&mtd_persist>;
#address-cells = <1>;
#size-cells = <1>;
partition@0 {
reg = <0x0 0x800000>;
label-base = "kernel";
};
partition@800000 {
reg = <0x800000 0x2a40000>;
label-base = "ubi";
};
};
- image-based partitions
(no additional properties)
example:
partition@5a0000 {
compatible = "mstc,boot";
label = "firmware1";
reg = <0x5a0000 0x3200000>;
mstc,bootnum = <1>;
mstc,persist = <&mtd_persist>;
};
Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18976
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Introduced with Linux 6.7, in commit:
5c2f7727d437 ("mtd: mtdpart: check for subpartitions parsing result"),
when a parser returns an error, this will be passed up, and
consequently, all parent mtd partitions get torn down.
Adjust the mtdsplit_uimage driver to only return an error if there is a
critical problem in reading from the mtd device or allocating memory.
Otherwise return 0 to indicate that no partitions were found.
Also add logging to indicate what went wrong.
E.g. on Realtek devices that are booted for the first time through
initramfs with OpenWrt never installed before boot log will show
[ 0.975518] Creating 7 MTD partitions on "spi0.0":
[ 0.981062] 0x000000000000-0x0000000e0000 : "u-boot"
[ 1.041320] 0x0000000e0000-0x0000000f0000 : "u-boot-env"
[ 1.060683] 0x0000000f0000-0x000000100000 : "u-boot-env2"
[ 1.080992] 0x000000100000-0x000000200000 : "jffs2-cfg"
[ 1.100988] 0x000000200000-0x000000300000 : "jffs2-log"
[ 1.120599] 0x000000300000-0x000000fe0000 : "firmware"
[ 1.157426] mtdsplit_uimage: no rootfs after uImage in "firmware"
[ 1.176456] mtdsplit_uimage: no rootfs after uImage in "firmware"
[ 1.200262] 0x000000fe0000-0x000001000000 : "log"
Similar issues were fixed before with commit ade045084b
("kernel: mtdsplit_minor: return 0 if not fatal") and
c78765213e ("kernel: mtdsplit_uimage: return 0 if not fatal")
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19163
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Introduced with Linux 6.7, in commit:
5c2f7727d437 ("mtd: mtdpart: check for subpartitions parsing result"),
when a parser returns an error, this will be passed up, and
consequently, all parent mtd partitions get torn down.
Adjust the mtdsplit_uimage driver to only return an error if there is a
critical problem in reading from the mtd device or allocating memory.
Otherwise return 0 to indicate that no partitions were found.
Also add logging to indicate what went wrong.
E.g. on Realtek devices that are booted for the first time through
initramfs with OpenWrt never installed before boot log will show
[ 0.932985] Creating 8 MTD partitions on "spi0.0":
[ 0.938412] 0x000000000000-0x000000080000 : "u-boot"
[ 0.990151] 0x000000080000-0x0000000c0000 : "u-boot-env"
[ 0.999907] 0x0000000c0000-0x000000100000 : "board-name"
[ 1.019971] 0x000000100000-0x000000e80000 : "firmware"
[ 1.051582] mtdsplit_uimage: no uImage found in "firmware"
[ 1.069365] 0x000000e80000-0x000001000000 : "kernel2"
[ 1.078959] 0x000001000000-0x000001040000 : "sysinfo"
[ 1.099747] 0x000001040000-0x000001c40000 : "rootfs2"
[ 1.119865] 0x000001c40000-0x000002000000 : "jffs2"
Similar issue was fixed before with commit ade045084b
("kernel: mtdsplit_minor: return 0 if not fatal")
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/19016
Signed-off-by: Robert Marko <robimarko@gmail.com>
RouterBOOT v7 on NOR devices no longer accepts the YAFFS kernel ELF
method of booting. It will accept an NPK image named bootimage.
Adjust mtdsplit_minor to accept this second possible boot file name.
Use the conservative value of 127 for YAFFS max name length (used when
YAFFS compiled with unicode support) vs 255.
Signed-off-by: John Thomson <git@johnthomson.fastmail.com.au>
Acked-by: Thibaut VARENE <hacks@slashdirt.org>
Link: https://github.com/openwrt/openwrt/pull/16780
Signed-off-by: Robert Marko <robimarko@gmail.com>
Introduced with Linux 6.7, in commit:
5c2f7727d437 ("mtd: mtdpart: check for subpartitions parsing result"),
when a parser returns an error, this will be passed up, and
consequently, all parent mtd partitions get torn down.
