Extract the minor version and modification level separately.
Previously only the modification level has been extracted, and was
reported as minor version.
Currently no one is checking the minor version or modification level,
so it does not hurt. But maybe in the future one will check, so report
it correctly.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
During plugin initialization, the external libraries such as the
EP11 or CCA host libraries are set up, if the configuration is
appropriate.
A secure identity key may be generated once the APQNs are configured,
but before the server connection is configured. Trying to re-encipher
the plugin's secure keys to a new HSM master key at that stage fails
with 'ERROR: Invalid ext lib type: 0' because the external libraries
have not been setup yet.
Change the code to setup the libraries once the APQNs have been
configured, and not only after the server connection has been
configured.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Increase initial update interval from 200ms to 1 seconds to avoid
fluctuations on the initial data output.
Signed-off-by: Mete Durlu <meted@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
pvattest is a tool to attest an IBM Secure Execution guest.
In a trusted environment, one can create a request using
`pvattest create`. To get a measurement of an untrusted
IBM Secure Execution guest call 'pvattest perform'.
Again in a trusted environment, call 'pvattest verify'
to verify that the measurement is the expected one.
The tool runs on s390 and x86.
It has the same requirements like libpv and therefore
requires openssl v1.1.1+, glib2.56+, and libcurl.
Additionally, to measure, the linux kernel must provide
the Ultravisor userspace interface `uvdevice` at /dev/uv
and must be executed on an IBM Secure Execution guest on
hardware with Ultravisor attestation support, like IBM z16 or later.
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
libpv is a collection of definitions and functions related to
Protected Virtualization (PV).
The functions cover mainly encryption (e.g. AES-GCM)
and certificates (X509). There are also helping functions for glib2.
Most of the code is extracted+refactored from `genprotimg`, which
will use this library in future.
Requires openssl v1.1.1+, glib2.56+, and libcurl.
libpv is not designed or intended to be dynamically linked or used
outside of this project. Its purpose is to avoid code duplication
as PV tools do very similar things regarding cryptography.
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
Acked-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This commit introduces a new module containing various helpers for
NGDump stand-alone dump. The purpose of these helpers is to facilitate
the implementation of DT and DFI interfaces for NGDump, share code
between both, reduce duplication and enable unit testing later on.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Return the partition number which contains the given block range,
before it was tested for exact match between a partition block range
and the one provided by user. The old behavior with exact match
should still work, this change just relaxes the partition matching
algorithm and allows one to find a partition which contains the given
block range.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This commit finalizes the implementation of the new stand-alone dump -
Next Gen Dump (NGDump).
NGDump stand-alone dump is a universal stand-alone dump which works for both
SCSI and NVMe disks. But currently can be used only with NVMe disks
and SCSI stand-alone dump remains the default for SCSI disks.
Currently, this stand-alone dump can be used only on IBM z15 or newer
machine generations because it requires larger amount of HSA memory offered
by firmware only on IBM z15 machines. Whereas SCSI stand-alone dump is able
to work with HSA memory of 32M, the new stand-alone dump requires 512M HSA
memory at the moment.
NGDump stand-alone dump installation is initiated by passing a path to
a NVMe disk partition to zipl via the command-line -d, similar to SCSI
stand-alone dump. zipl will then:
- build a dumper initramfs with either dracut (Fedora/RHEL/SLES) or
initramfs-tools (Ubuntu/Debian)
- format the given NVMe partition with ext4 file system
- create a bootmap file on the newly created file system using the built
initramfs and the currently active kernel image
- install a boot loader on the disk the given dump partition belongs to
The operations described above are destructive for the given NVMe dump
partition and the boot record(s) of its disk.
After the installation step, users can configure the dumpconf to IPL
the dumper automatically on panic or trigger a dump manually via HMC
interface.
When activated, the dumper will create a dump ELF file named "dump.elf"
on the given NVMe dump partition by using the makedumpfile tool. The kdump
compressed file format is not supported yet due to zgetdump not being able
to read such a file format. Therefore, the dumper is restricted to write
the dump only in ELF format but only kernel pages which are in use.
This will usually make the dump smaller than the original size of
/proc/vmcore and the whole dump process faster as well.
Example configuration for dumpconf:
ON_PANIC=dump # or dump_reipl
DUMP_TYPE=nvme
FID=0x00000001
NSID=0x00000001
BOOTPROG=0
BR_LBA=0
$ systemctl enable --now dumpconf
To install the dracut support on Fedora/RHEL/SUSE:
$ make -C zipl/dracut HAVE_DRACUT=1 install
To install the initramfs-tools support on Ubuntu/Debian:
$ make -C zipl/initramfs-tools HAVE_INITRAMFS=1 install
Example of NGDump console output on SLES during dump
----------------------------------------------------
Starting NGDump...
[ 8.932343] ngdump.sh[327]: NGDump started
[ 8.934739] ngdump.sh[336]: Checking for memory holes : [ 0.0 %] /
[ 8.967536] ngdump.sh[336]: Checking for memory holes : [100.0 %] |
[ 9.076976] ngdump.sh[336]: Excluding unnecessary pages : [100.0 %] \
[ 11.721157] ngdump.sh[336]: Copying data : [ 0.0 %] -
[ 12.317319] ngdump.sh[336]: Copying data : [ 22.5 %] / eta: 2s
[ 13.273000] ngdump.sh[336]: Copying data : [100.0 %] | eta: 0s
[ 13.273244] ngdump.sh[336]: The kernel version is not supported.
[ 13.273263] ngdump.sh[336]: The makedumpfile operation may be incomplete.
[ 13.273281] ngdump.sh[336]: The dumpfile is saved to /ngdump/dump.elf.
[ 13.273299] ngdump.sh[336]: makedumpfile Completed.
Example of NGDump console output on Ubuntu during dump
-------------------------------------------------------
[ 3.240078] zdump: The dump process started for a 64-bit operating system
Loading, please wait...
Starting version 245.4-4ubuntu3.17
Begin: Starting firmware auto-configuration ... done.
Begin: Loading essential drivers ... done.
Begin: Running /scripts/init-premount ...
Begin: NGDump ...
Checking for memory holes : [ 0.0 %] /
Checking for memory holes : [100.0 %] |
Excluding unnecessary pages : [100.0 %] \
Copying data : [ 0.0 %] -
Copying data : [ 40.7 %] / eta: 1s
Copying data : [100.0 %] | eta: 0s
The kernel version is not supported.
The makedumpfile operation may be incomplete.
The dumpfile is saved to /ngdump/dump.elf.
makedumpfile Completed.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Tested-by: Alexander Gordeev <agordeev@linux.ibm.com>
Reviewed-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Tested-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The purpose of the new zipl helper script is to build a dumper initramfs
for NGDump stand-alone dump. zipl executes the helper script when
preparing a NVMe dump partition for stand-alone dump.
The zipl helper script expects a single parameter - a path to the dump
partition to store a dump on. The helper script terminates with 0 on
success and a positive value otherwise.
The zipl helper script performs the following steps:
- It builds an initramfs image suitable for dumping. To perform this step,
the script employs the dracut tool on Fedora/RHEL/SLES and
the initramfs-tools on Ubuntu/Debian.
- It outputs the path to the newly built initramfs image and the currently
active kernel bzImage on the standard output.
- It prints the kernel command-line to be used by the dumper on
the standard output.
On success, zipl expects the helper script to return at least three lines
containing:
- initrd=<path to an initramfs image>
- kernel=<path to a kernel bzImage>
- cmdline=<kernel command-line parameters>
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Tested-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The NGDump initramfs-tools module is required to build an initramfs image
to be used with NGDump stand-alone dump. NGDump stand-alone dump does not
use a pre-built initramfs image in contrast to SCSI stand-alone dump.
Instead, an initramfs image is built with initramfs-tools if NGDump
stand-alone dump is installed on a NVMe partition.
The NGDump initramfs-tools module ensures that all necessary tools are
present within the built initramfs and a dump of /proc/vmcore is initiated
to the chosen NVMe partition when the installed dumper is IPLed.
The NGDump intramfs-tools module installs a configuration file and
a dump script into dumper initramfs. This dump script starts at boot
shortly after the initialization of the dump target device. It reads
the aforementioned configuration file that contains the name of a dump
partition to store a dump on. The dump script reads the configuration file,
mounts then the dump target device, creates a copy of /proc/vmcore with
the makedumpfile tool on it, and then shuts down the system.
Ubuntu and Debian are the main Linux distributions targeted by this
initramfs-tools module.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Tested-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The NGDump dracut module is required to build an initramfs image to be used
with NGDump stand-alone dump. NGDump stand-alone dump does not
use a pre-built initramfs image in contrast to SCSI stand-alone dump.
Instead, an initramfs image is built with dracut if NGDump stand-alone dump
is installed on a NVMe partition.
The NGDump dracut module ensures that all necessary tools are present
within the built initramfs and a dump of /proc/vmcore is initiated
to the chosen NVMe partition when the installed dumper is IPLed.
The NGDump dracut module installs a new systemd service and a dump script
into dumper initramfs. The systemd init process starts the NGDump service
at boot shortly after the initialization of the dump target device.
The NGDump systemd service, in its turn, starts the dump script provided
by the new dracut module. The dump script mounts then the dump target
device, creates a copy of /proc/vmcore with the makedumpfile tool on it,
and then shuts down the system.
The NGDump dracut module can be used on any Linux distribution which
supports dracut. But the primary targets are:
- Fedora
- RHEL
- SLES
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Tested-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change adds a new parameter to the function bootmap_create_file()
that requires the caller to specify a path to the directory where
a bootmap file will be created. The primary motivation for this change is
to support NVMe stand-alone dump, in that case the job's bootmap directory
is not set.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The purpose of this change is to simplify the function
bootmap_create_device() and make it more readable. Furthermore,
the dump partition check can be reused for NVMe dump partitions later on.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The purpose of this change is to simplify the function
bootmap_create_device() and make it more readable.
Furthermore, make bootmap_create_device() more robust by checking the return
value of disk_write_block_aligned().
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The purpose of this change is to simplify the function bootmap_create(),
make it more readable and prepare for new changes that will follow.
The original function bootmap_create() tried to handle two different cases
and, therefore, contained many conditional statements which had a negative
effect on readability.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The purpose of this change is to simplify the function bootmap_create(),
make it more readable and prepare for new changes that will follow.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Until now, NVMe disks were handled by zipl as SCSI disks. To support
NVMe stand-alone dump, it became necessary to further differentiate between
both types of disks.
There are two cases that must be handled:
1. A non-device-mapper device is a NVMe disk if it is assigned to the blkext
device driver.
2. A device-mapper disk is a NVMe disk if the ioctl NVME_IOCTL_ID succeeds.
This case is necessary to properly recognize NVMe disks which are
DM devices and, therefore, not directly handled by the blkext device
driver.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Stefan Haberland <sth@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use util_file_read_va() function to read out the event number
given the event file. Also use util_strdup() function to save
the event name. This function terminates on Out Of Memory condition.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
genprotimg is a user command and as such installed into /usr/bin. Adapt
the man page accordingly and move it from section 8 (system commnds) to
section 1 (user commands).
Reported-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The static analysis tool for Debian packages 'lintian', especially if called
like 'lintian -EvIL +pedantic', checks manual pages for correctness and typos.
This commit fixes typos that were identified by lintian and marked with
'typo-in-manual-page' while s390-tools version 2.20.0 was packaged.
Closes: https://github.com/ibm-s390-linux/s390-tools/pull/134
Signed-off-by: Frank Heimes <frank.heimes@canonical.com>
Acked-by: Steffen Maier maier@linux.ibm.com [ziomon]
Acked-by: Ingo Franzki ifranzki@linux.ibm.com [zkey]
[hoeppner@linux.ibm.com: fix commit message]
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The provider functions are passed to OpenSSL in the dispatch array
casted to 'void (*)(void)', so they can not be type checked by the
compiler. Add function prototypes using the OpenSSL provided
function typedefs to allow the compiler to check the function
signatures.
No functional change.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Gcc 12 produces a -Wmaybe-uninitialized warning that the content
of the tmp buffer may be uninitialized. This is a false positive.
Silence the warning by using calloc instead of malloc to ensure
that the allocated buffer is initialized.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Create an empty environment file /etc/ziplenv at make install
time if that file doesn't exist
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add definitions of sites, namespaces and sections.
Document the new options --site (-S) and --effective-site (-E)
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
. Support multiple namespaces in the installed environment block;
. Add option --site (-S) to specify a particular namespace
when operating on the installed environment block;
. Add option --effective-site (-E) to specify a particular
namespace. When using in a combination with the option -l
(--list), it displays zIPL environment that would take place if
the specified namespace was activated at boot time;
. Modify set/unset/reset/list operations over the installed
environment block in the case when the option --site is specified
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
. Parse LOADPARM specified by user to find out site-ID;
. Support multiple namespaces in the environment block. Use hash
function on "extended" strings (i.e. strings with logical
prefixes) for the key-value store implementation;
. Added logic to activate a particular namespace by site-ID passed
by user via LOADPARM;
. Modify the logic of evaluation of zIPL environment variables in
the kernel command line. Use the activated namespace for the
evaluation
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Support multiple namespaces in environment block. Use hash
function on "extended" strings for key-value store.
Support multiple sections in environment file.
Modify the import environment precedure to be "section-aware"
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Consolidate type definitions in order to avoid duplicated code, getting
better compiler support, and to avoid confusing namings for the same
thing - e.g. `blk_end` vs. `blockcnt` field name used in the old `struct
fba_dump_param` definitions. It also allows us to get rid of two
`uint64_t` casts.
While at it, fix the documentation of `struct eckd_blockptr`.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
...otherwise `.note.gnu.build-id` will be put at 0x0 (location for s390x
lowcore) if the linker option `--build-id` is used (which is the default
if gcc is used for linking).
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use the `tpi_info` struct definition, similar to the Linux kernel, in
the lowcore struct. This change allows us to use simple assignments
instead of using casts. Additionally, there is the advantage that the
lowcore definition from the s390-tools now looks more similar to the
lowcore definition used by the Linux kernel.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Until recently dumpconf might have been installed in two different
locations, /etc/init.d/ and /lib/s390-tools/. As dumpconf is now part of
the helper scripts in the scripts directory, extend the installation
rule and add the %S390_TOOLS_VERSION% wildcard to the script along the
way for correct version output.
The installation rule in the systemd/Makefile is removed, bringing it in
line with cpictl.
Acked-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Modern distributions use systemd and a corresponding systemd unit file
is provided that handles the dumpconf tool. The tool is also the last
script left in etc/init.d. Move it to scripts and get rid of the init.d
directory altogether.
Acked-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For cpacfstatsd, cpuplugd, and mon_statd there are still SysV daemon
scripts available. However, modern distributions use systemd and
corresponding systemd unit files are provided and in use for a long time
already.
Remove the daemon scripts for the obsolete SysV init system as they're
not used by modern systems anymore.
Acked-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The ap_tools utilities are intended to be used in conjunction with
the mdevctl utility for safely managing and inspecting vfio-ap
mediated devices. For now, this will consist of the ap-check tool
which will be driven via a call-out from mdevctl to validate a
proposed vfio-ap mediated device change.
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Reviewed-by: Tony Krowiak <akrowiak@linux.ibm.com>
Reviewed-by: Jason J. Herne <jjherne@linux.ibm.com>
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zdev uses a particular set of exit codes -- In preparation for sharing
some zdev udev code with other libraries, let's also create a libutil
include to specify reusable exit codes. For now, let's just initialize
it with the codes from zdev we care about.
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Reviewed-by: Jason J. Herne <jjherne@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add IBM z16 extended counter set. For each counter in this set
add the counter number, the short and long description and the
counter number.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
OpenSSL commit ee22a3741e3fc27c981e7f7e9bcb8d3342b0c65a changed the
OpenSSL provider's keymgmt_match() function to be not so strict with
the selector bits in regards to matching different key parts.
Adapt the secure key provider's match function accordingly.
This means, that if the public key is selected to be matched, and
the public key matches (together with any also selected parameters),
then the private key is no longer checked, although it may also be
selected to be matched. This is according to how the OpenSSL function
EVP_PKEY_eq() is supposed to behave.
OpenSSL function SSL_CTX_use_PrivateKey() calls the providers match
function to check if the private key specified matches the public key
of the certificate using EVP_PKEY_eq(). EVP_PKEY_eq() includes the
private key into the selector bits here, although the certificate
only contains the public key part.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The Processor Activity Instrumentation facility (PAI) Extension 1
adds support for the counter set for Neural Network Processing Assist
(NNPA) counters.
NNPA counter values are appended as raw data to the data report.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a check to detect Processor Activity Instrumentation (PAI)
Extension 1 facility unit. This covers the counter set for
Neural Network Processing Assist (NNPA) counters.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The Processor Activity Instrumentation facility (PAI) uses the
perf_event_open system call to report data and the crypto counter
values. The data can be inspected with the perf tool in a very
generic way using raw data report and a hex dump utility.
The pai tool also uses the perf_event_open system call to record the
data, but uses specialized event attributes for recording detailed
information such as the recording of fork, exit, exec system call
events and context switch events.
Cryptography counter values are appended as raw data attachments.
The pai command can be used for recording and reporting.
Also a list of CPUs can be specified.
The recording ring buffers are created per CPU
and are named paicrypto.XXX where XXX stands for the CPU number.
The numbering scheme has leading zeros, such as 007 for CPU number 7.
For invocation and possible command line options see the man page.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Since the switch to fuse3 setting the hard_remove option with the
FUSE_LIB_OPT() macro is no longer supported. See libfuse commit
8ee553dac029 ("fuse_new(): don't accept options that don't make sense
for end-users") for details. To fix this, add an appropriate init
function which sets this option.
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/131
Fixes: e679a88d88 ("Switch from fuse2 to fuse3")
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Fix the doxygen comments and add a description on the return values.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
By default, the Bash file globbing pattern '*' doesn't list files prefixed by a
dot (see https://linux.die.net/man/1/bash). Let's fix this by using the pattern
`.*` instead. While at it, add `--` option to remove possible problems with
leading dashes in filenames.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This work around fixes the gcc-12 false positive by disabling `Warray-bounds`:
CC genprotimg/boot/stage3a.o
In file included from stage3a.c:14:
In function ‘__test_facility’,
inlined from ‘test_facility’ at ../../include/boot/s390.h:428:9,
inlined from ‘start’ at stage3a.c:42:7:
../../include/boot/s390.h:418:17: error: array subscript 0 is outside array bounds of ‘void[0]’ [-Werror=array-bounds]
418 | return (*ptr & (0x80 >> (nr & 7))) != 0;
| ^~~~
Unfortunately, there is currently no better fix available that doesn't result
in larger boot loader code sizes. Given the importancy of the boot loader file
sizes the other fixes aren't acceptable. The Linux kernel shares the
problem (but for performance reasons), take a look at the discussion
https://lore.kernel.org/lkml/yt9dzgkelelc.fsf@linux.ibm.com/ for details.
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/130
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This patch implements an additional method of sorting bls entries
by version field and uses it as default one. If sorting by version
fails, then it falls back to the most robust method of sorting by
file names
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
It is used by the next patch.
Specifically, ->filter() and ->sort() callbacks of the scandir(3)
accept only short names of directory entries to be filtered/sorted,
while we need to know absolute names to perform filtering/sorting.
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The problem can happen when dfi_s390mv_init_gen() returns with an error
code to dfi_init() in dfi.c.
Double free condition occurs on zg_close() call at the end of the
while loop in dfi_init() if zg_close() has already been called for the
same file handle at the end of open_dump() function in scope of
dfi_s390mv_init_gen() processing.
This global file handle is not closed during init() call for any
other dump formats. Since it is not reopened/reused after open_dump() call
during multi-volume dump initialization, we should not close it at all.
The problem can be reproduced in the following steps:
1) Install multi-volume dump tool
# zipl -M mvdump.conf
Dump target: 2 partitions with a total size of 4732 MB.
Warning: All information on the following partitions will be lost!
/dev/dasdb2
/dev/dasdb3
Do you want to continue creating multi-volume dump partitions (y/n)?y
Done.
2) Run zgetdump -i using device (not partition) as a parameter without
taking actual dump.
# zgetdump -i /dev/dasdb
free(): double free detected in tcache 2
Aborted (core dumped)
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With "zpcictl --reset DDDD:BB:FF.F" now causing a fully Linux driven
reset where the Linux kernel does an explicit device driver unbind,
disable and re-enable, let's also expose a way to instead have firmware
perform a device reset by issuing an SCLP with SCLP_ERRNOTIFY_RESET.
When firmware is done resetting the device it will then issue an error
notification with PCI Error Code 0x3a indicating successful reset, which
will subsequently cause the new kernel based automatic recovery
mechanism to perform recovery in coordination with the device driver.
This allows resetting devices without unbinding them from their device
driver and thus without losing related block devices or network
interfaces. This may also be used to test the automatic recovery
mechanism.
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Currently "zpcictl --reset DDDD:BB:FF.F" issues an SCLP call with
SCLP_ERRNOTIFY_AQ_RESET followed by a Linux driven device reset via the
recover sysfs attribute. The latter was introduced with commit
bc0d40c580 ("zpcictl: Initiate recover after reset") because the
firmware driven reset leaves the device in the error state. Now with the
addition of transparent PCI recovery however the situation has changed
as Linux will not leave the device in the error state after the reset
but will instead initiate its automatic recovery flow. With that however
the two mechanisms, automatic PCI recovery and the zpcictl triggered
recovery attribute handler will race against each other.
In practice this is harmless as the automatic recovery is serialized
with the recover attribute and whichever wins the race will do the reset.
The losing side will detect that the original device was removed and
will refrain from causing a double reset.
Letting both mechanisms race against each other is not predictable
behavior though so instead of SCLP_ERRNOTIFY_AQ_RESET issue
a SCLP_ERRNOTIFY_AQ_REPORT_ERROR that will report a device error to the
Support Element but not cause a reset and then predictably reset via the
recover sysfs attribute. If instead a firmware driven reset followed by
the automatic recovery flow is desired the new "--reset-fw" option may
be used.
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The default issuer check may fail if the to-be-verified host key
document was issued and signed by an entity not known at the point
in time check_hostkeydoc was released.
In order to allow verification of the chain of trust for an unknown
but otherwise valid issuer, check_hostkeydoc can be called with
the -d command line option.
This commit also enhances the help text by briefly describing the
command line options and fixes a typo.
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
While the original default issuer's organizationalUnitName (OU)
was defined as "IBM Z Host Key Signing Service", any OU ending
with "Key Signing Service" is considered legal.
Let's relax the default issuer check by stripping off characters
preceding "Key Signing Service".
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The re-enciphering of EP11 asymmetric secure keys does not work.
First, the result of the re-encipher operation of the private key
part must be copied back into the user supplied key token buffer.
Second, the public key part, i.e. the MACed SubjectPublicKeyInfo
(SPKI) structure must also be re-enciphered (i.e. re-MACed), since
the MAC is calculated with the EP11 master key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Remove the DigiCert root CA pinning. The root CA used for the chain of trust can
change in the future therefore let's remove this check. If someone wants to
enforce the usage of a specific root CA it can be selected by the genprotimg
command line option `--root-ca $CA`. Make it transparent to the user which root
CA is actually being used by printing the subject name of the root CA to stdout
in verbose mode.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Acked-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Reviewed-and-tested-by: Nico Boehr <nrb@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some development kernel versions provide an 8-digit date number like
"20220325" in place of the second kernel sublevel, while the data
format used to send this information to the HMC only supports 16 Bit
numbers. As a result, the HMC displays a seemingly random sublevel
number.
Fix this by replacing excessive sublevel numbers with 0 to ensure that
these numbers are handled consistently.
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Hitting CTRL-D anywhere will cause fdasd to go into an endless loop
displaying the main menu over and over again. Killing fdasd (e.g.
via CTR-C) is the only way out.
The issue is that read_line() is just ignoring the resulting EOF
condition on stdin. Subsequent invocations of read_line() will return
immediately and thus cause the loop. A simple fix is to reset stdin
after EOF. A caller of read_line() will see the same behavior as for
EOL with no input.
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This can now include /lib/modprobe.d/10-unsupported-modules.conf
Strictly speaking, modprobe would only consider *.conf files inside
those directories, but for consistency with the already existing
collection of /etc/modprobe.d instead of /etc/modprobe.d/*.conf,
do the same for the added missing locations.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Recent multipath-tools can depend on this.
a1eabea75e
("multipathd.service: drop ExecStartPre for loading dm-multipath")
92f0893ac1
("multipath-tools: install modules-load.d/multipath.conf")
This can now also include /usr/lib/modules-load.d/s390-pkey.conf
Complements v2.8.0 commit
dffd41943e ("pkey: Support autoloading kernel pkey module").
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The target field is required for IPL sections, but the ones populated from
BLS snippets did not have this field. This was only working when using BLS
snippets because IPL sections inherit the target field with "defaultauto".
But that's not the case when using a menu, for example the following zipl
configuration will fail if it's used along with BLS defined IPL sections:
fail when running the zipl tool:
[defaultboot]
defaultmenu = menu1
:menu1
target = /boot
1 = linux
2 = test
default = 1
prompt = 1
timeout = 0
Because "linux" and "test" will be defined using BLS snippets and these
don't have an option to define target fields. Let's make these to always
inherit a target field, either from the [defaultboot] section or a menu
section if there's a defaultmenu defined.
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/111
Closes: https://github.com/ibm-s390-linux/s390-tools/pull/113
[hoeppner@linux.ibm.com: Fixed a couple of style issues]
Reported-by: Renaud Métrich <rmetrich@redhat.com>
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Tested-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Now that we've concluded that the compiler is used for calling the actual
linker, it doesn't make much sense to select a different compiler for linking.
Even worse, it's prone to error. A naive user might try to compile s390-tools
using clang as follows:
$ make -C genprotimg CC=clang
...
CC genprotimg/src/utils/curl.o
LINK genprotimg/src/genprotimg
/usr/bin/ld: genprotimg.o: `stderr@@GLIBC_2.2' non-PLT reloc for symbol defined in shared library and accessed from executable (rebuild file with -fPIC ?)
/usr/bin/ld: final link failed: bad value
collect2: error: ld returned 1 exit status
Therefore it makes sense to use the same compiler for linking as well.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Remove NO_PIE_LINKFLAGS and follow the naming convention by using
LDFLAGS for linker flags. Replace all occurrences accordingly.
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The "detour" with NUL terminated records ensures that sorting
works as expected for environment variables with values containing
one or more lines and thus '\n' as regular record separator.
Reviewed-by: Joern Siglen <siglen@de.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make use of multipath-tools 0.8.9 commit
("libmultipath: add %L path wildcard for 64-bit hex LUN")
260d7cb411 .
Multipath-tools versions that do not know the format wildcard simply
ignore it and expand to an empty string, so it's backwards compatible.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Eases mapping of multipath paths and FCP by-path information
without having to use different tool output and correlate,
such as "multipathd -k'show topo'" and "lszfcp -D".
Possibly the HBA is reported as "[undef]" for zfcp
[without multipath-tools 0.8.9 commit
("libmultipath: support host adapter name lookup for s390x ccw bus")
852a1dfd94 ]
and the FCP LUN needs to be manually converted from the SCSI LUN in h:c:i:l.
While at it, append correlation information for block device name and
major:minor, as well as all information from a standard 'show paths',
which is the path state 3-tuple and next_check,
plus the number of path failures.
Example output:
$ multipathd -k'show paths format "%w|%a|%r|%p|%i|%d|%D|%t|%T|%o|%0|%C"'
uuid |host adapter|target WWPN |pri|hcil |dev|dev_t|dm_st |chk_st|dev_st |failures|next_check
3600507640081818ab00000000000025e|[undef] |0x500507680b2581fa|10 |0:0:0:606|sdb|8:16 |active|ready |running|0 |XXXXXX.... 13/20
3600507640081818ab00000000000025e|[undef] |0x500507680b2581fb|50 |0:0:1:606|sdc|8:32 |active|ready |running|0 |XX........ 4/20
3600507640081818ab00000000000025e|[undef] |0x500507680b2681fa|10 |1:0:0:606|sdd|8:48 |active|ready |running|0 |XXXX...... 8/20
3600507640081818ab00000000000025e|[undef] |0x500507680b2681fb|50 |1:0:1:606|sde|8:64 |active|ready |running|0 |XXXXXXX... 15/20
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The new option -s, long --serial shows the serial numbers of
CCA and EP11 cards.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
New options:
--cardonly
--queueonly
which filter the output to show only card or queue information.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
New options:
--accelonly
--ccaonly
--ep11only
which restrict the output to only cards/queues with the given mode.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Newer kernel show a sysfs attribute chkstop which displays
the checkstop state of a crypto card. This patch enables
support for lszcrypt to display the checkstop state.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
If there is a max_msg_size attribute for the card
the lszcrypt -c option will show this limit as
AP bus max message size limit xx Kb
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
On several distributions `pandoc` is not available via the default
repositories, and thus not available in the build environments for the
distribution packages. That means, the man page can't be bundled along
with the packages generated for those distributions.
But since this is a valuable asset for some users, instead of requiring
`pandoc` in order to have a man page be generated during the build, so it
can be installed, bundle a pre-cooked version that corresponds to the
current version of the `README.md` file. This way, the man page can always
be packaged, and is now always installed, even if `ENABLE_DOC` is set to
`0` (the default).
This also means, whenever the `README.md` file is changed, the bundled
man page needs to be regenerated, so it stays in sync. As a safeguard
we also add a checksum of the `README.md` file that is regenerated along
with the man page. This checksum is tested even when `pandoc` is not
available, and the user is notified whenever it runs out of sync (with a
hint as to how to remedy it).
Nothing changes for the compile/installation workflow, if `ENABLE_DOC` is
set to `1`.
Suggested-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Sometimes dumping a virtual machine from the outside is the only way to
get the data that is needed. This can be the case if a dumping mechanism
like kdump hasn't been configured or data needs to be fetched at a
specific point. Dumping a protected guest from the outside without help
from FW/HW doesn't yield sufficient data to be useful. Hence we now
introduce Protected Virtualization (PV) dump support.
The PV dump support works by integrating the firmware into the dump
process. New Ultravisor calls are used to initiate the dump process,
dump cpu data, dump memory state and lastly complete the dump process.
The guest's data is fully encrypted and can only be decrypted by the
entity that owns the customer communication key for the dumped guest.
Also dumping needs to be allowed via a flag in the SE header.
This patch adds support for PV guest dumps to genprotimg. To prepare a
PV image in order that the PV guest can later be dumped from the
outside, the user has to provide a customer communication key used for
the dump process and he has to set the corresponding control flag. For
specifying the customer communication key a new command line option
`--comm-key` is added and for enabling/disabling the control flag the
command line options `--enable-dump` and `--disable-dump` are added.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Acked-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Declare `--enable-pckmo` and `--disable-pckmo` as mutually exclusive.
Let's define a helper macro for this which allows an easier definition
of mutually exclusive command line flags.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Introduce a private macro for the calculation of the control flags and
sort the flags by bit value.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Debian carried a patch forever that allowed zipl to run even if not all
menu items had files attached. If a required file is missing for an
entry (e.g. vmlinuz.old or initrd.img.old) and it is marked as
"optional" in the config, the section will be skipped. This allows
zipl to install after bootstrapping, as booting on s390 still relies
on the kernel/initrd symlinks in the root directory.
Closes: https://github.com/ibm-s390-linux/s390-tools/pull/2
Signed-off-by: Philipp Kern <pkern@debian.org>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
[sth@linux.ibm.com: adapted patches to latest changes, merged patches]
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This commit makes zgetdump work with /dev/mem again.
zg_seek() should not attempt to verify the given offset for devices before
issuing lseek() because special devices like /dev/mem report 0 in
stat.sb_size.
$ zgetdump -i /dev/mem
zgetdump: Trying to seek past file end "/dev/mem"
Fixes: 11e78cada5 ("zdump: catch attempts to seek past end of file in zg_seek()")
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change will group the commands executed for block and scsi devices
into a section with a comment.
By grouping, we intend to enhance the handling in our runtime.out file
and help to understand the purpose of command groups.
In addition we did get feedback for promoting some commands to the top.
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Sa Liu <saliu@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change will group the commands executed for Z device subsystem
into a section with a comment.
By grouping, we try to enhance the handling in our runtime.out file and
help to understand the purpose of command groups.
In addition we did get feedback for promoting some commands to the top.
lspci -t was added on request of SMEs during the discussions
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Sa Liu <saliu@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Acked-by: Nicklas Schnelle <Niklas.Schnelle@ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change will group the commands executed for overall state into a
first section with a comment.
By grouping, we intend to enhance the handling in our runtime.out file
and help to understand the purpose of command groups.
In addition we did get feedback for promoting some commands to the top.
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Sa Liu <saliu@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change will group the commands executed for crypto, special SW or
specific for a distro into sections with a comment.
By grouping, we intend to enhance the handling in our runtime.out file
and help to understand the purpose of command groups.
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Sa Liu <saliu@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change will group the commands with long output into two
sections with a comment.
By grouping, we intend to enhance the handling in our runtime.out file
and help to understand the purpose of command groups.
We did get feedback for moveing some independent long output files to
the end for speed up the scrolling in the upper parts.
Files which already have a separate output file are bundled in a
separate group at the end.
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Sa Liu <saliu@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change will regroup & reorder the commands executed for network.
By grouping, we intend to enhance the handling in our runtime.out file
and help to understand the purpose of command groups.
In addition we did get feedback for promoting some commands to the top.
"ip -br a" was added on request of SMEs during the discussions
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Sa Liu <saliu@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Acked-by: Sandy Winter <WINTERA@de.ibm.com>
Acked-by: Stefan Raspl <stefan.raspl@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When lshwc is invoked with
# ./lshwc -a :P
lshwc: ioctl S390_HWCTR_START: Invalid argument
#
it returns an error instead of listing all problem state counters
of all online CPUs. The reason is an empty CPU list when it is
omitted and only a counter set is specified. Fix this.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With the introduction of libcpumf/libcpumf.a library there is no need
for this file anymore. Its defines have been moved to
include/lib/libcpumf.h are obsolete.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a function to detect the existence of the CPU Measurement Sampling
Facility on a system.
Add a function which extracts the CPU Measurement Sampling Facility
characteristics on a system, such as
- minimum sample speed
- maximum sample speed
- current CPU speed
- basic sample size in bytes
- diagnostic sample size in bytes
Add a function to detect the existence of the CPU Measurement Sampling
Facility on a system and return the current sampling buffer management
characteristics, such as:
- minimum supported sampling buffer size
- maximum supported sampling buffer size
Add a function to detect the existence of the CPU Measurement Counting
Facility on a system.
Add a function which extracts the CPU Measurement Counting Facility
characteristics on a system, such as
- counter first version number
- counter second version number
- counter set authorization level
Add a function the returm the PMU type number of a CPU Measurement
Facility device driver. These numbers may vary between boots.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
'which' tool is packaged separately from other core tools and there will
likely be scenarios where this check will fail because 'which' is
unavailable.
The "type" is builtin itself and should solve this problem.
For a built in command, we will not suppress error out -> removing 2>&1.
On this replace - in call_run_command - get cmd_type as variable only once
Reported by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
a missing dump2tar command did show misleading error messages
now check it and give clear messages
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
as order itself does not matter for collection of log and config files,
we decided to use alphabetic order for better maintenance
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
critical paths in dump2tar data collection have been already excluded -
this update will sync the excludes in the fall back, running in case
dump2tar is missing or failing on a system
By this we also sync the filtering to match exact the same pattern.
Reported by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Sa Liu <saliu@de.ibm.com>
Fixes: 12c84469fd ("dbginfo.sh: exclude reading page_idle/bitmap sysfs
attribute")
Fixes: b627b8d8e1 ("Initial s390-tools-2.0.0 import")
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
adopt all indents to have the same setting overall - NO logic changes !
tooling: vim used with setting "shiftwidth=8",expandtab,autoindent
commands "gg=G",":retab"
some additional line brakes to stay below 100 chars
and one line added manual before "collect_procfs"
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
chzdev fails to generate correct cio_ignore udev-rules for a CCW device
with device number 0xffff. Also chzdev aborts due to a segmentation
fault when a CCW device with CSSID 0xff is configured. Both issues are
caused by off-by-one errors while interpreting CCW device ID limits
CSSID_MAX, SSID_MAX and DEVNO_MAX.
Fix these errors by correctly interpreting CCW device ID limits.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The endianness handling is already done before, so there is no need for
this function. Let's remove the useless `uint64_to_uint8_buf` function.
