When an entry in /dev does not exist, start_getty is returning directly.
As it is started from init in a respawn mode, it will loop infinitely.
In this case add a sleep inside start_getty to prevent the "Respawning
too fast" message popping up every 5 minutes.
This case is happening quite often when the system is started as an
hypervisor guest as the standard serial line is usually taken by it and
removed from the configuration.
This was triggered quite often running linux as dom0 on top of Xen on
arm as the serial line is taken by Xen and removed from the device tree.
Use the opportunity to replace one tab with spaces as the rest of the
file is using spaces.
(From OE-Core rev: 8b95138b09256df61578ddf7df16ef13bced1847)
Signed-off-by: Bertrand Marquis <bertrand.marquis@arm.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 5a704d0ff4d95b6d3a67b80a0db2be66253e7d61)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
An automated conversion using scripts/contrib/convert-spdx-licenses.py to
convert to use the standard SPDX license identifiers. Two recipes in meta-selftest
were not converted as they're that way specifically for testing. A change in
linux-firmware was also skipped and may need a more manual tweak.
(From OE-Core rev: ceda3238cdbf1beb216ae9ddb242470d5dfc25e0)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
refresh pidof-add-m-option.patch
Changelog:
==========
* Default to showing processes in the uninterruptable state (D).
The -z flag no longer affects whether processes in D state are shown.
The -z flag does still toggle whether zombie (Z) processes are shown.
* Removed unnecessary check which is always true from init tab parsing.
(From OE-Core rev: 5b5c1cd43923e552ae6bcfc50f1766981c44ae7b)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is the result of automated script conversion:
scripts/contrib/convert-overrides.py <oe-core directory>
converting the metadata to use ":" as the override character instead of "_".
(From OE-Core rev: 42344347be29f0997cc2f7636d9603b1fe1875ae)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This reverts commit ed69ef2016.
The console entry has already been added into /etc/inittab based
on the SERIAL_CONSOLES. So drop this redundant entry.
(From OE-Core rev: 633f0c6b74e3caa2bae52ca60c61b811b7b2215d)
Signed-off-by: Kevin Hao <kexin.hao@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The hvc tty driver doesn't populate a file like /proc/tty/driver/serial,
so the current implementation of start_getty doesn't work for the hvc
console. By checking the /sys/class/tty/ for the tty device existence,
it should support more console types and also make the codes more simple.
(From OE-Core rev: 670ceef0f6584ece5ce4176610255226a6148570)
Signed-off-by: Kevin Hao <kexin.hao@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The licenses were renamed to match their SPDX names, fix the
references in LIC_FILES_CHKSUM in OE-Core.
(From OE-Core rev: 9af48917cfe583d2db9e1e088c7e396fcf638949)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The psplash program contains a hidden text box immediately above the
progress bar. Any text sent via a "MSG" command through psplash's fifo will
be displayed, centred, above the progress bar. Add the ability to show
which startup script is currently running, in sync with updates to the
progress bar. If a startup script takes a bit longer than others and the
progress bar stops momentarily, this allows the user to know which script
is responsible.
This feature is added with a knob, default off, for enabling or disabling
this feature. The knob is in the form of a PACKAGECONFIG against the
sysvinit recipe: psplash-text-updates
NOTE: this knob can be changed in the filesystem at runtime by editing
/etc/default/rcS regardless of how it is set in the build.
(From OE-Core rev: 6a48631e24341895f13029fb7c69dd67031f53c1)
Signed-off-by: Trevor Woerner <twoerner@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The psplash program has a mechanism for showing updates graphically in the
form of a progress bar. The program is told when and how much to fill the
progress bar via text messages sent through a fifo. If the fifo doesn't exist
when the psplash program starts, it tries to create it. If the fifo doesn't
exist or can't be created, the psplash program will refuse to run.
In various circumstances when a system is booted for the very first time,
the filesystem is mounted, initially, read-only. As a result the psplash
program is not able to run. On systems where the root filesystem is not
meant to be read-only, it will eventually be mounted read-write. Therefore the
psplash program can run on shutdown, and all subsequent boots. Only the first
boot is affected.