Adjust the MiNOR mtdsplit driver to only return an error if there is a
critical problem in reading from the mtd device or allocating memory.
Otherwise return 0 to indicate that no partitions were found.
Also add logging to indicate what went wrong.
This mtdsplit parser makes a very limited check of the first YAFFS
header. For example, this will not match expectations when initially booting
an initramfs image with OEM on MTD.
Signed-off-by: John Thomson <git@johnthomson.fastmail.com.au>
Acked-by: Thibaut VARENE <hacks@slashdirt.org>
Link: https://github.com/openwrt/openwrt/pull/16780
Signed-off-by: Robert Marko <robimarko@gmail.com>
fit size should be rounded up instead of rounding down first and adding
a block. Otherwise the calculated size is one block more than needed
when fit size is exactly multiples of one block size.
Fixes: 9a863f803e ("kernel: mtdsplit: add support for FIT image")
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Signed-off-by: Qingfang Deng <qingfang.deng@siflower.com.cn>
This mtdsplit parser driver parses firmware partition on Internet
Initiative Japan Inc. (IIJ) SEIL series devices.
Structure of header:
0x0 - 0x7 : Identifier (hex)
0x8 - 0x57: Copyright (ascii)
0x58 - 0x5b: Data CRC (hex)
0x5c - 0x5f: Image Format Version (hex)
0x60 - 0x63: Image Major Version (hex)
0x64 - 0x67: Image Minor Version (hex)
0x68 - 0x87: Image Release Version (ascii)
0x88 - 0x8b: Xor value for Data? (hex)
0x8c - 0x8f: Data Length (hex)
Properties:
- compatible : "iij,seil-firmware"
- iij,seil-id : ID of SEIL firmware for the device (8 bytes)
- examples:
- SA-W2 : <0x5345494c 0x32303135> ("SEIL2015")
- SEIL/X1 : <0x5345494c 0x2F582020> ("SEIL/X ")
- iij,bootdev-name: boot device name assigned to the partition
(optional)
Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
Some devices rename cferam bootloader using specific patterns and don't follow
broadcom standards for renaming cferam files. This requires supporting
different cferam file names.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Kconfig docs say:
> The default value deliberately defaults to 'n' in order to avoid
> bloating the build.
Apply this rule everywhere, to avoid more cloning of bad examples
Signed-off-by: Tony Butler <spudz76@gmail.com>
Support devices that has vendor custom header before FIT image.
Some devices has vendor custom header before FIT image. In this case mtd-
split can not find FIT image and it results in rootfs mount failure.
Please refer iptime,a6004mx device for further examples.
Signed-off-by: Yoonji Park <koreapyj@dcmys.kr>
The bootloader on some H3C devices (for example HPE 1920 switches) only
supports booting from flash by reading an image from an "VFS" filesystem
which spans most of the available flash. The filesystem size is hard-
coded in the bootloader. However, as long as no write operations are
performed in the bootloader menu, it is sufficient if the start of the
partition contains a valid filesystem with the kernel image.
This mtdsplit parser reads the size and location of the kernel image and
finds the location of the rootfs stored after it. It assumes that the
filesystem image matches the layout of one generated by mkh3cvfs, with
a filename of "openwrt-kernel.bin" for the kernel image.
Signed-off-by: Jan Hoffmann <jan@3e8.eu>
U-Boot has recently added support for having data blobs in uImage.FIT
images stored at offsets after the FDT structure rather than embedding
the data into the FDT structure itself. This is useful as it allows
parts of the image to be mapped by the FIT partition parser, and it
allows the FIT structure itself to be parsed more easily as it usually
fits into single page.
mtdsplit_fit assumed that the total length of an image is identical
to the length of the FDT structure. For uImage.FIT with external data
this assumption no longer holds true.
Add support for uImage.FIT with external data to mtdsplit_fit and in
in that case only split-off rootfs_data -- selecting and mapping rootfs
partition is left to the block partition parser just like on UBI and
block/GPT based platforms.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
It's helpful for accessing booting data (DTS, kernel, etc.). It has to
be used carefully as CFE's JFFS2 support is quite dumb. It doesn't
recognize deleted files and has problems handling 0 inode.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
This is useful for dual-boot setups where the loader sets variables depending
on the flash boot partition.