Suggested-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Currently, a device type setting being --removed from both the active and
persistent configuration via 'chzdev -t' can result in a scenario where
the setting is not removed from the active configuration (and an error
message is presented) but chzdev still proceeds to remove the setting from
the persistent configuration.
Update this logic so that devtype_remove_settings behaves the same way as
device_remove_settings and only perform the removal when no errors are
encountered.
Reported-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Jason J. Herne <jjherne@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Extend cpictl to report Linux distribution type and version information,
and another sublevel of Linux kernel version data via the system level
field of the SCLP CPI firmware interface. This data is intended to be
decoded by the IBM Z HMC web-UI to provide users a more detailed view of
Linux software levels installed in LPARs.
For this purpose, the format of the 64-bit system level word is extended
in a compatible way, that is, the meaning of existing fields remains
unchanged while additional data is added to previously unused fields.
The new format is defined as 0xabccddeeeeffgghh, where:
- a=statistics flags
- b=distro id
- c=distro major version
- d=distro minor version(s)
- e=kernel sublevel 2
- f=kernel version
- g=kernel patchlevel
- h=kernel sublevel 1
This format will be automatically reported for supported distributions
based on data from /etc/os-release. As before, users can override the
data by specifying a custom system level word in /etc/sysconfig/cpi.
To support manual specification of new data fields, the format of
cpictl's -L command line option and the CPI_SYSTEM_LEVEL environment
variable are extended as follows:
- 0x<level>
Specifies the system level as 64 bit hex value
- [[[flags:]distro_id:distro_version:]kernel_version]
Specifies the system level as list of optional fields. Any field that
is omitted is initialized with automatically determined values.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Acked-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In the past s390 used a fixed command line length of 896 bytes. This has
changed with newer Linux kernels. There is now a parm area indicating
the maximum command line size. This parm area has always been
initialized to zero, so with old kernels this field would read zero and
we must then assume that only 896 bytes are available.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
EditorConfig (https://editorconfig.org/) defines a file format for
defining coding styles. The most common editor do either provide native
support or via plugin support for EditorConfig.
This patch adds an EditorConfig configuration for s390-tools.
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zdev provides path resolution logic to determine which z-specific
devices contribute to the file system mounted at a specific mount point.
This logic is used by command-line option --by-path, but also to
determine the list of devices needed to enable the root file system.
Path resolution fails when a device provides multiple mount points such
as, for example, when using btrfs subvolumes, or when mounting the same
file system at multiple mount points. The failure is caused by zdev
relying on the MOUNTPOINT attribute of lsblk's output which only
contains a single mount point.
Fix this by making use of lsblk's MOUNTPOINTS attribute that contains
the full list of mount points. Note that MOUNTPOINTS was only introduced
with util-linux v2.37, therefore a fall-back to the old format is
needed.
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/129
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Reviewed-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Eduard Shishkin <edward6@linux.ibm.com>
Reported-by: Dan Horák <dan@danny.cz>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
lshwc allows to specify a list of CPUs, for example
# ./lshwc -a 2-3,66-68:b
Date,Time,CPU,CPU_CYCLES(0),INSTRUCTIONS(1),L1I_DIR_WRITES(2),....
2022-01-21,12:51:54,CPU2,68762,20999,283,14157,601,87255
2022-01-21,12:51:54,CPU3,70514,21179,288,10654,586,90188
2022-01-21,12:51:54,CPU4,48504,21018,141,7831,438,78494
2022-01-21,12:51:54,Total,187780,63196,712,32642,1625,255937
#
However, as seen on the output, the CPU list is incorrect. It shows
CPUs 2, 3 and 4 even when only CPU 2 and 3 was requested. CPUs 66, 67
and 68 have not been online and can not be displayed.
What happens is a wrong parameter conversion for the device driver.
CPUs 66, 67 and 68 are passed to the device driver as CPUs 2 3 and 4
and that data is returned.
Fix this and submit a CPU list with correct bit ordering:
# ./lshwc -a 2-3,66-68:b
Date,Time,CPU,CPU_CYCLES(0),INSTRUCTIONS(1),L1I_DIR_WRITES(2),...
2022-01-21,12:59:18,CPU2,50753,18042,246,10972,494,60345
2022-01-21,12:59:18,CPU3,54002,20390,232,9219,511,66033
2022-01-21,12:59:18,Total,104755,38432,478,20191,1005,126378
#
Reported-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Fix Sparse findings like:
warning: symbol 'hash_table_find' was not declared. Should it be static?
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Since version 2.37.x, with the commit 58b510e58 ("libsmartcols: sanitize
variable names on export output"), util-linux changes the output
characters of lsblk, where the ":" is replaced with an "_". Align the
lsblk output parser function in lszdev as per this change.
Signed-off-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Suggested-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reported-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Tested-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a prophylactic check that @filename is not NULL before its
dereference.
This addresses the following warning (which is a false positive,
since stat(2) is never called with NULL file name):
job.c:514:13: warning: argument 1 null where non-null expected
[-Wnonnull]
514 | if (stat(filename, &stats)) {
| ^~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/include/features.h:488,
from /usr/include/errno.h:25,
from job.c:13:
job.c: In function 'check_job_ipl_data':
/usr/include/sys/stat.h:227:12: note: in a call to function
'stat' declared 'nonnull'
227 | extern int __REDIRECT_NTH (stat, (const char *__restrict __file,
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/127
Reported-by: Mario Held <mario.held@de.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Under certain conditions, e.g. in a KVM guest, a write operation
to /sys/firmware/cpi/set can result in an error message to
stderr while the script returns with a zero exit code, indicating
success.
This can cause confusion if cpictl is called by systemd due to
the existence of a writable /sys/firmware/cpi/. Let's silence
cpictl when writing to the /sys/firmware/cpi/set. This is OK, as
nothing can be inferred from failure or success from the perspective
of the invoking Linux instance.
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/120
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The maximum command line length is now dependent on the kernel image
that is loaded. Therefore move the check to check_common_ipl_data().
This function now reads the new kernel image, and check whether the
command line length is in the allowed range.
The command line size limit in zipl is now set to 64k, which is hopefully
enough.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
There are two places where we check whether the initrd and kernel image
is readably. Add one helper function that checks this. This is also a
preparation for the extended command line handling, so that we have to
add that code in only one location.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
There are quite a few places that do:
if (section == NULL) {
error_text("file '%s'", file);
} else {
error_text("file '%s' in section '%s'", file, section);
}
Add a helper function to deduplicate that.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Most of get_*_components() and finalize_*_address_data() is the same
for normal ip, tape and dump types, so we can unify that.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a small helper function to free the common ipl data to avoid
having the same code duplicated multiple times. Also remove the if()
check before the free as it is allowed to pass NULL pointers to free().
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Each ipl type takes at least a kernel image, parameter line and initrd
parameter. This is duplicated all over the place in the current
implementation. To simplify this, add struct job_common_ipl_data which
will hold this data.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This adds support for long command lines. The boot loader stages
will no longer use a fixed limit when copying the command line, but
will use the maximum length from the kernel image that is being loaded.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The assembly code in tape0.S loads the kernel from tape to offset 0x4000.
After loading the kernel it loads the parm file and stores it to the
address provided by zipl. zipl doesn't know about the 0x4000 load offset,
so this address is inside of the kernel image and will silently corrupt
the loaded image in memory.
Fix this by copying the kernel to the final destination before loading
the parm files and initrd. We can do this because we can strip the first
0x10000 bytes like the zipl C code does for non-tape IPL.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
No need to use memcpy() and carry the length around all the time.
Switch to using strlcpy() to simplify the code.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Compiling against fuse3 shows 'incompatible pointer type' warnings due
to the additional function arguments in the new API.
Therefore, adjust the declarations of zdsfs_getattr() and
zdsfs_readdir() to match.
GitHub-ID: https://github.com/ibm-s390-linux/s390-tools/pull/117
Signed-off-by: Graham Inggs <ginggs@debian.org>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Compiling against fuse3 shows 'incompatible pointer type' warnings due
to the additional function arguments in the new API.
Therefore, adjust the declarations of hmcdrv_fuse_getattr(),
hmcdrv_fuse_readdir() and hmcdrv_fuse_init() to match.
GitHub-ID: https://github.com/ibm-s390-linux/s390-tools/pull/117
Signed-off-by: Graham Inggs <ginggs@debian.org>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Compiling against fuse3 shows 'incompatible pointer type' warnings due
to the additional function arguments in the new API.
Therefore, adjust the declarations of cmsfs_getattr(), cmsfs_readdir(),
cmsfs_utimens(), cmsfs_rename() and cmsfs_truncate() to match.
GitHub-ID: https://github.com/ibm-s390-linux/s390-tools/pull/117
Signed-off-by: Graham Inggs <ginggs@debian.org>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When `make install` is called with parallel execution enabled (e.g.:
`-j2`), the dependencies for the necessary directories are not correct.
The `install` target depends on `install_dirs`, which creates the
necessary directories, and needs to run before any of the
`chreipl-fcp-mpath-install*` targets that have prerequisites on these
directories; but the `chreipl-fcp-mpath-install*` targets have
indirectly the same "hierarchy level" as `install_dirs`, and no direct
dependency on it.
Simplified it looks like this:
install: install_dirs chreipl-fcp-mpath-install
chreipl-fcp-mpath-install: chreipl-fcp-mpath-install-udev-rules ...
This works fine with only one job, as `install_dirs` will be satisfied
first, before recursing into any other branch of the dependency tree.
But if we have more than one job, there is nothing in the rules that
prevents Make from working on both branches - `install_dirs` and
`chreipl-fcp-mpath-install` - at the same time, and so
`chreipl-fcp-mpath-install-udev-rules` might run before `install_dirs`,
and end up with a unsatisfied dependency:
make[2]: *** No rule to make target '/usr/lib/udev/rules.d', needed by 'chreipl-fcp-mpath-install-udev-rules'. Stop.
Fix this by adding direct dependencies on `install_dirs` for all the
directory prerequisites of `chreipl-fcp-mpath-install*` targets. This
way Make will need to satisfy `install_dirs` before it can work on any
of the other targets, and so the necessary directories will be created
in the correct order.
Reported-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Tested-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This fixes the sparse warning:
utils/crypto.c:443:68: warning: Using plain integer as NULL pointer
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change allows mocking of print macros in unit tests.
Being able to do this in unit tests, enables us to catch output from
zgetdump functions and test it.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
_zg_err_errno() should save the current errno value before calling any
libc functions because they could change it. Failing to do so, may result
in _zg_err_errno() displaying an incorrect error message.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change allows mocking of error/abort macros in unit tests.
Being able to do this in unit tests, enables us to test error conditions w/o
terminating the unit test runner.
The new error functions do not have "noreturn" attribute because
this would make mocking of them in unit tests impossible. We must not
compile these functions as noreturn because we need to return from them
in unit tests and returning from a noreturn function is an undefined
behavior in the C++ standard!
For more details:
- ISO/IEC 14882:2017, Chapter 10.6.8 "Noreturn attribute""
- https://en.cppreference.com/w/cpp/language/attributes/noreturn.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
It is unusual to write the file output from a tool to stdout. Therefore
add a positional argument to the zgetdump convert action where the
output file can be specified. If no positional argument is given the
output is written to stdout as before and therefore there is no change
in the default behavior.
If the file output already exists an error is returned. The reason for
this is to avoid the situation where an existing dump is accidentally
overwritten by the user.
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The keyword extern before a function declaration is not required with
modern C compilers.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The primary goal of this change is to split DFO ELF module into
multiple smaller ones for better unit testability.
The change doesn't introduce any functionality changes, just shuffling code
around.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The primary goal is to separate DFO code into multiple smaller modules
and make it unit testable.
This refactoring only moved the code around w/o changing any functionality.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Without the fix, readers of syslog / systemd-journal can erroneously
think that dbginfo.sh was incomplete because the last logged step is
"17 of 18".
The fix also makes the missing line appear in dbginfo.log inside the
generated DBGINFO....tgz.
s390-tools v2.17.0 commit b7807d0195 ("dbginfo.sh: Add KVM commands
and rework data collection") turned create_package into a numbered
step.
Snippet of terminal output:
...
17 of 18: Postprocessing
18 of 18: Finalizing: Creating archive with collected data
Collected data was saved to:
>> /tmp/DBGINFO-....tgz <<
Please review all collected data before sending to your service organization.
Syslog before:
...
dbginfo.sh[...]: 17 of 18: Postprocessing
dbginfo.sh[...]: Data collection completed
Syslog after:
...
dbginfo.sh[...]: 17 of 18: Postprocessing
dbginfo.sh[...]: 18 of 18: Finalizing: Creating archive with collected data
dbginfo.sh[...]: Data collection completed
Fixes: b7807d0195 ("dbginfo.sh: Add KVM commands and rework data collection")
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Remove the call to `gzip` before installing the manpage during the
`make install` call. What and if compression is done should be handled by
the distribution tooling.
This also removes a dependency for the build process.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
`chreipl-fcp-mpath-common.sh.in` is never executed, only used as argument
for `source` in the udev helper scripts, so the shebang is unnecessary, and
might be confusing.
Also, tools like `rpmlint` from the rpm software management will complain
about this; e.g.:
s390-tools-chreipl-fcp-mpath.s390x: W: non-executable-script /usr/lib/chreipl-fcp-mpath/chreipl-fcp-mpath-common.sh 644 /bin/bash
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
...to make clear why `off` is initialized to 0 and that all calls to
`dfo_chunk_add` occur for the same reason. For consistency reasons adapt
the `off` value calculation as well.
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Rename the parameter `ptr` of `notes_init` to something meaningful. Since
it points to the start of the notes segment rename it to `segment_start`.
In addition, improve the documentation for the function.
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use `mem_chunk->size` instead of recalculate it. In addition, assign
`p_vaddr` to `p_paddr` to make clear that they must be equal and rename the
parameter `load_offset` to `elf_offset` to make clear that the offset in
the ELF vmcore file is meant. Finally rename `loads_init` to
`load_phdrs_init`.
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This cast makes the macros easier to use without getting compiler warnings.
e.g.
dfo_elf.c: In function 'dfo_elf_init':
dfo_elf.c:343:20: warning: assignment to 'Elf64_Phdr *' from incompatible pointer type 'char *' [-Wincompatible-pointer-types]
343 | phdrs_load = PTR_ADD(phdr_notes, sizeof(Elf64_Phdr) * phdrs_notes_count);
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With gcc < 8.1 the following compile error occurs:
lshwc.c:52:37: error: initializer element is not constant
static unsigned int read_interval = ioctlsleep, cfvn, csvn, authorization;
^~~~~~~~~~
Even though this gcc bug was fixed in gcc 8.1+, change the constant
definition to a #define to make the tool compile with older compilers as
well.
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69960
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66618
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a README.md for chreipl-fcp-mpath and infrastructure to build and
install a man page generated from the README.md. Building and installing
of the man page is only done when supplying the make option `ENABLE_DOC=1`
(per default disabled). The `pandoc` utility is used to build the man page.
The README.md is reused since there is no other tool or command as part of
chreipl-fcp-mpath that needs to be documented in a separate man page.
Also add information about chreipl-fcp-mpath to the main README.md and
CHANGELOG.md.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a dracut configuration file that might optionally be installed along
with the toolset (per default to: $(USRLIBDIR)/dracut/dracut.conf.d). It
causes the udev rules and helper scripts to be included into an initial
ramdisk, when (re-)built with dracut; along with any dependency.
This is only done when passing `HAVE_DRACUT` set to 1 during the
installation (default is 0), e.g.: make install HAVE_DRACUT=1.
Enabling the toolset during the initial ramdisk phase has the advantage
that paths can be audited earlier in the life cycle of this Linux
runtime, e.g. to record the WWID in the ID-file.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In the final step of the udev rules toolset, we either know that the
current event subject is the re-IPL target, a dm-multipath device with
the re-IPL target, or represents the same volume based on its WWID. As
such it is a candidate to replace the current re-IPL target.
The new helper `chreipl-fcp-mpath-try-change-ipl-path` will use the
subject itself - in case it is a single scsi disk -, or the dm-multipath
device, to test whether there is a path in a good state.
"Good state" is based on the zfcp device driver state
(<SDEV>/zfcp_failed, <SDEV>/zfcp_in_recovery), the scsi_transport_fc
port state (<SDEV>/../../fc_remote_ports/rport-*/port_state), and the
scsi device state (<SDEV>/state). A path is only considered if they all
indicate a device in good conditions.
If such a path is found, the helper will try to set it as new re-IPL
target regardless of whether the current re-IPL is still in good shape
or not. This is by design, and done to reduce complexity in further
state checking, and prevention of races with overlapping events in udev
(when executing in parallel workers).
Whenever a new re-IPL target is selected and set in
/sys/firmware/reipl/fcp/{device,wwpn,lun}, the helper will also update
the records in the ID-file - otherwise they might appear as stale, when
they aren't.
This step in the udev rule processing might also result in log messages
written to the syslog (using the utility `logger` from util-linux;
writing to /dev/log).
In case the re-IPL target is changed, a message with level notice is
logged, informing about the new target.
In case no good path is found as part of a dm-multipath device, a
message with level critical is logged, as it might result in a failed
re-IPL if no path is available.
Lastly, when changing the information in /sys/firmware/reipl/fcp/ fails
for whatever reason, a message with level alert is logged, as the
written information might be inconsistent and must be audited/corrected
manually by an operator.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When the first test, whether the current event subject represents the
re-IPL target failed, we attempt a second test, using the recorded WWID
of a previously successful attempt.
The new helper `chreipl-fcp-mpath-is-ipl-vol` utilizes the information
previously stored in the ID-file (read under lock via `flock`).
The recorded WWID is compared with the one of the event subject; if it
matches we know that it addresses the same volume, and might be used as
alternative re-IPL target.
Additionally we also compare the current re-IPL triplet with the one
recorded in the ID-file. If it doesn't match anymore, we assume the
recorded WWID is stale - e.g., because the operator changed the re-IPL
target manually - and the previous check invalid.
If both the WWID match, and the re-IPL triplet is still the same as
previously recorded, the toolset may continue with the subject, and use
it as replacement for the current re-IPL target.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When built with `D=1` chreipl-fcp-mpath automatically creates debug data
for the helpers that run during udev event processing. This is the only
way to effectively debug what is happening during event processing in
chreipl-fcp-mpath.
Collect this data under its default path in /run/udev/.
Also collect the chreipl-fcp-mpath statefile which contains WWID, Device
Bus-ID and LUN of the last path that was detected to represent a FCP
re-IPL device.
Suggested-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When we have identified the current event subject to represent the
re-IPL target, we record its WWID for future identification in a
stateful ID-file (per default: /run/udev/chreiplzfcpmp-ipl-volume-id). In
addition to the WWID, we also record the current re-IPL triplet
(<Dev-Bus-ID>:<WWPN>:<LUN>), so that when that changes - e.g. due to an
operator manually changing the re-IPL target -, we know that the
recorded WWID is stale.
This record may be used in cases when the current re-IPL target is
completely gone from the system, so we can't used it as comparison
object for when events arrive for paths that go to the same volume, but
don't have the same I_T_L nexus. They however have the same WWID. We
may use these (new) paths as replacement for the one that is completely
gone.
The new helper `chreipl-fcp-mpath-record-volume-identifier` uses the
kernel scsi-device attribute `wwid` as source for the WWID (verbatim).
As with reading the re-IPL firmware information, when writing to the
ID-file, a lock is taken via `flock`, to prevent overlapping
writes/reads to the file.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Now that we know that the system uses s390x IPL and the current re-IPL
target is from FCP, whenever we get a udev event that indicates a path
state changed, we have to figure out whether it affects the path that is
currently set as re-IPL target, or goes to the same volume (so we might
use it as alternative re-IPL path).
Add a new helper `chreipl-fcp-mpath-is-ipl-tgt` for this task.
Based on the information provided in
/sys/firmware/reipl/fcp/{device,wwpn,lun} it figures out whether the
current event subject has the same triplet <Dev-Bus-ID>:<WWPN>:<LUN> (in
T10 SCSI: I_T_L nexus), or whether one of its parts (in case of
dm-multipath) has.
If true, we know for sure, that we deal with an event affecting the
current re-IPL setting, and continue.
When accessing the re-IPL firmware information a lock is taken via the
`flock` utility (from util-linux). This is done so concurrent changes
from the toolset, that are added in a later commit, don't result in
inconsistent reads.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add second filter stage to the udev rules: only continue with the
toolset, if the system uses s390x IPL and if the current re-IPL target
is from FCP.
For the test, whether the next re-IPL target is planned to be done from
FCP, a helper `chreipl-fcp-mpath-is-reipl-zfcp` is added, and the
information in /sys/firmware/reipl/reipl_type is used.
If either one or both of the tests are not true, the udev rules will
exit.
For debugging, the toolset can be built with the Make variable `D` set
to 1 (e.g.: make D=1; make install D=1). This causes the toolset to
write shell traces of the helper to $(DEBUGOUTDIR) (per default:
/run/udev); otherwise no output is generated. `dbginfo.sh` will be
changed to collect this data automatically.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a new tool to s390-tools: chreipl-fcp-mpath. A toolset based on udev
to change the active re-IPL target if the current target goes into an
error-state, due to for example a pulled cable.
Add the base infrastructure consisting of a Makefile and a base set of
udev rules. Integrate the Makefile into s390-tools Make machinery.
The rules filter events base on the event subject type (dm-multipath,
scsi disk), udev action (change, or add), and device mapper action
(path failure, or path reinstated). No further processing is done right
now in case of a match; if neither of these criteria matches, the
rules exit, and the toolset won't be invoked.
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
running mpio data disks with offline paths might cause dbginfo
to hang up on the command "blockdev --report" -> we remove
it to ensure a non blocking data collection.
In addition we add a second kill step with the same timeout
value to retry stopping a command
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Reviewed-by: Mario Held <mario.held@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Always clear the COMMAND_LINE_EXTRA region, also in case sclp_setup()
fails. If the region is not cleared properly, there might be junk
in there after a reboot, causing confusion for the kernel later.
While we're at it, also switch to COMMAND_LINE_EXTRA_SIZE instead
of COMMAND_LINE_SIZE, since that constant is more appropriate here.
Closes: https://github.com/ibm-s390-linux/s390-tools/pull/122
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With an upcoming patch the output target can be different to `stdout`.
Therefore, the term `ZG_ACTION_STDOUT` is no longer generic enough. In addition,
rename `do_write` to `do_copy` to match the enum name.
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Currently, for some commands (e.g. `zgetdump unmount`) unused positional
arguments are silently ignored, for some commands (`zgetdump mount`) they're
not. Therefore let's fix this by always returning an error message and exit code
if too many positional arguments are given.
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In case the hsci tool was upgraded on a running system, it can happen
that an hsci interface exists, that was created by an older version of
the tool (and thus has no veth pair). Add support to show and delete
such an hsci interface.
Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Without kernel commits
f7936b7b2663 s390/qeth: Update MACs of LEARNING_SYNC device
4e20e73e631a s390/qeth: Switchdev event handler
60bb1089467d s390/qeth: Register switchdev event handler
HSCI cannot be used with multiple MACs, but the single MAC usecase needs
to be supported even with old kernel. So manually setting the same single
MAC on hsci, HiperSockets and the external interface is still required.
The hsci itself is not a bridgeport, so a static forwarding rule in the
bridge is also required. It seems cleaner to use the initial MAC of the
veth hsci as single static MAC, than the initial MAC of the HiperSockets
interface.
Even with an updated kernel a single static MAC interface, that is not
subject to ageing and re-learning, is beneficial when hsci is used as
a single MAC interface.
Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The first implementation of HSCI (HiperSockets Converged Interface)
supports only a single static MAC address per HSCI and thus constructs
like MACVTAP or OpenVSwitch cannot be used on top of it.
Kernel commits
f7936b7b2663 s390/qeth: Update MACs of LEARNING_SYNC device
4e20e73e631a s390/qeth: Switchdev event handler
60bb1089467d s390/qeth: Register switchdev event handler
changed the behaviour of a LEARNING_SYNC HiperSockets (HS) device
such that any MAC address that is added or learned on a non-isolated
bridgeport of the HSCI bridge is set on the LEARNING_SYNC HS device and
its isolated siblings.
So no new trigger is required for a HS device with the bridge attribute
LEARNING_SYNC to support multiple MACs on a northbound HSCI interface.
However the first implementation of the hsci tool uses the bridge itself
as northbound HSCI interface. This is not treated as a real bridgeport.
So we add a veth bridgeport as northern link for HSCI.
Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Instead of checking for ideal starting conditions, try to achieve
the desired outcome and accept conditions that do not prevent it.
This allows the user to get to a defined state, even if the
hsci interface was manipulated by other means than the hsci tool,
e.g. by ip commands or bridge commands
Minor cleanup:
hsci show:
- display PNET_ID, even if only one hsci port has one
(e.g. if External is a bond)
- only 1 header for mulitple hscis
hsci add:
- sanitize error messages
- allow nonsense HSCI of 2 HiperSockets
Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Don't include dfi_mem_chunk.h in dfi.h because not every module including
dfi.h needs it.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Separate the definition of a struct from the declaration of a global
variable of the type. To improve readability.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Create a separate C header for STDOUT to separate independent things and
make it more readable.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Set DFO format after the parsing of command-line arguments is done.
The command-line argument parsing should be independent of DFO.
This improves testability of the command-line argument parsing and
separates independent things from each other.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make the C module responsible for the parsing of command-line arguments
independent of other global variables. This improves its testability.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Get rid of C headers which include everything. This is bad for testing and
maintainability. C files should include only what they need and no more.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Create a separate C header for FUSE to separate independent things and
make it more readable.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Create a separate C header for DFI VMCOREINFO to separate
independent things and make it more readable.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Replace magic values with descriptive names to improve readability.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To prevent segmentation faults when the function dfi_mem_phys_read() is
given invalid parameters, check that the given memory range falls within
a known physical memory chunk, just like it is done for dfi_mem_virt_read().
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The new name describes the purpose of the function better and
emphasizes its difference with the function dfi_mem_phys_read().
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Always use the function dfi_mem_read_rc() which verifies that the given
address and size fall within a valid memory chunk. We cannot trust
user's input.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The function dfi_mem_chunk_virt_add() is used only internally and should
not be exported to prevent misuse.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The keyword extern is redundant for function declarations in C headers.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Set stage3_parms.envblk_len to 0, if add_envblk is false.
This fixes segmentation fault when performing job with
not defined zIPL environment (e.g. installing SCSI dump)
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Tested-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
PCI and crypto devices defined using the IBM Z Dynamic Partition Manager
(DPM) will start in an offline/unconfigured state that requires manual
intervention before the associated Linux function can be used. This
results for example in PCI networking interfaces being unavailable in a
distribution installer system.
Fix this by providing a udev rule and initial RAM-disk logic that
automatically enables PCI and crypto devices either during boot, or when
they are defined at run-time. This processing can be suppressed by
specifying the "rd.zdev=no-auto" parameter on the kernel command line.
Auto-configuration is limited to Linux running in DPM LPARs because
PCI-functions and crypto devices defined by DPM are always intended for
use by a single LPAR only.
For Linux running in classic-mode LPARs or virtual machines, leaving PCI
and crypto devices in an offline state may be useful to allow defining a
device as available to multiple systems, where only the system that
intends to make use of the device should enable it.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add an internal tool intended to be used by udev rules to determine
zdev and hypervisor related system information in key=value format.
Supported keys are:
ZDEV_NEST_LEVEL=n
Virtualization nesting level of running system
ZDEV_HYPERVISOR_<n>=LPAR|z/VM|KVM/Linux
Type of hypervisor that provides virtualization at nesting level <n>
ZDEV_IS_DPM=0|1
Indicator if top-level LPAR is managed by Dynamic Partition Manager
ZDEV_NO_AUTO=0|1
Indicator if auto-configuration is requested
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zgetdump is for analyzing VM coredumps and not user space coredumps therefore
return an error in case of a user space coredump.
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add new tool named 'sclpdbf'. This script displays the contents of
s390dbf sclp debug logs in human-readable format.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Added a tool to operate with zIPL environment, currently
installed in the boot record. All modifying operations result
in atomic update of the installed environment block, bypassing
the file system driver.
Operations currently supported by zIPL environment editor:
. set/unset environment variables,
. print/reset environment
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
from disk.c to a dedicated source file fs-map.c, so that the new
zipl-editenv tool will be also able to use it.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Before starting the kernel:
1. parse environment block (if any) and create a hash table
of all found pairs (NAME, VALUE);
2. scan the command line, which already contains extra-parameters
(if any), and for each found ${NAME} replace it with VALUE, as
found in the hash table by NAME.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When installing a boot record for "ipl" and "menu" job, always add
an environment block as a boot component located in bootmap file
at offset alinged on file system block size boundary. When adding,
first try to import environment from a file at location specified
by newly introduced "--environment" zipl option, or by default at
"/etc/ziplenv". If nothing was imported, then add a blank environment
block. Optionally print the content of the environment block.
Store environment block size and address (as of boot component) in
stage3_parms.
Change interface of add_ipl_program(): add 2 additional arguments:
a pointer to bootmap file name and a predicate indicating if we
need to add environment block as a boot component.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Provide helper functions disk_write_block_buffer_align()
and add_component_buffer_align() to align offset of the record
on specified block size boundary and to save the offset, where
the record was made at.
This will be used by a later patch.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
. expand interface of disk_get_blocklist_from_file() to get
block pointers for a specific range of data within a file.
. provide a helper function add_component_file_range() to add a
specific range of data within a file as a boot component.
This will be used by a later patch.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This fixes the following Valgrind finding:
==28040== Syscall param rt_sigaction(act->sa_flags) points to uninitialised byte(s)
==28040== at 0x48C98B0: __libc_sigaction (sigaction.c:58)
==28040== by 0x10D507: sig_handler_init (zgetdump.c:58)
==28040== by 0x10D507: main (zgetdump.c:179)
==28040== Address 0x1ffefffc84 is on thread 1's stack
==28040== in frame #0, created by __libc_sigaction (sigaction.c:43)
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Fix the following incorrect help text line:
-T, --file-timeout (null) Stop reading file after SEC seconds
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Sanity check ELF notes descriptor size before reading it. This prevents
stack buffer overflows in case a dump contains invalid ELF notes.
Instead of reading a note's descriptor into a temporary buffer on stack,
read it directly into the buffer given to nt_read() but also provide
a maximum length of the given buffer to nt_read() in order to prevent
overflows.
This problem was found with valgrind and AFL fuzzing + ASAN.
AFL + ASAN findings:
[root@t83lp49 s390-tools]# ./zdump/zgetdump -iVVVVV ~/zgetdump-fuzzing/findings/crashes/id\:000008\,sig\:06\,src\:000007\,op\:arith8\,pos\:67\,val\:+3
TRACE: DFI initialization
DEBUG: DFI trying s390tape
DEBUG: DFI s390tape returned with rc -19
DEBUG: DFI trying devmem
DEBUG: DFI devmem returned with rc -19
DEBUG: DFI trying s390mv_ext
DEBUG: DFI s390mv_ext returned with rc -19
DEBUG: DFI trying s390mv
DEBUG: DFI s390mv returned with rc -19
DEBUG: DFI trying s390_ext
DEBUG: DFI S390 extended initialization
DEBUG: DFI s390_ext returned with rc -19
DEBUG: DFI trying s390
DEBUG: DFI S390 initialization
DEBUG: DFI s390 returned with rc -19
DEBUG: DFI trying lkcd
DEBUG: DFI lkcd returned with rc -19
DEBUG: DFI trying elf
DEBUG: DFI ELF initialization
DEBUG: DFI ELF e_phnum 11
DEBUG: DFI ELF p_type[0] 0x4
DEBUG: DFI ELF n_type 0x0
AddressSanitizer:DEADLYSIGNAL
=================================================================
==208548==ERROR: AddressSanitizer: stack-overflow on address 0x03ffef05d000 (pc 0x0000010051b0 bp 0x03fff107dc40 sp 0x03ffef05dac8 T0)
#0 0x10051b0 (/root/s390-tools/zdump/zgetdump+0x10051b0)
SUMMARY: AddressSanitizer: stack-overflow (/root/s390-tools/zdump/zgetdump+0x10051b0)
==208548==ABORTING
valgrind findings:
==56423== Source and destination overlap in memcpy(0x4a86d38, 0x4a875e0, 4096)
==56423== at 0x4839F86: memcpy (in /usr/lib/s390x-linux-gnu/valgrind/vgpreload_memcheck-s390x-linux.so)
==56423== by 0x114253: memcpy (string_fortified.h:34)
==56423== by 0x114253: nt_read (dfi_elf.c:95)
==56423== by 0x1145FF: nt_s390_prefix_read (dfi_elf.c:195)
==56423== by 0x1145FF: pt_notes_add (dfi_elf.c:259)
==56423== by 0x1145FF: dfi_elf_init (dfi_elf.c:326)
==56423== by 0x112A57: dfi_init (dfi.c:1212)
==56423== by 0x10D663: do_dump_info (zgetdump.c:127)
==56423== by 0x10D663: main (zgetdump.c:182)
==56423==
==56423== Invalid write of size 8
==56423== at 0x4839E28: memcpy (in /usr/lib/s390x-linux-gnu/valgrind/vgpreload_memcheck-s390x-linux.so)
==56423== by 0x114253: memcpy (string_fortified.h:34)
==56423== by 0x114253: nt_read (dfi_elf.c:95)
==56423== by 0x1145FF: nt_s390_prefix_read (dfi_elf.c:195)
==56423== by 0x1145FF: pt_notes_add (dfi_elf.c:259)
==56423== by 0x1145FF: dfi_elf_init (dfi_elf.c:326)
==56423== by 0x112A57: dfi_init (dfi.c:1212)
==56423== by 0x10D663: do_dump_info (zgetdump.c:127)
==56423== by 0x10D663: main (zgetdump.c:182)
==56423== Address 0x4a86ee0 is 0 bytes after a block of size 912 alloc'd
==56423== at 0x483675E: calloc (in /usr/lib/s390x-linux-gnu/valgrind/vgpreload_memcheck-s390x-linux.so)
==56423== by 0x10E71D: zg_alloc (zg.c:93)
==56423== by 0x114793: nt_prstatus_read (dfi_elf.c:123)
==56423== by 0x114793: pt_notes_add (dfi_elf.c:234)
==56423== by 0x114793: dfi_elf_init (dfi_elf.c:326)
==56423== by 0x112A57: dfi_init (dfi.c:1212)
==56423== by 0x10D663: do_dump_info (zgetdump.c:127)
==56423== by 0x10D663: main (zgetdump.c:182)
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Suggested-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reported-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Had a confusing decoded trace record:
Timestamp : ...
Area : SCSI
...
Tag : rsl_err
...
SCSI result : 0x00000002
SCSI opcode : 1a...
FCP rsp inf cod: 0x00
FCP rsp IU : 00000000 00000000 00000a02 000000ff
00000012 00000000
^^
and was wondering where that sense data would be or if the
corresponding PAY record "fcp_riu" was lost due to trace area wrapping.
So the fix here adds the following two lines to the same trace record:
FCP rsp IU len : 42
FCP rsp IU all : record not available anymore.
Similarly, the following LUN reset Task Management Function response
was missing the last two lines regarding the full FCP response IU
and its optional part with the FCP response info:
Timestamp : ...
Area : SCSI
...
Tag : lr_okay
...
FCP rsp inf cod: 0x00
FCP rsp IU : 00000000 00000000 00000100 00000000
00000000 00000008
^^
FCP rsp IU len : 32
FCP rsp IU all : record not available anymore.
As of today, depending on the kernel version we have
either fcp_sns or fcp_riu PAY trace records. See Linux kernel v4.14 commit
12c3e5754c80 ("scsi: zfcp: fix payload with full FCP_RSP IU in SCSI trace
records").
Always print non-zero payload length independent of payload type.
Actually this also fixes a regression for wrapped away fcp_sns.
If there is no fcp_sns it either wrapped away or it's a kernel with fcp_riu.
Only print fcp_riu if it contains more content than the preceding
unconditional fixed-size field "FCP rsp IU".
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
If the while loop in mem_chunks_add_ext() is never executed, then
the stack variable containing the dump segment header will never be
initialized.
clang's static code analyzer reports the following problem:
$ make CC="clang --analyze" -C zdump
dfi_s390.c:157:6: warning: Branch condition evaluates to a garbage value [core.uninitialized.Branch]
if (!dump_segm.stop_marker)
^~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reported-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use util_log from libutil to output various log messages that can be helpful
during problem analysis.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Replace the incorrect %d format specifier with %u one for unsigned integer
types.
