Age | Commit message (Collapse) | Author |
|
This is particularly problematic since qemu images switched to ext4 by
default and now cannot work properly with UIs like hob.
This patch adds in ext4 to the appropriate IMAGE* variables fixing this.
[YOCTO #7426]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
[YOCTO #7392]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
[YOCTO #6804]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
The newer btrfs-utils needs an empty file to build the filesystem in, so
create an empty file and use it for the mkfs to build the fs in.
[YOCTO #6804]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
For cpio images, do_rootfs() can operate on a dirty '${WORKDIR}/cpio_append'
directory which contains e.g. files from previous builds. This can cause
unwanted files in the image or can break the build.
E.g. when there is a cpio_append/init -> /sbin/init symlink symlink, the
'ln -sf' can fail due to SELinux restrictions:
| $ ls -la cpio_append/init
| lrwxrwxrwx. 1 ensc ensc 10 22. Jan 16:26 cpio_append/init -> /sbin/init
|
| $ strace ln -sf /sbin/init cpio_append/init
| ...
| stat("cpio_append/init", 0x7fffbb9ca310) = -1 EACCES (Permission denied)
| exit_group(1) = ?
Patch cleans up 'cpio_append' before executing the 'do_rootfs' task by
adding it to 'cleandirs'. An alternative implementation (which avoids
creation of this empty dir for non-cpio images) might remove it within
IMAGE_CMD_cpio, but this might break builds where people rely on the
existence of this directory (e.g. to add local files).
Signed-off-by: Enrico Scholz <enrico.scholz@sigma-chemnitz.de>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
When /init is a dangling symlink or a symlink to a file which can not be
stated on the build system (e.g. due to SELinux restrictions), the '[ !
-e .../init ]' test will succeed which causes the manual creation of
/init.
E.g. here:
| $ ls -la cpio_append/init
| lrwxrwxrwx. 1 ensc ensc 10 22. Jan 16:26 cpio_append/init -> /sbin/init
|
| $ strace /bin/test -e cpio_append/init
| stat("cpio_append/init", 0x7fff374a9db0) = -1 EACCES (Permission denied)
| exit_group(1) = ?
To test for the existence of a file, both '-L' and '-e' checks must be
executed and to prevent SELinux noise, the '-L' should happen before
'-e'.
Signed-off-by: Enrico Scholz <enrico.scholz@sigma-chemnitz.de>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The IMAGE_TYPEDEP dependencies also need to be taken into account when
building an IMAGE_FSTYPE.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
For cpio/ramfs the kernel will first attempt to execute /init and will
emit the following error as the file is empty:
Failed to execute /init (error -13)
If /sbin/init exists symlink to it so the kernel can immediately start
the correct init executable instead of an empty file.
Signed-off-by: Nathan Rossi <nathan.rossi@xilinx.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
Use a multi-line value to set IMAGE_TYPES and put all the compressed
versions of a format on one line. Other than moving ext2.lzma this is
just whitespace.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
The ubi command assumes the ubifs file is present.
This makes sure this is really the case.
Signed-off-by: Pascal Bach <pascal.bach@siemens.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
Add "squashfs-lzo" to the image types. LZO compression support has been
in both kernel and squashfs tools for many years, but OE never enabled
it.
Signed-off-by: Mike Looijmans <mike.looijmans@topic.nl>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If the first command returns an error, it will not cause the image generation
step to fail. Simply split up the statement into multiple lines to avoid
this issue, they no longer need to be one line expressions.
[YOCTO #6391]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The Linux kernel requires that initrd images contain a /init file for
the image to be used as an initrd, even if it is empty. Adding it into
the rootfs directory creates a race, that can upset tar when building
both a .tar and .cpio image file ("tar: .: file changed as we read it").
Additionally, whether or not the tar file will contain the /init file is
also up to the race condition.
To avoid this problem, move the /init addition out from the rootfs
directory, and thus only include it in the .cpio image.
Signed-off-by: Jonas Eriksson <jonas.eriksson@enea.com>
Signed-off-by: Josep Puigdemont <josep.puigdemont@enea.com>
Cc: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The image not correctly created if 'ptest-pkgs' is in IMAGE_FEATURES,
this is because there is no free inode left. We can use 4096 instead of
8192 bytes-per-inode to fix the problem, and most of the distributions
us 4096, such as Ubuntu, Suse, Fedora and CentOS.
There are another problems:
* There are error message when there is no free inode left if we run the
mke2fs command manually, but they are not in log.do_rootfs.
* The image generation doesn't stop when error happens because mke2fs
doesn't return failed for this case.
Will fix them in other threads.