For example the Linksys E8450 sets mtdparts=master for the first partition
and mtdparts=slave for the second one.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Broadcom CFE bootloader relies on a tag for identifying the current firmware,
such as version, image start address, kernel address and size, rootfs size,
board id, signatures, etc.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
The "edimax,uimage"" parser can be replaced by the generic
parser using device specific openwrt,partition-magic and
openwrt,offset properties.
Signed-off-by: Bjørn Mork <bjorn@mork.no>
Some devices prepend a standard U-Boot Image with a vendor specific
header, having its own magic. Adding two new properties will support
validation of such images, including the additional magic.
Signed-off-by: Bjørn Mork <bjorn@mork.no>
The "netgear,uimage" parser can be replaced by the generic
parser using device specific openwrt,ih-magic and
openwrt,ih-type properties.
Device tree properties for the following devices have not
been set, as they have been dropped from OpenWrt with the
removal of the ar71xx target:
FW_MAGIC_WNR2000V1 0x32303031
FW_MAGIC_WNR2000V4 0x32303034
FW_MAGIC_WNR1000V2_VC 0x31303030
FW_MAGIC_WPN824N 0x31313030
Tested-by: Sander Vanheule <sander@svanheule.net> # WNDR3700v2
Tested-by: Stijn Segers <foss@volatilesystems.org> # WNDR3700v1
Signed-off-by: Bjørn Mork <bjorn@mork.no>
Convert users to the generic "openwrt,uimage" using device specific
"openwrt,ih-magic" properties, and remove "allnet,uimage".
Signed-off-by: Bjørn Mork <bjorn@mork.no>
The only difference between the "openwrt,okli" and the generic
parser is the magic. Set this in device tree for all affected
devices and remove the "openwrt,okli" parser.
Tested-by: Michael Pratt <mcpratt@protonmail.com> # EAP300 v2, ENS202EXT and ENH202
Signed-off-by: Bjørn Mork <bjorn@mork.no>
Convert users of the "fonfxc" and "sge" parsers to the generic
"openwrt,uimage", using device specific "openwrt,padding" properties.
Tested-by: Stijn Segers <foss@volatilesystems.org> [DIR-878 A1]
Signed-off-by: Bjørn Mork <bjorn@mork.no>
It's required for BCM4908. It cannot use "bcm-wfi-fw" parser because
that one requires *two* JFFS2 partitions which is untested / unsupported
on the BCM4908 architecture. With a single JFFS2 partition "bcm-wfi-fw"
parser will:
1. Fail to find "vmlinux.lz" as it doesn't follow "1-openwrt" file
2. Create partitions that don't precisely match bootfs layout
The new parser is described in details in the MTD_SPLIT_CFE_BOOTFS
symbol help message.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Add support for uimage headers from ALLNET and provide support for the
SG8208M and SG8310PM devices' magic bytes.
Signed-off-by: Birger Koblitz <git@birger-koblitz.de>
To parse the ELF kernel loader, a small ELF parser is used that can
handle both ELF32 or ELF64 class loaders. The splitter assumes that the
kernel is always located before the rootfs, whether it is embedded in
the loader or not. If the kernel is located after the rootfs on the
firmware partition, then the rootfs splitter will include it in the
dynamically created rootfs_data partition and the kernel will be
corrupted.
The kernel image is preferably embedded inside the ELF loader, so the
end of the loader equals the end of the kernel partition. This is due to
the way mtd_find_rootfs_from searches for the the rootfs:
- if the kernel image is embedded in the loader, the appended rootfs may
follow the loader immediately, within the same erase block.
- if the kernel image is not embedded in the loader, but placed at some
offset behind the loader (OKLI-style loader), the rootfs must be
aligned to an erase-block after the loader and kernel image.
In case section header table is empty, determine the elf loader size by
finding the end of the last segment, as defined by the program header
table.
Signed-off-by: Sander Vanheule <sander@svanheule.net>
This is useful when booting OpenWrt from ramdisks in order to have both
images partitions defined.
Furthermore, instead of always using img2 for the inactive image, let's use
img1 or img2 accordingly.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