This issue was found with AFL fuzzing.
$ ./zdump/zgetdump -i ~/input.bin
General dump info:
Dump format........: s390
Version............: -2147483647
System arch........: s390x (64 bit)
CPU count (online).: 32768
Dump memory range..: 1 MB
Memory map:
0000000000000000 - 00000000000f423f (1 MB)
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
We found situations, where the debug commands from dbginfo.sh
did overwrite some debug files (e.g. OSA s390dbf) - so we change
the collection order
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Before reading data in dfi_vmcoreinfo_init(), check the validity
of the memory range. Otherwise this can result in a segmentation fault when
zgetdump is given a very small dump, e.g. S390 DASD single-volume dump of
size 0x10 bytes.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Before reading data in os_info_get(), check the validity of the memory
range. Otherwise this can result in a segmentation fault when zgetdump is
given a very small dump, e.g. S390 DASD single-volume dump of size 0x10
bytes.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Before reading data at addresses 0x10418 and 0x10420 in kdump_init(),
validate the validity of the memory range. Otherwise this can result in
a segmentation fault when zgetdump is given a very small dump, e.g.
S390 DASD single-volume dump of size 0x1000 bytes.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The dfi_vmcoreinfo_init() function might be called on a dump of a non-ELF
format because the DFI goes through all supported dump formats when first
trying to identify of what dump format the given input is. Therefore,
we must be very careful in interpreting read data and ensure that it
makes sense before accessing or using it.
This commit prevents a potential overflow of a stack buffer in
dfi_vmcoreinfo_init() if note.n_namesz is bigger than the stack buffer.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
- adding new print function for simpler output customizing
- consolidate some output files
- sync structures and use of checks
- some formatting updates
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The function mem_chunks_add_ext() allocates but does not release memory
if it encounters unexpected errors during dump segment reading.
Allocate memory passed to dfi_mem_chunk_add() right before calling
the function. This way there is no need for a cleanup anymore.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reported-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
- clenaup outdated kernel - we no longer have 2.x kenrel in support
- add timeout on command excution - ensure to not miss all data for a
single command to hang
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When integrity support is enabled for a LUKS2 volume, then the LUKS2
volume key may include the integrity key concatenated to the secure key
(dependent on the integrity type).
Get the integrity key size from the LUKS2 header (if integrity support
is enabled) and calculate the size of the secure key when performing
operations with the secure key only.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
According to the ELF specification, the file size of a PT_LOAD
memory segment might be less than its memory size. In that case,
the remaining bytes must be filled with zeros. The ELF DFI implementation
didn't handle this case correctly. Such ELF core files might be produced
by the makedumpfile tool.
From elf.5 man page:
------------------------------------------------------
PT_LOAD
The array element specifies a loadable segment,
described by p_filesz and p_memsz. The bytes
from the file are mapped to the beginning of the
memory segment. If the segment's memory size
p_memsz is larger than the file size p_filesz,
the "extra" bytes are defined to hold the value
0 and to follow the segment's initialized area.
The file size may not be larger than the memory
size. Loadable segment entries in the program
header table appear in ascending order, sorted
on the p_vaddr member.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Before attempting to read ELF program headers, the file must be positioned
at the offset of the first program header as specified in the ELF header.
Furthermore, read all ELF program headers with one call to zg_read(). This
has the advantage of not having to restore the file offset every time
after it is changed in pt_notes_add() and other functions which need to
read the same file as well.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This is a false positive because IBM Z architecture supports unaligned
memory accesses.
dfo_s390.c: In function ‘df_s390_dump_init’:
dfo_s390.c:218:29: warning: taking address of packed member of ‘struct df_s390_hdr’ may result in an unaligned pointer value [-Waddress-of-packed-member]
218 | timeval2tod(&dh->tod, dfi_attr_time());
| ^~~~~~~~
dfo_s390.c:219:29: warning: taking address of packed member of ‘struct df_s390_em’ may result in an unaligned pointer value [-Waddress-of-packed-member]
219 | timeval2tod(&em->tod, dfi_attr_time());
| ^~~~~~~~
dfo_s390.c:222:29: warning: taking address of packed member of ‘struct df_s390_em’ may result in an unaligned pointer value [-Waddress-of-packed-member]
222 | timeval2tod(&em->tod, dfi_attr_time_end());
| ^~~~~~~~
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When creating a bootmap, look for zipl components (stage3.bin and
zipl helpers) in a directory specified by S390TOOLS_LIBDIR shell
environment variable. If that variable is not set, then look for
them at the default location.
In verbose mode print the name of s390-tools library directory,
where the search was implemented.
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add generic functions
util_sysdir();
util_sysdir_path()
and their special cases supplied for users:
util_libdir();
util_libdir_path();
util_datadir();
util_datadir_path()
The function util_sysdir() determines the absolute name of a
s390-tools system directory. It could be data, or library directory.
The function util_sysdir_path() determines the absolute name of a
file installed in the s390-tools system directory.
The function util_libdir() determines the absolute name of a
s390-tools library directory. By default that name is defined
by the compile-time macro TOOLS_LIBDIR (/lib/s390-tools).
Users can specify an override by setting environment variable
S390TOOLS_LIBDIR.
The function util_libdir_path() determines the absolute name of a
file installed in the s390-tools library directory.
The function util_datadir() determines the absolute name of a
s390-tools system data directory. By default the name is defined
by the compile-time macro TOOLS_DATADIR (/usr/share/s390-tools/).
Users can specify an override by setting environment variable
S390TOOLS_DATADIR.
The function util_datadir_path() determines the absolute name of a
file installed in the s390-tools data directory.
The ability to override the setting for TOOLS_LIBDIR and
TOOLS_DATADIR is required for implementing tests on tools that are
not installed in their default system path locations.
Signed-off-by: Eduard Shishkin <edward6@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
lscss shows device number as 'None' if the corresponding device in the
subchannel is non-operational or not connected. Instead, show the
device-no derived from the new dev_busid attribute which provides the
value of device_id irrespective of the device availability.
i.e the current lscss shows
Device Subchan. DevType CU Type Use PIM PAM POM CHPIDs
----------------------------------------------------------------------
0.0.0000 0.0.0000 3390/0c 3990/e9 f0 a0 ff 32333435 00000000
none 0.0.00c7 f0 a0 5f 32333435 00000000
none 0.0.00c8 f0 a0 5f 32333435 00000000
here the subchannels 0.0.00c7 and 0.0.00c8 do not have an operational
device on them. Hence the device is shown as none.
With this patch,
Device Subchan. DevType CU Type Use PIM PAM POM CHPIDs
----------------------------------------------------------------------
0.0.0000 0.0.0000 3390/0c 3990/e9 f0 a0 ff 32333435 00000000
0.0.74c8 0.0.00c7 f0 a0 5f 32333435 00000000
0.0.74c9 0.0.00c8 f0 a0 5f 32333435 00000000
The device-no is shown as 0.0.74c8; Where the DevType field still
indicates that the device is not present.
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>
lszcrypt now evaluates and displays the 'status' in the
'STATUS' column if the vfio provides this sysfs attribute.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The 'STATUS' field was within the code used as 'online'
record because it showed the online state. But now it
displays some more info like 'deconfig' and maybe even
more in the future. So now the code also uses a 'status'
record for this info.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Slight rework of the lszcrypt code with:
- a queue or card in deconfigured mode is displayed
as 'deconfig' regardless which driver is bound to
the device.
- a queue or card in 'deconfig' mode does not show
any counters any more but displays a '-' instead.
- a queue not bound or bound to something else as
the 'default' zcrypt dd does not show any counters
but displays a '-' instead.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The command
# vmur punch -r rhcos-live-kernel-s390x -u etpgxku -N kernel.img
Reader file with spoolid E YO created and transferred to ETPGXKU.
#
did not indicate an error when executed. However the file was not
transferred and this error message went unnoticed:
RPIMGR032E YOU ARE NOT AUTHORIZED TO CLOSE TO TATAC05.ETPGXKU
HCPCSL007E Invalid userid - ETPGXKU
Fix this by checking the CP return code and print the error message
returned by CP. This is the behavior now:
# vmur punch -r rhcos-live-kernel-s390x -u etpgxku -N kernel.img
vmur: CP command failed with rc=7
RPIMGR032E YOU ARE NOT AUTHORIZED TO CLOSE TO TATAC05.ETPGXKU
HCPCSL007E Invalid userid - ETPGXKU
#
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/119
Reported-by: Vance Morris <vmorris@us.ibm.com>
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Tested-by: Vance Morris <vmorris@us.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Complements v2.9.0 commit 4036e80b26 ("lszfcp: add new output marker
for non-good FCP devices (hosts)").
Otherwise "linkdown" shows as "failed" because linkdown happens to be
the one special case that also sets the failed status flag of the
FCP device (host). Linkdown is kind of benign compared to other cases
where the FCP device is in failed status.
Examples:
$ lszfcp -He
0.0.194a host3 linkdown
$ lszfcp -HeV
/sys/devices/css0/0.0.0004/0.0.194a linkdown
/sys/class/fc_host/host3
/sys/class/scsi_host/host3
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The DFI memory chunk manager cannot handle PT_LOAD segments of size 0.
This can lead to a NULL pointer access in mem_chunk_find() in case
zgetdump is given an invalid VMCORE ELF file as input which contains
only zero-sized ELF PT_LOAD segments. Instead of crashing, show an error
message and terminate gracefully.
The TELA test dump_with_zero_sized_segments.sh can reproduce this bug.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Zipl.conf is collected via /etc/*.conf in CONFIGFILES but BLS snippets
do not appear in /etc/zipl.conf.
Complements v2.5.0 commit d280e5b1fa ("zipl: Add BootLoaderSpec support")
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some libraries depend on other libraries, so split the libraries
into two groups: base libraries that do not depend on any others, and
the remaining libraries, that may depend on the base libraries.
Ensure that the base libraries are built first, before starting to
build the remaining libraries. The tools are only built once the base
libraries and the remaining libraries have been built.
This fixes a possible build problem with libseckey when parallel build
is used. The libseckey library is used by libekmfweb and the zkey KMIP
plugin (zkey-kmip). With parallel build both are built in parallel,
and thus both trigger the build of libseckey. This can lead to the
situation that libseckey is built twice at the same time, which can lead
to build failures (corrupted archive, etc).
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
overlay.o was being recompiled during the install step because
it depended on the target check-dep-fuse which is phony and,
therefore, always outdated. The solution is to create an empty file
for the target check-dep-fuse after its successful completion. This
prevents make from rebuilding overlay.o during installation.
Closes: https://github.com/ibm-s390-linux/s390-tools/pull/118
Fixes: 5a7d2a58c8 ("hsavmcore: Fix fuse dependency checking")
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Suggested-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use LONG_BIT provided by C standard in <limits.h>.
Fixes the following compile errors with buildroot:
lshwc.c: In function ‘parse_cpulist’:
lshwc.c:295:15: error: ‘__BITS_PER_LONG’ undeclared (first use in this function)
295 | no_a = i % __BITS_PER_LONG;
| ^~~~~~~~~~~~~~~
Fixes: 27a562da0a ("cpumf/lshwc: Program to extract complete counter sets")
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use standard 32- and 64-bit C types.
Fixes the following compile errors with buildroot:
In file included from lshwc.c:41:
lshwc.h:61:2: error: unknown type name ‘__u64’
61 | __u64 version; /* Version of interface */
| ^~~~~
lshwc.h:62:2: error: unknown type name ‘__u64’
62 | __u64 data_bytes; /* # of bytes required */
| ^~~~~
lshwc.h:63:2: error: unknown type name ‘__u64’
63 | __u64 cpumask_len; /* Length of CPU mask in bytes */
| ^~~~~
lshwc.h:64:2: error: unknown type name ‘__u64’
64 | __u64 *cpumask; /* Pointer to CPU mask */
| ^~~~~
lshwc.h:65:2: error: unknown type name ‘__u64’
65 | __u64 counter_sets; /* Bit mask of counter set to get */
| ^~~~~
lshwc.h:69:2: error: unknown type name ‘__u32’
69 | __u32 set; /* Counter set number */
| ^~~~~
Fixes: 27a562da0a ("cpumf/lshwc: Program to extract complete counter sets")
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
KMIP values of type BYTE-STRING are represented as hex values when XML
or JSON encoding is used. Do not drop any leading zero bytes, if the
value has them.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
util_path_sysfs() allocates memory for the created path which must be
freed by the user.
Fixes: 8023a72b11 ("libdasd: Provide function to read ese sysfs attribute")
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
List (find) and import a key that exists in a KMIP server into the zkey
repository. Retrieve the key attributes of a key stored in a KMIP server,
and extract the zkey properties from it.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When a key is removed from the zkey repository, it is usually not removed
from the KMIP server. Optionally, the state of a KMIP key can be changed
when it is removed by zkey.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Change the attributes of a key at the KMIP server when the properties
of a key are changed in zkey.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow to generate a new AES key at the KMIP server, and retrieved it.
Keys are always retrieved in wrapped format, wrapped by the public RSA
wrapping key used by the KMIP plugin. The KMIP plugin can then unwrap
the retrieved key with its secure private RSA key, and can then import
the retrieved AES key into the zkey repository as secure key. That way,
the retrieved AES key is never in clear, once it leaves the KMIP server.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When the master keys of the configured APQNs are changed, the secure
identity key as well as the secure wrapping key must be re-enciphered.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Keys retrieved from a KMIP server are wrapped with an RSA key. The
RSA wrapping key is a secure key generated by the KMIP plugin. The
public RSA key is registered at the KMIP server, and the KMIP server
can then wrap the keys with it. The KMIP plugin can unwrap the retrieved
key with the private RSA key and import it as secure key into the zkey
repository.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Configure a connection to a KMIP server, establish trust between
server and client, query the KMIP server and verify if it supports
the required KMIP operations. Also, the profile to be used with this
KMIP server is tried to matched. If not specified in the profile,
the KMIP protocol version to be used is discovered.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
KMIP server profiles (not to be confused with profiles defined by
the KMIP standard) define how the KMIP plugin talks to the KMIP
server. Profiles can be used to configure certain KMIP usages for
certain KMIP servers.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Lines with '#' as first character are treated as comments, and are
skipped. Empty lines are also skipped.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For the client certificate to authenticate with the KMIP server,
a certificate signing request (CSR) or self signed certificate can be
generated. The generated CSR or certificate is signed with the secure
identity key of the KMIP plugin.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Authentication with KMIP servers is based on TLS client authentication.
To establish trust between zkey and the KMIP server, a client certificate
must be registered with the KMIP server before a client can communicate
with it.
The signature key of the client's certificate is a secure key created by
zkey. It can either be an RSA or an ECC key, using either CCA or EP11
type crypto adapters configured with the KMIP plugin.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The KMIP plugin supports CCA and EP11 APQNs, but only it can only be
configured with one type. Once configured with one type of APQN, it accepts
only APQNs of the same type.
It supports key types matching to the APQn type it is configured with.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support to integrate the external key management systems based
on the KMIP protocol (KMIP: OASIS Key Management Interoperability
Protocol) into zkey by providing a zkey KMS plugin for KMIP. The
plugin communicates with KMIP servers using libkmipclient over a
TLS or HTTPS connection.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some functions are likely to be used by multiple plugins. Move them
into zkey common code (plugin-utils.c), this module is then linked
into the plugins.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
KMS plugins that support multiple key types need to know which
key type to produce with the kms_import_key() function. Extend
the plugin API to allow a plugin to provide a kms_import_key2()
function, that accepts the key type as additional parameter.
Existing plugins do not need to be changed, as the old function
kms_import_key() still exists, and is still used when the plugin
does not provide a kms_import_key2() function.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
libkmipclient is a shared library that provides an KMIP client to
communicate with an KMIP server. KMIP stands for Key Management
Interoperability Protocol, and is an extensible communication protocol
that defines message formats for the manipulation of cryptographic
keys on a key management server.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The libseckey is a secure key library to perform secure key operations
with OpenSSL. It provides a framework to create OpenSSL PKEYs with a
secure key attached. Such a PKEY contains the public key parts in clear,
but the private key as secure key blob. Only the private key operations
are actually performed with the secure key, public key operations are
performed in software by OpenSSL. It supports CCA and EP11 secure
keys for RSA and ECC crypto operations.
Because many PKEY method related functions are deprecated since OpenSSL
3.0, two versions of the OpenSSL secure key support are needed. One (using
a PKEY method override) for OpenSSL 1.1.1, and another one (using an own
OpenSSL provider) for OpenSSL 3.0 and later. The desired implementation is
selected automatically at compile time, using OpenSSL version defines. The
interface of both implementations is the same, so a user does not need to
care which one is used.
Reviewed-by: Juergen Christ <jchrist@linux.ibm.com>
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The new user-defined function allows to check for C headers in
make's conditional statements.
Example:
ifeq ($(call check_header_prereq,"systemd/sd-daemon.h"),yes)
$(warning "systemd support enabled")
else
$(warning "systemd support disabled")
endif
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The callback function used for CURLOPT_SSL_CTX_FUNCTION requires
that OpenSSL is used as SSL backend.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add the genprotimg flags `--(enable|disable)-pckmo` flag to enable/disable the
support for PCKMO-encrypt-DEA/TDEA/AES/ECC-key functions.
Acked-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow the PCKMO-encrypt-DEA/TDEA/AES/ECC-key functions to be used in a protected
virtualization guest by default.
Acked-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The new name reflects that it's a plaintext control flag. While at it, use a
left shift operation to calculate the value.
Acked-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The HSA size, which limits the address space in kdump case, is not constant
and depends on the underlying architecture.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The util_arch module is supposed to provide general information about
the underlying architecture of the machine in use.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
... to avoid vulnerabilities that could occur if set-user-ID or
set-group-ID programs accidentally trusted the environment.
zkey is usually NOT a set-user-ID or set-group-ID program, but its
safer to use secure_getenv() anyway.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The only user of util_proc_mnt_get_entry() so far was util_path_sysfs().
With the simplified version there is no user left. Remove
util_proc_mnt_get_entry() and related code.
Reviewed-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Using util_path_sysfs always leaves 5 bytes of memory unfreed as the
value for the sysfs mount point is stored in a static variable to avoid
multiple queries of /proc/mount.
$ valgrind ./util_path_example sysfs
==3629315== Memcheck, a memory error detector
==3629315== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==3629315== Using Valgrind-3.15.0 and LibVEX; rerun with -h for copyright info
==3629315== Command: ./util_path_example sysfs
==3629315==
Path for cpu: "/sys/devices/system/cpu"
Path for memory: "/sys/devices/system/memory"
==3629315==
==3629315== HEAP SUMMARY:
==3629315== in use at exit: 5 bytes in 1 blocks
==3629315== total heap usage: 22 allocs, 21 frees, 18,435 bytes allocated
==3629315==
==3629315== LEAK SUMMARY:
==3629315== definitely lost: 0 bytes in 0 blocks
==3629315== indirectly lost: 0 bytes in 0 blocks
==3629315== possibly lost: 0 bytes in 0 blocks
==3629315== still reachable: 5 bytes in 1 blocks
==3629315== suppressed: 0 bytes in 0 blocks
==3629315== Rerun with --leak-check=full to see details of leaked memory
==3629315==
==3629315== For lists of detected and suppressed errors, rerun with: -s
==3629315== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
As per the Kernel rules for accessing sysfs information [1], searching
for the sysfs mount point is a waste of time and systems that don't have
sysfs mounted at /sys are considered broken.
With those things in mind, util_path_sysfs() and especially
sys_mount_point() can be simplified. sys_mount_point() will always
return '/sys' unless the environment variable SYSFS_ROOT is set.
With SYSFS_ROOT still being present, special container setups or test
case scenarios are still possible but might need to be modified if they
previously relied on util_path_sysfs() automatically finding the correct
sysfs mount point.
To make things more secure against malicious strings in SYSFS_ROOT,
secure_getenv() is being used and the ordering of creating the formatted
path string in util_path_sysfs() is changed slightly.
Furthermore, the static variable is removed as no complicated query of
the /proc fs is required anymore. Memory for the sysfs mount point value
is properly freed now at the end of util_path_sysfs().
[1] https://www.kernel.org/doc/html/latest/admin-guide/sysfs-rules.html
Reviewed-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
A compile with sparse (make C=1) shows a couple of warnings:
like:
chzcrypt.c:36:3: warning: symbol 'l' was not declared.
Should it be static?
lszcrypt.c: In function 'show_device':
lszcrypt.c:556:9: warning: 'card' may be used uninitialized
[-Wmaybe-uninitialized]
Fix those warnings.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
A compile with sparse (make C=1) shows the following warnings:
util_rec.c:211:6: warning: symbol 'rec_print_wide' was not declared.
Should it be static?
util_rec.c:383:6: warning: symbol 'rec_print_csv_hdr' was not declared.
Should it be static?
util_rec.c:404:6: warning: symbol 'rec_print_csv' was not declared.
Should it be static?
These functions are only used internally from function util_rec_print_hdr(),
declare them as static.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
A compile with sparse (make C=1) shows the following warnings:
util_opt.c:42:19: warning: symbol 'util_opt_l' was not declared.
Should it be static?
util_prg.c:28:19: warning: symbol 'util_prg_l' was not declared.
Should it be static?
util_opt_l and util_prg_l are not used anywhere, and also not declared in
any header file. Looks like they are superfluous, remove them.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
hsavmcore is designed to make the dump process with kdump more efficient.
With hsavmcore, the HSA memory that contains a part of the production
kernel's memory can be released early in the process. Depending on the size
of the production kernel's memory, writing the dump to persistent storage
can be time consuming and prevent the HSA memory from being reused
by other LPARs.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The CCA Key Translate2 verb needs a pre-built AES CIPHER key token skeleton
with XPRTCPAC set in the output buffer, to produce an AES CIPHER key
that can be exported to CAPCF.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Build:
$ make -C libutil/util_log_example
* Every time a -V option is passed, the verbosity level increases and extra
messages are enabled.
Test:
$ ./libutil/util_log_example
# No output
$ ./libutil/util_log_example -V
ERROR: This is an ERROR message
$ ./libutil/util_log_example -VV
ERROR: This is an ERROR message
WARN: This is a WARN message
$ ./libutil/util_log_example -VVV
ERROR: This is an ERROR message
WARN: This is a WARN message
INFO: This is an INFO message
$ ./libutil/util_log_example -VVVV
ERROR: This is an ERROR message
WARN: This is a WARN message
INFO: This is an INFO message
DEBUG: This is a DEBUG message
$ ./libutil/util_log_example -VVVVV
ERROR: This is an ERROR message
WARN: This is a WARN message
INFO: This is an INFO message
DEBUG: This is a DEBUG message
TRACE: This is a TRACE message
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The goal of util_log is to provide a facility for a multi-level message
logging on stderr. This allows to selectively enable/disable log messages
via a log level which can be adjusted at runtime.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The tmp buffer is one byte too short missing space for the final \0 byte
leading to out of bound access.
Fix by increasing the buffer by one.
Reported-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add transparent codepage conversion feature to zdsfs.
Add options to specify that codepage conversion should be done using
default codepages from 'CP1047' to 'UTF-8' or using user defined codepage
tables.
Also add a new config file allowing the user to specify codepage conversion
options on a per dataset basis.
When a dataset is opened and codepage conversion settings match this
dataset an iconv handle is created and the conversion itself is performed
by the lzds_dshandle_read() function.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add codepage conversion capabilities to libzds.
Whenever a dataset is read and an appropriate iconv handle is set for the
dataset the conversion is automatically performed.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Replace function linux_error() by err/warn family.
Delete function linux_error() which is not used anymore.
Also fix some checkpatch warnings regarding comparison to zero.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Simplify function read_sfb(). Convert it to type void
and return only when no errors occurred.
On error print error message and exit instead of return.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Program reads complete counter sets from any CPU by opening
device /dev/hwctr which is available with linux version 5.12.
The counter sets and CPUs can be specified on the command line.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Reviewed-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use function util_opt_print_parse_error() instead of fprintf to stderr.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Simplify comparison against 0 or NULL and fix a checkpatch.pl
warning.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Using "chzdev -e" to configurate a device persistently, the
corresponding udev rule is also created, which will trigger a uevent
that sets the device online as soon as the group device is created.
If 'znetconf -r/R' is first used to remove such a configured device
(which won't remove the corresponding udev rule) and then
'znetconf -a/A' afterwards on the same device, either the error
"Failed to make <device> online" or "Failed to configure layer2=1"
will occur.
For the first error, the udev queue might not have been fully processed
and hence the check whether a device is online may fail due to that
race. Call 'udevadm settle' to ensure a fully processed queue before
checking whether a device is online.
The second error occurs, because configure_ccwgroupdev_option() does
not check whether the device is online. Add the check and set the
device offline if necessary. Also, log any details for debugging
purposes.
Signed-off-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Alexandra Winter <wintera@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add virsh commands for KVM debug data collection and rework the domain
data collection and step numbering. Update the man page accordingly.
[hoeppner@linux.ibm.com: Reword commit message]
Signed-off-by: Joern Siglen <siglen@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Performance evaluation showed that using the mq-deadline scheduler for DASD
devices leads to a significantly higher CPU consumption compared to using
none scheduler while having the same amount of throughput.
Setting none scheduler as default in the dasd udev rule.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When the compiler flags are overridden, the build of zkey may fail with:
kms.c:44:2: error: #error KMS_PLUGIN_LOCATION must be defined
44 | #error KMS_PLUGIN_LOCATION must be defined
| ^~~~~
The Makefile uses CFLAGS variable for defining the KMS_PLUGIN_LOCATION,
but it should rather use ALL_CFLAGS.
Also use ALL_CPPFLAGS for defining HAVE_LUKS2_SUPPORT.
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/108
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For maximum performance the MTU of an HSCI interface was set to the value
of the MTU of its HiperSockets interface. However, the MTU of a
HiperSockets interface can be very large, and when the data packets are
not sent over the HiperSocket but over the corresponding ethernet
interface (which usually has a smaller MTU size) then it can happen that
the large packets are silently dropped. One example would be communication
from an HSCI in one CEC to another HSCI in another CEC via Ethernet.
To solve the problem, the MTU setting in this tool is removed, and the
default MTU value is used. Thus, the minimum value of the MTUs of the both
converged interfaces is used as the MTU of the HSCI interface.
Fixes: 0566a492ae ("s390-tools: add hsci tool")
Signed-off-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Since crash 7.2.9 commit
f1f43bf355
in support of kernel v5.9 commit 0990d836cecb
("s390/debug: debug feature version 3"),
the symbolic caller function name is left aligned instead of right
aligned, so now we alternatively have to remove whitespace between
function name and offset.
Otherwise, zfcpdbf output contains wrong warnings about ambiguous payload
records with wrong values for reqid and area, as well as incorrectly
decoded trace records with wrong content values and potential error
messages on not defined record id, and finally a potential list of
occurrences of unknown tags which should empty:
Loading trace records ...(this might take a while)
Warning: Ambiguous PAYload records. reqid:>>>wrong<<< area:V^D^@fcp_
Warning: Ambiguous PAYload records. reqid:>>>wrong<<< area:W^D^@fsf_
Warning: Ambiguous PAYload records. reqid:>>>wrong<<< area:i^B^@san_
...
Timestamp : ...
Area : SAN
Subarea : 00
Level : 1
Exception : -
CPU ID : 0001
Caller : <zfcp_dbf_san_in_els
Record ID : 0
Tag : 6^B^Cfcie
Request ID : >>>wrong<<<
Destination ID : >>>wrong<<<
SAN record id=0 not defined.
Timestamp : ...
Area : SAN
Subarea : 00
Level : 1
Exception : -
CPU ID : 0002
Caller : <zfcp_dbf_san_in_els
Record ID : 0
Tag : 6^B^Cfcie
Request ID : >>>wrong<<<
Destination ID : >>>wrong<<<
SAN record id=0 not defined.
...
occurrences of unknown tags:
...
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For the stage2 of zipl's boot loaders only 3 heap pages are available
[0x6000, 0x6000 + 0x3000]. Therefore, stage2 code needs to be very careful
when and for how long it allocates a heap page.
The heap OOM problem in stage2 was hidden until the commit 252be376
("zipl/boot: fix potential heap overflow in stage2"). Before this commit,
stage2 assumed that heap is 4 pages large which is apparently very wrong
but get_zeroed_page() still allowed allocating an extra 4th page by
overriding the memory segment following the heap area of stage2 (stage3
parameter area).
Try to avoid allocating a heap page and keeping it allocated over a long
period of time in the DASD dumpers. Free a heap page as soon as possible
if not required anymore.
And be extra careful with printf() because it always tries to allocate
a free heap page.
Fixes: 252be376 ("zipl/boot: fix potential heap overflow in stage2")
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
RFC 7517 requires that the x and y coordinates of a ECC JSON Web Key
(JWK) are specified in its full size of a coordinate for the curve used.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Deleting an HSCI interface, the connected HiperSockets interface's name
and the connected Ethernet interface's name are extracted from "ip link
show". Currently the extracted names are directly used in "bridge -d link
show dev ifname" for further checking. If "iflink" and "ifindex" of an
interface are not matched, ip commands will print @ifXX appended to the
real interface's name. Thus, it will be failed if the interface's name
with the ending @ifXX is used in "bridge -d link show dev ifname" for
further checking.
To solve the problem, the ending @ifXX must be removed before the
interface name is used further.
Signed-off-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In the current code, after the function check_pnetids() and the function
verify_precon() are called, $? from each function is used twice
successively. It is used for the first time in the condition of the IF
statement, while for the second time as return code of the current
function. In bash, $? can be used to get the return code, but only for
the first time directly after the previous function is called.
Afterwards, $? will be reset to "0". Thus, the right return code can not
be get, if $? is used second time.
To solve the problem, using $? for the second time should be avoided.
Signed-off-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Install libasan first, e.g. on Fedora 32:
dnf install -y libasan libasan-static.
To compile with address sanitizer, pass ASAN=1 to make.
AdressSanitizer helps to catch various memory bugs for which C/C++ is
infamous.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This fixes the following compiler errors when ASAN is enabled:
CC cmsfs-fuse/cmsfs-fuse.o
cmsfs-fuse.c: Assembler messages:
cmsfs-fuse.c:310: Error: operand out of range (0xffffffffffffff00 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:310: Error: operand out of range (0xffffffffffffff00 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:310: Error: operand out of range (0xffffffffffffff00 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:310: Error: operand out of range (0xffffffffffffff00 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:310: Error: operand out of range (0xffffffffffffff00 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:310: Error: operand out of range (0xffffffffffffff00 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:302: Error: operand out of range (0xfffffffffffffda0 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:302: Error: operand out of range (0xfffffffffffffda0 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:302: Error: operand out of range (0xfffffffffffffda0 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:302: Error: operand out of range (0xfffffffffffffda0 is not between 0x0000000000000000 and 0x0000000000000fff)
cmsfs-fuse.c:302: Error: operand out of range (0xfffffffffffffda0 is not between 0x0000000000000000 and 0x0000000000000fff)
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When keys are generated concurrently in EKMF Web from multiple clients,
and the key labels use a sequence number tag, then the assignment logic
of the sequence number may cause a duplicate sequence number to be used,
because another key might have already been generated with the same
sequence number, since the next-to-use sequence number has been retrieved
from EKMF Web. This results in a HTTP response code of 409 and a -EEXIST
return code from ekmf_generate_key().
Instead of getting the last used sequence number from EKMFWeb, tell
EKMFWeb to automatically use the next available sequence number when
generating a key. When the sequence number label tag value is 'next'
then EKMFWeb will automatically assign the next available sequence number
to the label tag and increment the sequence number in an atomic way.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
If a memory chunk is added to mem_phys as well as mem_virt
in dfi_mem_chunk_add_vol() then an illegal memory access might occur
when accessing mem_chunk->data e.g. in dfi_elf_mem_chunk_read_fn()
because the data block pointed to by the data field is now being referenced
by two memory chunks, one in mem_phys and one in mem_virt. If it happens
that the memory chunk from mem_virt is freed in mem_unmap() then
the memory chunk in mem_phys still points to the common data block
which has been already freed. This leads to all sort of bad behavior
in dfi_elf_mem_chunk_read_fn() and other places where mem_chunk->data
might be accessed.
Fixes the following bug:
zgetdump: Unexpected end of file for "dump.all.elf"
And this was found by AddressSanitizer:
=================================================================
==81170==ERROR: AddressSanitizer: heap-use-after-free on address 0x602000000570 at pc 0x00000101ac10 bp 0x03ffd897e250 sp 0x03ffd897e248
READ of size 8 at 0x602000000570 thread T0
#0 0x101ac0f in dfi_elf_mem_chunk_read_fn s390-tools/zdump/dfi_elf.c:27
#1 0x100d8a5 in mem_read s390-tools/zdump/dfi.c:339
#2 0x100d8a5 in dfi_mem_phys_read s390-tools/zdump/dfi.c:616
#3 0x100d8a5 in mem_chunk_map_read_fn s390-tools/zdump/dfi.c:353
#4 0x100fd29 in mem_read s390-tools/zdump/dfi.c:339
#5 0x100fd29 in dfi_mem_read s390-tools/zdump/dfi.c:608
#6 0x1018e89 in os_info_get s390-tools/zdump/dfi_vmcoreinfo.c:65
#7 0x1018e89 in dfi_vmcoreinfo_init s390-tools/zdump/dfi_vmcoreinfo.c:86
#8 0x10175b3 in dfi_init s390-tools/zdump/dfi.c:1215
#9 0x1006e71 in do_stdout s390-tools/zdump/zgetdump.c:161
#10 0x1006e71 in main s390-tools/zdump/zgetdump.c:180
#11 0x3ffb07abb89 in __libc_start_main (/lib64/libc.so.6+0x2bb89)
#12 0x1007e8d (s390-tools/zdump/zgetdump+0x1007e8d)
0x602000000570 is located 0 bytes inside of 8-byte region [0x602000000570,0x602000000578)
freed by thread T0 here:
#0 0x3ffb0bc961b in free (/lib64/libasan.so.6+0xc961b)
#1 0x100d2d9 in mem_unmap s390-tools/zdump/dfi.c:1050
previously allocated by thread T0 here:
#0 0x3ffb0bc9aa9 in calloc (/lib64/libasan.so.6+0xc9aa9)
#1 0x100a271 in zg_alloc s390-tools/zdump/zg.c:93
SUMMARY: AddressSanitizer: heap-use-after-free s390-tools/zdump/dfi_elf.c:27 in dfi_elf_mem_chunk_read_fn
Shadow bytes around the buggy address:
0x100c0400000050: fa fa 00 fa fa fa 00 fa fa fa 00 fa fa fa 00 fa
0x100c0400000060: fa fa 00 fa fa fa 00 fa fa fa 00 fa fa fa 00 fa
0x100c0400000070: fa fa 00 fa fa fa 00 fa fa fa 00 fa fa fa 00 fa
0x100c0400000080: fa fa 00 fa fa fa 00 fa fa fa 00 fa fa fa 00 fa
0x100c0400000090: fa fa 00 fa fa fa 00 fa fa fa 00 fa fa fa 00 fa
=>0x100c04000000a0: fa fa 00 fa fa fa 00 fa fa fa 00 fa fa fa[fd]fa
0x100c04000000b0: fa fa fd fa fa fa fd fa fa fa fd fa fa fa fd fa
0x100c04000000c0: fa fa fd fa fa fa fd fa fa fa fd fa fa fa 04 fa
0x100c04000000d0: fa fa 00 fa fa fa 00 fa fa fa 00 fa fa fa 00 fa
0x100c04000000e0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
0x100c04000000f0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
Shadow byte legend (one shadow byte represents 8 application bytes):
Addressable: 00
Partially addressable: 01 02 03 04 05 06 07
Heap left redzone: fa
Freed heap region: fd
Stack left redzone: f1
Stack mid redzone: f2
Stack right redzone: f3
Stack after return: f5
Stack use after scope: f8
Global redzone: f9
Global init order: f6
Poisoned by user: f7
Container overflow: fc
Array cookie: ac
Intra object redzone: bb
ASan internal: fe
Left alloca redzone: ca
Right alloca redzone: cb
Shadow gap: cc
==81170==ABORTING
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Verify the given ELF header before adding any memory chunks otherwise
a memory chunk might be added even if the given ELF header is invalid.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some Linux distributions always include a copy of all persistent device
configuration data when updating the initial RAM-disk. This makes
chzdev's persistent device configuration changes ineffective because
device configuration directives applied in the RAM-disk take precedence
over those stored in the root filesystem.