[YOCTO #5957]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We used populate-extfs.sh which invoked the debugfs to create the image,
now the mke2fs' option "-d root-directory" can do the same thing, and
which is more faster, for example, the core-imag-sato:
* In the past:
$ time mke2fs -t ext4 rootfs.ext4
real 0m0.249s
user 0m0.036s
sys 0m0.132s
$ time populate-extfs.sh /path/to/rootfs/ rootfs.ext4
real 0m29.355s
user 0m10.637s
sys 0m5.544s
* Now:
$ time mkfs.ext4 -F -d /path/to/rootfs/ rootfs.ext4
real 0m6.338s
user 0m3.824s
sys 0m1.356s
More than 25 seconds are saved.
[YOCTO #4083]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The working directory is changed in a subshell when executing cpio to
preserve the working directory for any subsequent commands. This is to
keep the working directory consistent when generating multiple image
types.
Signed-off-by: Jonathan Liu <net147@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The following dependencies were manually added in the image creation
code. However, in order to have an image dependency mechanism in place,
use a new variable, IMAGE_TYPEDEP, to declare that an image type depends
on another being already created.
The following dependencies are added by this commit:
elf -> cpio.gz
live -> ext3
vmdk -> ext3
iso -> ext3
hddimg -> ext3
This commit adds also another new variable: IMAGE_TYPES_MASKED. Currently,
masking out certain types from IMAGE_FSTYPES was hardcoded in the image
creation code.
[YOCTO #5830]
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Since generic compression solution has been introduced
(see revision b7e4ed41ee480f00b7265341e9e2d2c2b9135143),
CPIO_TOUCH_INIT() is only called by IMAGE_CMD_cpio.
Signed-off-by: Matthieu Crapet <Matthieu.Crapet@ingenico.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
below commit
commit 9d9bca8785911e8ae06d507bbfb99d6a811f072e
Author: Jonathan Liu <net147@gmail.com>
Date: Mon Feb 17 21:46:28 2014 +1100
lib/oe/image.py: fix working directory
The working directory needs to be changed before the image creation
commands instead of afterwards.
exposed an issue when generating tar images, where it would cd into
rootfs and then create the rootfs tarball and remain there since the
cmd to cd into deploy dir moved up the immediate following cmd like
tar.bz2 or tar.gz were still looking for tarball in current directory
which esentially was not deploy dir but IMAGE_ROOTFS instead
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Previous commit added support for sum 'compression' (rather
postprocessing) of jffs2 images. Drop support for sum.jffs2 image type.
Signed-off-by: Dmitry Eremin-Solenikov <dmitry_eremin@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
I got hit by a missing dependency in generation of sum.jffs2 images - at
runtime, when images are generated using bitbake pool, it is possible
for sum.jffs2 image to be scheduled for generation before jffs2 image is
fully generated.
Insted of adding additional hacks, reimplement sum.jffs2 generation as a
compression scheme on top of plain jffs2 images. For now support for
sum.jffs2 images is left in, it will be removed by the subsequent
commit.
Signed-off-by: Dmitry Eremin-Solenikov <dmitry_eremin@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This commit will:
* remove old bash code common to all backends;
* create a new do_rootfs() python function that will use the new
rootfs/image creation routines;
* allow creation of dpkg based images;
* fail for rpm/opkg (not implemented yet);
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
|
|
Signed-off-by: Koen Kooi <koen.kooi@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
We're seeing a pattern of one image type needing to depend on another
type. A good example is jffs2 and sum.jffs2. This patch makes sum.jffs2
depend on jffs2 which will then allow a EXTRA_IMGAGECMD to be set for
sum.jffs2 individually without changing the jffs2 command. This allows the
-pad option to be configured differently.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
For mkfs.jffs2 endianness can be be expressed in the long or short form
with the optional size argument:
--little-endian
-l
Strangely the sumtool has a different syntax and does accept the forms:
--littleendian
-l
Prefer the short form valid for both tools.
Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Changes to split live into iso and hddimg without
adding a new image type class.
This patch has only a visible effect on HOB and solves
part 2 of #3197
[YOCTO #3197]
Signed-off-by: Valentin Popa <valentin.popa@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The former is provided by util-linux and the latter is
now to be removed for meta-filesystems.
This allows to generate cramfs images whithout extra layers.
Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* The benefits:
- Really support ext4
- Support the sparse file (we lost the sparse file in the image in the
past, the sparse file became into the common file)
- Fix the error reported by fsck: (ext2/ext3)
Inode 1025, i_size is 16384, should be 17408.
- Have a uniform code for ext2/3/4 generation
* Comments from Darren Hart:
Basically, genext2fs doesn't support creating ext4 filesystems. It
creates, as I understand it, an ext2 filesystem, then adds a journal,
and sets some bits. It can't support the newer features like extents. So
what we end up with is a bit of a hack for a filesystem.
