The SITE-ID parameter in the --site section is supposed to be formatted
as italic. Instead of .I .ID is used leading to misinterpretation and
the mentioning of the parameter missing in the rendered man page.
Found via:
$ man --warnings -E UTF-8 -l -Tutf8 -Z zdev/man/lszdev.8
troff:<standard input>:455: warning: macro 'ID' not defined
Use .I correctly to fix the issue.
Github-ID: https://github.com/ibm-s390-linux/s390-tools/issues/170
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add new command-line option --ipldev. When specified and IPL was done
from a supported device type, the IPL device is selected for the
requested tool operation.
Example to list IPL device information:
$ lszdev --ipldev
Example to create a persistent configuration for the IPL device:
$ chzdev --enable --persistent --ipldev
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Reduce code complexity by adding a utility function for reading text
files based on a variable path.
Note: chzdev/lszdev tools apply special path-handling via command line
option --base to allow redirecting hard-coded paths therefore libutil's
util_file functions cannot easily be used.
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The qeth-udev-rule execution logs a warning as below in the journalfs.
Failed to write ATTR{/sys/bus/ccwgroup/drivers/qeth/group}, \
ignoring: Invalid argument
This occurs because the udev rule for qeth devices is triggered by
multiple ccw devices and can be invoked in parallel. This can lead to a
situation where each of the three parallel-running udev rules tries to
group the devices simultaneously. As a result, only the first rule
succeeds, while the others fail with an -EINVAL error because the CCW
devices have already been grouped.
To prevent this, schedule a new change uevent for the base device, which
will verify the sysfs before attempting the grouping. The change event is
triggered by all three devices, ensuring that the actual write to the
attribute occurs only once and only when the sysfs attribute is ready.
Hence the add events on different devices are serialized to change event
on the base device, which then prevent duplicate concurrent grouping
attempts.
Suggested-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
All files created by chzdev feature a common header; however, this
header is absent in temporary files. It is necessary to incorporate
the consistent "Generated by chzdev" header into temporary files
generated by chzdev as well, so as to properly identify these files
via option --is-owner.
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Enhance the 'chzdev' tool by introducing a new option to discern
files created by 'zdev-tools.' The command usage is as follows:
$ chzdev --is-owner <file-name>
When executed, the command will return an exit code of 0 for all
files generated by zdev-tools. In the case of an unknown file, the
tool will return the exit code 'EXIT_UNKNOWN_FILE i.e 33.'
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add zdev-specific definitions into a lightweight header file. Rather
than creating a new one, transform the existing "site.h" into "zdev.h"
to house all generic zdev-specific definitions that needs to be shared
between chzdev, lszdev and zdev_id going forward.
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use similar `install` mode option in Makefile as in commit
9b2fb1d4d2 ("zdev: add helper to convert from dasd_mod.dasd to zdev
config").
Fixes: 73c46a3056 ("zdev/dracut: fix kdump by only activating required devices")
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
s390-tools doesn't define a common C/C++ standard at the moment.
Set the standard to gnu11/gnu++11, which is also used by the Kernel, and
establish a common baseline for all tools.
The -std flag is added to ALL_CFLAGS and ALL_CXXFLAGS to avoid losing it
in case CFLAGS are set by an outside entity.
It is also added to CLAGS_FOR_BUILD for this one special cross build
case.
The -std flag is removed from all tools that set it manually until now.
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Converts zdev configuration into the syntax of the dracut cmdline option
rd.znet. Only znet options with non-default values are emitted. The result
string occurs on stdout. It represents one device-specification for the
given network interface name.
Example:
/lib/s390-tools/zdev-to-rd.znet persistent encbdf0
qeth,0.0.bdf0,0.0.bdf1,0.0.bdf2,layer2=1
User:
https://github.com/rhinstaller/anaconda/pull/5250
to generate dracut cmdline entries such as
rd.znet=qeth,0.0.bdf0,0.0.bdf1,0.0.bdf2,layer2=1
Github-ID: https://github.com/ibm-s390-linux/s390-tools/pull/158
Acked-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Useful for debugging (what happened during early boot).