This patch introduces a new build-time switch which allows distributions
to specify that whenever there is a persistent device configuration
change, the RAM-disk is updated automatically.
This feature can be enabled by adding 'ZDEV_ALWAYS_UPDATE_INITRD=1' as
a zdev build option. Where, by default ZDEV_ALWAYS_UPDATE_INITRD is 0.
Co-developed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-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>
After a master key change, it can happen that the AF_ALG setkey
operation returns EINVAL, although the key is valid. This is a
temporary situation and the operation will succeed, once the firmware
has completed some internal processing related with the master key
change. Delay 1 second and retry up to 10 times.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
If fdasd is used right after dasdfmt, it may fail with the following
error due to udev still processing events for the device:
```
fdasd error: IOCTL error: Error while rereading partition table.
```
Call dasd_reread_partition_table() with an increased retry value to
circumvent this issue.
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/103
Signed-off-by: Nikita Dubrovskii <nikita@linux.ibm.com>
[hoeppner@linux.ibm.com: Reword commit message]
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Extend zkey to allow to store a (dummy) LUKS2 passphrase together with a
secure key, for use with LUKS2 volumes. That way, when the repository is
backed up, or archived, also the passphrases of the associated LUKS2
volumes are backed up, or archived, and thus can not be forgotten.
This passphrase is then used in generated commands to format and open
the LUKS2 volumes.
Because of the use of secure keys to encrypt the volumes, the LUKS2
passphrase is of no or less relevance for security. It is therefore OK to
store the passphrase in clear text in the key repository, and also use
them in /etc/crypttab. Therefore, the passphrase could actually be a
trivial passphrase, or a dummy passphrase.
Note: Such a dummy passphrase is NOT considered a secret that needs to
be protected. If for a certain usage the passphrase is of relevance for
security, then the zkey dummy passphrase option must not be used to store
the passphrase.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Juergen Christ <jchrist@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
gcc's -Werror switch also enables -Werror=inline which leads to a compilation
failure due to inline warnings caused by GLIB macros. To fix it, disable
the conversion of inline warnings into errors.
In file included from buildroot/qemu-s390x/host/bin/../s390x-buildroot-linux-gnu/sysroot/usr/lib/glib-2.0/include/glibconfig.h:9,
from buildroot/qemu-s390x/host/bin/../s390x-buildroot-linux-gnu/sysroot/usr/include/glib-2.0/glib/gtypes.h:32,
from buildroot/qemu-s390x/host/bin/../s390x-buildroot-linux-gnu/sysroot/usr/include/glib-2.0/glib/galloca.h:32,
from buildroot/qemu-s390x/host/bin/../s390x-buildroot-linux-gnu/sysroot/usr/include/glib-2.0/glib.h:30,
from pv/pv_hdr.c:10:
buildroot/qemu-s390x/build/s390-tools-2.15.1/genprotimg/src/utils/buffer.h: In function ‘pv_hdr_new’:
buildroot/qemu-s390x/host/bin/../s390x-buildroot-linux-gnu/sysroot/usr/include/glib-2.0/glib/gmacros.h:1028:49: error: inlining failed in call to ‘glib_autoptr_clear_Buffer’: call is unlikely and code size would grow [-Werror=inline]
1028 | #define _GLIB_AUTOPTR_CLEAR_FUNC_NAME(TypeName) glib_autoptr_clear_##TypeName
| ^~~~~~~~~~~~~~~~~~~
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Current heap size in stage2 is three pages long [0x6000-0x9000] but
get_zeroed_page() assumes it is one page more which might lead to
a heap overflow which will corrupt data located at 0x9000 (stage3
parameters). Calculate the heap size of a stage at run-time
by using the symbols provided by linker.
Furthermore, validate the given address in free_page() to stop illegal
memory accesses.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Prevent access to freed memory still pointed to by entry.
entry still points into load_page that has been released with free_page.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When some of the key properties are skipped due to null_values_only
being true or false, then the resulting EKMF tag list might be built
incorrectly.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To remove a KMS property function kms_set_key_properties() can be
called with the property value set to NULL. Do not assert for non-NULL
property values.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
`-Winline` can cause problems with the cleanup functions defined by
glib. Therefore let's remove the cflag for now.
DEBUG: /usr/include/glib-2.0/glib/gmacros.h:1031:43: error: inlining failed in call to 'glib_autoptr_cleanup_STACK_OF_X509_CRL.isra.0': call is unlikely and \
code size would grow [-Werror=inline]
DEBUG: 1031 | #define _GLIB_AUTOPTR_FUNC_NAME(TypeName) glib_autoptr_cleanup_##TypeName
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This should be no problem (if OpenSSL works correctly), because
`crls_download_cb` is only called out of the verification context of
OpenSSL. It's used to look up a CRL and therefore
X509_STORE_CTX_get_current_cert should always return a certificate
since X509_STORE_CTX_get_current_cert returns NULL only if an error
has occurred during the verification (see
https://www.openssl.org/docs/man1.1.0/man3/X509_STORE_CTX_get_current_cert.html).
Fixes: 074de1e14e ("genprotimg: add host-key document verification support")
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add host-key document verification support to genprotimg. This ensures
that a host-key document is genuine and provided by IBM. For this the
user must provide the IBM Z signing key, the intermediate CA
certificate (signed by the root CA used) so a chain of trust starting
from the host-key document and ending in the root CA can be
established.
By default, genprotimg tries to download all revocation lists needed
by looking up in the corresponding certificate on how CRL information
can be obtained (see https://tools.ietf.org/html/rfc5280#section-4.2.1.13
for details).
Acked-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
A argument is required for the optional options 'ramdisk' and
'parmfile'.
Fixes: 65b9fc442c ("genprotimg: introduce new tool for the creation of PV images")
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add the -v short option for version information to the mk-s390image and
mk-pxelinux-ramfs. Adjust the Makefile accordingly to set the correct
s390-tools version during installation.
Acked-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use '--' for rm calls to distinguish between options and files. This
fixes, for example, the following error:
make[1]: Entering directory '/root/git/s390-tools/genprotimg/src'
rm -f *.gcda *.gcno *.gcov
rm: invalid option -- '.'
Try 'rm ./-.gcno' to remove the file '-.gcno'.
Try 'rm --help' for more information.
make[1]: *** [../../common.mak:369: clean_gcov] Error 1
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
chcpumf must be executed as root. A non-root user sees this error
message:
[tester@t35lp46 ~]$ chcpumf -m 1000
Error: /sys/module/kernel/parameters/cpum_sfb_size: Permission denied
[tester@t35lp46 ~]$
Enhance the error message and be clear about the root cause:
[tester@t35lp46 ~]$ chcpumf-new -m 1000
Error: Must run as root
[tester@t35lp46 ~]$
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Provide the status of the FC Endpoint Security information via the long
output of lsdasd for online Base and Alias devices.
New Output:
$ lsdasd -l c600
0.0.c600/dasdd/9412
status: n/f
type: ECKD
blksz: 512
size:
blocks:
extent_size: 1113
logical_capacity: 1113
space_allocated: 1113
use_diag: 0
readonly: 0
eer_enabled: 0
erplog: 0
hpf: 1
uid: IBM.750000000KMV11.c600.00
fc_security: Encryption
paths_installed: 38 39 3a 3b
paths_in_use: 38 39 3a 3b
paths_non_preferred:
paths_invalid_cabling:
paths_cuir_quiesced:
paths_invalid_hpf_characteristics:
paths_error_threshold_exceeded:
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Provide information for the DASD device attribute 'fc_security'.
The help output for this attribute can be queried like this:
chzdev dasd --help-attribute
The lszdev tool can be used to display the current state of the FC
Endpoint Security information:
lszdev dasd -a -c TYPE,ID,ATTR:fc_security,ATTRPATH:fc_security
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When many domains per card are available, then the results of the SCDMD
CHSC call may not fit into the output area, and a partial result is
returned. The further results must be retrieved with another CHSC call.
Fix the code to pass the correct next-domain to the subsequent CHSC call
of a partial response. Otherwise the same set of domains 1 to n are
retrieved again, resulting in an infinite loop, because this will always
produce a partial result.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Noticed on RHEL CoreOS that the CPI service failed to start with:
"Cannot access lock file: /var/lock/cpictl.lock"
This was a timing issue where /var was not created yet and the symlink
to /run/lock was not present. Add sysinit.target to fix this.
This fix has been tested with RHEL CoreOS.
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/82
Signed-off-by: Prashanth Sundararaman <psundara@redhat.com>
Suggested-by: Colin Walters <walters@verbum.org>
Reviewed-by: Dan Horák <dan@danny.cz>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Modern systems have systemd manage system mount points like sysfs which
specify 'sysfs' as a keyword for the device as there is no device
associated with this special filesystem. However, any arbitrary string
could be specified here and the determination of the sysfs mount point
would fail in such a case.
To make sure that the mount point of the sysfs is still found when
mounted with a device keyword specified other than 'sysfs', check for
the filesystem type instead, which is more specific.
Fixes: https://github.com/ibm-s390-tools/s390-tools/issues/91
Suggested-by: Mark Post <mpost@suse.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a proper definition to the dasdfmt header file of the error()
function that lets the compiler check the format string against the
provided parameters.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
hsci is used to control and show HSCI (HiperSockets Converged Interfaces)
settings. A HiperSockets interface and an external network interface are
converged to an HSCI interface.
Signed-off-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
==1005844== HEAP SUMMARY:
==1005844== in use at exit: 18,907 bytes in 14 blocks
==1005844== total heap usage: 82 allocs, 68 frees, 32,529 bytes allocated
==1005844==
==1005844== 136 (104 direct, 32 indirect) bytes in 1 blocks are definitely lost in loss record 12 of 14
==1005844== at 0x483885A: calloc (vg_replace_malloc.c:760)
==1005844== by 0x48C950D: g_malloc0 (gmem.c:132)
==1005844== by 0x100EC41: pv_args_new (pv_args.c:364)
==1005844== by 0x100587F: main (genprotimg.c:122)
==1005844==
==1005844== LEAK SUMMARY:
==1005844== definitely lost: 104 bytes in 1 blocks
==1005844== indirectly lost: 32 bytes in 1 blocks
==1005844== possibly lost: 0 bytes in 0 blocks
==1005844== still reachable: 18,771 bytes in 12 blocks
==1005844== suppressed: 0 bytes in 0 blocks
==1005844== Reachable blocks (those to which a pointer was found) are not shown.
==1005844== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==1005844==
==1005844== For lists of detected and suppressed errors, rerun with: -s
==1005844== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Abort compilation as soon as one of the recursive targets is failing.
Fixes: 65b9fc442c ("genprotimg: introduce new tool for the creation of PV images")
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Adding one to the pointer per block value results in a wrong
calculation of pointer block levels. When writing a file that
would result in about 454MB size, the code would calculate that
it needs a two level pointer directory but correct would be a three
level pointer directory. This causes an invalid filesystem state:
besides missing all the records of one top level pointer directory
entry, it would also record a higher record number in the FST than
what is present in the pointer blocks.
Note that this bug only hits when the file is about 454MB and the write
is stopped (means the file is closed) - if file writing continues the
problem doesn't happen because at some point in time cmsfs-fuse would
switch to three levels of pointer blocks anyways as the file grows.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When specifying an incorrect program option, dasdfmt segfaults as the
format string for the corresponding error message has no parameter.
Add the missing parameter to fix this.
Fixes: 732b3dddab ("dasdfmt: Replace ERRMSG_EXIT macro with an error handling function")
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
util_sys_get_dev_addr() returns the device address for a given blockdevice.
This does not work for partitions but some tools rely on the ability to get
the device address for partitions.
Add code that first determines the base device for a partition.
Fixes: 6014d07cb1 ("dasdview/libdasd/zipl: Use util_sys_get_dev_addr() instead of u2s_getbusid()")
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some operations are only possible on base devices not on partitions.
Add functions to determine if a given device is a partition or a base
device and to get the base device to a given partition.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When using zgetdump on a multi-volume device dump, the sysfs path to
determine the device node is incomplete, resulting in the following
error:
$ zgetdump -i /dev/dasdb1
zgetdump: Could not open "/sys/bus/ccw/devices/0.0.9300/dasdb/dev" (No
such file or directory)
A simple fix would be to add the missing "block" part in the sysfs path.
However, the logic still assumes sysfs links like "block:" that have
been deprecated a decade ago and are no longer present on modern
systems anyway.
Therefore, the logic can greatly be improved. Create a separate function
to determine whether a device is online, remove the logic for "block:"
entries, and use libutil functions to reduce the complexity even further.
Fixes: 70a79fab3c ("zgetdump: Avoid using PATH_MAX")
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In the process of switching to glibc defined error functions it becomes
apparent that self-defined error functions that do special things are
not in line with the output of the glibc functions.
To address this, set the prog_name variable to the last component of the
program name (stored in argv[0]) and guarantee a uniform error output.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In check_track_format() the base error message is the same for all cases
but has different additional content depending on which mode dasdfmt is
running.
Currently the message is compiled by using different outputs. To make it
a little bit cleaner and for a better error message handling, construct
the message string completely before passing it to the error() function
for output.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The ERRMSG_EXIT macro is a bit clunky in its usage and a change is
necessary in order to free memory in error cases.
Create a new function error() that takes only a format string and adds
all other relevant information by itself. This function frees memory
before terminating the program with the EXIT_FAILURE exit code.
This simplifies the error handling for pretty much all calls and makes
the code a bit cleaner.
While at it, the defines EXIT_MISUSE and EXIT_BUSY don't provide any
value. Neither are they documented nor are these standardized. Also, a
parent process is mostly interessted in success or failure only anyway.
Replace these by using only EXIT_FAILURE in the error() function.
Also, change multiline output to a combination of warnx and error to
have a uniformed output. So this:
WARNING: Device is formatted with a different blocksize (4096).
Use --mode=full to perform a clean format.
becomes this:
dasdfmt: WARNING: Device is formatted with a different blocksize (4096).
dasdfmt: Use --mode=full to perform a clean format.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When calling dasdfmt on device node symlinks like
/dev/disk/by-id/ccw-0X9300, dasdfmt fails with "dasdfmt: the ioctl to get
the blocksize of the device failed (Bad file descriptor)"
This is because before the actual formatting process starts, the disk
will be disabled calling the BIODASDDISABLE ioctl, resulting in the
removal of the symlink. Trying to open this file later in the process to
retrieve e.g. blocksize information results in the mentioned error, as
the file doesn't exist any longer.
This incorrect behaviour was introduced with commit 8826028bdc
("dasdfmt: Use libdasd provided ioctl functions") at which the use of a
globally available file descriptor was omitted. Instead, the ioctl
library functions require a device node to be passed as a function
parameter and will open a file descriptor themselves.
So, before that change, the global file descriptor was always valid even
when the file was removed.
In order to fix this without modifying the behaviour of libdasd,
introduce the two global variables dev_node and dev_path. dev_path is
the original device path entered by the user. dev_node on the other hand
is the reliable device node under /dev/block/ using the major and minor
numbers and is determined in get_device_name().
The dev_path is used for message output only and the dev_node variable
is used for the actual disk operations.
As these two variables are global, the device parameters for several
functions are removed to make the code a bit cleaner.
Fixes: 8826028bdc ("dasdfmt: Use libdasd provided ioctl functions")
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The global program options (and other actually globally used
information) are currently passed through via function parameter
whenever this information is needed. However, in reality, this
information is used almost everywhere across the program.
Make this information globally available and get rid of all the
unnecessary function parameters for dasdfmt_info_t. This makes the code
a little bit cleaner and more readable. Also, rename the global struct
to a more suitable name.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
dasdfmt_format() does nothing but calling process_tracks() with the same
parameters. Get rid of it, call process_tracks() directly, and avoid
confusion.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Since commit 75e3afb6a0 ("libdasd: Move get_host_access_count() to
libdasd") dasd_get_host_access_count() reports always 0 as the check for
unsuccessful execution of util_sys_get_dev_addr() is incorrect.
Fix the behaviour by turning the check around.
Fixes: 75e3afb6a0 ("libdasd: Move get_host_access_count() to libdasd")
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zkey-ekmfweb.so requires libekmfweb.so. If libekmfweb.so can not be built
due to missing build requirements, then zkey-ekmfweb.so can't be built
either.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Building zkey-ekmfweb.so will also trigger to build libekmfweb.so,
which requires OpenSSL. So, skip zkey-ekmfweb.so if HAVE_OPENSSL=0.
zkey-cryptsetup also has a dependency on OpenSSL, so skip it also
if HAVE_OPENSSL=0
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
chcpumf does not print out version (option -v) or help text
(option -h) when the CPU Measurement Sampling facility is not
installed. This was different up to and including release 4.10.
Fix this and hounour option -v and option -h even when
the sampling facility is not installed.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Reviewed-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
lszcrypt now shows a card/queue which is in 'deconfigured' state as
'deconfig' in the STATE column (verbose and non verbose mode).
chzcrypt became two new options: --config-on and --config-off to
switch one or more card devices to 'configured' or 'deconfigured'
state.
Both applications are able to handle older kernels which do now
provide the config sysfs attribute required for this new feature.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This patch adds support for the "clear" sysfs attribute for nvme re-IPL,
if available. This attribute allows to control whether the memory should
be cleared on re-IPL.
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
During using of ziorep_utilization utility for generating of CSV format output,
there is one extra comma added after Bus-ID column in virtual adapter report
for each line. This creates one extra column with empty content for each
line. As a result in CSV viewer programs 'qdio utilization max %' column has
no content, all subsequent columns have shifted content from previous
column and the last column has no header. Avoid this situation by deleting
of extra comma.
Before patch application typical line of virtual adapter report looks like:
2020-05-14 13:55:43,0,60,0.0.1906,,0.000000,0.782567,0,0,0.0,321.8,1,1251
After patch application typical line of virtual adapter report looks like:
2020-05-14 13:55:43,0,60,0.0.1906,0.000000,0.782567,0,0,0.0,321.8,1,1251
Signed-off-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Julian Wiedmann <jwi@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Import existing keys that are stored in EKMF Web into the zkey repository.
The keys are imported including their associated information, such as
textual description, volumes, sector size, etc.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add the possibility to list keys stored in EKMF Web that are eligible to
be used with zkey. The list of keys can be filtered by name, label,
associated volumes, state and export-ability.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When a key is removed from the zkey repository, the state of the key in
EKMF web can optionally be changed. Keys are usually not removed in EKMF
Web, but the state of the keys are changed to e.g. DEACTIVATED.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add functions to get and set key properties (i.e. custom tags) of
keys stored in EKMF Web. The key properties are set when the information
associated with a key in the zkey repository is changed, e.g. using the
'zkey change' command.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Generate and import volume encryption keys of type CCA-AESCIPHER in
EKMF Web and import them int the zkey repository. Additional information
can be associated with the keys, such as a textual description, volumes
encrypted with the key, sector size of the volumes, etc. This is
also stored in EKMF Web with the keys as custom tags.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To register the zkey client with EKMF Web, the X.509 certificate
generated with the identity key is sent to EKMF Web, and an identity
key object is generated in EKMF Web with the public key from the
certificate. This identity key is then used to control the export of
keys stored in EKMF Web, and to verify the cryptographically signed
requests sent by the plugin.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To identify the zkey client with EKMF Web, an X.509 certificate must be
generated using the identity key, and must be made known to EKMF Web.
Either a self signed certificate can be generated, or a certificate
signing request (CSR) that is then passed to a certificate authority
(CA) to have a certificate issued.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The identity key of the plugin is a secure key. When the APKA master key
of the associated APQNs is changed, the identity key must be re-enciphered
under the new master key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To identify the zkey client with EKMF Web, the plugin generates a secure
ECC or RSA identity key, and stores it in its configuration directory.
This key is then used to cryptographically sign requests sent to EKMF Web.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Retrieve several settings from EKMF Web after the connection to
EKMF Web has been configured. This includes the EKMF Web server's
public key, which is user later on to verify cryptographically signed
responses. Also the key templates used by EKMF Web to generate keys
for zkey are retrieved, and it is checked if the require feature
'Pervasive Encryption' is available.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To perform operations in EKMF Web, the user must log in with a EKMF Web
user id and a time based one time passcode. The passcode can be obtained
by the user from the EKMF Web user interface, after logging in into EKMF
Web. That way a two factor authentication is performed. The plugin passes
the passcode to EKMF Web retrieves a bearer token from EKMF Web which it
then uses on subsequent requests to authenticate with EKMF Web. Such a
bearer token is valid for several minutes, thus no re-login is required
for zkey commands run during that time.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow the user to configure the connection to the EKMF Web server
using the 'zkey kms configure' command. The communication is based
on a RESTful API via HTTPS. Besides the URL of the EKMF Web server,
other TLS related settings can be configured.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The EKMF Web plugin requires APQNs of one or multiple IBM cryptographic
adapters in CCA coprocessor mode to operate. It makes use of secure
RSA and ECC keys, and thus requires the APKA master keys of the CCA APQNs
to be set.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support to integrate the external key management system EKMF Web
(IBM Enterprise Key Management Foundation - Web Edition) into zkey by
providing a zkey KMS plugin for EKMF Web. The plugin communicates
with EKMF Web using libekmfweb via a RESTful API.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The C-XLATE flag determines if the secure key can be used with the
CCA Cipher Text Translate2 verb. Cipher Text Translate2 deciphers
encrypted data under one key and re-enciphers it under another key
without having the data appear in the clear outside the cryptographic
coprocessor. Since this is not a security risk, the check can be
removed.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Retrieve EKMF Web settings such as the template names used by
EKMF Web for generating identity keys, as well as volume encryption
keys (XTS and non-XTS). Also check that the EKMF Web server has the
required 'Pervasive Encryption' feature installed.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To perform operations in EKMF Web, the user must log in with a EKMF Web
user id and a time based one time passcode. The passcode can be obtained
by the user from the EKMF Web user interface, after logging in into EKMF
Web. That way a two factor authentication is performed. The client passes
the passcode to EKMF Web retrieves a bearer token from EKMF Web which it
then uses on subsequent requests to authenticate with EKMF Web. Such a
bearer token is valid for several minutes, thus no re-login is required
for zkey commands run during that time.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow to add, change and delete custom tags of a key. Custom tags
can be used store any kind of textual information with a key. Tags
have a name and a value.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
EKMF Web maintains a key state for each key. Keys can be in state
PRE-ACTIVATION, ACTIVE, DEACTIVATED, COMPROMISED, DESTROYED, and
DESTROYED-COMPROMISED. Key states can be changed as defined in NIST
Special Publication 800-57 Part 1.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow to request EKMF Web to generate a new key in its repository.
The newly generated key is generated based on the key template used.
Also, custom properties can be specified that are stored together with
the key, as well as export control information defining who is allowed
to request an export of the key later on.
Besides generating new random keys in EKMF Web, you can also register
the client's public identity key with EKMF Web by generating a new
identity key using an X.509 certificate containing the client's public
key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Retrieve information about keys managed by EKMF Web, such as the
key algorithm, the key size and type. Additional, the key's tags
are retrieved. Label tags are used to build the label name of a key.
Custom tags can be used to store any kind of textual data together with
a key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Key templates are used by EKMF Web to generate new keys. The template
specifies the key algorithm, the key size, and type (e.g. CCA DATA or
CIPHER). It also determines how a newly generated key is named through
a key label template containing label tags.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To retrieve a secure key from EKMF Web, a Elliptic-curve Diffie-Hellman
(EC-DH) protocol is used to securely transport the key, without revealing
the key to be transported in clear. The key to be transported remains
encrypted all the time, either encrypted with the master key of the
cryptographic adapter on the source or target side, or with a transport
key, that is derived using the EC-DH key agreement protocol. The transport
key is also a secure key, itself encrypted with the master key of the
cryptographic adapters on both sides.
To generate the transport key, each side generates a new secure ECC
session key with its cryptographic adapter. This session keys are then
used with EC-DH to derive the secure transport key. The secure key to
be transported is then exported by EKMF Web under the transport key, is
sent to the client, and is then imported using the transport key.
The key retrieval request is signed using the identity key of the client.
EKMF Web knows the public key of the client's identity key through a one
time registration process, and can therefore verify the signature with
the client's public key. The response is also signed by the EKMF Web
server's identity key, and the client can verify the signature with the
server's public key that it retrieved once during registration.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To identify the client with EKMF Web, an X.509 certificate must be
generated using the identity key, and must be made known to EKMF Web.
Either a self signed certificate can be generated, or a certificate
signing request (CSR) that is then passed to a certificate authority
(CA) to have a certificate issued. The certificate is then used to
register the client with EKMF Web, so that EKMF Web knows the public
part of the client's identity key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The identity key of the client is a secure key enciphered with the
master key of a cryptographic adapter (APQN). When the master key of
the used APQNs is changed, the identity key must be re-enciphered
under the new master key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To identify the client with EKMF Web, the client generates a secure
ECC or RSA identity key. This identity key is then used to
cryptographically sign certain requests sent to EKMF Web.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When login in with EKMF Web, a bearer token is retrieved from EKMF Web
which is then used on subsequent requests to authenticate with EKMF Web.
Such a bearer token is valid for several minutes, thus no re-login is
required during that time. The bearer token contains a JSON Web Token
(JWT, see RFC7519).
Allow to check such a token if it is still valid, or already expired.
That way a client application can check the token before issuing the next
request, and re-login if needed.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The EKMFWeb client library provides functions to communicate
with an EKMF Web server via REST calls over HTTPS. EKMF Web stands
for IBM Enterprise Key Management Foundation - Web Edition and is
used to manage keys in an enterprise.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow to refresh or update a KMS-bound key in the repository.
The secure key is re-imported from the KMS, and optionally also
its properties are updates from the KMS.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow to import keys that exist in a key management system into the
repository. The key together with its properties is imported.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow to list the keys managed by a key management system. The keys
are displayed together with the key properties stored in the KMS.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When removing a KMS-bound key from the local repository, call the
KMS plugin and let it perform an action in the KMS. Usually keys are
not removed from key management systems, instead the key state is
changed to non-active or similar.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When key properties are changed with the 'change' command, also update
the key properties in the KMS, if the key is bound to a KMS. Do not
allow to change the associated APQNs for KMS bound keys. KMS bound keys
inherit the APQNs from the KMS plugin.
When a key is renamed in the repository, also update the key name
property in the KMS if the key is KMS bound.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Do not allow to convert a KMS bound key. The key type of a KMS
bound key is determined by what the KMS plugin supports, thus it
can not be converted.
Do only allow to copy a KMS bound key, if the '--local' option is
specified. The result of the copy is a local key, that is not bound
to a KMS.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When a repository that is bound to a key management system, the
keys are generated by the key management system by default. To
generate a local key, add option '--local'.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add the following sub-commands for key management system plugin
handling:
- configure: Configure a KMS plugin
- reencipher: Re-encipher secure keys used by a KMS plugin
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some commands requires additional options that are to be determined
by the KMS plugin. Add support to let the KMS plugin provide an
option vector per command, and add the options to the already existing
command options.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add command 'kms' and the following sub-commands of it to handle
key management system plugin binding and unbinding:
- plugins: List available/configured KMS plugins
- bind: Binds a KMS plugin to the repository
- unbind: Unbinds a KMS plugin from the repository
- info: Displays information about the currently bound KMS plugin
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To support keys that are bound to a key management system (via a KMS
plugin) add new key properties that reflect this. Display if a key is
bound to a KMS with the 'zkey list' and 'zkey validate' commands.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support to integrate external key management systems into zkey
by using a plugin concept. Add helper functions to load, bind, and
unbind a key management system plugin.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for sub-commands to zkey. Sub-commands allow to group a
set of commands within a sub command.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add the -fPIC option when building the libutil objects to generate
position-independent code, and allow them to be used in dynamically
loaded shared libraries.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add description for coordinated read access to man page as well as some
editorial changes after ID review.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For data consistency it is essential that datasets are not modified by
z/OS applications during zdsfs access. This is currently ensured by
manually setting the device offline in z/OS before mounting it in Linux.
This patch improves the usability and data security by making this manual
step obsolete.
Before opening a dataset zdsfs will obtain an ENQ to mark it as in use for
z/OS and release the ENQ when closing the dataset.
A timer is set up that pings the z/OSMF REST services with the ENQ to
prevent it from a timeout after 10 minutes.
The behavior of zdsfs can be configured in a new config file or with new
command line options.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add library functions that allow to communicate with z/OSMF REST services
using libcurl.
Following three functions are added:
lzds_rest_get_enq()
to obtain an ENQ that will mark a dataset as in use to z/OS until
it is released again
lzds_rest_release_enq()
to release an ENQ and mark a dataset as no longer in use to z/OS
lzds_rest_ping()
to ping a z/OSMF REST server to check if it accessible
or to refresh an ENQ and prevent it from a timeout after 10 minutes
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Stage 3 flags is a 64bit integer but function add_ipl_program
used a 32bit integer instead. Due to this bug
STAGE3_FLAG_SCSI and STAGE3_FLAG_KDUMP were not correctly set.
Fixes: 412d3e8c ("zipl: consolidate stage3_params structs and stage3 flags")
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
libpfm does not yet include counters introduced with z15 in a release.
cpacfstatsd now implements its own counter logic to become independent
of the slow release cycles of libpfm.
Reviewd-By: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewd-By: Harald Freudenberger <freude@de.ibm.com>
Signed-off-by: Juergen Christ <jchrist@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
+ remove outdated comment that the revocation list checking is not implemented
+ improve usage example
Reviewed-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
We cannot rely on what $SHELL says as it describes what the users
default shell is and not which kind of shell /bin/sh is. So instead of
checking the environment variable $SHELL, we can test whether the
environment variable $BASH is set. If this is the case, we can be
pretty sure that the script is executed by the bash shell.
This fixes the error on Ubuntu 20.04:
./check_hostkeydoc se-certs/host-key.crt se-certs/ibm-z-host-key-signing.crt -c se-certs/DigiCertCA.crt -r se-certs/ibm-z-host-key.crl
./check_hostkeydoc: 42: set: Illegal option -o pipefail
Reviewed-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Set the counter name for z15 counter numbered 265. It is either named
DFLT_CCERROR or DFLT_CCFINISH, depending on the linux version. The
counter was renamed from CCERROR to CCFINISH in linux version 5.8.
Check for existence of file /sys/devices/cpum_cf/events/DLFT_CCERROR.
f this file exists the counter is named DFLT_CCERROR else
DFLT_CCFINISH (only for z15), other machines do not have this counter.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Modern distributions do not provide the symlink /sbin/udevadm any more.
Use /usr/bin/udevadm instead for the example init script / systemd units.
Note that those are meant for example purpose and would need to be adjusted
by distributions, so no extra effort is taken to determine the actual
location of udevadm.
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When udevadm is going to be used, try to locate it
in $PATH and if it fails, try to locate it in well-known
binary paths.
Signed-off-by: Guevenc Guelce <guvenc@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Commit 86856f98db ("zipl: Make use of __noreturn macro") moved the
'noreturn' attribute from declaration to definition. With this the
compiler can no longer optimize when the function is called in a
separate source file. Move the attribute back to the declaration
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Function util_opt_init() build the option string for getopt_long().
If a command has been set via util_opt_set_command(), then util_opt_init()
must honor that command and only add those options that match the command,
or are command independent.
That way the same option can be used in different commands with different
flags and different argument settings. E.g. for command 'a' option '-x'
might require an argument, for command 'b' the same option '-x' might not
require an argument.
The behavior of util_opt_init() is unchanged if no command is set, and
also if different commands use the same option, but with the same flags
and argument settings. Currently only the zkey tools set a command, but
use unique options per command.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Remove obsolete and unused constants for zcore memory
character devices '/sys/kernel/debug/zcore/mem' and
'/sys/kernel/debug/zcore/memmap'.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Fix that a misleading BLS error message is printed in any case for missing
files.
The scan_check code in zipl already does proper error handling for missing
file. Remove the duplicated error handling from scan_check_bls and just use
the function to update the file with the target path if needed.
To indicate the intention of this function also rename it.
Fixes: #84
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Tuan Hoang <tmhoang@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When vmcp fails to execute a CP command with both error conditions
- response buffer is too small
- CP command failed
then the vmcp program exits with 'response buffer too small' indication.
However, an exit code indicating 'CP command failed' would be more
important in this case.
So change the vmcp exit code and return 'CP command failed' for above
error scenario.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
z15 architecture supports counters for ECC operations. We now add and evaluate
these counters on supported architectures. If ECC counters are not supported,
we simply print "unsupported" for ECC counter values.
Manpages adapted accordingly.
Signed-off-by: Juergen Christ <juergen.christ@ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Reviewed-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Hard-coding udevadm's location isn't robust enough - for instance,
Ubuntu 20.04 moved it to /usr/bin.
First see if we can reach it through $PATH, then fall back to hard-coded
locations that we know about.
Also when finally falling back to udevsettle, check that it exists.
Otherwise we end up throwing tons of "No such file or directory" error
messages at the user.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Rewrite lscpumf and chcpumf in C. The previous version
habe been written in perl which require a lot of instructure.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Reviewed-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Copying the SCSI scheme wasn't good enough due to how major/minor
numbers are assigned by the blkext driver. We now query the relevant
information via sysfs.
Signed-off-by: Jason J. Herne <jjherne@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This attribute is a no-op and never provided the described functionality.
It will be removed in the 5.7 kernel release with
commit 5f4019a80839 ("s390/qeth: remove fake_broadcast attribute").
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Support is added to lsreipl to detect the nvme reipl type and list the
appropriate parameters.
Support is added to chreipl to specify and populate sysfs reipl entries
with nvme device targets.
Signed-off-by: Jason J. Herne <jjherne@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The genprotimg tool doesn't yet contain code to verify IBM Secure Execution
host key documents. The sample script check_hostkeydoc outlines the steps
that need to be followed by users to verify a host key before it should
be used with genprotimg to build secure Linux images.
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.ibm.com>
Reviewed-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Printing the number zero (e.g. printf("%u", 0)) currently only gives you
an empty string. This is because the while-do loop to map the number to a
string is only entered when the value is 'true', i.e. non-zero. Fix this
by using do-while instead.
Fixes: 6fe9e6c ("zipl/libc: Introduce vsnprintf")
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reported-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
nvme devices use the same on-disk bootmap as scsi devices. When the target
device is nvme (using the blkext driver) we simply treat it as scsi.
Signed-off-by: Jason J. Herne <jjherne@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Let's assume the original `command_line` is 'aaaaaaaaaaaaaaaaaa' and
`cextra` is ' =blub'. With the old code this would result in the
following new command line:
'blubaaaaaaaaaaaaa' instead of the expected result 'blub\0aaaaaaaaaaaa'
Let's fix this by setting the NULL-termination for the new command
line. While at it, use the `command_line` variable instead of casting
an address.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The inserted blank must also be taken into account for the calculation
of `cmdline_len`. In general it's only possible to insert a blank if
the command line length is not already the maximum.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
`src` parameter for `memcpy` is set to `cextra + begin` but the offset
begin is not taken into account for the calculation of the length that
should be copied.
Fix this by increment the cextra pointer for each skipped whitespace
of the extra parameter.
While at it, rename `end` variable to `cmdline_len`, `command_line`
variable to `cmdline`, remove useless void pointer cast, and use
strlen instead of a while loop to determine the cmdline string length.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
1. `begin` is used as array index and should therefore never be greater
than COMMAND_LINE_SIZE - 1. Therefore let's fix the while condition.
2. `length` describes the string length and should therefore also never
be greater than COMMAND_LINE_SIZE - 1.
The 1. off-by-two error can lead to a out-of-bounds read and the 2.
to a buffer overflow.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This fixes an error when genprotimg is compiled with clang and the
sanitizers address and undefined behaviour enabled.
pv/pv_ipib.c:47:33: runtime error: member access within misaligned address 0x62100000a100 for type 'IplParameterBlock' (aka 'struct ipl_parameter_block'), which requires 4096 byte alignment
0x62100000a100: note: pointer points here
01 00 00 45 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
^
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior pv/pv_ipib.c:47:33 in
This change should be no problem as genprotimg is the only user.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use STAGE2_ENTRY and PSW_LOAD instead of hard-coded values. While at
it, also move the libc.h to the top in tape2dump.c.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
While at it, rename PSW_SHORT_ADDR_MASK and PSW_ADDRESS_MASK to
PSW32_ADDR_MASK. As a side effect this removes the zipl.h dependency
of the boot loaders. While at it,
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use C pre-processor for linker script generation and refactor some
constants. This allows the usage of the constants in our "linker
scripts" `*.lds.S` (actually, these are assembler files, so we can
make us of the C pre-processor and its capabilities).