If a fifo is created and included in the filesystem as part of the recipe,
then filesystems that are meant to be read-only will have psplash work, as
well as the cases where (on first boot) a read-write filesystem is initially
mounted read-only.
NOTE: this is only an issue with sysvinit, and non-qemu machines.
systemd-based systems don't suffer from this first-boot issue, and neither
do the qemu machines.
NOTE 2: when psplash is done, it removes the fifo. Therefore the fifo used
for communicating with psplash doesn't hang around unnecessarily in the
filesystem.
(From OE-Core rev: d20978f9bce3caf473f5b4f55d645dbeabc1642d)
Signed-off-by: Trevor Woerner <twoerner@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add an entry for the psplash fifo directory to /etc/default/rcS and have the
pieces of code that need it source it from there rather than duplicating the
definition in multiple places throughout the code.
(From OE-Core rev: dc4065b6f101e6418301e0cb8d73ae3a1b2bdfb2)
Signed-off-by: Trevor Woerner <twoerner@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This will help in defining init system specific portions of initscripts
which are shared
(From OE-Core rev: 12380bccce7b87b755d1e24620c397641e88f656)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Cc: Sinan Kaya <okaya@kernel.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Replace the equality operator '==' with '=' inside of '[]' to be
compatible with bash and dash.
(From OE-Core rev: b7f0ec6eafb35117eaf4eeef281162080f0ca79a)
Signed-off-by: Fedor Ross <fedor.ross@ifm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The VERSION variable is not set unless the
top level Makefile is used as the version is
derived from the Changelog.
If VERSION is not set, none of the binaries built
in the sysvinit recipe will have a valid version
output.
Before:
INIT: version booting
After:
INIT: version 2.96 booting
Using the top level Makefile does not cause any
additional builds to occur.
(From OE-Core rev: 10992ad4f343d0f92d9d6246cc2e58effd3f3eeb)
Signed-off-by: Michael Gloff <mgloff@emacinc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
psplash expects the fifo to be in /run or specified
by PSPLASH_FIFO_DIR. This patch allows psplash to
quit normally. Also, fix the work around of using
echo directly into the fifo and use psplash-write.
(From OE-Core rev: 2b48bbd0a8de17df74d1fae12b5e4eb3c2c0fb90)
Signed-off-by: Michael Gloff <mgloff@emacinc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
"start_getty" support for tty devices with under 10 number.
When SERIAL_CONSOLES has tty devices with 10 or more number,
do not't start getty and output the following message.
----------
sh 1: unknown operand
----------
(From OE-Core rev: 39dc49523f5d24db8f8053444c543c371c32d10c)
Signed-off-by: Yuki Hoshino <yuki.hoshino@miraclelinux.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
logsave is a duplicate from e2fsprogs, which was already removed,
but the man page remained which is also a duplicate. However it
does not get modifed on install so it conflicts with the e2fsprogs
version.
(From OE-Core rev: 90ee5557c0e1d7088b5d9b7e66d70ff7fccac9a3)
Signed-off-by: Jeremy A. Puhlman <jpuhlman@mvista.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
It is possible for sysvinit to be installed before the passwd/group
files are installed from base-passwd. This results in an warning that
the shutdown group can't be found.
Adding a dependeny on base-passwd will resolve this issue.
This has similar symptoms to the problem that oe-core commit
0227e929021263c51d2e7db36224000fecb01f1c fixed. The fix is simular in that
it ensures that base-passwd is installed first, just at runtime vs setscene
time.
(From OE-Core rev: c4fb899b19a905a2b027f33947b6ec56f55b81a6)
Signed-off-by: Mark Hatle <mark.hatle@kernel.crashing.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With a sequence like:
bitbake sysvinit
bitbake sysvinit -c clean
bitbake sysvinit -c package_write_ipk -f
then the resulting package has two files with group "root/70" rather
than "root/shutdown". The issue is that of do_package is a setscene
task, base-passwd isn't present. This patch fixes that dependency
but there may be other cases of this problem around.
[YOCTO #13776]
(From OE-Core rev: 0227e929021263c51d2e7db36224000fecb01f1c)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Remove 0001-This-fixes-an-issue-that-clang-reports-about-mutlipl.patch,
the problem has been fixed upstream.