Also useful for distro installers, which can import the early config into
their own environment early after starting:
chzdev --import /run/zdev.initrd.config --persistent --yes \
--no-root-update --force --verbose
After that, distro installers can modify/add the device config based
on interactive or unattended installation choices using
chzdev --enable --active --persistent ...
Finally, distro installers can likewise transfer the entire device config
to the installed system mounted under $SYSROOT:
chzdev --export /tmp/zdev.config --all --type --persistent --verbose
Github-ID: https://github.com/ibm-s390-linux/s390-tools/pull/158
Acked-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add parsing of dracut cmdline option "rd.zfcp=..." at initrd runtime.
It delegates configuration to chzdev.
Implement `dracut --print-cmdline` and `dracut --hostonly-cmdline` for
initrd build time. Emit an rd.zfcp option for each zfcp-attached SCSI disk
in dracut's device dependency graph (to mount the root-fs, or to access the
kdump target).
This allows a distribution independent device configuration.
Configuration is consistent by using chzdev as backend.
It also prevents duplicate activations of the same device.
Along with the existing functionality of zdev/dracut, it makes the
following dracut modules superfluous:
https://github.com/dracutdevs/dracut/tree/master/modules.d/95zfcp
[rd.zfcp.conf is no longer needed and thus ignored here; the preceding
("zdev/dracut: fix marking hostonly files so delete option works")
makes rd.hostonly=0 work as a generic replacement]
https://github.com/dracutdevs/dracut/tree/master/modules.d/95zfcp_rules
Github-ID: https://github.com/ibm-s390-linux/s390-tools/pull/158
Acked-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The existing code can already determine the required devices for the
actively running system, but this does not work for execution environments
in different (disaster recovery) sites.
Kdump likely does not notice when running in a different site and does not
re-generate the kdump initrd (because the kdump config itself does not
change).
The new code allows users to explicitly specify devices required for early
boot in initrd on different sites. Assuming this is a small number of
devices and could even overlap between the root-fs and the kdump case, do
not further distinguish those two sets of devices, but simply use the same
marker attribute "zdev:early=1" for both cases. With all this site
information available at the (initial) kdump initrd generation, the
resulting kdump initrd can work on any site without having to re-generate.
Complements commit 3c7adcc3c81d ("zdev: dracut: modifiy the
module-setup.sh").
Github-ID: https://github.com/ibm-s390-linux/s390-tools/pull/158
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The existing code can already determine the required devices for the
actively running system. In case users would have a need to configure
additional devices for early boot in kdump initrd, import those devices
marked with "zdev:early=1" for completeness. Assuming this is a small
number of devices and could even overlap between the root-fs and the kdump
case, do not further distinguish those two sets of devices.
Github-ID: https://github.com/ibm-s390-linux/s390-tools/pull/158
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Additional debugging output can be generated with e.g. dracut option
"--stdlog 5" (or short -L5). It shows the chzdev export result, the output
of chzdev imports, and an overview of the resulting persistent config
within the initrd.
On systems, which default to using dracut option "--quiet", you might need
an additional "--verbose" to counter "--quiet" so -L5 has effect.
Typically combined with "--debug" to get a shell trace from building an
initrd (Note: --debug does not increase the log levels).
Github-ID: https://github.com/ibm-s390-linux/s390-tools/pull/158
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Otherwise, dracut skips modules that are "not hostonly", i.e. not used /
loaded on the system when running dracut to build an initrd.
Without the fix, zdev device auto configuration only works for device
types for which a device driver happened to be loaded when building an
initrd. Likewise for specifying rd.* boot options.
Hostonly mode is often the default in Linux distributions.
Make zdev device auto configuration work nonetheless without users
having to know about and tweak dracut modes for initrd generation.
On the typical sloppy hostonly mode, the code disables hostonly for
installing kernel modules. This covers building regular initrds, where zdev
device auto configuration should even work for device types, for which
device drivers were not loaded when an initrd was generated. This can
happen when new devices of new types are configured for a DPM logical
partition. Also, users could want to start using dracut cmdline options
rd.{dasd,zfcp,znet} for a device type that was not used when the initrd was
built.