While at it, enforce the described memory layout in the generated
linker script.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Move COMMAND_LINE_EXTRA to `boot/loaders_layout.h` so it can be reused
by `menu.c`and `stage3.c`. Additionally, reuse COMMAND_LINE_SIZE from
`linux_layout.h` in `menu.c`.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Introduce a struct definition for the stage3 parameters used by
stage3.c. This makes it easier to maintain the stage3 linker script
and it allows us to consolidate the stage3_params structs defined in
`stage3.h` and `include/boot.h` as well as the stage3 flag macros.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This allows the usage of stage3.h in files where other libc
implementations than our minimal libc is used, e.g. glibc.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This fixes cpuplugd.h:157:12: warning: redundant redeclaration of ‘debug’ [-Wredundant-decls]
157 | extern int debug; /* is verbose specified? */
| ^~~~~
cpuplugd.h:155:12: note: previous declaration of ‘debug’ was here
155 | extern int debug;
| ^~~~~
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In case secure boot is enabled but no components in the IPL report
are found it ends up in an endless loop in the component verification.
To prevent this check if components are found and exit if not.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The lowcore parmblock pointer is not valid in every case. For example
it is invalid for CCW type IPL.
To have an indication if the pointer is valid do a diag308 to store the
parmblock and check if secure boot is enabled.
If it is enabled the lowcore pointer is valid and the ipl report that is
needed for secure boot can be found right behind the ipl parmblock.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Variable names are case sensitive in Makefiles, therefore fix the typo
in the variable reference.
Fixes: 2d600570df ("genprotimg: boot: use C pre-processor for linker script generation")
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This fixes the false positive
CC zipl/src/misc.o
misc.c: In function ‘misc_temp_dev’:
misc.c:313:27: warning: ‘%04d’ directive writing between 4 and 11 bytes into a region of size 5 [-Wformat-overflow=]
sprintf(filename, "zipl%04d", retry);
^~~~
misc.c:313:22: note: directive argument in the range [-2147483648, 999]
sprintf(filename, "zipl%04d", retry);
^~~~~~~~~~
misc.c:313:4: note: ‘sprintf’ output between 9 and 16 bytes into a destination of size 9
sprintf(filename, "zipl%04d", retry);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
While at it, add an assertion that protects against a format-overflow,
e.g. if TEMP_DEV_MAX_RETRIES changes.
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some drivers provide per-queue IRQ coalescing information, collect it.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Protected VMs (PVM) are KVM VMs, where KVM can't access the VM's state
like guest memory and guest registers anymore. Instead the PVMs are
mostly managed by a new entity called Ultravisor (UV), which provides
an API, so KVM and the PV can request management actions.
PVMs are encrypted at rest and protected from hypervisor access while
running. They switch from a normal operation into protected mode, so
we can still use the standard boot process to load an encrypted image
and then move it into protected mode.
This commit adds the tool 'genprotimg'. It takes a kernel, key files,
optionally an initrd, optionally a file with the kernel command line,
and it generates a single, loadable image file. The image consists of
a concatenation of a plain text boot loader, the encrypted components
for kernel, initrd, and cmdline, and the integrity-protected PV
header, containing metadata necessary for running the guest in PV
mode.
It's possible to use this image file as a kernel for zipl or for a
direct kernel boot using QEMU.
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Acked-by: Patrick Steuer <patrick.steuer@de.ibm.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The pod2usage function from module Pod::Usage always emits this warning
when the module is loaded, even if the funtion is not invoked at all:
The script_run feature is experimental in regex;
marked by <-- HERE in m/(*script_run: <-- HERE ^ .* $ )/
at (eval 14) line 2.
To avoid this warning every time the cpumf_helper script is invoked,
do not use function pod2usage. The use of this function has no
benefit. cpumf_helper is invoked under the covers from lscpumf and
chcpumf, so it is sufficient to print the help text in case a wrong
parameter has been supplied.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
`linux/asm/types.h` also defines the struct `__vector128` so in order
to avoid definition conflicts add the macro guard _S390_TYPES_H, which
is defined in `linux/asm/types`. `linux/asm/types.h` is included by
glib2, which is used by the PV tooling.
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add IBM Z15 extended counter definition file to Makefile
DATA_FILES list.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Format and print HBA FC Endpoint Security trace records with a record ID
of ZFCP_DBF_HBA_FCES (5). Align output to the one of FSF response trace
records for readability. Note that the FSF protocol status is most likely
FSF_PROT_GOOD (0x00000001).
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support to report Fibre Channel (FC) Endpoint Security
related information for zfcp-host and zfcp-lun devices.
Allow to query (lszdev) and provide help for (chzdev) the FC Endpoint
Security capability of a FCP device (zfcp-host). Provide this
information by defining the FCP device attribute fc_security in zdev
zfcp-host.
Allow to query (lszdev) and provide help for (chzdev) the FC Endpoint
Security in use on a connection between a FCP device and a FC remote
port. Provide this information by introducing a virtual zdev zfcp-lun
attribute fc_security. It returns the fc_security attribute value of
the zfcp port object representing the FC remote port used to access the
LUN.
This approach allows users of lszdev to easily query the FC Endpoint
Security in use on the connections to one or more FCP LUNs. It also
does not require the effort to introduce a zfcp-port device type in
zdev.
The fc_security attribute help for both zfcp-host and zfcp-lun can be
queried for example as follows:
chzdev zfcp --help-attribute
The fc_security attribute values and path of all active zfcp devices
can be queried for example as follows:
lszdev zfcp -a -c TYPE,ID,ATTR:fc_security,ATTRPATH:fc_security
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
As preparation for a following commit reorder the path logic in function
zfcp_lun_st_get_active_attrib_path(). This then allows to handle
subsequent special cases more easily. Additionally it also enhances the
readability of the expression.
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for defining read-only attributes in the chzdev/lszdev
tools. These attributes can be used to provide online documentation
for specific attributes via the --help-attribute and --list-attributes
tool functions.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for the placement of the stage3b loader at other addresses
than 0xa000. For this add a position independent relocator that first
copies the original stage3b code to the memory location 0xa000 and
then starts it.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use C pre-processor for linker script generation. This allows the
usage of constants in our "linker scripts" `*.lds.S` (actually, these
are assembler files, so we can make us of the C pre-processor and its
capabilities).
Suggested-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a boot loader for protected virtualization (PV) that can be
combined with a kernel/initrd/parmfile to form a single bootable file.
This file must be constructed in a way that it can be used (1) for a
QEMU direct kernel boot and (2) it can be zipl'ed by the normal,
unmodified zipl program.
This new boot loader consists of two parts:
1. stage3a boot loader (cleartext), this loader is responsible for
the transition into the protected mode by doing diag308 subcode 8
and 10 calls.
2. stage3b boot loader (encrypted), this loader is very similar to the
normal zipl stage3 boot loader. It will be loaded by the Ultravisor
after the successful transition into protected mode. Like the zipl
stage3 boot loader it moves the kernel and patches in the values
for initrd and parmline.
The requirements for (1) and (2) result in the following constraints:
1. It must be possible to place stage3a and stage3b at a location >=
0x10000 because the zipl stage3 loader zeroes out everything at
addresses lower than 0x10000 of the image.
2. As the stage3 loader of zipl assumes that the passed kernel image
looks like a normal kernel image, the zipl stage3 loader modifies the
content at the memory area 0x10400 - 0x10800, therefore we leave this
area unused in our stage3a loader.
3. The default entry address used by the zipl stage3 loader is 0x10000
so we add a simple branch to 0x11000 at 0x10000 so the zipl stage3
loader can modify the area 0x10400 - 0x10800 without affecting the
stage3a loader.
The stage3b loader is linked at address 0x9000, therefore it will not
work at another address. The relocation support for the stage3b
loader, so that it can be placed at addresses != 0x9000, is added in
the next patch. This loader with relocation support has the name
'stage3b_reloc'.
The memory layout of the single bootable file looks like:
+-----------------------+-----------+------------------------+
|Start |End |Use |
+=======================+===========+========================+
|0 |0x7 |Short PSW, starting |
| | |instruction at 0x11000 |
+-----------------------+-----------+------------------------+
|0x10000 |0x10012 |Branch to 0x11000 |
+-----------------------+-----------+------------------------+
|0x10013 |0x10fff |Left intentionally |
| | |unused |
+-----------------------+-----------+------------------------+
|0x11000 |0x12fff |Stage3a |
+-----------------------+-----------+------------------------+
|0x13000 |0x13fff |IPIB used as argument |
| | |for the diag308 call |
+-----------------------+-----------+------------------------+
|0x14000 |0x1[45]fff |UV header used for the |
| | |diag308 call (size can |
| | |be either 1 or 2 pages) |
+-----------------------+-----------+------------------------+
|NEXT_PAGE_ALIGNED_ADDR | |Encrypted Kernel |
+-----------------------+-----------+------------------------+
|NEXT_PAGE_ALIGNED_ADDR | |Encrypted Cmdline |
+-----------------------+-----------+------------------------+
|NEXT_PAGE_ALIGNED_ADDR | |Encrypted Initrd |
+-----------------------+-----------+------------------------+
|NEXT_PAGE_ALIGNED_ADDR | |Encrypted Stage3b_reloc |
+-----------------------+-----------+------------------------+
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Consolidate `ALIGN, __ALIGN_MASK, ARRAY_SIZE` macros and add them to
lib/zt_common.h. While at it, adapt coding style.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for new deflate counters:
- Counter 247: cycles CPU spent obtaining access to Deflate unit
- Counter 252: cycles CPU is using Deflate unit
- Counter 264: Increments by one for every DEFLATE CONVERSION CALL
instruction executed.
- Counter 265: Increments by one for every DEFLATE CONVERSION CALL
instruction executed that ended in Condition Codes
0, 1 or 2.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add `sclp_print_ascii` function that can be used to print output on
the SCLP ASCII console. This would increase the size of the
loaders (e.g. eckd2.bin and fba2.bin) and therefore might break the
compilation. In order to avoid that add a macro 'ENABLE_SCLP_ASCII'
which must be defined by the users of the `sclp_print_ascii` function.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This allows the reuse of the code later in sclp.c. While at it, also
declare @source parameter of `ebcdic_to_ascii` function as `const` and
rename all `ebc_` function name prefixes into `ebcdic_`. Move
conversion tables to separate file so it only gets linked into loaders
that need it.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Errors fixed:
../../include/boot/s390.h: In function '__stfle_asm':
../../include/boot/s390.h:424:41: error: conversion to 'long unsigned int' from 'int' may change the sign of the result [-Werror=sign-conversion]
register unsigned long reg0 asm("0") = size - 1;
../../include/boot/s390.h: In function 'stfle':
../../include/boot/s390.h:453:31: error: conversion to 'long unsigned int' from 'int' may change the sign of the result [-Werror=sign-conversion]
nr = MIN((nr + 1) * 8, size * 8UL);
../../include/boot/s390.h: In function 'diag308':
../../include/boot/s390.h:296:9: error: conversion from 'long unsigned int' to 'int' may change value [-Werror=conversion]
return _rc;
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Now that we made sure that s390.h can be used with our minimal libc
implementation and glibc move s390.h to `include/boot/s390.h`. While
at it, make sure that s390.h is assembler compatible as it will be
used later in the PV boot loader and include s390.h in ipl.h as
PAGE_SIZE is used there.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This allows the use of s390.h in combination with other libc
implementations than our minimal libc, e.g. with glibc.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Rename `inline` macro into `__always_inline` so one can differentiate
between the macro and the C keyword. While at it, undefine the
previous `__always_inline` macro so s390.h can be used in combination
with glibc.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Move the loaders (stage2/stage3) layout values to
`include/boot/loaders_layout.h` and use the `_AC` macro for the
constants. This allows the reuse of the definitions, e.g. in assembler
files, and later for the creation of linker scripts.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use STAGE3_ENTRY for STAGE3_LOAD_ADDRESS as they have the same value
and this makes it more clear that the load address of stage3 is also
its entry point.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use the constants defined in `linux_layout.h`. Therefore move the
`PARMAREA` address offset to `linux_layout.h` and include the header.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Move the Linux layout values to `include/boot/linux_layout.h`. This
allows the reuse of the definitions, e.g. in assembler files, and
later for the creation of linker scripts.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make `IPL_DEVICE` definition consistent with the kernel definition and
the definition in tape0.S. This allows us to refactor the code later.
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Move the IPL related definitions into `include/boot/ipl.h`. This
allows the reuse of the definitions, e.g. in the boot loader for
protected guests.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Support `lib/zt_common.h` to be used in assembler files. In addition,
add the macro `_AC` that can be used to make constant macros usable in
both assembler and C code.
Suggested-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This can be helpful when diagnosing network performance issues.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The secure= option is not supported in the defaultboot section when a
menu is used. It should be placed in the menu section in this case.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The use of sprintf can easily result in buffer overflows as it assumes
that the buffer it writes to is large enough to contain the formatted
string. Thus replace sprintf by snprintf and update its users.
This removes the last user of vsprintf. Thus also remove vsprintf and
its dependencies.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Per definition vsprint assumes that the provided buffer it writes to is
large enough to contain the formatted string. As printf uses a fixed
sized buffer (81 bytes) and has no size checks the use of vsprintf can
easily cause buffer overflows. Protect against these buffer overflows by
using vsnprintf instead.
While at it fix a typo in the comment.
Reported-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zipl boot loaders have their own minimalistic libc implementation.
In it printf and sprintf use vsprintf for string formatting. Per
definition vsprintf assumes that the buffer it writes to is large enough
to contain the formatted string and performs no size checks. This is
problematic for the boot loaders because the buffer they use are often
allocated on the stack. Thus even small changes to the string format can
potentially cause buffer overflows on the stack with the well known
consequences. Protect against such errors by implementing vsnprintf.
Later patches will make use of it.
This implementation of vsnprintf only supports a small subset of format
options defined in the C standard. In particular it allows the
specifiers:
* %s (strings)
* %o (unsigned int octal)
* %u (unsigned int decimal)
* %x (unsigned int hexadecimal)
Integer specifiers (o, u, and x) always use the long form, i.e. assume the
argument to be of type 'unsigned long int'. The length modified 'l' can
be given but is ignored.
Furthermore, it is possible to provide the optional field width (aligned
to the right only) and precision as decimal integer (i.e. not via '*')
as well as the flag for zero padding integers (i.e. '0').
The implementation was heavily inspired by the implementation in
lib/vsprintf.c from the Linux kernel tree.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
A simple comment above the struct declaration to indicate that the
structure must not have any padding is prone to error. Therefore let's
add a check for the structure size at compile time.
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This also makes sure that we don't try to write to the
/sys/bus/pci/device/<dev>/recover attribute if reset failed.
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
After a zpcitctl --reset the PCI function is currently left in an error
state. This seems unexpected, so follow the SCLP reset with an OS level
recovery using /sys/bus/devices/<dev>/recover.
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use after free causes the XTS attribute to be displayed wrongly.
Fixes: 298fab68fe ("zkey: Preparations for introducing a new key type")
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Different crypto card types use different ways to re-encipher a secure
key with a new master key. Generalize the handling of re-enciphering,
so that the majority of the code does not have to care about the card
type when dealing with it.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
EP11 secure keys require a certain firmware version. Enhance
the APQN cross check to not only check for a minimum card level,
but also for a minimum firmware version.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a new source file that contains EP11 specific helper routines.
These routines require to load the EP11 host library.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Different crypto card types use different serial number formats.
Generalize the handling of serial numbers so that the majority of the
code does not have to care about the card type when dealing with it.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Different crypto card types use different master key verification patterns
(MKVPs). Generalize the handling of MKVPs so that the majority of the
code does not have to care about the card type when dealing with MKVPs.
Also generalize messages about the crypto card's master keys to not
mention a specific card type.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
EP11 cards provide the firmware version and API ordinal number as
a sysfs attribute on the crypto card level. Add a helper function
that obtains the firmware version and API ordinal number from there.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
As preparation for adding support for EP11 secure keys,
generalize the code to support different crypto card types.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zkey list function uses scandir() to look for files in the
zkey key repository directory. It checks the dirent.d_type field
to consider only regular files, but skip all others.
Unfortunately, not all file systems have full support for returning
the file type in d_type. When the zkey repository is located in a file
system that does not support d_type, such as xfs, zkey list shows no
keys, although the key repository contains keys.
Fix this by also considering directory entries with d_type = DT_UNKNOWN.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zipl creates a temp bootmap file to keep the original file in case of an
error. Delete this temp file in case of an error.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The hierarchy of the secure boot settings in the config file should be:
defaultboot > menu > section
This patch implements this hierarchy and adds a check if a valid option is
specified and prints an error message otherwise.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow that --secure or -S is specified on command line without the need to
allow all options on the command line.
Also ensure that the command line option overrules the config option and
correctly ensure that secure boot is only set for SCSI devices.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zkey convert command must select an CCA adapter out of the
list of associated APQNs (if any) to perform the key conversion.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
As the definition of `stfle_fac_list` in the lowcore uses uint64_t, we
should also use uint64_t for the `stfle_fac_list` parameter of the
`stfle/__stfle_asm` function.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use '$(MAKE)' to pass down the makefile flags.
This fixes the warning:
make[4]: warning: jobserver unavailable: using -j1. Add '+' to parent make rule.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
dasd_is_ro() uses the O_RDWR flag when opening the file descriptor.
However, an attempt to open the file descriptor of a device that is in
read-only state will fail and the BLKROGET ioctl that is used to
actually determine the read-only state can't be issued.
Change the flag to O_RDONLY to fix this.
Fixes: a5c1923d32 ("libdasd: Add common DASD ioctl's to the lib")
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The help output for 'zkey generate' has a typo in the description
of the '-N, --name' option mentioning a non-existing '-M' option.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
An invalid APQN like '01.00xx' is not treated as invalid, but results
in APQN '01.0000'. Fix the checking to treat such invalid APQNs as
invalid.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In error cases, dasdview returns -1 which results in the return value
255. This is due to the fact that only the low-order 8 bits are used for
the status value. See 2.13 Status Information [1] in the POSIX standard
and the exit() POSIX man page [2] for more details.
Instead of returning -1, use the EXIT_FAILURE constant to indicate
unsuccessful termination properly. This change also makes the exit
status consistent for all error cases in dasdview, as some exit() calls
already use EXIT_FAILURE.
[1]: https://pubs.opengroup.org/onlinepubs/9699919799/functions/V2_chap02.html
[2]: https://pubs.opengroup.org/onlinepubs/9699919799/functions/exit.html
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Increase consistency with the other macros by moving and renaming
the STAGE{2,3} macros in zipl.h.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make use of the pre-defined __packed macro throughout zipl. This
requires adding the global include dir to ALL_CFLAGS.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When adding new header from zipl/include to a .c file within zipl/boot
a compiler error appears
stage3.c:16:10: fatal error: zipl.h: No such file or directory
#include "zipl.h"
^~~~~~~~
compilation terminated.
CC zipl/boot/stage3.o
This is because the rule to generate dependencies (*.o.d) does not use
the CFLAGS_BOOT. Thus it cannot find the header and fails. Note this
only applies to the dependency generation, the actual build succeeds.
To fix this rename the CFLAGS_BOOT to ALL_CFLAGS. Using ALL_CFLAGS
instead of e.g. ALL_CPPFLAGS is important to also overwrite flags given
on the commandline via OPT_FLAGS, e.g.
make V=1 OPT_FLAGS="-D__FOO__"
While at it also remove the unused and wrong '-D__ASSEMBLY__'.
Fixes: 5a6605fe ("zipl: Ensure that boot loader CFLAGS are not overwritten")
Fixes: aa913b1e ("build process: Add automatic dependency generation")
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Currently zipl doesn't differentiate between the load address and the
entry point of an image, causing stage3 to strip away the entry point at
0x10000 for stand-alone kdump. This breaks the kdump kernel as it jumps
to 0x10000 after the special handling needed for kdump has been
performed.
Fix this by differentiating between the load address and the entry point
of an image.
Fixes: d142fbd5 ("zipl: Do not strip kernel image IPL header")
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The __packed__ attribute is not needed for these structures as they
don't need any padding to meet the size and alignment constraints
defined in the Linux for zSeries ABI.
Reported by GCC 9.2.1 when building with '-Waddress-of-packed-member'.
stage3.c: In function 'is_verified_address':
stage3.c:241:26: warning: taking address of packed member of 'struct ipl_rb_components' may result in an unaligned pointer value [-Waddress-of-packed-member]
241 | for_each_rb_entry(comp, comps) {
| ^~~~~
stage3.c:18:15: note: in definition of macro 'for_each_rb_entry'
18 | for (entry = rb->entries; \
| ^~
CC zipl/boot/kdump3.o
CC zipl/boot/sclp_stage3.o
sclp_stage3.c: In function '__sclp_hsa_copy':
sclp_stage3.c:75:34: warning: converting a packed 'struct sdias_sccb' pointer (alignment 1) to a 'struct read_sccb' pointer (alignment 4096) may result in an unaligned pointer value [-Waddress-of-packed-member]
75 | if (sclp_hsa_copy_wait((struct read_sccb *)sccb))
| ^~~~~~~~~
In file included from sclp_stage3.c:13:
sclp_stage3.h:43:8: note: defined here
43 | struct sdias_sccb {
| ^~~~~~~~~~
In file included from sclp_stage3.c:12:
sclp.h:149:8: note: defined here
149 | struct read_sccb {
| ^~~~~~~~~
sclp_stage3.c: In function 'sclp_hsa_get_size':
sclp_stage3.c:126:34: warning: converting a packed 'struct sdias_sccb' pointer (alignment 1) to a 'struct read_sccb' pointer (alignment 4096) may result in an unaligned pointer value [-Waddress-of-packed-member]
126 | if (sclp_hsa_copy_wait((struct read_sccb *)sccb))
| ^~~~~~~~~
In file included from sclp_stage3.c:13:
sclp_stage3.h:43:8: note: defined here
43 | struct sdias_sccb {
| ^~~~~~~~~~
In file included from sclp_stage3.c:12:
sclp.h:149:8: note: defined here
149 | struct read_sccb {
| ^~~~~~~~~
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Reported by GCC 9.2.1 when building with '-Waddress-of-packed-member'.
menu.c: In function 'menu_read':
menu.c:30:22: warning: taking address of packed member of 'struct boot_stage2_params' may result in an unaligned pointer value [-Waddress-of-packed-member]
30 | uint16_t *configs = __stage2_params.config;
| ^~~~~~~~~~~~~~~
menu.c: In function 'menu_list':
menu.c:83:22: warning: taking address of packed member of 'struct boot_stage2_params' may result in an unaligned pointer value [-Waddress-of-packed-member]
83 | uint16_t *configs = __stage2_params.config;
| ^~~~~~~~~~~~~~~
menu.c: In function 'menu':
menu.c:139:22: warning: taking address of packed member of 'struct boot_stage2_params' may result in an unaligned pointer value [-Waddress-of-packed-member]
139 | uint16_t *configs = __stage2_params.config;
| ^~~~~~~~~~~~~~~
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Reported by GCC 9.2.1 when building with '-Wmaybe-uninitialized'.
job.c: In function 'job_get':
job.c:1817:14: warning: 'filename' may be used uninitialized in this function [-Wmaybe-uninitialized]
1817 | scan_size = scan_file(filename, &scan);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
job.c:1791:14: note: 'filename' was declared here
1791 | const char *filename;
| ^~~~~~~~
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Reported by GCC 9.2.1 when building with '-Wdiscarded-qualifiers'.
job.c: In function 'get_job_from_config_file':
job.c:1810:14: warning: assignment discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When calling ziomon --version (or --help) the tool already checks
whether blktrace is present and prevents any output for those command
options when it isn't. This is unnecessary as no actual command is
issued during the parameter check and all other dependencies are checked
at a later point as well.
Move the dependency check to ensure a working --version and --help
option at any time.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In the BLS files, the 'linux' and 'initrd' fields are relative to the
$BOOT directory. If $BOOT is a mount point of boot partition, 'linux'
and 'initrd' fields are relative to boot partition. If not, the paths
are relative to $BOOT directory of root partition.
zipl always starts searching at / regardless of partitions and mount
points.
This commit extends the semantics of target= field in zipl.conf to make
it as a second search path, besides /, for those 2 BLS fields.
See issues #69 for more details.
Fixes: https://github.com/ibm-s390-tools/s390-tools/issues/69
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/74
Signed-off-by: Tuan Hoang <tmhoang@linux.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zipl config file (zipl.conf) and the BootLoaderSpec (BLS) fragments
in /boot/loader/entries define a set of tokens that are parsed by zipl.
These are stored in an array of tokens whose size is calculated to make
sure that there is enough memory allocated for all the scanned tokens.
But the size calculation logic was wrong, since it was checking if the
current size was enough to store a single token per BLS fragment, while
up to 4 tokens can be defined in a BLS file: a section heading and the
image, ramdisk and parameter keywords.
This led to zipl being killed by a SIGABRT signal when trying to parse
more tokens than the ones that could fit in the scanned tokens array:
Using config file '/etc/zipl.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-9.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-8.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-7.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-6.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-5.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-4.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-3.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-2.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-1.conf'
Using BLS config file '/boot/loader/entries/vmlinuz-0.conf'
double free or corruption (out)
Aborted (core dumped)
Fixes: https://github.com/ibm-s390-tools/s390-tools/issues/68
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/73
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This can be helpful in diagnosing network-related issues.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The basic statistics are already collected via ifconfig. But that tool
is deprecated, and provides no way of collecting the extended stats.
So also gather data from 'ip link'.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The duplicate statement does not have any effect. Therefore remove it.
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When a secure key has been converted from type CCA-AESDATA to type
CCA-AESCIPHER, the secure key stored in the LUKS2 header of a volume
encrypted with that key should also changed.
Command 'zkey-cryptsetup setkey' allows to set (replace) the volume
key in the LUKS2 header. It now accepts keys to be set that have
a different size of the original volume keys. CCA-AESCIPHER keys
are larger than CCA-AESDATA keys.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a new 'convert' command. It allows to convert a secure key from
one key type to another. Currently only keys of type CCA-AESDATA can be
converted to CCA-AESCIPHER.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Perform extended checks on a secure key that is imported into
the key repository. Warn the user if the imported key is by
any means insecure, e.g. has been originally created in an
insecure way. Prompt the user to continue the import if a
potential insecurity is detected.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The helper function performs a deep check of the AES CIPHER key
token and checks for any potentially insecure attributes.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Secure keys of type CCA-AESCIPHER can be export restricted, so that
these keys can not be exported by another key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some future functions are dependent on the firmware version of the
CCA crypto adapter. This helper function allows to query the version
of the currently selected CCA adapter.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Secure keys of type CCA-AESCIPHER require a CEX6C or newer crypto
card. Also check for the minimum required card level during cross
checking of APQNs. Also display the card level in the APQN report.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For secure keys of type CCA-AESCIPHER the CCA verb CSNBKTC2
(Key Token Change2) is used. CCA-AESDATA keys will continue
to use CCA verb CSNBKTC (Key Token Change).
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for validating secure keys using the new pkey
IOCTLs. This allows to validate secure keys of type CCA-AESDATA
as well as CCA-AESCIPHER.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for generating secure keys using the new pkey
IOCTLs. This allows to generate secure keys of type CCA-AESDATA
as well as CCA-AESCIPHER, either by random inside the crypto
card, or from a given clear key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The new IOCTLs are based on list of APQNs that they try to send
the request to. Add some helper functions to build such lists
of APQNs based on the key type, and optionally a given mkvp.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
By default the new pkey IOCTL are used. In case the pkey device does not
support the new IOCTLs (i.e. errno ENOTTY is returned), then the wrapper
falls back to the old IOCTLs. The old IOCTLs only support secure keys of
type CCA-AESDATA.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add definitions and helper functions to support the new
CCA-AESCIPHER key type. Also enhance existing helper functions
to support CCA-AESCIPHER keys.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Introduce helper functions and definitions to allow key type
independent code in the keystore implementation
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zkey generate command allows to specify the --key-type|-K
option to specify the key type. If not specified, then the
default is CCA-AESDATA.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zkey list command now accepts option --key-type|-K type
to filter the displayed keys by key type. If not specified,
then all key types are displayed.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For the 'zkey list', 'zkey validate' and 'zkey-cryptsetup validate'
commands, display the key type.
As of today there is only one possible key type (CCA-AESDATA),
but in the future there might be additional key types.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This patch adds CEX7S exploitation support to lszcrypt and chzcrypt.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To allow better control about the secure AES volume key re-enciphering
with 'zkey-cryptsetup reencipher', add options '--to-new' and '--from-old'
to specify if a re-enciphering from CURRENT to NEW, or OLD to CURRENT master
key registers is to be performed. If these options are not specified, then
it is auto-detected, based on the master key that the secure key is currently
re-enciphered with.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When re-enciphering secure AES keys, select the correct APQN for used
with the CCA host library. Re-enciphering a secure key requires the use
of the CCA host library. The APQN is selected based on the master key
verification pattern obtained from the secure key to re-encipher.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a utility function to select an APQN that is set up with
a specific master key for use with the CCA host library. The
selection is based on the master key verification pattern, which
is typically obtained from an existing secure AES key.
The function iterates over a set of APQNs to find one that is setup
with the desired master key in the CURRENT or OLD master key register,
and optionally has a new master key loaded. It then selects the found
APQN for use with the CCA host library.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Some operations require the CCA host library to be used, such as
re-enciphering a secure key. The CCA host library uses a different
approach to select the APQN it operates with. To ensure that the
desired APQN is used for an operation, a utility function is added
to select a specific APQN for usage with the CCA host library.
The CCA host library allows to set environment variables to override
the default CCA APQN selection. The environment variables are inspected
during CCA host library initialization only. To select a specific
domain for CCA, the CSU_DEFAULT_DOMAIN environment variable is set,
and then the CCA host library is un-loaded and re-loaded again.
Furthermore, the 'Cryptographic Resource Allocate' verb of the CCA
host library is used together with the 'Cryptographic Facility Query
function' verb to iterate over the crypto cards known by the CCA host
library, and to identify the desired crypto card based on its serial
number. That way, a specific APQN can be selected for use with
subsequent CCA verbs.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Perform a cross check of the APQNs when the APQN association of a
secure AES key in the key repository is changed. When adding new APQNs,
or associating a new set of APQNs to a secure key, then the APQNs are
cross checked. If all associated APQNs are removed, then all currently
available APQNs are cross checked. If a master key mismatch is detected,
then the change is rejected.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Perform a cross check of the APQNs when an existing secure AES key is
imported into the key repository. When a set of APQNs are associated to
the imported secure key, these APQNs are cross checked. If no APQNs are
associated to imported secure key, then all currently available
APQNs are cross checked. If a master key mismatch is detected, then
the key import is rejected.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Perform a cross check of the APQNs when a secure AES key is validated.
When a set of APQNs are associated to a secure key, these APQNs are
cross checked. If a secure key is validated outside of the key repository,
or no APQNs are associated to a secure key inside the key repository,
then all currently available APQNs are cross checked. If a master key
mismatch is detected, then an error message is issued.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Perform a cross check of the APQNs when a new secure AES key is
generated. When a set of APQNs are associated to a new secure key,
these APQNs are cross checked. If a new secure key is generated
outside of the key repository, or no APQNs are associated to a secure
key generated inside the key repository, then all currently available
APQNs are cross checked. If a master key mismatch is detected, then
the key generation is rejected.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Display the master key verification pattern of a secure key while
'zkey validate' and 'zkey-cryptsetup validate'
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
A secure AES key token contains the master key verification pattern
of the master key it is encrypted with. Add a function to obtain the
master key verification pattern of a secure key token.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a utility function to cross check the master keys of a set of
APQNs. It checks for valid master keys in the CURRENT and OLD
master key registers, as well as newly loaded master keys in the NEW
register. It issues information and warning messages for various
findings and also indicates improper master key setup to the caller.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a utility function to print the master key verification patterns
of a set of APQNs. This allows the user to visually check which
master keys are set on which APQNs.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a utility function to iterate over all available APQNs of
type CCA-Coprocessor. This function is required for various
future enhancements.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With recent changes in the zcrypt device driver, the master key verifi-
cation patterns of the AES master key of am APQN can be obtained by
reading the sysfs attribute 'mkvps' of an APQN device of type CCA-
Coprocessor. The sysfs attribute can be found under
'/sys/devices/ap/cardnn/nn.mmmm/', where nn specifies the card number
in hex, and mmmm specifies the domain number on hex.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With recent changes in the zcrypt device driver, the serial number of
a crypto card can be obtained by reading the sysfs attribute 'serialnr'
of a crypto card device of type CCA-Coprocessor. The sysfs attribute
can be found under '/sys/devices/ap/cardnn/', where nn specifies the
card number in hex.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
As preparation for future changes, move a sysfs specific functions
into a separate source file (utils.c).
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
As preparation for future changes, rework the loading of the
CCA host library so that the exported symbols are not passed
individually to the functions that use it. Pass a structure
that contains all entry points of all loaded CCA functions
instead. This will make it easier to add further CCA functions
at a later time.
Also add a version query for the CCA host library since some
future functions might be dependent on the library version.
While at it, separate the CCA related functions and definitions,
and move them into a separate source file (cca.h/cca.h).
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
...
==9448== 32 bytes in 1 blocks are definitely lost in loss record 1 of 3
==9448== at 0x4831010: malloc (vg_replace_malloc.c:309)
==9448== by 0x48E2205: __vasprintf_internal (vasprintf.c:71)
==9448== by 0x1005061: misc_asprintf (misc.c:48)
==9448== by 0x10131F3: disk_get_info (disk.c:239)
==9448== by 0x1013C41: disk_get_info_from_file (disk.c:517)
==9448== by 0x1011C87: bootmap_create (bootmap.c:1127)
==9448== by 0x1004C73: main (zipl.c:192)
...
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This patch adds ethtool commands for network interfaces.
Especially the offload settings are important.
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This patch adds the base lspci command to list pci devices
and the smc_dbg command from the smc-tools package to show
SMC socket related information.
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zdev incorrectly reports an invalid udev rule for zfcp-lun type rules.
This is caused by a faulty check for an empty udev rule file. This
commit fixes the check to only report invalid rules when there was no
valid configuration directive found in a zfcp-lun type udev rule.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reported-by: Steffen Maier <maier@linux.ibm.com>
Reported-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Show '[y/N]' on all user confirmation prompts to give the user a
hint what to reply. Also indicate that the operation was aborted
on a negative confirmation.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The volume type of a secure key is not really case sensitive,
but for better usability store and display it in uppercase always,
regardless in whatever case it was specified.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zfcp dumper fails with the following error:
MLOPDM003I: Machine loader finished, moving data to final storage
location.
uncompression error
--- System halted
HCPGIR450W CP entered; disabled wait
PSW 00020001 80000000 00000000 DEADBEEF
The zipl command shows overlapping components when installing the dumper
to a zfcp disk:
zipl -d /dev/sda1
Building bootmap directly on partition '/dev/sda1'
Adding dump section
kernel image......: /lib/s390-tools/zfcpdump/zfcpdump-image
kernel parmline...: 'root=/dev/ram0 dump_mem=1 possible_cpus=1
cgroup_disable=memory '
component address:
heap area.......: 0x00002000-0x00005fff
stack area......: 0x0000f000-0x0000ffff
internal loader.: 0x0000a000-0x0000dfff
parameters......: 0x00009000-0x000091ff
kernel image....: 0x00010000-0x005761ff
^^^^^^
parmline........: 0x00567000-0x005671ff
^^^^^^
Preparing boot device: sda.
Done.
With the secure IPL patchset the offset of the kernel image has been
removed for the normal IPL case but it has not been removed for the dump
image which leads to the overlap of 0x10000.
Fix by removing the offset for the dump case.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewd-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Tested-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make dasdfmt aware of thinly provisioned (Extent Space Efficient (ESE))
DASD volumes.
If an ESE volume is recognised a QUICK format is performed, formatting
only the first two tracks. The mode can always be overwritten by --mode.
Previously allocated space is always released before formatting, if not
specified otherwise. The option --no-discard (-D) is provided to omit
the space release.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The BIODASDRAS (Release Allocated Space) ioctl allows to release
track-wise or full space of an Extent Space Efficient DASD volume.