Rebase the other patches.
License-Update: copyright years
(From OE-Core rev: 5372e1acdd68367e64d13e19d1beb2d7b3e3c224)
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Besides checking DISTRO_FEATURES for required or conflicting features,
being able to check MACHINE_FEATURES and/or COMBINED_FEATURES may also
be useful at times.
Temporarily support the old class name with a warning about future
deprecation.
(From OE-Core rev: 5f4875b950ce199e91f99c8e945a0c709166dc14)
Signed-off-by: Denys Dmytriyenko <denys@ti.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fixes musl issue
implicit declaration of function 'minor' is invalid in C99 [-Wimplicit-function-declaration]
which eventually ends up with a linker error
(From OE-Core rev: 6b603924e50a5694421b9142494315799422928d)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
when both installed, do_rootfs report error like:
file /etc/inittab conflicts between attempted installs of
busybox-inittab-1.30.1-r0.qemux86 and sysvinit-inittab-2.88dsf-r10.qemux86
so add each other to rconflicts
(From OE-Core rev: d66114b07ef3e8482fee4c3c93e4f6324c38da3b)
Signed-off-by: Changqing Li <changqing.li@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Busybox' getty has code to try to make itself a session leader,
whereas util-linux' agetty doesn't. It expects this to happen
from outside.
When getty is not a session leader, many things don't work on
the serial console, e.g. setting the terminal process group,
job control doesn't work, etc.
Executing image tests also fails with AssertionErrors, because
Feb 5 16:12:55 qemuarm getty[590]: /dev/ttyAMA1: cannot get controlling tty: Operation not permitted
Feb 5 16:12:55 qemuarm getty[590]: /dev/ttyAMA1: cannot set process group: Inappropriate ioctl for device
Update the start_getty script to invoke getty via the setsid
utility if needed, i.e. if /sbin/getty is not busybox getty.
[YOCTO #13058]
(From OE-Core rev: 37be77565d323fc543427ad47399996119f59ab1)
Signed-off-by: André Draszik <andre.draszik@jci.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Rationale - excerp from `info chown`
====================================
OWNER‘:’GROUP
If the OWNER is followed by a colon and a GROUP (a group name or
numeric group ID), with no spaces between them, the group ownership
of the files is changed as well (to GROUP).
Some older scripts may still use ‘.’ in place of the ‘:’ separator.
POSIX 1003.1-2001 (*note Standards conformance::) does not require
support for that, but for backward compatibility GNU ‘chown’ supports
‘.’ so long as no ambiguity results. New scripts should avoid the use
of ‘.’ because it is not portable, and because it has undesirable
results if the entire OWNER‘.’GROUP happens to identify a user whose
name contains ‘.’.
(From OE-Core rev: 185918234a07cb506d7d7464a49ac33972c7d963)
Signed-off-by: Kosta Zertsekel <zertsekel@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Instead, first check if we need to do anything at all during first boot,
and if so, either postpone to first boot via pkg_postinst_ontarget()
when running on host, or run the necessary setup code when running on target.
(From OE-Core rev: 16df1717c3813ba773e0dfa2d1db471816d8b99b)
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Use distro_features_check so that we can have a uniform controller.
(From OE-Core rev: 658c59c90092f15c026fa3c72399f481c7241f65)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Stop using update-alternatives for managing /etc/init.d/functions. Also,
make the initscripts-functions subpackage to (runtime) conflict with
lsbinitscripts.
[YOCTO #10944]
(From OE-Core rev: cdcebd81c872cb7386c658998e27cf24e1d0447c)
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Unbreak serial console when driver name contains spaces (PXA serial).
Fix commit ac0e954
"start_getty: Over added SERIAL_CONSOLE cause error in userspace log"
(From OE-Core rev: 8b98302c30efb7073f61dc2a166f7414f050ef65)
Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Error log will be logged into /var/log/message.
Added in more condition checking on the script. Check
/proc/tty/drivers and /proc/tty/driver/*
file system to retrieve active targeted serial.
Only establish getty with active serial in runtime.