The special strict hostonly mode is used by some kdump implementations. In
that case, hostonly remains in effect intentionally because only support
for the really required devices as determined by dracut module 95zdev-kdump
should be included in a kdump initrd due to the memory-constrained kdump
environment. Cf. commit 73c46a3056 ("zdev/dracut: fix kdump by only
activating required devices"), which also provides more references on
strict hostonly mode. Even for non-kdump cases, let strict hostonly mode be
effective here in 95zdev in case some future use case appears for this
special mode beyond kdump.
Github-ID: https://github.com/ibm-s390-linux/s390-tools/pull/158
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Currently ZDEV_SITE_ID is derived with the help of an additional
udev-rule, 40-zdev-id.rules. The sole purpose of this rule is to
determine the ZDEV_SITE_ID environment value with the help of zdev_id
binary. This solution is minimal, but this has some unwanted side-
effects. The zdev_id logic get executed for all the events, even
those completely unrelated to zdev/or site, and imports the unneeded
envionment values to the udev-db.
Instead of having an additional rule file, add this logic as part of
the udev-rule of those devices which are configured with site-support.
The logic will then be available on all those rules with the
site-supported devices only.
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Introduce dev_site_configured macro,which can be used to find the
availability of site configurations for the device during udev rule
creation.
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Previously SITE_FALLBACK and other site-specific configuration support
macros were defined in device.h. Instead, move them to a relatively
smaller header file which is exclusive for site-related definitions.
This way, light-weight zdev_id also can use the same header file.
Reported-by: Steffen Maier<maier@linux.ibm.com>
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Behaviour of the qeth performance_stats sysfs attribute has changed
with kernel commit
b0abc4f5df76 ("s390/qeth: overhaul ethtool statistics")
that went into kernel v5.1.
Before the kernel commit
- collection of statistics was turned on and off by writing 1 or 0
- default after device activation was 0
- statistics were reset by writing 0
After the kernel commit:
- collection is always on
- attribute always reads 1
- statistics is reset by writing 1; writing 0 is a no-op
Problems of chzdev on new kernels:
chzdev cannot reset statistics ('performance_stats=1' does nothing).
'chzdev --export' always lists performance_stats.
'chzdev qeth --help-attribute performance_stats' reflects old behaviour.
This patch will do the following:
'chzdev qeth --help-attribute performance_stats' reflects new behaviour.
'chzdev --export' does not list performance_stats on new kernels.
'chzdev performance_stats=1' resets statistics on new kernels.
'chzdev performance_stats=0' still resets statistics on old kernels,
does nothing on new kernels.
Suggested-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This complements v2.27.0 commit 73c46a3056 ("zdev/dracut: fix kdump by
only activating required devices"). On older distributions, the absence of
zdev_id can cause the following harmless error messages for each udev
event:
(spawn)[387]: failed to execute '/lib/s390-tools/zdev_id' \
'/lib/s390-tools/zdev_id': No such file or directory
Kdump is still functional nonetheless.
As of v2.24.0 commit 2e89722ef0 ("zdev: make site specific udev-rule for
ccw"), the invocations of chzdev within
zdev/dracut/95zdev-kdump/module-setup.sh generate
/etc/udev/rules.d/40-zdev-id.rules. And so even though zdev-kdump
intentionally does not install zdev_id and its previous singular user
zdev/udev/81-dpm.rules into the kdump initrd, because DPM device auto
configuration is not desired in the kdump environment, zdev_id meanwhile
has an additional functionality for site-support and the generated
40-zdev-id.rules calls /lib/s390-tools/zdev_id. By installing zdev_id into
the kdump initrd, 40-zdev-id.rules can work without error.
Fixes: 73c46a3056 ("zdev/dracut: fix kdump by only activating required devices")
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The compiler doesn't fully understand the code block that precedes the
usage of `site` in the condition and thus it thinks it could be
uninitialized. Silence the warning with an explicit initialization.