Provide a library function to utilise this functionality.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Provide a function to determine whether a DASD is an Extent Space
Efficient (ESE) volume.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Move definitions mostly used by IOCTLs to libdasd. Clean up unused and
old definitions while at it as well.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add basic checking when reading udev rules and print a warning when an
invalid udev rule file is found. This addresses the current inconsistent
behavior:
- CCW devices: invalid udev rules are considered valid
- CCW group devices: invalid udev rules are silently ignored
- zFCP LUNs: invalid rules cause chzdev/lszdev to silently terminate
with non-zero exit code
Checks include a test for empty or truncated rule files that might be
the result of an interrupted chzdev operation, or a file system or I/O
error.
Note: The recommended way to correct invalid udev rules is to either
remove the offending rules file, or to repeat the associated persistent
configuration step.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reported-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zipl internal loader adds a variable BOOT_IMAGE= to the commandline
so that it is visible in the operating system which menu entry has been
chosen.
This entry was overwritten by the stage3 parameter page.
Fix by re-arranging the internal memory layout and putting the command
line extra param, which contains the BOOT_IMAGE entry, at 0xe000.
This location is available because less than one page is used for the
stack.
Fixes: https://github.com/ibm-s390-tools/s390-tools/issues/67
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When 'zkey generate' is used with options '--clearkey <file>' and
'--xts', but without the '--keybits <bitsize>' option, then the
auto-detection of the bitsize of the specified clear key fails.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Dependent on kernel config option CHSC_SCH the device node /dev/chsc may
exist per default or not. In case the kernel is built with CHSC_SCH=M,
then the device driver is built as separate kernel module, and must
therefore be loaded using modprobe chsc_sch to make device node /dev/chsc
available.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Currently the zipl tool doesn't remove trailing spaces from the BLS field
values. So for example if a 'title' field has trailing spaces and is used
as the default, zipl will complain that there's no section with that name:
Using config file '/etc/zipl.conf'
Using BLS config file '/boot/loader/entries/f871a0cf218348c5ba921f61c92b7eac-4.18.0-80.20.el8.s390x.conf'
Using BLS config file '/boot/loader/entries/f871a0cf218348c5ba921f61c92b7eac-0-rescue.conf'
Error: Config file '/etc/zipl.conf': Line 6: no such section 'Red Hat Enterprise Linux (4.18.0-80.20.el8.s390x) 8.1 (Ootpa)'
Since the trailing spaces are also removed from the fields defined in the
zipl.conf file, do the same for the ones that are defined in the BLS file.
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/62
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Acked-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for CPU-Measurement facility counter second version
number 6. This adds some more counters to the crypto counter set.
Extended counter set is the same as for z14.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The util_opt_example allows to specify '-l' and '-m, --manual' options,
but does not handle them. This leads to error message 'PANIC: The
application terminated due to an unrecoverable error' with 'Option 'l'
should not be handled here' and the program is aborted.
Add the required case statements in the switch to handle those options.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zfcp kernel module was changed to introduce separate parameters for
selecting DIF and DIF&DIX. This commit implements the corresponding
changes in chzdev and lszdev.
Signed-off-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When directory $SYSFSDIR/bus/ccwgroup does not exist znetconf issues an
error message and terminates. Its not even possible to get the output of
--help or --version. Fix that by moving the responsible check behind the
parameter evaluation.
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The validate command should allow to specify the --no-apqn-check
option even when --apqns is not specified.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The stfle inline assembly returns the number of double words written
(condition code 0) or the double words it would have written
(condition code 3), if the memory array it got as parameter would have
been large enough.
The current stfle implementation assumes that the array is always
large enough and clears those parts of the array that have not been
written to with a subsequent memset call.
If however the array is not large enough memset will get a negative
length parameter, which means that memset clears memory until it gets
an exception and the system crashes.
To fix this simply limit the maximum length. Move also the inline
assembly to an extra function to avoid clobbering of register 0, which
might happen because of the added MIN invocation together with code
instrumentation.
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The header exit_code.h is required for the exit_code_t type and
the header misc.h for the err_t type.
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The function util_path_exists() consumes its first parameter as format
string. Passing the user input specified to the chzdev option --export
as-is to the function opens up the possibility for a format string
injection.
Fixes: d542138868 ("zdev: use libutil provided path functions")
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Testing whether a file is a directory by comparing the struct stat
field st_mode against S_IFDIR is wrong. If st_mode has any access
permission bits set along with the file type code the test will always
fail.
The file type encoded in the struct stat field st_mode is actually an
enumeration. To test whether a file is a directory either extract the
file type from st_mode using the mask S_IFMT and compare it against
S_IFDIR or simply use the macro S_ISDIR().
Fixes: b627b8d8e1 ("Initial s390-tools-2.0.0 import")
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Testing whether a file is a directory by masking the struct stat field
st_mode with S_IFDIR is wrong. Depending on the values of the macros
S_IF* block special devices might be considered as directories.
The file type encoded in the struct stat field st_mode is actually an
enumeration. To test whether a file is a directory either extract the
file type from st_mode using the mask S_IFMT and compare it against
S_IFDIR or simply use the macro S_ISDIR().
Fixes: c944f23d7e ("zkey: Add keystore implementation")
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Acked-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The functions use the C99 _Bool type in form of the alias bool from
stdbool.h as return type. Scalars are implicitly converted to _Bool.
Values equal to zero to 0 (false). Nonzero values to 1 (true).
Therefore the explicit conversion using the ternary operator ?: with
true and false is not required.
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Testing whether a file is a directory or a regular file by masking the
struct stat field st_mode with S_IFDIR or S_IFREG is wrong. Depending
on the values of the macros S_IF* sockets and symbolic links might
erroneously be considered as regular files and block special files as
directories.
The file type encoded in the struct stat field st_mode is actually an
enumeration. To test whether a file is a directory or a regular file
either extract the file type from st_mode using the mask S_IFMT and
compare it against S_IFDIR or S_IFREG or simply use the macros S_ISDIR()
and S_ISREG().
Fixes: b627b8d8e1 ("Initial s390-tools-2.0.0 import")
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Since commit d542138868 ("zdev: use libutil provided path functions")
lszdev fails to correctly report most read-only sysfs attributes as such
in its detailed information output (options -i or -ii). This is because
the libutil function util_path_is_writable() is not equivalent to the
former function file_writable().
util_path_is_writable() always returns true (the underlying function
access() indicates the file is writable), if the path exists and the user
is root, regardless of the path's effective access permissions. This is
correct for most use cases, as root can effectively read/write any path
regardless of its effective access permissions). util_path_is_writable()
behaves exactly like Bash -w in this regard.
The former function file_writable() examined the file's access permissions
to check if one or more of user, group, and other have write permission.
For lszdev it is irrelevant whether a file is effectively writable by
the current user. Instead it needs to determine whether sysfs attributes
access permissions contain write access in any form. If not it lists
those sysfs attributes in a separate read-only section.
Use the new function util_path_is_readonly_file(), which does explicitly
check the file's effective read and write access permissions. It behaves
similar to the former function file_writable(), except that it does test
whether any of user, group, and other have read permission.
Fixes: d542138868 ("zdev: use libutil provided path functions")
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For root the functions util_path_is_readable() and util_path_is_writable()
do not take the path's permissions into account. The function access()
succeeds as root is allowed to read/write any file regardless of its
permissions.
Introduce the functions util_path_is_readonly_file() and
util_path_is_writeonly_file() to test whether a path is a regular file and
is either read-only (neither user, group, nor other have write permission,
but one or more of user, group, and other have read permission) or
write-only (neither user, group, nor other have read permission, but one
or more of user, group, and other have write permission).
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The file pointer in the functions is_online() and cpu_is_configured() is
only freed in the good case. Make sure it is always freed.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
bootmap.c: In function ‘check_secure_boot_support’:
bootmap.c:131:2: warning: ignoring return value of ‘fscanf’, declared
with attribute warn_unused_result [-Wunused-result]
fscanf(fp, "%d", &val);
^~~~~~~~~~~~~~~~~~~~~~
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zpcictl.c: In function ‘sysfs_read_value’:
zpcictl.c:154:2: warning: ignoring return value of ‘fscanf’, declared with
attribute warn_unused_result [-Wunused-result]
fscanf(fp, "%x", &val);
^~~~~~~~~~~~~~~~~~~~~~
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add option --no-apqn-check to the generate, import, change, and
validate commands to disable checking of the specified APQNs.
With this option a currently non-existing APQN can be associated
with a key. This is useful to associate APQNs that exist only on
other systems, such as disaster recovery systems, but not on the
current system. When generating keys, at least one of the specified
APQNs must be available to generate the key.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Function get_payload_records() reading from disk, blindly adds payload
record (junks) to a list in a 2-stage hierarchical hash keyed with
request ID and PAY area.
On adapter offline/online, zfcp resets the request ID to re-start with 1.
Hence it's not a unique foreign key in the lifetime of the PAY area.
The last payload record (junks) read for the same request ID and PAY area
is the one printed as named PAY field in the tool output.
This is not necessarily the chronologically last record due to PAY area
wrapping.
With this workaround, we:
* warn on reading PAY records if ambiguous entries occurred,
* print additional PAY timestamps which the user can compare
with the actual trace record timestamps to determine if they are close
enough to indicate a match.
We only found this by accident because we kept "short" "payload" being a
built-in part of some actual trace record types in areas HBA, SAN, SCSI.
For above cases, the full PAYload did not match the
built-in "short" "payload".
While at it, also add a comment that PAY output can also be wrong
in another case when junks of multi-part PAY records are missing due
to PAY area wrapping.
Below is an abbreviated example output with this workaround:
Loading trace records ...(this might take a while)
Warning: Ambiguous PAYload records. reqid:0000000000000037 area:san_res
...
Timestamp : 2017-07-07-18:01:34:574827
Area : SAN
...
Tag : fssct_1
Request ID : 0x0000000000000037
Destination ID : 0x00fffffc
SAN req short : 01000000 fc020000 01720ffc 00000000
00000008
SAN req length : 20
Timestamp : 2017-07-07-18:01:34:619310 ------------------+
Area : SAN |
... |
Tag : fsscth2 |
Request ID : 0x0000000000000037 |
Destination ID : 0x00fffffc |
SAN resp short : 01000000 fc020000 80020000 00000000 |
00671a07 00000000 c05076ff d6801850 |
00671a7e 00000000 c05076ff d6801968 |
SAN resp length: 16384 |
Payload time : 2017-07-07-18:16:17:165708 <== does not match
San resp info : 01000000 fc020000 80020000 00000000 <== wrong
4a49424d 20202020 20323936 34202020 <== wrong
20202020 20202020 20303230 30303030 <== wrong
30303745 43383720 20504348 49443a20 <== wrong
30313938 204e5049 5620556c 7049643a <== wrong
20303536 30303330 30202020 00000000 <== wrong
00000000 00000000 00000000 00000000 <== wrong
... <== wrong
...
Timestamp : 2017-07-07-18:16:17:165713
Area : SAN
...
Tag : fsscth2
Request ID : 0x0000000000000037 <== same reqid & pay "area"
Destination ID : 0x00fffffc
SAN resp short : 01000000 fc020000 80020000 00000000
4a49424d 20202020 20323936 34202020
20202020 20202020 20303230 30303030
SAN resp length: 273
Payload time : 2017-07-07-18:16:17:165708 <== 2nd & last, match
close to Timestamp
San resp info : 01000000 fc020000 80020000 00000000
4a49424d 20202020 20323936 34202020
20202020 20202020 20303230 30303030
30303745 43383720 20504348 49443a20
30313938 204e5049 5620556c 7049643a
20303536 30303330 30202020 00000000
00000000 00000000 00000000 00000000
...
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Reading DASD specific sysfs attributes should be collected in one
library. Move get_host_access_count() to libdasd/dasd_sys.
Remove the old implementation and update any user accordingly.
Also, fix the build order for zdsfs.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use the newer and more robust libutil provided function
util_sys_get_dev_addr() to identify a device address for any former user
of u2s_getbusid().
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Many tools need to identify the device address of a given device. So
far, either the tool had its own implementation, or u2s_getbusid() was
used. Though, u2s_getbusid() was mainly designed for and used by the
DASD tools.
Introduce util_sys with a first function util_sys_get_dev_addr() which
provides a more universal way to identify the device address which is
not limited to one particular device type. The device address represents
either a busid (e.g. DASD), slot address (NVMe), H:C:T:L tuple (SCSI),
or other id types associated with a device.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
dasdview and libdasd deal with DASD bus ids and should use a definition
for the size. libu2s already provides a definition, which is used by
dasdview. However, the size of 32 is a bit much and the definition
should be part of libdasd.
Including the terminating null byte ('\0') and considering a DASD bus id
length of 8 characters (e.g. 0.0.4711), this leads to a size of 9.
Provide such a definition via libdasd and update any user.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Don't rely on rather arbitrary values for the string buffer size of
different sysfs paths. Instead use util_path_sysfs(), which allocates
the exact amount of memory necessary for a certain path string.
As these path strings are dynamically constructed, potential buffer
overflows can hereby be avoided and make the library functions more
robust.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add zipl and zipl.conf man page updates.
The zipl man page will look like:
-S <SWITCH> or --secure <SWITCH>
Control the zIPL secure boot support. <SWITCH> can take one of three values:
auto (default)
Write signatures if available and supported by the system.
1
Signatures are written independent of support indicated by the local
system. Also missing signatures for stage 3 and kernel IPL files
will result in an error.
0
No signatures will be written.
The zipl.conf man page will look like:
secure = auto/1/0 (configuration only)
Configuration section:
Control the zIPL secure boot support. Set this option to one of the following:
- auto: Write signatures if available and supported by the system.
- 1: Signatures are written independent of support indicated by the local system.
Also missing signatures for stage 3 and kernel IPL files will result in an error.
- 0: No signatures will be written.
The default value for 'secure' is auto.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To ensure that only verified code is executed the stage 3 loader has
to check if the load psw points to an address that has previously been
verified by the machine loader.
If secure boot is enabled the IPL process will only continue if the
verification was successful.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The firmware needs to know which memory locations are used by
the stage 3 loader so that it can allocate its own memory.
To indicate the usage of the heap and stack area of the stage 3 loader
add a dummy component to block this memory area for the firmware.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The stage 3 loader size has to be a multiple of the filesystem
blocksize so that the signature maps with the later in memory
stage 3 loader. The lowest common blocksize is 4k so the image
is padded to a 4k boundary.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a command line option and configuration section keyword to control
the zIPL secure boot support. This option is named "secure" and can take
one of three values:
auto (default)
Write signatures if available and supported by the system.
1
Signatures are written independent of support indicated by the local
system. Also missing signatures for stage 3 and kernel IPL files
will result in an error.
0
No signatures will be written.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for parsing IPL file signatures as produced by the Linux
kernel's sign-file tool. When preparing a disk for IPL, the signatures
will be added as separate data components that can be used by the Secure
Boot firmware feature to validate the integrity of the IPL files.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Change the fixed component entry text array to a more dynamic approach
to allow easier extension and make the order changeable.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The first component that gets control after IPL is the zIPL-provided
stage 3 boot loader. This boot loader is based on a binary image linked
into the zIPL executable. When preparing a disk for IPL, zIPL updates
runtime parameters in the boot loader image and writes the resulting
data to disk.
To support the Secure Boot firmware feature, the stage 3 code must be
signed, and can therefore no longer contain variable parameter portions.
Also the boot loader image should be accessible as a separate file to
enable the use of external programs for creating the signature.
This patch moves the stage 3 boot loader code to an external file
location and splits out the parameter portion into a separate, unsigned
on-disk component.
The new memory layout of the stage 3 loader during its execution looks
as follows:
* 0x0000-0x1fff Lowcore
* 0x2000-0x5fff Memory allocation (heap)
* 0x7000-0x8fff free
* 0x9000-0x9fff Stage3 parameter
* 0xa000-0xcfff Stage3 code
* 0xd000-0xefff Section: bss, rodata, data
* 0xf000-0xffff Stack
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The binary Linux kernel image is built to be loaded to memory address
0x0 but the first 64 kbyte contain an IPL header that is not used for
disk IPL. zIPL strips away this IPL header when writing IPL records to
disk, loads the remaining data to memory address 0x10000 and uses the
memory area below that for its own boot loader code.
The Secure Boot firmware feature checks the integrity of an installed
image during IPL using a checksum that was generated for the full image.
Since the checksum becomes invalid if the IPL header is removed, zIPL
must be changed to write the full image to disk.
This patch modifies the zIPL logic to no longer strip away the IPL
header. Instead the full image is loaded to a higher memory address and
relocated by the stage 3 boot loader code to its final location.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When a 'zkey change' command is used to change (i.e. set) the volume
association of a key, and the new volume association contains a volume
that is already assigned to the key to be changed, the command fails
with an error that the volume is already associated to the key.
This commit fixes the problems and allows to set such a volume
association. When setting a new association the current association of
the key to be changed is of no relevance, since the set operation
replaces the current association with the new association.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
s390-tools-1.9.0 introduced a new way of recognizing of multipath
device paths with using of sed command invocation in ziomon script.
With this new way of recognizing, if there are paths, related to
multipath device, with SCSI host ID longer, than one digit,
ziomon incorrectly parses the multipath -l command output. It
erroneously cuts off all but the least significant digit of the
SCSI host ID (H) of paths in H:B:T:L format (Host:Bus:Target:Lun).
This leads to passing of hosts (-a) and paths (-l) with
non-existing SCSI host ID to ziomon_util. In turn ziomon_util
cannot recognize hosts with non-existing SCSI host ID and issues
an error.
Also, wrong sed command invocation could lead to receiving of
duplicate LUNs by ziomon after parsing of multipath -l command
output. Then ziomon excludes duplicates from WRP_LUNS, which
leads to mismatch between number of LUNs and number of detected
block devices and issues ziomon script error, without starting
ziomon_util and without writing to specified log file.
The regular expression to match a path in H:B:T:L format started
with a greedy ".*", which erroneously consumed parts of the SCSI
host ID (H). This patch replaces the greedy ".*" by "[^0-9]*",
so that sed command does not consume parts of the SCSI host ID
any more.
Test example with unique SCSI host IDs:
$ multipath -l
...
mpathc (36005076307ffc5e300000000000083f5) dm-2 IBM ,2107900
size=20G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='service-time 0' prio=0 status=active
|- 10:0:0:1089814659 sdb 8:16 active undef running
`- 11:0:0:1089814659 sdf 8:80 active undef running
...
Behavior without fix applied:
$ ziomon -d 5 -o log /dev/mapper/mpathc
Check devices...done
NOTE: No size limit specified, run without a limit.
Estimated maximum disk space required for log data: approx. <1 MBytes
Collecting configuration data...done
Start data collection processes...ziomon_util: Path does not exist: /sys/class/scsi_host/host0/utilization - correct kernel version?
ziomon_util: Path does not exist: /sys/class/scsi_host/host0/utilization - correct kernel version?
ziomon_util: Path does not exist: /sys/class/scsi_host/host0/queue_full - correct kernel version?
ziomon_util: Path does not exist: /sys/class/scsi_host/host1/utilization - correct kernel version?
ziomon_util: Path does not exist: /sys/class/scsi_host/host1/queue_full - correct kernel version?
failed
ziomon: Failed to determine ziomon_util pid
Shutting down
Shutting down blktrace process
Shutting down blkiomon process
Shutting down ziomon_zfcpdd process
blkiomon: terminated by signal
Shutting down data manager
User can see more information, when using ziomon with -V option:
...
=== WRP_LUNS : 0:0:0:1089814659 1:0:0:1089814659
=== WRP_HOST_ADAPTERS: host0 host1
...
=== starting ziomon_util: ziomon_util -V -a 0 -a 1 -l 0:0:0:1089814659 -l 1:0:0:1089814659 ...
...
User can also see this type of errors, while using
ziorep_traffic or ziorep_utilization tools:
$ ziorep_traffic -t1 log.log
Extracting config data...done
ziorep_traffic: Could not retrieve initial data - data files corrupted or broken, or the .agg file is missing.
Behavior with fix applied:
$ ziomon -d 5 -o log /dev/mapper/mpathc
Check devices...done
NOTE: No size limit specified, run without a limit.
Estimated maximum disk space required for log data: approx. <1 MBytes
Collecting configuration data...done
Start data collection processes...done
Collecting data...done
Shutting down
Shutting down data manager
User can see more information, when using ziomon with -V option:
...
=== WRP_LUNS : 10:0:0:1089814659 11:0:0:1089814659
=== WRP_HOST_ADAPTERS: host10 host11
...
=== starting ziomon_util: ziomon_util -V -a 10 -a 11 -l 10:0:0:1089814659 -l 11:0:0:1089814659 ...
...
Test example with duplicate SCSI host IDs:
$ multipath -l
...
mpathc (36005076307ffc5e300000000000083f5) dm-1 IBM ,2107900
size=20G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='service-time 0' prio=0 status=active
|- 0:0:0:1089814659 sdb 8:16 active undef running
`- 10:0:0:1089814659 sdf 8:80 active undef running
...
Behavior without fix applied:
$ ziomon -d 5 -o log /dev/mapper/mpathc
Check devices...done
ziomon: Number of LUNs does not match number of devices: 2 devices and 1 LUNs
User can see more information, when using ziomon with -V option:
...
=== #Devices total : 2
=== WRP_DEVICES : /dev/sdb /dev/sdf
=== WRP_LUNS : 0:0:0:1089814659
=== WRP_HOST_ADAPTERS: host0
...
Behavior with fix applied:
$ ziomon -d 5 -o log /dev/mapper/mpathc
Check devices...done
NOTE: No size limit specified, run without a limit.
Estimated maximum disk space required for log data: approx. <1 MBytes
Collecting configuration data...done
Start data collection processes...done
Collecting data...done
Shutting down
Shutting down data manager
User can see more information, when using ziomon with -V option:
...
=== #Devices total : 2
=== WRP_DEVICES : /dev/sdb /dev/sdf
=== WRP_LUNS : 0:0:0:1089814659 10:0:0:1089814659
=== WRP_HOST_ADAPTERS: host0 host10
...
=== starting ziomon_util: ziomon_util -V -a 0 -a 10 -l 0:0:0:1089814659 -l 10:0:0:1089814659 ...
...
Signed-off-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In case a regular directory was specified, rather than a device node,
the check if the device exists will pass. The following code paths then
assume a slot id was specified. This in turn may lead to a buffer
overflow when the device data is copied to to the zpci_device struct.
Check if the specified path is a regular directory and prevent a
possible later buffer overflow and copying wrong data respectively.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Also mark FCP devices with CHPID that is logically varied off,
e.g. after running "chchp -v 0 60".
Such FCP devices can be online, not failed, and are not defunct,
so the other existing markers do not appear.
Shows potential issues with FCP devices at one glance.
Examples:
$ lszfcp -He
0.0.1900 host0 NotAvailable
$ lszfcp -HeV
/sys/devices/css0/0.0.010d/0.0.1900 NotAvailable
/sys/class/fc_host/host0
/sys/class/scsi_host/host0
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Complements: 4036e80b26 ("lszfcp: add new output marker for non-good FCP devices (hosts)")
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zcryptstats tool displays usage statistics of IBM Crypto Express
adapters. It obtains cryptographic performance measurement data
periodically and displays the data for each cryptographic device for
each interval. A cryptographic device can be either a card device or
a queue device (APQN).
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
I often use "systool -v -m zfcp" but that's a bit longish to type
and contains more output than just the parameters I'm interested in.
"tail /sys/module/zfcp/parameters/*" is also not so nice.
"lszfcp -Z" makes it very easy and can be combined with the other
existing output options of lszfcp.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Since zfcp lazily allocates Scsi_Host et al. on setting an FCP device
online for the first time since the last driver core bind to zfcp
(such as zfcp kernel module load),
lszfcp has only been showing FCP devices (vHBAs) that have been online
at least once including those that have been set offline again.
This is somewhat confusing.
Also it would be nice if lszfcp could be an easier to type replacement for
"lscss -t 17032/03" to simply show all devices bound to the zfcp device
driver.
Of course those FCP devices that have never been online, won't have any
zfcp-specific attribute details, only CIO attributes from the CCW bus.
They also do not have a Scsi_Host nor fc_host.
In order not to break any scripting or expectations around lszfcp,
make the new feature depend on the extended output command line option.
Example:
$ lszfcp -He
0.0.1880 - offline
$ lszfcp -HeV
/sys/devices/css0/0.0.010b/0.0.1880 offline
-
-
$ lszfcp -Hea
0.0.1880 - offline
Bus = "ccw"
availability = "good"
cmb_enable = "0"
cutype = "1731/03"
devtype = "1732/03"
modalias = "ccw:t1731m03dt1732dm03"
online = "0"
uevent = "DRIVER=zfcp"
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
s390-tools-1.8.1 introduced a new error message.
It gave the same and thus misleading error message as if no FCP device
(vHBA) found.
In order not to break any scripting built around lszfcp output,
only print the fixed error message depending on the extended output
command line option.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
s390-tools-1.8.0 introduced a regression skipping all FCP devices
that are currently "defunct".
Changing
$SYSFS/devices/css0/*/*/host*
to
$SYSFS/devices/css0/[0-9]*/[0-9]*/host*
was too strict as it won't match
$SYSFS/devices/css0/defunct/[0-9]*/host*
anymore.
Fix:
$SYSFS/devices/css0/[0-9d]*/[0-9]*/host*
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
s390-tools-1.8.1 introduced a regression skipping all fc_rports
with port_state != Online. While we need to skip some output parts
for a zfcp_port which does not exist anymore due to port_remove,
there are other cases with port_state != Online for which we should
not skip an fc_rport.
Such as port_state "Blocked" after a cable pull beyond host-adjacent switch
or port_state "Not Present" after dev_loss_tmo ran out.
This way the user can see the previously discovered ports,
even if they might currently not be usable for traffic.
Non-good ports are marked in the short output with suffix "NotOnline".
In order not to break scripting or expectations around lszfcp, make the new
feature depend on the extended output command line option.
Sysfs path globbing does not work for a removed zfcp_port
to obtain the FCP device bus-ID. Instead, properly walk the sysfs.
Examples with a removed zfcp_port:
$ lszfcp -Pe
0.0.50c0/0x500507680b2481fa rport-6:0-2
$ lszfcp -PeV
- (NoMoreZfcpPort:0.0.50c0/0x500507680b2481fa)
/sys/class/fc_remote_ports/rport-6:0-2
Examples with an existing zfcp_port but pulled fibre:
$ lszfcp -Pe
0.0.3c40/0x500507680b2181fb rport-7:0-4
$ lszfcp -PeV
/sys/devices/css0/defunct/0.0.3c40/0x500507680b2181fb
/sys/class/fc_remote_ports/rport-7:0-4
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Used in subsequent patches to guard their changes in order not to break
any scripting or expectations around the lszfcp output format.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When re-enciphering or setting LUKS2 volume keys using zkey-cryptsetup,
allow to specify the --batch-mode|-q option to suppress confirmation
questions.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For LUKS2 volumes the cryptsetup command currently generates
'cryptsetup luksFormat' commands, but for plain mode volumes it
generates 'cryptsetup plainOpen' commands.
With the --open option it can now also generate 'cryptsetup luksOpen'
commands for opening LUKS2 volumes. With the --format option you can
limit the generated commands to only generate 'cryptsetup luksFormat'
commands for LUKS2 volumes, and skip plain mode volumes.
The default behavior (i.e. if none of the two options specified),
remains the same as before.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When generating crypttab entries for LUKS2 volumes, allow to
specify common passphrase options like --key-file, --keyfile-offset,
--keyfile-size and --tries and pass those to the generated crypttab
entries.
Note that not all distributions support the keyfile-offset and
keyfile-size options in crypttab entries.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For some device driver SysFS attributes, values read may not be
acceptable input values for that attribute.
An example would be the group of qeth VNICC attributes that return "n/a"
when VNICC setup is not supported, but only accept "0" and "1" as valid
values that can be written to it.
This leads to errors such as the following when data for such attributes
is imported:
# chzdev f500 --import test.conf
Importing configuration data from test.conf
QETH device 0.0.f500:0.0.f501:0.0.f502 configure failed
Error: Invalid value for qeth attribute: vnicc/flooding=n/a (*)
Acceptable values:
- Integers in the range 0 - 1
Use 'chzdev qeth --help-attribute vnicc/flooding' for more information
Note: You can use --force to override safety checks (*)
To fix this, change chzdev's --export function to skip any attribute
value that is not acceptable for that attribute.
Fixes: e831269e74 ("zdev: Add support for VNIC Characteristics")
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Remove 'hash=plain' option for plain type entries, add 'none luks'
for LUKS type entries.
Note that the format of a crypttab entry is distribution specific.
On some distributions only the first 2 fields are required, the
remaining are optional, on other distributions all 4 fields are
required. With this patch correct crypttab entries are generated
for all distributions.
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/55
Fixes: https://github.com/ibm-s390-tools/s390-tools/issues/57
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With libcryptsetup version 2.1 a new debug level was introduced
to log the JSON structures when adding key slots, digests, etc
into the LUKS2 header.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Messages from libcryptsetup for logging level CRYPT_LOG_DEBUG
do not include an EOL.
Prior to libcryptsetup version 2.1 debug messages were not printed
through the log callback, but using printf directly within
libcryptsetup. Now they are also printed through the log callback,
but without an EOL terminating the text.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For zkey-cryptsetup commands reencipher and setkey, use the
same password-based key derivation function (PBKDF) when
creating unbound key-sots that the key slot uses, which was
unlocked through the specified passphrase. That way the
unlocked key slots created by these commands inherit the
PBKDF from the existing key slot.
This feature requires libcryptsetup version 2.1 or later.
If an older libcryptsetup version is available at compile
time, then PBKDF2 is used for newly created unbound key slots.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The output of lsof command is sometimes too long and makes it difficult
to search in runtime.out. Move the output to a separate file.
Signed-off-by: Sa Liu <saliu@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Running lscpumf -i as regular user causes a warning when trying to
disply sampling buffer sizes. The root cause is in the cpumf_helper
that returns "0,0" instead of a list reference "[0,0]".
This happened when running as regular user because the sampling buffer
size sysfs attribute is not readable for them.
Remove the shortcut and always return a list reference.
Signed-off-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
LUKS2 encrypted volumes use Argon2i as default password based key
derivation function (PBKDF). Argon2i is a so-called memory-hard
function. It requires a certain amount of physical memory to make
dictionary attacks more costly.
Unfortunately, when multiple encrypted volumes are unlocked
automatically during system startup via /etc/crypttab, the use of
Argon2i will most likely cause an out-of-memory error in systemd.
To avoid the out-of-memory error, use PBKDF2 instead. Because PAES
uses secure keys as volume keys, the security of the key derivation
function used to derive the key to encrypt the volume key in the LUKS
key slots is of less relevance. Thus it is safe to use a weaker key
derivation function.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
LUKS2 encrypted volumes use Argon2i as default password based key
derivation function (PBKDF). Argon2i is a so-called memory-hard
function. It requires a certain amount of physical memory to make
dictionary attacks more costly.
Unfortunately, when multiple encrypted volumes are unlocked
automatically during system startup via /etc/crypttab, the use of
Argon2i will most likely cause an out-of-memory error in systemd.
To avoid the out-of-memory error, use PBKDF2 instead. Because PAES
uses secure keys as volume keys, the security of the key derivation
function used to derive the key to encrypt the volume key in the LUKS
key slots is of less relevance. Thus it is safe to use a weaker key
derivation function.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When a user that is in the zkeyadm group, but is not the owner
of the key file changes a key in the zkey repository, then
error "Operation not permitted" occurs. This is due to a superfluous
call to chmod on the key file. When key files are changed then the
permissions of that file are retained anyway.
Also fix a typo in the error message when chmod fails.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Enable zdsfs to access datasets that were created after zdsfs was
mounted without the need to remount zdsfs.
This is done by re-reading the VTOC with every readdir system call.
To ensure a consistent VTOC state the DASD device is reserved for
every VTOC read and released afterwards.
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Added some explanations about the columns shown with the
lszcrypt verbose output.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Show how to use util_rec_set_indent() and util_rec_print_separator()
in util_rec_example.c
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In wide format the header is separated from the rest of the
records by a separator line. Add support to print such a
separator line also between some records.
Add function util_rec_print_separator() that prints the
separator line for wide format only. For other formats this
is a NOP.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Functions rec_print_xxx allows to print records in
different formats. Add support to optionally indent
the output by a specific number of characters.
The indention is set using a new function util_rec_set_indent().
The default indention is zero, thus existing applications
will behave the same as before.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The cmsfs_write function only writes a single record in line mode
and returns the number of bytes it consumed from the input buffer.
This is valid behaviour as the write system call can always return
with a partial write. But there are tools that ignore the return
value of the write and just assume that a single write call is
sufficient to write many lines aka records on the cmsfs filesystem.
Enhance the cmsfs_write function to loop until all lines from the
input buffer have been consumed.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
GCC8 shows the following compile warning:
cmsfs-fuse.c: In function ‘convert_text’:
cmsfs-fuse.c:2558:19: warning: passing argument 2 to restrict-qualified
parameter aliases with argument 4 [-Wrestrict]
rc = iconv(conv, &data_ptr, &in_count, &data_ptr, &out_count);
^~~~~~~~~ ~~~~~~~~~
Fix this by using a separate buffer for iconv. The buffer size depends on
the record format: variable-length records have a maximum length of
MAX_RECORD_LEN (plus 1 byte for linefeed), while fixed-length records
could be larger in theory. This patch takes that into account for the iconv
buffer, but the rest of the code currently cannot handle fixed-length records
larger than MAX_RECORD_LEN.
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Commit aefc8cd3ff ("libutil/libu2s: Move strlcpy() implementation to
libutils") added a dependency to libutil. This is not necessary and
could potentially lead to parallel build errors. Remove the dependency.
Fixes: aefc8cd3ff ("libutil/libu2s: Move strlcpy() implementation to libutils")
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
There are several reasons why an operation might fail. Either the
running kernel, the PCI function itself, or the hypervisor lacks support
for a certain operation. In such cases the kernel returns either with
EIO or EOPNOTSUPP.
The corresponding warning, however, just says the file couldn't be
closed. Display a warning for EIO and EOPNOTSUPP to tell the user about
the missing support.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make the error message when the CCA library is missing more
user friendly. Also display an URL where the CCA package
can be downloaded.
The CCA library is used for the REENCIPHER commands of
zkey and zkey-cryptsetup and must be installed separately
as part of the IBM CCA Host Libraries and Tools package.
This packages is not provided by the distributors, but must
be downloaded from an IBM web page and installed separately.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Get rid of the following GCC8 compile warning by using different buffers
for input/output:
In file included from vmdump_convert.cpp:13:
vm_dump.h: In member function ‘void VMDump::ebcAsc(char*, size_t)
const’:
vm_dump.h:145:26: warning: passing argument 2 to restrict-qualified
parameter aliases with argument 4 [-Wrestrict]
iconv(ebcdicAsciiConv, &inout, &len, &inout, &len);
^~~~~~ ~~~~~~
vm_dump.h:145:34: warning: passing argument 3 to restrict-qualified
parameter aliases with argument 5 [-Wrestrict]
iconv(ebcdicAsciiConv, &inout, &len, &inout, &len);
^~~~ ~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Get rid of the following GCC8 compile warning by using different buffers
for input/output:
qethqoat.c: In function ‘ebctoasc’:
qethqoat.c:51:30: warning: passing argument 2 to restrict-qualified
parameter aliases with argument 4 [-Wrestrict]
iconv(l_iconv_ebcdic_ascii, &inout, &len, &inout, &len);
^~~~~~ ~~~~~~
qethqoat.c:51:38: warning: passing argument 3 to restrict-qualified
parameter aliases with argument 5 [-Wrestrict]
iconv(l_iconv_ebcdic_ascii, &inout, &len, &inout, &len);
^~~~ ~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make sure there are two different buffers passed to iconv() to get rid
of the following GCC8 compile warning:
helper.c: In function ‘ht_ebcdic_to_ascii’:
helper.c:103:30: warning: passing argument 2 to restrict-qualified
parameter aliases with argument 4 [-Wrestrict]
iconv(l_iconv_ebcdic_ascii, &inout, &len, &inout, &len);
^~~~~~ ~~~~~~
helper.c:103:38: warning: passing argument 3 to restrict-qualified
parameter aliases with argument 5 [-Wrestrict]
iconv(l_iconv_ebcdic_ascii, &inout, &len, &inout, &len);
^~~~ ~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Commit 851f685993 ("Makefile: Add dependency between libvmcp and
libutil to fix -j builds") introduced the following regression
$ make clean
CLEAN /home2/holzheu/src/s390-tools
...