[YOCTO #10844]
Reviewed-by: Saul Wold <sgw@linux.intel.com>
(From OE-Core rev: ac0e9541fe93e866e42914f65a0516b993f0cffe)
Signed-off-by: Choong YinThong <yin.thong.choong@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
getty devices were not being removed in some cases because device name
was not at the end of the line, for example a ttyS1 device:
S1:12345:respawn:/bin/start_getty 115200 ttyS1 vt102
Removing this limitation allows sed to remove any line containing
the device.
(From OE-Core rev: 2e0b36981c1f91ed0d3d457c370df10a099407af)
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This makes more sense than the default TERM=linux (as set
by the linux kernel).
In addition, when using busybox init, it tries to achieve
the same (in a different way).
Both agetty, and busybox getty support the terminal type as
the last argument.
(From OE-Core rev: a23004f32dda9310c7efc29720fa9c3027a7c329)
Signed-off-by: André Draszik <adraszik@tycoint.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With some hardware the name of the device node and the name in
/proc/console differ. This causes SERIAL_CONSOLES_CHECK to not enable
working consoles in these cases. This patch changes SERIAL_CONSOLES_CHECK
to have an optional alias for the checked consoles. The new format is:
<device>:<alias to check(optional)>
Fixes [YOCTO #9440].
(From OE-Core rev: 91d9f3271c12fb755ab332637b17650d5fe75ce2)
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The current recipe creates inittab labels based off the device node name
of TTYs used as consoles. If those names exceed the 4 character label
limit of inittab, it will break. This change takes the last 4 chars of
the device names in order to avoid any errors.
[ YOCTO #9529 ]
(From OE-Core rev: 30acc7a6b9e6d1c42ba1df6e5a362d10b43cb4eb)
Signed-off-by: Stephano Cetola <stephano.cetola@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
util-linux has an alternative for it. Add it to sysvinit too.
(From OE-Core rev: 43069c2bd9947357970bd48504987bda514f5219)
Signed-off-by: Dan McGregor <dan.mcgregor@usask.ca>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
sysvinit is objectively less maintained than util-linux or busybox, each
of which may supply its own mountpoint implementation. Adjust the
ALTERNATIVE_PRIORITY to select the sysvinit implementation as the last
resort.
(From OE-Core rev: 92c6b896e3b11283e2c313e6a2ac3497a99a13a0)
Signed-off-by: Richard Tollerton <rich.tollerton@ni.com>
Signed-off-by: Ioan-Adrian Ratiu <adrian.ratiu@ni.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When this file is in ${sysconfdir}/init.d, SELinux labels it as a generic
init script (initrc_t). This causes problms at runtime because SELinux
doesn't let the login process execute generic init script. Moving this
helper script to base_bindir results in it being labeled as a generic
binary (bin_t). Nearly every SELinux domain is allowed to execute
generic binaries and the login process is one of them.
(From OE-Core rev: 826bfea4b7018c7974ef388facc99ca70bb3654f)
Signed-off-by: Philip Tricca <flihp@twobit.us>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add a wrapper script around getty to check if a given console exists, this
allows for multiple Console ttys to be specified for various boards without
having additional BSP types just for different console types.
[YOCTO #8689]
(From OE-Core rev: b93369a7943949f51057e0a704f5524ab7682fe6)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This is due to specifying .h files on linker cmdline
clang driver is picky about it, and its not entirely correct
either
(From OE-Core rev: de45b5e68faeefe3d68818d456f280b98f397634)
(From OE-Core rev: 44dcee60e343944199ee766ec0886931ff2b0699)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
On qemuarm64 the second serial port in SERIAL_CONSOLES will be hvc0.
Since that doesn't have tty in the name, a correct label didn't get
added to inittab.
This change makes both names with tty and without work.
(From OE-Core rev: 4cca92b34defae425929d92671edde621f8a5e80)
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Similarly to systemd, only enable the recipe in builds where
sysvinit is configured in DISTRO_FEATURES. This allows the new
cleanup mechanism to handle it correctly in existing builds.
(From OE-Core rev: 7f85e74d5c53b34e5f470967fdbdbd19fed1929a)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