In function 'get_site_from_pers',
inlined from 'dev_table_get_value' at lszdev.c:1079:10:
lszdev.c:258:20: warning: 'site' may be used uninitialized [-Wmaybe-uninitialized]
258 | if (site == SITE_FALLBACK)
| ^
lszdev.c: In function 'dev_table_get_value':
lszdev.c:234:13: note: 'site' was declared here
234 | int site, i, num = 0;
| ^~~~
Closes: https://github.com/ibm-s390-linux/s390-tools/pull/152
Signed-off-by: Dan Horák <dan@danny.cz>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Using option --by-interface with a non-existent network interface name
results in an unexpected duplicate warning message:
$ lszdev --by-interface xx
Could not open directory /sys/class/net/xx: No such file or directory
Could not open directory /sys/class/net/xx: No such file or directory
lszdev: Could not determine device that provides xx (xx)
Fix this by checking for the existence of the associated /sys/class/net
directory before initiating the associated directory traversal.
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zdev's --by-node option returns exit code 15 (EXIT_RUNTIME_ERROR) when
the specified device node cannot be found. This is inconsistent with
other selection options --by-path and --by-interface which both return
the more appropriate exit code 3 (EXIT_DEVICE_NOT_FOUND).
Fix this by changing --by-node to also return exit code 3 in case the
specified device node does not exist.
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The udev-rules generated with the current version of chzdev command
is missing the configuration label, incase of auto configuration,
resulting in an ineffective configuration logic.
Add the missing configuration start label for autoconfig.
Fixes: 2e89722ef0 ("zdev: make site specific udev-rule for ccw")
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
During the boot, the ZDEV_SITE_ID is derived with the help
of loadparm and will be saved in ZDEV_SITE_ID_FILE, which
will be the used by the udev-rules.
ZDEV_SITE_ID_FILE creation can have a surface of symlink attack
as we are directly using the fopen and fprintf on it. To avoid
this, make sure that we are writing the ZDEV_SITE_ID to a temporary
file, which will then be renamed to ZDEV_SITE_ID_FILE, which will
remove all the existing symlinks associated with the target file.
Reported-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
udev does not allow an empty value for keys when importing output
from an external program. Providing an empty value for any key
invokes a warning during the parsing. Currently, ZDEV_SITE_ID for
fallback sites are not assigned any value. Add an empty double
quotes as the value in case of failover sites.
This modification is tested on udevadm version 253 on fedora38.
Also verify that the ZDEV_SITE_ID is properly written, if not log
the error.
Fixes: c8ad5f57d0 ("zdev: modify zdev_id to read the site_id from loadparm")
Reported-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Autoquiesce is a mechanism that tells Linux to stop issuing I/Os to a
specific DASD after certain events.
Add support for configuring related DASD device attributes
that govern the following aspects of autoquiesce:
aq_mask - Configure which events lead to autoquiesce.
aq_requeue - Configure if autoquiesce will requeue all I/O to blocklayer.
aq_timeouts - Configure the number of timeouts before autoquiesce.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
With the introduction of zdev for device configuration in v1.33.0,
it used a hardcoded "/" to refer to the root-fs.
Dracut has its own mechanism to find the root file system and to walk its
device dependencies.
Mkdumprd also has its own mechanism to interpret the kdump target
and invokes dracut so that dracut only considers arguments from mkdumprd
without any root-fs dracut would default to [--no-hostonly-default-device].
Use the dracut mechanism to walk device dependencies and individually
have chzdev treat each leaf-node in the dependency tree.
See also https://github.com/dracutdevs/dracut commit
83e0dc7a3dca ("Add for_each_host_dev_and_slaves for device only checking").
This enables the use of dracut options such as --add-device or
--mount used in turn by mkdumprd.
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This avoids exporting explicit settings if it's the default anyway.
I.e. avoid the last line in the example below:
$ chzdev --export - --active --type --all
[active zfcp-lun 0.0.1941:0x500507630904d...:0x4055408f00000000]
scsi_dev/queue_depth=32
Suggested-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