CLEAN libccw
CC libutil/util_base.o
CC libutil/util_path.o
For the "clean" target the new library dependency triggers a build
of libutil.
So remove the dependency again.
To fix the parallel build issue, no longer build the examples with
"make all" and add a new target "examples" that can now be used for building
the example programs.
Fixes: 851f685993 ("Makefile: Add dependency between libvmcp and libutil to fix -j builds")
Signed-off-by: Michael Holzheu <holzheu@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Most bootloaders use the BootLoaderSpec "title" field to name the entries
in their boot menu. The zipl bootloader used the "version" field instead,
since it was wrongly assumed that the zipl boot menu didn't support names
that contained spaces, which are usually present in a BLS "title" field.
But this is not the case, names with space characters are supported by the
IPL and is just a constraint of the section heading in the zipl.conf file.
So to be consistent with all the other bootloaders, use the "title" field
also on zipl when populating the boot menu entries from BLS files.
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/47
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Stefan Haberland sth@linux.ibm.com
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use util_strlcpy() to copy log data correctly and get rid of the
following GCC8 compile warning:
zpcictl.c: In function ‘sclp_issue_action’:
zpcictl.c:312:3: warning: ‘strncpy’ specified bound 4054 equals
destination size [-Wstringop-truncation]
strncpy(report.data.log_data, sdata, sizeof(report.data.log_data));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use util_strlcpy() to copy strings correctly and get rid of the
following GCC8 compile warning:
In function ‘hmcdrv_parse_line’,
inlined from ‘hmcdrv_cache_dir’ at hmcdrvfs.c:981:11:
hmcdrvfs.c:862:4: warning: ‘strncpy’ specified bound 192 equals
destination size [-Wstringop-truncation]
strncpy(symlink, arrow + 4, HMCDRV_FUSE_MAXPATH);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
By using util_strlcpy() and correctly copying strings we can get rid of
the following GCC8 compile warnings:
In function ‘path_to_ds_name.constprop’,
inlined from ‘zdsfs_readdir’ at zdsfs.c:282:2:
zdsfs.c:78:2: warning: ‘strncpy’ specified bound 45 equals destination
size [-Wstringop-truncation]
strncpy(normds, path, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~
In function ‘path_to_ds_name.constprop’,
inlined from ‘zdsfs_open’ at zdsfs.c:339:2:
zdsfs.c:78:2: warning: ‘strncpy’ specified bound 45 equals destination
size [-Wstringop-truncation]
strncpy(normds, path, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~
In function ‘path_to_member_name.constprop’,
inlined from ‘zdsfs_open’ at zdsfs.c:372:3:
zdsfs.c:94:3: warning: ‘strncpy’ specified bound 45 equals destination
size [-Wstringop-truncation]
strncpy(normds, path, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~
In function ‘path_to_ds_name.constprop’,
inlined from ‘zdsfs_getxattr’ at zdsfs.c:527:2:
zdsfs.c:78:2: warning: ‘strncpy’ specified bound 45 equals destination
size [-Wstringop-truncation]
strncpy(normds, path, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~
In function ‘path_to_member_name.constprop’,
inlined from ‘zdsfs_getxattr’ at zdsfs.c:547:4:
zdsfs.c:94:3: warning: ‘strncpy’ specified bound 45 equals destination
size [-Wstringop-truncation]
strncpy(normds, path, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~
In function ‘path_to_ds_name.constprop’,
inlined from ‘zdsfs_getattr.part.1’ at zdsfs.c:134:2,
inlined from ‘zdsfs_getattr’:
zdsfs.c:78:2: warning: ‘strncpy’ specified bound 45 equals destination
size [-Wstringop-truncation]
strncpy(normds, path, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~
In function ‘path_to_member_name.constprop’,
inlined from ‘zdsfs_getattr.part.1’ at zdsfs.c:164:3,
inlined from ‘zdsfs_getattr’:
zdsfs.c:94:3: warning: ‘strncpy’ specified bound 45 equals destination
size [-Wstringop-truncation]
strncpy(normds, path, size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use the more robust util_strlcpy() to copy strings instead of strncpy().
This fixes the following GCC8 compile warnings:
qethqoat.c: In function ‘main’:
qethqoat.c:616:2: warning: ‘strncpy’ specified bound 16 equals
destination size [-Wstringop-truncation]
strncpy(ifr.ifr_name, opts.ifname, IFNAMSIZ);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make sure strings are copied correctly by using util_strlcpy().
Also, rename the NIC_{OWNER,NAME}_LEN definitions to *_SIZE to make it
more clear that the terminating null byte ('\0') is included in this
buffer size. Using *_LEN would mean only the number of characters in the
string fit in.
Furthermore, use these definitions rather than sizeof().
This fixes the following GCC8 compile warning:
nic.c: In function ‘nic_data_get’:
nic.c:58:2: warning: ‘strncpy’ specified bound 9 equals destination size
[-Wstringop-truncation]
strncpy(data.owner, argv[11], sizeof(data.owner));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
nic.c:59:2: warning: ‘strncpy’ specified bound 9 equals destination size
[-Wstringop-truncation]
strncpy(data.name, argv[12], sizeof(data.name));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Change the buffer size to what we actually need and use util_strlcpy()
to correctly copy strings.
This gets rid of the following GCC8 compile warnings:
fcp.c: In function ‘fcp_wwpn_get’:
fcp.c:44:2: warning: ‘strncpy’ output may be truncated copying 20 bytes
from a string of length 4095 [-Wstringop-truncation]
strncpy(wwpn, buf, 20);
^~~~~~~~~~~~~~~~~~~~~~
fcp.c: In function ‘fcp_lun_get’:
fcp.c:65:2: warning: ‘strncpy’ output may be truncated copying 20 bytes
from a string of length 4095 [-Wstringop-truncation]
strncpy(lun, buf, 20);
^~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use util_asprintf() to allocate memory for the string buffers
dynamically and get rid of the following GCC8 compile warnings:
system.c: In function ‘print_fw_str’:
system.c:86:46: warning: ‘%s’ directive output may be truncated writing
up to 4095 bytes into a region of size 4082 [-Wformat-truncation=]
snprintf(path, sizeof(path), "/sys/firmware/%s", file);
^~
system.c:98:19:
read_fw_str(str, path, sizeof(str));
~~~~
system.c:86:2: note: ‘snprintf’ output between 15 and 4110 bytes into a
destination of size 4096
snprintf(path, sizeof(path), "/sys/firmware/%s", file);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cmd_chreipl.c: In function ‘set_reipl_type_helper’:
cmd_chreipl.c:316:19: warning: ‘%d’ directive writing between 1 and 11
bytes into a region of size between 0 and 4095 [-Wformat-overflow=]
sprintf(cmd, "%s %d:%d", chreipl_helper, major(dev), minor(dev));
^~
cmd_chreipl.c:316:15: note: using the range [-2147483648, 2147483647]
for directive argument
sprintf(cmd, "%s %d:%d", chreipl_helper, major(dev), minor(dev));
^~~~~~~~~~
cmd_chreipl.c:316:15: note: using the range [-2147483648, 2147483647]
for directive argument
cmd_chreipl.c:316:2: note: ‘sprintf’ output between 5 and 4120 bytes
into a destination of size 4096
sprintf(cmd, "%s %d:%d", chreipl_helper, major(dev), minor(dev));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use util_strlcpy() to copy strings correctly and get rid of the
following GCC8 compile warnings:
main.c: In function ‘main’:
main.c:34:2: warning: ‘strncpy’ specified bound 256 equals destination
size [-Wstringop-truncation]
strncpy(g.prog_name, argv[0], sizeof(g.prog_name));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In function ‘set_bootprog’,
inlined from ‘parse_chreipl_options’ at cmd_chreipl.c:510:4:
cmd_chreipl.c:185:2: warning: ‘strncpy’ specified bound 11 equals
destination size [-Wstringop-truncation]
strncpy(l.bootprog, bootprog, sizeof(l.bootprog));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cmd_chreipl.c: In function ‘set_reipl_type’:
cmd_chreipl.c:297:2: warning: ‘strncpy’ specified bound 15 equals
destination size [-Wstringop-tr uncation]
strncpy(l.dev, dev_name, sizeof(l.dev));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Copy strings correctly by using util_strlcpy() over strncpy() and get
rid of the following GCC8 compile warning:
In function ‘dinfo_extract_dev’,
inlined from ‘dinfo_get_dev_from_blockdev’ at dasdinfo.c:365:6:
dasdinfo.c:337:2: warning: ‘strncpy’ specified bound 80 equals
destination size [-Wstringop-trunc ation]
strncpy(tmp, str, RD_BUFFER_SIZE);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
IFNAMSIZ is the maximum buffer size for an interface name, including its
terminating zero byte. [1] strncpy() should therefore only copy a size
of 'IFNAMSIZ - 1' and the destination string should then be
null-terminated properly.
Use util_strlcpy() to correctly copy the string and get rid of the
followin GCC8 compile warning:
In function ‘ethtool_checksumming’,
inlined from ‘process_sysfs_attribute’ at lsqeth.c:333:5,
inlined from ‘print_device’ at lsqeth.c:468:3:
lsqeth.c:171:2: warning: ‘strncpy’ specified bound 16 equals destination
size [-Wstringop-truncat ion]
strncpy(ifr.ifr_name, if_name, IFNAMSIZ);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[1]: https://www.gnu.org/software/libc/manual/html_node/Interface-Naming.html
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Get rid of fixed buffers and avoid the following GCC8 compile warnings:
lszcrypt.c: In function ‘main’:
lszcrypt.c:642:28: warning: ‘%04x’ directive writing between 4 and 8
bytes into a region of size between 7 and 13 [-Wformat-overflow=]
sprintf(sub_dev, "%02x.%04x", id, dom);
^~~~
lszcrypt.c:642:22: note: directive argument in the range [0, 2147483647]
sprintf(sub_dev, "%02x.%04x", id, dom);
^~~~~~~~~~~
lszcrypt.c:642:5: note: ‘sprintf’ output between 8 and 18 bytes into a
destination of size 16
sprintf(sub_dev, "%02x.%04x", id, dom);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
DEV_NAME_SIZE is defined as IFNAMSIZ. IFNAMSIZ is the maximum buffer size
for an interface name, including its terminating null byte. [1] The
buffers dev_name and ifr_name are both defined with a size of
DEV_NAME_SIZE and IFNAMSIZ respectively.
Given these facts, only a size of 'IFNAMSIZ - 1' should be copied and
the destination string should then be null-terminated properly.
Use util_strlcpy() to correctly copy the strings and rid of the followin
GCC8 compile warnings:
xcec-bridge.c: In function ‘open_incoming_socket’:
xcec-bridge.c:94:2: warning: ‘strncpy’ specified bound 16 equals
destination size [-Wstringop-tru ncation]
..strncpy(if_req.ifr_name,dev_name,DEV_NAME_LEN);
..^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
xcec-bridge.c: In function ‘read_sys’:
xcec-bridge.c:284:4: warning: ‘strncpy’ output may be truncated copying
16 bytes from a string of length 255 [-Wstringop-truncation]
....strncpy(is->dev_name, if_name, DEV_NAME_LEN);
....^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[1]: https://www.gnu.org/software/libc/manual/html_node/Interface-Naming.html
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The term 'len' only represents the number of characters in a string,
as opposed to 'size' which means the number of bytes including the
terminating null byte ('\0').
DEV_NAME_LEN is defined as IFNAMSIZ, so we are clearly interested in the
size. Rename to DEV_NAME_SIZE to make this more clear.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The strlcpy() implementation in libu2s is beneficial for other tools as
well. Move the implementation to libutils and replace misc_strlcpy() in
libu2s accordingly.
Change the link order in zipl to make it build again.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Safely replace strncpy() with memcpy() as we don't want the terminating
null byte anyway. Get rid of the following warning:
dasdview.c: In function ‘dasdview_print_volser’:
dasdview.c:621:3: warning: ‘strncpy’ output truncated before terminating
nul copying 6 bytes from a string of the same length
[-Wstringop-truncation]
strncpy(volser, " ", 6);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
While assembling the volume label, strncpy() is used. As we don't care
about NUL termination of the strings here, safely replace strncpy() with
memcpy() and get rid of the following compile warnings:
fdasd.c: In function ‘fdasd_write_vtoc_labels’:
fdasd.c:1324:4: warning: ‘strncpy’ output truncated before terminating
nul copying 44 bytes from a string of the same length
[-Wstringop-truncation]
strncpy(ch, "LINUX.V "
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
" ", 44);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
fdasd.c:1332:4: warning: ‘strncpy’ output truncated before terminating
nul copying 5 bytes from a string of the same length
[-Wstringop-truncation]
strncpy(c1, ".PART", 5);
^~~~~~~~~~~~~~~~~~~~~~~
In function ‘fdasd_write_vtoc_labels’,
inlined from ‘fdasd_write_labels’ at fdasd.c:1399:3:
fdasd.c:1329:4: warning: ‘strncpy’ output may be truncated copying 6
bytes from a string of length 6 [-Wstringop-truncation]
strncpy(c1, volser, VOLSER_LENGTH);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
fdasd.c:1336:4: warning: ‘strncpy’ output may be truncated copying 5
bytes from a string of length 5 [-Wstringop-truncation]
strncpy(c1, dsno, 5);
^~~~~~~~~~~~~~~~~~~~
fdasd.c:1340:4: warning: ‘strncpy’ output truncated before terminating
nul copying as many bytes from a string as its length
[-Wstringop-truncation]
strncpy(c1, dsname, strlen(dsname)); /* We don't want \0 */
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
fdasd.c:1313:4: warning: ‘strncpy’ output may be truncated copying 31
bytes from a string of length 44 [-Wstringop-truncation]
strncpy(c1, s2, 31);
^~~~~~~~~~~~~~~~~~~
In function ‘fdasd_change_part_type’,
inlined from ‘main’ at fdasd.c:3000:4:
fdasd.c:1639:3: warning: ‘strncpy’ output may be truncated copying 6
bytes from a string of length 19 [-Wstringop-truncation]
strncpy(ch, str, 6);
^~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
We can safely replace strncpy() with memcpy() here and get rid of the
following compile warnings:
dfi_s390mv.c: In function ‘set_magic_numbers’:
dfi_s390mv.c:570:3: warning: ‘strncpy’ output truncated before
terminating nul copying 7 bytes fr om a string of the same length
[-Wstringop-truncation]
strncpy(l.dumper_magic, DF_S390_DUMPER_MAGIC_MV_EXT, 7);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
dfi_s390mv.c:573:3: warning: ‘strncpy’ output truncated before
terminating nul copying 7 bytes fr om a string of the same length
[-Wstringop-truncation]
strncpy(l.dumper_magic, DF_S390_DUMPER_MAGIC_MV, 7);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Using inappropriate sizes for snprintf() leads to the following compile
warnings:
dfi_s390mv.c: In function ‘dev_from_busid’:
dfi_s390mv.c:116:34: warning: ‘/online’ directive output may be
truncated writing 7 bytes into a region of size between 1 and 4096
[-Wformat-truncation=]
snprintf(tmp_file, PATH_MAX, "%s/online", dev_file);
^~~~~~~
dfi_s390mv.c:116:2: note: ‘snprintf’ output between 8 and 4103 bytes
into a destination of size 4096
snprintf(tmp_file, PATH_MAX, "%s/online", dev_file);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
dfi_s390mv.c:143:35: warning: ‘%s’ directive output may be truncated
writing up to 255 bytes into a region of size between 0 and 4095
[-Wformat-truncation=]
snprintf(tmp_file, PATH_MAX, "%s/%s/dev", dev_file, direntp->d_name);
^~
dfi_s390mv.c:143:2: note: ‘snprintf’ output between 6 and 4356 bytes
into a destination of size 4096
snprintf(tmp_file, PATH_MAX, "%s/%s/dev", dev_file, direntp->d_name);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
A simple fix would be to use different suitable fixed sizes for tmp_file
and dev_file. However, the code can be improved here a little.
Use the libutil function util_path_sysfs() to build the sysfs path
string. This is more robust, as the function will figure out the correct
mount point of the sysfs. util_path_sysfs() will also terminate
execution, if no sysfs mount point could be found. That means, we can
get rid of check_sysfs().
Furthermore, the two variables (tmp_file, dev_file) can be combined to
one. Also, check the return value of open() and act accordingly.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use misc_asprintf() for concatenating strings and get rid of the
following GCC8 compile warnings:
disk.c: In function ‘disk_get_info’:
disk.c:247:43: warning: ‘%d’ directive output may be truncated writing
between 1 and 11 bytes int o a region of size between 0 and 79
[-Wformat-truncation=]
snprintf(ppn_cmd, sizeof(ppn_cmd), "%s %d:%d",
^~
disk.c:247:39: note: using the range [-2147483648, 2147483647] for
directive argument
snprintf(ppn_cmd, sizeof(ppn_cmd), "%s %d:%d",
^~~~~~~~~~
disk.c:247:39: note: using the range [-2147483648, 2147483647] for
directive argument
disk.c:247:4: note: ‘snprintf’ output between 5 and 104 bytes into a
destination of size 80
snprintf(ppn_cmd, sizeof(ppn_cmd), "%s %d:%d",
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
script_file, major(stats.st_rdev),
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
minor(stats.st_rdev));
~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Using the DATA_SIZE() macro gets rid of the following GCC8 compile
warning:
boot.c: In function ‘boot_init_fba_stage0’:
boot.c:120:2: warning: ‘memcpy’ forming offset [2, 152] is out of the
bounds [0, 1] of object ‘_binary_fba0_bin_start’ with type ‘char’
[-Warray-bounds]
memcpy(stage0, DATA_ADDR(fba0), sizeof(*stage0));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from boot.c:18:
../boot/data.h:24:13: note: ‘_binary_fba0_bin_start’ declared here
extern char _binary_fba0_bin_start;
^~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With GCC8 the following warning is displayed:
In file included from cio.h:15,
from stage2.h:15,
from menu.h:15,
from stage2.c:14:
s390.h:193:1: warning: alignment 1 of ‘struct _lowcore’ is less than 8
[-Wpacked-not-aligned]
} __packed;
^
Fix by aligning struct _lowcore to double page size.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The message buffer of DumpException doesn't account for additional
strings such as strerror(), device nodes, or filenames. Make it a bit
larger to be safe.
This fixes the following GCC8 compile warnings:
In file included from vm_dump.h:62,
from vm_dump.cpp:17:
dump.h: In static member function ‘static Dump::DumpType
VMDump::getDumpType(const char*)’:
dump.h:46:16: warning: ‘ (’ directive writing 2 bytes into a region of size
between 1 and 200 [-W format-overflow=]
sprintf(msg, "%s (%s)", m, strerror(errno));
^~~~~~~~~
dump.h:46:10: note: ‘sprintf’ output 4 or more bytes (assuming 203) into a
destination of size 200
sprintf(msg, "%s (%s)", m, strerror(errno));
~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
dump.h:46:16: warning: ‘ (’ directive writing 2 bytes into a region of size
between 1 and 200 [-W format-overflow=]
sprintf(msg, "%s (%s)", m, strerror(errno));
^~~~~~~~~
dump.h:46:10: note: ‘sprintf’ output 4 or more bytes (assuming 203) into a
destination of size 200
sprintf(msg, "%s (%s)", m, strerror(errno));
~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
dump.h:46:16: warning: ‘ (’ directive writing 2 bytes into a region of size
between 1 and 200 [-W format-overflow=]
sprintf(msg, "%s (%s)", m, strerror(errno));
^~~~~~~~~
dump.h:46:10: note: ‘sprintf’ output 4 or more bytes (assuming 203) into a
destination of size 200
sprintf(msg, "%s (%s)", m, strerror(errno));
~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To assemble things like the volume id or fields of the format1 label
struct, strncpy() is used. Data copied here is later written to disk and
we don't want any terminating null byte ('\0') there. Therefore, we can
simply use memcpy() instead and get rid of the following GCC8 compile
warnings:
vtoc.c: In function ‘vtoc_init_format_1_8_label’:
vtoc.c:747:2: warning: ‘strncpy’ output truncated before terminating nul
copying 6 bytes from a s tring of the same length [-Wstringop-truncation]
strncpy((char *) f1->DS1DSSN, " ", 6);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
vtoc.c:746:2: warning: ‘strncpy’ output may be truncated copying 44 bytes from
a string of length 79 [-Wstringop-truncation]
strncpy(f1->DS1DSNAM, str, 44);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
vtoc.c:761:2: warning: ‘strncpy’ output may be truncated copying 13 bytes from
a string of length 79 [-Wstringop-truncation]
strncpy((char *)f1->DS1SYSCD, str, 13);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
vtoc.c: In function ‘vtoc_volume_label_set_volser’:
vtoc.c:401:2: warning: ‘strncpy’ output may be truncated copying 6 bytes from a
string of length 6 [-Wstringop-truncation]
strncpy(vlabel->volid, s, VOLSER_LENGTH);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
fread()/fwrite() return the unsigned count of elements read/written or a
short count or zero in case of an error. Check if the number of elements
read/written matches the expectation. In case of fread(), if it does not
match, check if an error occurred and otherwise assume the file to be too
small. In case of fwrite(), if it does not match, assume an error.
Resolves Cppcheck style warnings:
[zkey/pkey.c:161]: (style) Checking if unsigned variable 'count' is less than zero.
[zkey/pkey.c:209]: (style) Checking if unsigned variable 'count' is less than zero.
[zkey/pkey.c:299]: (style) Checking if unsigned variable 'count' is less than zero.
Cc: Ingo Franzki <ifranzki@de.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When realloc() fails to allocate memory it returns NULL and the original
block is left untouched. In that case return the potentially larger
original block.
Found using Cppcheck:
[zdev/src/misc.c:1106]: (error) Memory leak: name
Cc: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The modified flag was actually never copied during merge of setting lists.
Copy the modified flag if requested.
Found using Cppcheck:
[zdev/src/setting.c:553]: (warning) Redundant assignment of 'n->modified' to itself.
Cc: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The computation of blk in functions fdasd_check_volume() and
fdasd_write_vtoc_labels() contained the possibility for an undetected
unsigned integer overflow. Unconditionally subtracting one from the
return value of function cchhb2blk() may cause an unsigned integer
overflow, as cchhb2blk() may return zero if cc, hh, and b are all zero.
Resolves Cppcheck style warning:
[fdasd/fdasd.c:1260]: (style) Checking if unsigned variable 'blk' is less than zero.
Cc: Stefan Haberland <sth@linux.ibm.com>
Cc: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Only check if size is zero and not if equal or less than zero. This is
safe, as size is only decremented by rlen, which is either less than or
equal to size. So size may never overflow.
Resolves Cppcheck style warning:
[cmsfs-fuse/cmsfs-fuse.c:3746]: (style) Checking if unsigned variable 'size' is less than zero.
Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
For linking the "libvmcp" example program the "libutil" is required.
Therefore when the "libvmcp" directory is entered in a parallel build
and the "libutil.a" file has not yet been created, a second "libutil"
build is triggered which can then lead to build errors like the following:
ar: libutil.a: File format not recognized
ar: libutil.a: No such file or directory
make[2]: Leaving directory '/builddir/build/BUILD/s390-tools-2.6.0/libutil'
make[2]: *** [../common.mak:246: libutil.a] Error 1
make[1]: *** [../common.mak:246: libutil.a] Error 1
make[1]: *** Waiting for unfinished jobs....
make[1]: *** [../common.mak:315: ..//libutil/libutil.a] Error 2
make[1]: Leaving directory '/builddir/build/BUILD/s390-tools-2.6.0/libvmcp'
make: *** [Makefile:57: libvmcp] Error 2
make: *** Waiting for unfinished jobs....
ving directory '/builddir/build/BUILD/s390-tools-2.6.0/libutil'
make: *** [Makefile:57: libutil] Error 2
error: Bad exit status from /var/tmp/rpm-tmp.wCulxB (%build)
So fix this and add a dependency between libvmcp and libutil.
Signed-off-by: Michael Holzheu <holzheu@linux.ibm.com>
Reviewed-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When using 'zkey cryptsetup' with the '--run' option,
it might not find the executable to execute when those
are in /sbin.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Change all error messages from printf() to warnx().
While at it, remove the punctuation mark for some of the error messages.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The .o.d make targets in common.mak do not expect that
header files are generated by a make target. When a new header
file is generated, the .o.d targets will be rebuilt on the
next make invocation, because that new header file is then
detected, and is then treated as a new dependency of all .o.d
targets.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The linux.vnet.ibm.com addresses will be become disfunct by eoy.
Signed-off-by: Stefan Raspl <raspl@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add the missing identifiers to report the card name and link speed.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Print a meaningful reason when the keystore can not be
accessed, e.g. because the user is not allowed to access
the keystore directory.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Dump compression was a feature of the file system dumper which was removed
several years ago. This include is a leftover. Remove it.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Acked-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Remove unused code because query_OSA_EXP does actually never return -1.
So make its return value unsigned and remove all checks for negative
return values.
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Osasnmpd would start without any devices, but refuses to start when all
devices are no real OSA devices.
Start osasnmpd in this case because it re-scans via update_mib_info
whenever the interface data changes.
Reviewed-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The util_scandir_* functions may return 0 or -1, in which case no vector
is allocated in libutil/util_scandir.c. util_ptr_vec_free, called by
util_scandir_free or directly from lschp.c and lsscm.c, does always
call free for the vector which might be not initialized.
Fix this by always initializing the vector with NULL in __scandir and
add some api hardening by checking the vector and count in
util_ptr_vec_free before iterating over the vector.
And update the comment for util_scandir to indicate that -1 may be
returned in error cases plus that the vector is initialized with NULL.
Fixes: https://github.com/ibm-s390-tools/s390-tools/issues/43
Reported-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Unfortunately zdev hook already exists in Ubuntu, from an unrelated
project. ZFS uses zdev/zpool names, and ships a zdev hook to do ZFS
specific initialisation. It is available on s390x and thus results in
file-conflict upon installing both. Thus renaming this zdev hook to
s390-tools-zdev.
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/41
Signed-off-by: Dimitri John Ledkov <xnox@ubuntu.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
At the moment, if we specify the slot address of an NVMe device but
can't find the corresponding device node, the execution is terminated.
This is a bit harsh as the device node is rather optional and only
necessary to collect S.M.A.R.T. data. We should still be able to issue
the error reporting, even if we couldn't determine the device node.
Therefore, make sure the device node for NVMe devices is optional by
changing various error messages to warnings.
Change sysfs_get_slot_addr() to have a return value and work with that
accordingly.
Also make sure, that execution is terminated when a valid device node
was specified but no matching slot address was determined. The slot
address is necessary to issue the error reporting commands.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The address sysfs attribute might not be present on some older kernel
levels. Read the device link instead using readlink() to obtain the
address.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Be consistent with the rest of the code and use fopen() rather than
open().
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This patch fixes false positives when attributes are checked for
conflicts between BridgePort and VNICC.
Signed-off-by: Hans Wippel <hwippel@linux.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The minor()/major() function definitions are moved to sys/sysmacros.h
and will be removed from sys/types.h. To correct below warning, simply
include sys/sysmacros.h.
zpcictl.c: In function ‘sysfs_get_slot_addr’:
zpcictl.c:184:13: warning: In the GNU C Library, "major" is defined
by <sys/sysmacros.h>. For historical compatibility, it is
currently defined by <sys/types.h> as well, but we plan to
remove this soon. To use "major", include <sys/sysmacros.h>
directly. If you did not intend to use a system-defined macro
"major", you should undefine it after including <sys/types.h>.
major = major(dev_stat.st_rdev);
^~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.ibm.com>
zpcictl is used to manage PCI devices on z Systems. In this first
version it is mainly used to handle erroneous PCI devices by changing
their state and make those changes known to the SE. Log data, such as
S.M.A.R.T. data for NVMe devices, is sent alongside those state changes.
The state change is issued by sending data via the PCI 'report_error'
sysfs attribute. It's a binary attribute which will cause the host to
send an Adapter Notification Event.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With the latest kernel comes an extension for the zcrypt
device driver to support multiple zcrypt device nodes.
Here is the userspace part of this which adds a new
application zcryptctl for user friendly management of
this feature.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zipl currently uses the FIBMAP ioctl to map blocks for the bootloader;
on XFS, if FIBMAP is requested on a reflinked file, it will fail -
and FIBMAP returns 0 in this case, which is indistinguishable from a
hole. This causes boot to fail because the file is not mapped.
We can use the FIEMAP ioctl instead, which is able to map reflinked
files. While FIEMAP is able to map entire extents at once, here we
simply use it to obtain the mapping block-by-block so that it fits
in with the current FIBMAP calls.
Fixes: https://github.com/ibm-s390-tools/s390-tools/issues/34
Closes: https://github.com/ibm-s390-tools/s390-tools/pull/36
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Tested-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
With kernel 4.19 there comes an extension to the existing
AP bus which supports alternate zcrypt drivers. For details
about this see kernel patch "s390/zcrypt: AP bus support for
alternate driver(s)". So now lszcrypt displays the driver
name in verbose mode. As some of the information displayed
by lszcrypt was based on sysfs attributes, which are only
available when the default zcrypt driver is bound to the
device, this also needed some rework. If a sysfs attribute
is not available because of an alternate driver binding
(or no driver) a question mark is printed into the field.
Together with this a slight rework of the displayed information
has been done. The two columns for pending requests and pending
replies has been merged to one pending column and the column
sizes have been adjusted.
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
zdsfs writes output for --help and --version to stderr. This is
likely because zdsfs calls into a FUSE function for additional
usage output that also writes to stderr (prior to FUSE 3.0.0).
To be consistent with other s390-tools and GNU coding guide lines,
fix this by using stdout for zdsfs generated output, and by
redirecting stderr to stdout before calling the FUSE function.
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
tape390_display writes output for --help to stderr. To be consistent
with other s390-tools and GNU coding guide lines, fix this by using
stdout for tool-generated usage help.
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add missing long options for --help and --version in util_prg_example.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Both iucvconn and iucvtty write output for --help to stderr. To be
consistent with other s390-tools and GNU coding guide lines,
fix this by using stdout for tool-generated usage help.
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
hmcdrvfs writes output for --help and --version to stderr. This is
likely because hmcdrvfs calls into a FUSE function for additional
usage output that also writes to stderr (prior to FUSE 3.0.0).
To be consistent with other s390-tools and GNU coding guide lines,
fix this by using stdout for hmcdrvfs generated output, and by
redirecting stderr to stdout before calling the FUSE function.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
cmsfs-fuse writes output for --help and --version to stderr. This is
likely because cmsfs-fuse calls into a FUSE function for additional
usage output that also writes to stderr (prior to FUSE 3.0.0).
To be consistent with other s390-tools and GNU coding guide lines,
fix this by using stdout for cmsfs-fuse generated output, and by
redirecting stderr to stdout before calling the FUSE function.
Also remove an inactive call to add FUSE version output on --version to
be consistent with other FUSE-based s390-tools.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Building zipl from outside the zipl/ sub-directory fails due to a
missing dependency:
# cd s390-tools
# make -C zipl/src/ zipl
boot.c:18:10: fatal error: ../boot/data.h: No such file or directory
#include "../boot/data.h"
This is because boot/data.h needs to be generated before src/boot.c can
be compiled. Fix this by adding the missing dependency to the zipl
Makefile.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Assume "N/A" for non-zfcp SCSI devices, such as iSCSI or virtio-scsi-ccw,
to not erroneously access absent zfcp-specific sysfs attributes.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Simply skip non-zfcp SCSI devices, such as iSCSI or virtio-scsi-ccw,
to not erroneously access absent zfcp-specific sysfs attributes.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The default bash setting have nullglob disabled so if
$SCSI_DEV/scsi_generic*
aka
/sys/bus/scsi/devices/*:*:*:*/scsi_generic*
does not match anything, it leaves the glob pattern unmodified and
SG_DEV=$(basename $SG_DEV/*)
results in the literal:
*
If $SG_INQ exists, it invoked sg_inq with more than the one allowed
positional argument for a SCSI generic device node:
sg_inq /dev/*
Causing error messages and the usage of sg_inq to land in $TAPE_SERIAL.
To not have to rely on the nullglob setting, explicitly check for the
existence of $SCSI_DEV/scsi_generic before evaluating
SG_DEV=$(basename $SG_DEV/*).
Also handle availability of sg_inq but absence of scsi_generic
individually to provide the user with a hint if only sg is missing.
Suggested-by: Eric Farman <farman@linux.vnet.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Cc: Eric Farman <farman@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This should be more stable than parsing the IBM lin_tape proc-fs output.
If sysfs is not available or provides no match, fall back to proc-fs.
An alternative to "ls -1d ... | head -n1" would have been to use
"shopt -s extglob; ls -d $SCSI_DEV/lin_tape/$DEV_NAME+([0-9])"
but I did not want to have to rely on extglob potentially influencing
other old code of lstape.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
https://www.ibm.com/developerworks/linux/linux390/s390-tools-1.8.0.html
before the first upstream commit b627b8d8e1 ("Initial s390-tools-2.0.0
import") introduced SCSI tape/changer output for lstape. It used the SCSI
device serial number as lookup key to find a match in IBM lin_tape device
driver proc-fs output for a given SCSI device name. Since multiple paths
to the same tape/changer have the same serial number, multiple matches can
occur and distort the output:
% lstape
...
SCSI tape devices (found 12):
...
Generic Device Target Vendor Model Type State
sg0 IBMtape0 3 0:0:0:0 IBM ULT3580-TD6 tapedrv
running
sg1 IBMchanger0 1 2 3 0:0:0:1 IBM
3573-TL changer running
...
There are unexpected additional Device suffix numbers for each additional
path of the same tape/changer (independent of actual path failover
enablement in lin_tape). The multiple matches above cause excess arguments
to printf. Explaining the resulting output, the bash man page says:
"The format is reused as necessary to consume all of the arguments."
This also causes a wrong number of found devices.
Current lin_tape (v3.0.31) uses the printf format string
"%-8d%-12s%-25s%-16s%-16s%-10s" for fixed column proc-fs output.
This causes the following example input to lstape:
==> /proc/scsi/IBMchanger <==
lin_tape version: 3.0.31
lin_tape major number: 251
Attached Changer Devices:
Number model SN HBA SCSI FO Path
0 3573-TL 00L4U78W6497_LL0 Virtio SCSI HBA 0:0:0:1 Primary
1 3573-TL 00L4U78W6497_LL0 Virtio SCSI HBA 0:0:1:1 Alternate
2 3573-TL 00L4U78W6497_LL0 Virtio SCSI HBA 1:0:1:1 Alternate
3 3573-TL 00L4U78W6497_LL0 Virtio SCSI HBA 1:0:0:1 Alternate
==> /proc/scsi/IBMtape <==
lin_tape version: 3.0.31
lin_tape major number: 251
Attached Tape Devices:
Number model SN HBA SCSI FO Path
0 ULT3580-TD6 10WT037733 Virtio SCSI HBA 0:0:0:0 Primary
1 ULT3580-TD6 10WT037701 Virtio SCSI HBA 0:0:1:0 Primary
2 ULT3580-TD6 10WT037701 Virtio SCSI HBA 1:0:1:0 Alternate
3 ULT3580-TD6 10WT037733 Virtio SCSI HBA 1:0:0:0 Alternate
If that was reliable and stable, we could use fixed column parsing:
IBM_IDX=$(
awk 'BEGIN {
FIELDWIDTHS="8 12 25 16 16 10"
}
gensub(" ","","g",$5) == "'$SCSI_ID'"{
print gensub(" ","","g",$1)
}' /proc/scsi/$DEV_NAME
)
However, we can hardly guarantee a field value always fits into the left-
adjusted columns (printf won't cap to field width!). Also, the fixed column
format changed between lin_tape v3.0.20 and v3.0.18 from
"%-8d%-12s%-18s%-16s%-16s%-10s" with less width for the 'SN' field.
The column format also changed between v2.2.0 and v1.37.0 from
"%-8d%-12s%-18s%-32s%-10s" where the 'SCSI' field did not even exist.
Likely that was the reason for the original Serial lookup in lstape.
As can be seen in the example, at least the HBA device driver string can
contain whitespace and breaks the field numbers with tokenized parsing.
We want to match column 5 (SCSI) but above we would match against the
string 'SCSI' as part of "Virtio SCSI HBA" instead of e.g. '0:0:0:0'.
The following works currently knowing that the last two columns do not
contain white space. But it can also fail if the value of field
'HBA' does not fit into the left-adjusted columns so there is no white
space between columns 'HBA' and 'SCSI' making them collapse:
awk '$(NF-1) == "'$SCSI_ID'" { print $1 }' /proc/scsi/$DEV_NAME
Grep for the SCSI device name as word (to skip names with same substring,
such as 0:0:1:1 also matching 0:0:1:10) and cut the first field 'Number'.