The ext tools (e2fsprogs) unfortunately don't provide an integrated
solution for generating prepopulated filesystem images as many other
mkfs* tools do. One thing missing was symlink support in libext2fs. I
added that support and demonstrated a script which uses the e2fsprogs
debugfs tool that can populate the newly formatted filesystem from a
directory and without root privileges.
[YOCTO #3848]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Generating filesystems with this flag allows more efficient directory
traversals because getdents() returns the filetype in 'd_type' which
allows to avoid an extra lstat() call.
Creating ext4 filesystems with 'mkfs.ext4' sets this flag by default
too.
Signed-off-by: Enrico Scholz <enrico.scholz@sigma-chemnitz.de>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
* the IMAGE_TYPES variable is used exclusively by hob
* IMAGE_CMD_ubifs is already present in the class
* ubifs images can be flashed on existing ubi volumes with 'ubiupdatevol'
Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Remove the old image before the new one generated to save disk
space when RM_OLD_IMAGE is set to 1, this is an easy way to keep
the DEPLOY_DIR_IMAGE clean.
[YOCTO #4391]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Generation of elf images fails because kernel images are no longer staged under
${STAGING_DIR_HOST}/kernel, but rather ${STAGING_DIR_HOST}/usr/src/kernel. This
patch fixes the path to point to the correct location.
Signed-off-by: Tomas Frydrych <tomas@sleepfive.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Booting an image generated with squashfs-lzma results in a kernel
error: "Filesystem uses "lzma" compression. This is not supported".
Currently (well at least in Linux 3.8) the officially supported
decompressors are:
* LZO
* XZ
* ZLIB
This change makes sure we use a supported compression algorithm for
squashed root filesystems.
Signed-off-by: Marc Ferland <ferlandm@sonatest.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When creating images containing many partitions, rootfs.$type
may not be available.
Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Explain the elf/cpio.gz snippet of code, and update a comment to
reflect the correct variable name. No functional change.
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
This removes the dependency on a seperate binary which we
don't seem to have. So, use mksquashfs's -comp lzma to
replace that functionality
[YOCTO #3126]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If we round up ROOTFS_SIZE to IMAGE_ROOTFS_ALIGNMENT before checking if
base_size is greater then IMAGE_ROOTFS_SIZE, we can end up adding an
unaligned value to IMAGE_ROOTFS_SIZE. Obviously, if
IMAGE_ROOTFS_EXTRA_SPACE was overwritten with an unaligned value. So
let's add the round up code after the base_size calculus and it's
comparison.
Signed-off-by: Andrei Gherzan <andrei@gherzan.ro>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Add a new variable "IMAGE_EXTENSION_live" for Hob to map image type
"live" with real image file extension names.
This is for Hob to remove the hard-coded maps.
Signed-off-by: Kang Kai <kai.kang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Having XZ_COMPRESSION_LEVEL on -e -9 and -T 0 will make xz eat more
than 6Gb memory. Reduce this to -6 to make xz to use about 471Mb
on the tested machine.
Signed-off-by: Koen Kooi <koen@dominion.thruhere.net>
Signed-off-by: Andrei Gherzan <andrei@gherzan.ro>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
while compressing
Default this variable to 0. This will make xz use as many threads as there are CPU
cores on the system.
Signed-off-by: Koen Kooi <koen@dominion.thruhere.net>
Signed-off-by: Andrei Gherzan <andrei@gherzan.ro>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Having -c modifier makes xz to output the compressed data to stdout. In this
way the needed data will be in the do_rootfs log.
Redirect data to ${IMAGE_NAME}.rootfs.${type}.xz .
Signed-off-by: Andrei Gherzan <andrei@gherzan.ro>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
On x86, an ELF image file may be stored as a coreboot payload.
The image file is constructed, using the mkelfimage utility,
from a kernel and an initrd.
Signed-off-by: Raymond Danks <ray.danks@se-eng.com>
Acked-by: Darren Hart <dvhart@linux.intel.com>
|
|
Signed-off-by: Ting Liu <b28495@freescale.com>
|
|
inode count is already set in the EXTRA_IMAGECMD definition
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Define runnable image type and machine pattern.
Define deployable image type.
Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
In the case of self-hosted-image.bb, IMAGE_FSTYPES = "vmdk", so the
variables alltypes and subimages don't contain ext3, and .rootfs.ext3
won't be created, and finally the generated .hddimg and .vmdk don't have
an actual rootfs -- the size of the .vmdk file is only about 9MB.
[YOCTO #2067]
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
|