If there is no SCSI column at all (and no SCSI LLDD or other column with a
name accidentally matching an existing SCSI device name), we get no match
and better bail out with the initialized "N/A" for the lstape column
"Device". If the SCSI column exists, the SCSI device names are unique and
we always get at most one match.
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When zfcpdump is built you get a compile warning
zfcpdump.c: In function ‘init_sig’:
zfcpdump.c:307:24: warning: cast between incompatible function types from ‘void (* (*)(int, siginfo_t *, void *))(int)’ {aka ‘void (* (*)(int, struct <anonymous> *, void *))(int)’} to ‘void (*)(int)’ [-Wcast-function-type]
g.sigact.sa_handler = (__sighandler_t)dump_sig_handler;
Furthermore the man pages for sigaction says
$man 2 sigaction
[...]
If SA_SIGINFO is specified in sa_flags, then sa_sigaction (instead
of sa_handler) specifies the signal-handling function for signum.
This function receives three arguments, as described below.
[...]
Because SA_SIGINFO is set, using sa_handler at all is wrong in this case.
Fix this by giving dump_sig_handler the correct return type and assign it
to sa_sigaction.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zfcpdump kernel is built without module support. Furthermore the initrd
doesn't contain modprobe. So trying to load modules using modprobe in
zfcpdump userspace doesn't really make sense.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Since version 198 (Feb 2013) systemd contains kernel-install, a script
managing kernel installs. This script allows execution of drop-in scripts
for customization. Add such a drop-in script to s390-tools to handle the
installation of zfcpdump kernels.
It's main purpose is to manage a link to the latest installed zfcpudump
kernel at ZFCPDUMP_IMAGE, where zipl expects to find the image to install
for a SCSI dumper.
The script supports two installation modes. One recommended by the
BootLoaderSpecs [1] to /boot/<machine-id>/<kernel-version> directories and
one directly to /boot. In the second case files are renamed during
installation to <original-name>-<kernel-version> to guarantee unique names.
Because the zfcpdump kernel is so special make the script stand-alone and
prevent any other script from being executed (exit 77) when a zfcpdump is
installed. Especially avoid functionality like creating an initrd (already
provided by s390-tools) or creating a boot entry (the zfcpdump kernel
should not be used for any other purpose than dumping).
The script requires systemd >= 203.
[1] https://www.freedesktop.org/wiki/Specifications/BootLoaderSpec/
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In the zipl Makefile macros are defined with the sole purpose to define
other macros in zipl.h. Remove this unnecessary indirection.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The file system dumper no longer exists. So there is no need in keeping
'PART' to distinguish between the file system and the partition dumper.
Adjust the macro name to the usage in zipl by removing 'PART'.
While at it also adjust the file names the macros stand for.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The macro names are a mixture of the original file system dumper (FSDUMP)
and its extension to cut the file system and write directly to a partition
(PART). The original dumper no longer exists and today the feature is
called zfcpdump. Prevent confusion by renaming the macros to fit todays
usage.
Signed-off-by: Philipp Rudo <prudo@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Allow to specify a volume-type for a key. This applies to all
associated volumes. The volume type can be either 'plain' or
'luks2'. New keys created will default to 'luks2', but existing
keys that do not have a volume-type property default to 'plain'
for compatibility reasons.
The volume type 'luks2' is only available when the define
HAVE_LUKS2_SUPPORT is set in the makefile. This is set only
when libcryptsetup version 2.0.3 or newer is available
at build time. If the define is not set, the volume-type
option is not available to the user, and the volume-type of
a key defaults to 'plain'.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Store a verification pattern in the properties file along
with the secure key. The verification pattern allows to identify
the inner key even when the secure key is no longer valid.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The zkey-cryptsetup tool is used to validate and re-encipher
secure AES volume keys of volumes encrypted with LUKS2 and
the paes cipher.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The IPAPPEND option in the pxelinux config file can be used
to pass information about the boot interface to the booted
Linux kernel.
The support for this option was missing and could lead to
installation failures, e.g. using kickstart.
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Client architecture aware DHCP configurations may send a
bootfile for the wrong architecture, if the DHCP request
doesn't include the client architecture identifier 0x001f.
Further, the DHCP reply may include the pxepathprefix option,
which points to the base directory to use for TFT requests.
If the value is not honored by the client, the wrong
configuration file, kernel, etc. could be loaded.
Both issues are fixed by this commit.
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Compressed kernel modules (filename *.ko.xz) would not be included
since the script was looking for files with .ko suffix only.
This can result in a pxelinux.0 image that can't access the
network, which is not good for a network boot loader...
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The output of /proc/<pid>/stat will show the process name in parentheses.
The parsing code in read_stat() tries to filter out the parentheses, which
will go wrong when the process name itself also contains parentheses, e.g.
in an output like this: "2421 ((sd-pam)) S 2420 2420 2420 ..."
In this case, the first closing parentheses will be taken as end marker,
and the sscanf() on the remaining string will silently fail, leaving its
values in uninitialized state and producing wrong data.
Fix this by using strrchr() instead of strchr() to find the last closing
parentheses. Also add return value checking for sscanf() and initialize the
values to 0.
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
mk-pxelinux-ramfs, mod_fsstatd|procd.service, zfcpdump all use
/sbin/modprobe path. Adjust zdev to use that path as well. This works
universally on both usr-merge systems (e.g. Debian/Ubuntu with
usrmerge package installed), and split-usr systems alike
(Debian/Ubuntu current stable releases default).
Closes: #31
Link: https://bugs.launchpad.net/ubuntu/+source/s390-tools/+bug/1777600
Signed-off-by: Dimitri John Ledkov <xnox@ubuntu.com>
Acked-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When a volume or APQN association is made while generating or
importing a key, and a duplicate association is detected, then
this may leave the key in an inconsistent state.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When a zkey generate or change command is used to associate one
or multiple APQNs the command succeeds, but no key is generated
and no APQNs are associated, because the return code returned by
_keystore_apqn_check() is wrong.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The APQN check routine as well as the properties helper functions
do not free all memory that they allocated.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a zipl-switch-to-blscfg script that can be used to switch the zipl
configuration in a system to use the BootLoaderSpec (BLS) config files
to define the IPL sections instead of having them defined in zipl.conf.
Closes: #28
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The BootLoaderSpec (BLS) defines a file format for boot configurations,
so bootloaders can parse these files and create their boot menu entries
by using the information provided by them [0].
This allow to configure the boot items as drop-in files in a directory
instead of having to parse and modify a bootloader configuration file.
If the /boot/loader/entries exists and there are BLS files there, then
these are parsed and configuration sections are added without the need
to have these in a zipl.conf file.
A different BLS directory can be specified from the command line using
the --blsdir option.
[0]: https://www.freedesktop.org/wiki/Specifications/BootLoaderSpec/
GitHub-ID: #28
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The function returns 0 on success and a negative number on error but is
useful to know how many tokens were allocated. This will be used by the
BLS parsing code to determine if needs to allocate mor tokens or not to
parse the BLS fragments.
GitHub-ID: #28
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
lsluns lists nothing and returns with return code 0 if no adapter (FCP
device) or target port exists in sysfs. This can be confusing for the user.
Print a message if no adapter exists. Print a message if no port exists on
any adapter.
Reported-by: Thorsten Diehl <thorsten.diehl@de.ibm.com>
Suggested-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Changing the cmb_enable attribute of a CCW device can cause a blocked
chzdev process or udev rule. This problem occurs when the device is
offline, cmb_enable is 0, and chzdev is used to set cmb_enable=1 and
online=1 at the same time.
The following call will block:
chzdev 190d -a cmb_enable=1 online=1
The following call will generate a udev rule that will block subsequent
calls to 'udevadm settle':
chzdev 190d -p online=1 cmb_enable=1
Fix this by ensuring that cmb_enable is only set when the CCW device
is not online.
Reported-by: André Wild <wild@linux.vnet.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Enable user to specify that a device should be configured early, that is
during the initial RAM-disk boot phase. This may be necessary, e.g. to
override auto-configuration for a device which is also applied during
that boot phase. It may also be used to manually specify devices that
are required to access the root file system, such as networking devices.
Users can mark devices as requiring early configuration by specifying
a value of 1 for the newly added internal attribute zdev:early:
# chzdev dasd-eckd 0.0.1234 -p zdev:early=1
This can be changed back by removing the attribute setting, or by
setting the attribute value to 0:
# chzdev dasd-eckd 0.0.1234 -p -r zdev:early
or
# chzdev dasd-eckd 0.0.1234 -p zdev:early=0
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This change adds base infrastructure for implementing internal device
attributes. In the context of the zdev tools, an internal device
attribute is a new type of device attribute with the following
characteristics:
- Can be set and removed like normal device attributes
- Affects zdev-internal handling only
- Does not correspond to an actual device attribute, that is
it has no representation in SysFS
- Can not be set in the active configuration
- Name starts with "zdev:" to prevent conflicts with actual
device attributes
Values for internal device attributes are stored in udev rules alongside
the normal persistent configuration of a device. They are encoded as
udev environment variables. Note that they have no further effect on
udev processing.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add initramfs-tools scripts that apply firmware-provided I/O
configuration data as auto-configuration during boot. This way, all I/O
devices configured by DPM are automatically brought online without
further user interaction.
This mechanism is active by default. It can be deactivated by specifying
the following parameter on the kernel command line:
rd.zdev=no-auto
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a dracut hook that applies firmware-provided I/O configuration data
as auto-configuration during boot. This way, all I/O devices configured
by DPM are automatically brought online without further user
interaction.
This mechanism is active by default. It can be deactivated by specifying
the following parameter on the kernel command line:
rd.zdev=no-auto
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Auto-configuration is the name of a new configuration target that is
supported by chzdev and lszdev besides the existing active and
persistent configuration targets. Directives created in this new
configuration are stored as udev rules in the /run/udev/rules.d
directory.
Auto-configuration directives are only in effect if there are no
directives for the same device in the user-provided persistent
configuration. This allows users to override auto-configuration
directives if necessary.
Due to the volatile nature of the /run directory, auto-configuration
directives are cleared on reboot. Therefore mechanisms that generate
auto-configuration directives must recreate them on every boot.
The lszdev tool displays auto-configuration data both in list view
as well as in detail view. Users can specify the new option --auto-conf
to only show data from this configuration target.
Mechanisms that generate automated configuration directives can use
chzdev together with the --auto-conf option to create the corresponding
udev rules.
Note: This change does not include a mechanism that generates
auto-configuration directives.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Change chzdev's udev rule generation from the previous approach of
combining all zfcp-lun udev rules associated with an FCP device into a
single file to storing zfcp-lun udev rules in one file per zfcp-lun.
This is done to enable per-device udev rule masking.
With udev rule masking, if a udev rule file by the same name exists in
both /etc and /run, the udev daemon will only consider the rules found
in /etc.
The auto-configuration feature will make use of per-device udev rule
masking to introduce a new class of configuration data (stored in /run)
that is only active if no user-provided configuration data (in /etc)
exists.
In addition, change chzdev to allow the regeneration of udev rules by
using the --force command line like in the following example:
# chzdev zfcp-lun --configured --enable --persistent --force
This can be used to convert all existing zfcp-lun udev rules from the
old format to the new one.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
There are some situations where running "udevadm settle" can result in
a deadlock, such as in the early stages of initial RAM-disk processing.
Introduce a new command-line option --no-settle that can be used to
suppress calling "udevadm settle" to allow chzdev to be run in such
situations.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add support for reading firmware-provided I/O configuration data files.
Such configuration files are generated by the Dynamic Partition Manager
and made available via a kernel interface for consumption by Linux.
To read a firmware configuration file, use the existing --import option:
# chzdev --import /sys/firmware/sclp_sd/config/data
This will apply all I/O configuration data found in the specified file
to the persistent configuration.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Apply some changes to existing functions and data structures to simplify
the firmware configuration file support implementation.
- Make qeth and dasd subtype objects non-static
- Change the existing helper functions for reading file contents into
memory to also support binary functions
- Move some configuration file import functions to make them available
for use in other source files
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The udev rule for SCSI devices is still using the WAIT_FOR keyword,
which was used to make sure that a file does in fact exist before any
rule is applied. That keyword, however, was removed with systemd-220 and
is therefore no longer valid, which in turn leads to errors on newer
systems.
Use the TEST keyword instead to check the existence of a file before we
apply any rules.
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Create the default keystore directory '/etc/zkey/repository'
and the user group 'zkeyadm' during make install.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add new commands to the zkey utility to store secure AES keys in
the secure key repository provided by the keystore implementation
introduced in the previous commit.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a keystore implementation that stores secure AES keys in a
key repository, located in a directory, e.g. '/etc/zkey/repository'.
The keystore allows you to generate, validate, re-encipher, modify,
list, delete, etc secure keys.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
To reduce the size of the zkey.c source file, all routines that
deal with secure keys are moved to a new source file pkey.c.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Comma separated strings are used in property values to store
multiple values in one property. These helper functions allow to
work with such comma separated strings.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The integrity support for the properties file routines use
SHA-256 to build a hash of the keys and values of a property file.
The codes uses the EVP_DigestInit_ex, EVP_DigestUpdate, and
EVP_DigestFinal from the libcrypto library (OpenSSL).
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
In preparation for a new feature, introduce property file
handling routines. A property file stores key value pairs
in a text file. Optionally a hash of all keys and values
contained in the properties file can be generated to
ensure integrity of the properties file and to detect
manual modifications.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add 'Dump file size' field for zgetdump -i output to show the actual size
of dump file in s390 extended format on disk in megabytes.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add verbose option for zgetdump to display the detailed layout of memory
map when printing dump info for s390_ext or ELF dump format.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Adjust ELF formatted dump processing by zgetdump tool in order to handle
so-called zero loads using zero memory chunks introduced with s390
extended dump format.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Update zgetdump tool to process dumps of the new s390 extended format.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Update 'zgetdump -d' to process the new DASD dump tool installed (single
volume or multi-volume).
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Update df_s390_dumper structure and dumper reading functions to read
certain dumper fields instead of the entire dump tool.
Update other s390 dump common functions in preparation for the new dumper
and the new s390 extended dump format.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Adjust progress_print() function to print dump progress message based on
the time interval thus printing the message with a regular rate.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Update zipl DASD standalone dumpers for single-volume, multi-volume and FBA
to drop the sets of zero pages when dumping memory to disk. Thus, all
continuous sets of zero-megabytes in the memory are dropped during the dump
stage.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Old implementation was using Test Protection (TPROT) instruction to
validate the memory page. Given that it is a rather slow instruction and
the fact that we only need to validate the read-access for a page, we can
use simple Load or Insert Character instruction to test that.
The new version of page_is_valid() without a TPROT is introduced.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.vnet.ibm.com>
Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The program check handler has a bug in calculating the address of the target
to jump to. Furthermore the use of relative addresses in the exception table
can lead to situations where the (calculated) fault is not unique. Storing
the absolute address of the fault and target in the exception table solves
both problems.
This patch is intended to:
- Modify exception table to store the absolute address of 'fault' and
'target' points
- Adjust program check handler function pgm_check_handler_fn() accordingly
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.vnet.ibm.com>
Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
chreipl could only be used as root (effective userid 0), which is
explicitly checked upon invocation.
"access" call is a wrong method to check super user read/write access
for sysfs files, because it is simply always returns 0 (success), despite
actual sysfs file permissions or underlying sysfs file callbacks setup.
The only guaranteed way of checking sysfs file super user access is an
actual open call with the corresponding access mode.
The problem is that chreipl always tries to update some sysfs attributes
even through they are not specified as command line arguments. Together
with a broken sysfs file access checks this leads to inability to use
the tool, when some sysfs attributes are read-only (which is the case
on older systems where diag308 set does not work).
$ chreipl ccw -d 0.0.ec5a
chreipl: Could not open "reipl/ccw/parm" (Permission denied)
The change fixes access checks, which are in place to handle "diag308
set does not work" case (presence of read-only sysfs attributes).
Also replaces R_OK with F_OK in "set_target_type_auto" to underline that
only file presence is checked, not an actual read access.
Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Make write_str_errno return errno in all error cases, so that it would
be usable in scenarious where open call could fail and we just want to
get an errno, not to exit.
This function is currently only used by chshut in a context where
such a change wouldn't hurt debug ability.
159 if (write_str_errno(argv[2], path))
160 ERR_EXIT_ERRNO("Could not set \"%s\"", path);
Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
kdump reipl implementation relies on os_info kernel structure
residing in old memory being dumped. os_info contains reipl block, which
is used (if valid) by the kdump kernel for reipl parameters.
The problem is that the reipl block and its checksum inside
os_info is updated only when /sys/firmware/reipl/reipl_type is
written. This sets an offset of a reipl block for "reipl_type" and
re-calculates reipl block checksum. Any further alteration of values
under /sys/firmware/reipl/{reipl_type}/ without subsequent write to
/sys/firmware/reipl/reipl_type lead to incorrect os_info reipl block
checksum. In such a case kdump kernel ignores it and reboots using
default logic.
This change makes sure that all fcp values are written before
reipl_type, to avoid potential problem with incorrect checksum.
Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
When dump is not complete or dump memory limit is set, zgetdump might end up
with the Segfault on reading Vector Registers. That might take place when
the Vector Registers save area is beyond the dump memory limit.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.vnet.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The short descriptions for a counter in the extender counter
set should be identical for all models.
This patch fixes minor differences.
Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
double free or corruption (out)
Program received signal SIGABRT, Aborted.
0x000003fffdd40404 in raise () from /lib64/libc.so.6
(gdb) bt
#0 0x000003fffdd40404 in raise () from /lib64/libc.so.6
#1 0x000003fffdd41ec2 in abort () from /lib64/libc.so.6
#2 0x000003fffdd88d8e in __libc_message () from /lib64/libc.so.6
#3 0x000003fffdd905c8 in malloc_printerr () from /lib64/libc.so.6
#4 0x000003fffdd98dfa in free () from /lib64/libc.so.6
#5 0x0000000001005994 in util_ptr_vec_free (count=<optimized out>, ptr_vec=0x101de90) at ../include/lib/util_base.h:46
#6 util_scandir_free (de_vec=0x101de90, count=<optimized out>) at util_scandir.c:174
#7 0x00000000010043ce in print_defunct_devices (rec=rec@entry=0x10114f0,
path=path@entry=0x101cc80 "/sys/devices/css0/defunct") at lscss.c:678
#8 0x0000000001004cfe in print_subchannels_of_type (type_requested=type_requested@entry=SUBCHANNEL_TYPE_IO,
rec=rec@entry=0x10114f0) at lscss.c:726
#9 0x0000000001003888 in cmd_lscss () at lscss.c:761
#10 main (argc=<optimized out>, argv=<optimized out>) at lscss.c:932
Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
The cpumf_helper.in is called to parse counter definition
files stored in /usr/share/s390-tools/cpumf directory and
returns perl references to be examined by other perl tools
such as lsctrdef.
cpumf_helper.in expects several keywords in the counter
definition files, one is 'shortdesc' which contains
a short description of the purpose of the counter.
However this short description not is stored in the hash
field shortdesc keyed by counter number 'ctr' because the
wrong key is used.
Output before:
{
'64' => {
'desc' => 'Total number of the PRNG functions issued by the CPU',
'name' => 'PRNG_FUNCTIONS',
'set' => 8
},
'65' => {
'desc' => 'Total number of CPU cycles when the DEA/AES coprocessor is busy performing PRNG functions issued by the CPU',
'name' => 'PRNG_CYCLES',
'set' => 8
},
....
'ctr' => { <-------- WRONG
'shortdesc' => 'AES Blocked Cycles'
}
}
Output with this patch:
{
'64' => {
'desc' => 'Total number of the PRNG functions issued by the CPU',
'name' => 'PRNG_FUNCTIONS',
'set' => 8,
'shortdesc' => 'PRNG Functions'
},
'65' => {
'desc' => 'Total number of CPU cycles when the DEA/AES coprocessor is busy performing PRNG functions issued by the CPU',
'name' => 'PRNG_CYCLES',
'set' => 8,
'shortdesc' => 'PRNG Cycles'
},
...
}
Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Use the new libvmcp.a to send commands to z/VM CP.
The error messages remain unchanged.
Local header file vmcp.h now only contains variable definitions
and some defines used by vmcp.c. Move them to the source file
and remove the now empty header file.
Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This patch prevents code duplication in vmcp related programs.
Some applications such as vmur, vmcp, and lsqeth
use z/VM CP commands to query information. Each program
uses its own defines and code.
Supply a library and common header file to use common code.
Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Fixed typo and added one sentence to explain more details
about online switching of queue devices.
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
This patch
- moves function util_strstrip to libutil/util_libc.c and deletes the
extra files util_strstip.[ch]
- adds doxygen comments for util_strstrip()
- provides an example on how to use util_strstrip().
Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Fix some warnings found by sparse.
Declare one function static and change static variables to defines.
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Remove unused include to fix an error reported by make C=1.
CHECK zipl/src/zipl_helper.device-mapper.o
../../include/lib/dasd_base.h:159:20: error: redefinition of struct hd_geometry
../../common.mak:234: recipe for target 'zipl_helper.device-mapper.o' failed
make: *** [zipl_helper.device-mapper.o] Error 1
The structure hd_geometry is defined in dasd_base.h.
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
hmcdrvfs will ignore files with a link count >= 1000, due to an inconsistency
in the input data. The parsing code relies on having spaces between the
different fields in its input data. When a file has a link count >= 1000,
there will be no space, and the "link count" field will be placed directly
after the previous "mode" field. This will confuse the parser, and all such
files will not be accesible.
The "mode" field will never contain digits, so in order to fix this issue,
the parser can recognize the "link count" field by the presence of a digit.
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Function strstrip strips leading and trailung
spaces from a given string.
During review it was decided to move this function
to the libutil library to make it available for other tools.
Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Compilers in Debian and Ubuntu default to enabling PIE for all binaries.
This is a poor fit for boot loader code that has strict requirements
for its layout. Hence always forcefully disable PIE.
As older compilers still need to be supported, no-pie usage is
conditional on CC's -dumpspecs output.
Fixes: #3
Signed-off-by: Philipp Kern <pkern@debian.org>
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Update the cpumf helper program to read the split counter set
definition files. Changes to higher-level program like lscpumf
are not necessary.
Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
With z14, the counters in the problem-state are reduced resulting
in an increased first version number of the CPUM CF. To adapt to
this change, split the counter sets according to their counter
first and second version number. The second version number controls
the crypto-activity and extended counter set. Treat the crypto-activity
counter set as generic, as the extended counter set is already handled
based on hardware models.
Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
When 'make C=1' reports a fatal error for a source file, re-running
'make C=1' will no longer report this error because the target object
file was successfully created before the check was run.
Fix this by moving the check before the compilation step so that the
object file is not built in case of a fatal check error.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Reviewed-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Since now util_list_new() internally fails for out-of-memory situations,
we can also remove some NULL pointer checks.
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Use a character, that can't be used in rpm Version and Release tag, as
the delimiter in sed expressions.
Fixes: #25
Signed-off-by: Dan Horák <dan@danny.cz>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
When calculating partition start offset during multi-volume dump processing,
the result of multiplication is truncated to 32-bits thus resulting in
invalid behavior in case the real offset is beyond 4Gig boundary.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.vnet.ibm.com>
Reviewed-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Running dracut when the root device is not known to zdev (for example
because it is located on a virtio block device) will cause the zdev
dracut module to incorrectly return an error in the installkernel()
function. As a result dracut aborts with an error.
Fix this by ensuring that the non-zero exit code resulting from lszdev
not being able to determine the root device is not passed on to the
calling function. Also remove unnecessary error output in this case
by leaving the install() function early when the root device is not
known to zdev.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
When a persistent device configuration is changed, chzdev tries to
find out if it needs to perform additional steps to make this change
persistent. If this check fails, for example because the root device
is located on a RAM-disk, or on a device type not managed by chzdev,
the tool reports an error and exits with non-zero exit code:
chzdev: Could not determine device that provides /
or
chzdev: Could not determine device that provides loop0
This behavior unnecessarily restricts chzdev from being used in
scripted environments like an installation initial RAM-disk.
Fix this by removing the non-zero exit code and moving the message to
verbose output mode only.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Commit 61b60baf57 ("dasdinfo: Fix GCC 7 overflow warning") erroneously
introduced line breaks to the UID (-u) output. To be consistent with the
extended uid ouput code, keep the newley indroduced line breaks and
remove the additional one from the string split above instead.
Fixes: 61b60baf57 ("dasdinfo: Fix GCC 7 overflow warning")
Signed-off-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
When building libutil in parallel, it can happen that the archive
file is created twice:
$ cd libutil
$ make -j
...
CC libutil/util_rec_example.o
AR libutil/libutil.a
AR libutil/libutil.a
LINK libutil/util_base_example
This can lead to the following build error:
$ make -j OPT_FLAGS="-pipe"
..//libutil/libutil.a(util_path.o): In function `sys_mount_point':
util_path.c:56: undefined reference to `util_proc_mnt_get_entry'
util_path.c:60: undefined reference to `util_proc_mnt_free_entry'
collect2: error: ld returned 1 exit status
../common.mak:232: recipe for target 'util_path_example' failed
The problem is that the following rule from common.mak is triggered in
the libutil directory:
$(rootdir)/libutil/libutil.a: $(rootdir)/libutil
$(MAKE) -C $(rootdir)/libutil/ libutil.a
.PHONY: $(rootdir)/libutil
To prevent this rule, use the local "libutil.a" as dependency for the
example binaries.
Reported-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
For GitHub and git beginners explain in more detail how to submit s390-tools
contributions as pull requests, how to update already submitted code
and how to rebase a contribution branch.
Based on the initial proposal from: Jens Remus <jremus@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.vnet.ibm.com>
Fix possible c&p mistake when checking the presence of `queue_full` attribute
in old kernels. There should be no impact on how the tool works other than
exiting gracefully in case the attribute does not exist.
Fixes: #21
Signed-off-by: Rafael Fonseca <r4f4rfs@gmail.com>
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
On Fedora 27 "net-snmp-config --agent-libs" returns
"--specs=/usr/lib/rpm/redhat/redhat-hardened-ld" which introduces "-pie":
# cat /usr/lib/rpm/redhat/redhat-hardened-ld
*self_spec:
+ %{!static:%{!shared:%{!r:-pie}}}
*link:
+ -z now
This leads to the following link error:
/usr/bin/ld: osasnmpd.o: `getopt_long@@GLIBC_2.2' non-PLT reloc for symbol
defined in shared library and accessed from executable
(rebuild file with -fPIC ?)
/usr/bin/ld: final link failed: Bad value
Add -fPIC to CLFAGS as a workaround to enbable linking again.
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
For systems that do not have glibc-static installed, print the
following error message when building zfcpdump:
s390-tools/zfcpdump # make
REQCHK zfcpdump (stdio.h)
****************************************************************
* Missing build requirement for: zfcpdump *
* Install package..............: glibc-static or libc-dev *
* You can skip build with......: make HAVE_LIBC_STATIC=0 *
****************************************************************
To allow the check add a new compiler/linker option to the "check_dep"
macro and do full compile/link instead of a simple compile.
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
This fixes the same issue as in commit 82c8148983 ("cpuplugd: Improve
systemctl start error handling") for mon_tools (mon_procd and mon_fsstatd).
Currently "systemctl start mon_procd/fsstatd" does not report any errors
in case the startup fails.
Example (with mon_procd):
(change interval in /etc/sysconfig/mon_procd to an invalid value "abc")
# systemctl start mon_procd
The reason is that for type=simple systemd forks/execs mon_procd and if
that is successful immediately returns. There is no way to find out if the
initial startup fails.
Fix this by using type=fork and running the process in the background. In
this case systemd waits until the initial process returns.
In addition use PIDFile and ensure that the pid file is already available
when the initial process returns. To achieve this, use startup
synchronization via pipe. Without that systemd would print the following
warning:
systemd[1]: mon_procd.service: PID file /var/run/mon_procd.pid not readable
(yet?) after start: No such file or directory
With this patch, an early startup error like in the example above, is now
reported correctly in "systemctl start":
# systemctl start mon_procd
Job for mon_procd.service failed because the control process exited...
See "systemctl status mon_procd.service" and "journalctl -xe" for ...
# journalctl -xe | grep mon_procd
mon_procd[3184]: Error: Invalid interval (needs to be greater than 0)
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Currently "systemctl start cpuplugd" does not report any errors in
case the startup fails.
Example:
# mv /etc/cpuplugd.conf /etc/cpuplugd.conf.xxx
# systemctl start cpuplugd
The reason is that for type=simple systemd forks/execs cpuplugd and if
that is successful immediately returns. There is no way to find out
if the initial startup fails.
Fix this by using type=fork and running cpuplugd in the background. In
this case systemd waits until the initial process returns.
In addition use PIDFile and ensure that the pid file is already available
when the initial cpuplugd process returns. To achieve this, replace the
daemon() function by our own implementation that introduces startup
synchronization via pipe. Without that systemd would print the following
warning:
systemd[1]: cpuplugd.service: PID file /var/run/cpuplugd.pid not readable
(yet?) after start: No such file or directory
With this patch, an early startup error like in the example above, is now
reported correctly in "systemctl start":
# systemctl start cpuplugd
Job for cpuplugd.service failed because the control process exited...
See "systemctl status cpuplugd.service" and "journalctl -xe" for ...
# journalctl -ex | grep cpuplugd
Nov 16 15:52:27 ... cpuplugd[5096]: Opening configuration file failed:
No such file or directory
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Acked-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
During the GPL to MIT conversion process from v1.39.0 to v2.0.0 one line has
been lost by accident. Fix this and re-add that line again.
Fixes: #17
Reported-by: Mikhail Obirvalger
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
So far it infinitely retried the bootscript, that doesn#t seem
reasonable. Per IRC discussion this is now changed to a 600 sec timeout
waiting for a config to appear to then run once on that config.
In case the pxe boot fails the it is now no more a retry loop, but
instead considered a crash as init exits.
Closes: #16
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
Acked-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Currently multiple entries in a single append is broken by how the
shell in busybox escapes them. No usual grouping with " or ' will
work but calling them directly but without the --append in the
variable will let busybox do the right thing.
GitHub-ID: #16
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
Acked-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Currently for a user the pxe boot seems unclear and it
is hard to follow what is going on.
- Report on when (re-)starting the bootscript.
- Report configs fetched (Without e.g. on a virsh console
one only sees a loop of timeouts.)
- Report which config is missing on the PXE_CONF check
- Report kexec calls made
GitHub-ID: #16
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
Acked-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
lsmem/chmem moved to util-linux >= 2.30, remove them from s390-tools.
For kernels >= 4.13, the default memory zone for hotplug memory was
changed from zone MOVABLE to zone NORMAL. In order to keep using zone
MOVABLE for memory hotplug, either use the lsmem/chmem tools from
s390-tools version 2.1.0, or an updated version of the util-linux
version of lsmem/chmem that contains the following util-linux git
commits (included in util-linux 2.32):
60a7e9e94e49 "lsmem/chmem: add memory zone awareness"
0a4320f5e785 "tests/lsmem: update lsmem test with ZONES column"
afee3f204247 "lsmem/chmem: add memory zone awareness to bash-completion"
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The zdev tools use a hard-coded path to locate the vmcp binary. Since
this path may differ between distributions, the vmcp binary may not be
successfully located in all cases. Fix this by using the BINDIR macro
to determine the correct path. This macro is also used during
installation of the vmcp binary itself.
Reviewed-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
If SYSTEMDSYSTEMUNITDIR is defined, the systemd units are automatically
installed. Therefore consequently also install the required config files
in that case.
Example:
$ make install SYSTEMDSYSTEMUNITDIR=/usr/lib/systemd/system/
Reviewed-by: Benjamin Block <bblock@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
dasdview command does not accept the '-f' option so because of that the
command will always fail.
Fixes: #10
Signed-off-by: Rafael Fonseca <r4f4rfs@gmail.com>
Acked-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
When doing parallel builds, we currently get the following error for
iucvterm:
~/s390-tools/iucvterm (bb/holzheu/install-configs)# make install -j
INSTALL iucvterm
CC iucvterm/src/iucvconn.o
INSTALL iucvterm/src
CC iucvterm/src/getopt.o
install: cannot stat 'iucvconn': No such file or directory
install: cannot stat 'iucvtty': No such file or directory
Makefile:19: recipe for target 'install' failed
Fix this and "all" as explicit dependency to the install target.
Reviewed-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
The cpi service unit requires an Install section to tell systemd
when to start the cpi service. Add an install section and start
cpi for the multi-user target.
Reported-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
The man page date was AUG 2008. Changed to OCT 2017.
A previous commit had a wrong indentation on following
options text for lszcrypt. Fixed.
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Commit 3c80f7e025 ("dasdinfo: fix buffer overflow warning") changed a
sprintf call to snprintf to avoid a buffer overflow warning. However,
GCC 7 now warns about a potential truncation with snprintf:
dasdinfo.c: In function 'main':
dasdinfo.c:577:18: warning: '%s' directive output may be truncated
writing up to 255 bytes into a region of size 69 [-Wformat-truncation=]
"/sys/block/%s/device/uid", dir_entry->d_name);
^~
dasdinfo.c:576:4: note: 'snprintf' output between 23 and 278 bytes into
a destination of size 80
snprintf(*uidfile, RD_BUFFER_SIZE,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"/sys/block/%s/device/uid", dir_entry->d_name);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
We could get around this by increasing the buffer. Though, the current
buffer size is already plenty and we know better anyway.
Avoid the warning by simply checking the return value of snprintf and
display an error in case data was truncated nonetheless.
Fixes: 3c80f7e025 ("dasdinfo: fix buffer overflow warning")
Signed-off-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
With z14 there comes a new crypto card 'CEX6S'.
This patch introduces the s390-tools changes needed
to list the new card and show the capabilities correctly.
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
get_device_name() contains a chunk of unreachable code, as the
'name' argument is never filled with any value.
So turn things around to have get_device_name() always fill the
'name' argument with the real device name, and remove the
devname entry from the dasdfmt_info_t structure.
GitHub-ID: #8
Signed-off-by: Hannes Reinecke <hare@suse.com>
Acked-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
dasdfmt_get_volser() opens its own private filedescriptor
despite the global filedes already being opened.
And we should be passing in only the bits of the info structure
that we actually need.
GitHub-ID: #8
Signed-off-by: Hannes Reinecke <hare@suse.com>
Acked-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The wording in the lsluns usage statement and man page was misleading
in several aspects:
* lsluns does not list all LUNs, just those discovered in the FC SAN
* lsluns -a should only be used to display the LUN encryption status
Fix filter option arguments. Clarify filter option usage. Refer to
lszfcp and lsscsi.
Reported-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.vnet.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The formatting in the SYNOPSIS, OPTIONS, and EXAMPLES sections was flawed
in the lsluns(8) man page:
* The comma between short and long options were erroneously formatted in
bold, which would indicate to be typed exactly as shown. [see man(1)]
* The option parameters (e.g. adapter and port) were erroneously formatted
in bold instead of italic text (usually displayed as underlined on the
console), which would again indicate to be typed exactly as shown instead
of to be replaced with the appropriate argument. [see man(1)]
* Ellipses were missing after the the options that may be specified multiple
times (e.g. adapter and port). [see man(1)]
* Dashes in options in the SYNOPSIS and OPTIONS section needed to be
escaped. [see man-pages(7)]
* User input in example shell sessions should have been formatted in bold.
[see man-pages(7)]
Correct formatting based on man(7) and man-pages(7) man pages as reference.
Add proper spacing between options and their surrounding square brackets and
between the three periods of ellipses.
Signed-off-by: Jens Remus <jremus@linux.vnet.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.vnet.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
lsluns printed potentially confusing messages when a filter or combination
of filters matched nothing:
No valid combination found for adapter '0.0.1906'. Removing from
resource list.
No valid combination found for port '0x50050763071845e3'. Removing from
resource list.
...
To the user it is potentially unclear which 'combination' is actually being
referred to, as only one part of the combination is mentioned, and what the
ominous 'resource list' is. The later information is merely useful for a
developer to debug the script.
Such a message was written for every user supplied filter that did not
contribute anything to the resulting subset that is being listed, although
the filter actually might match something when used standalone.
Additionally those messages were printed to stdout instead of stderr. As
there is no debug or verbose switch and the information level of those
messages is low, we may simply discard them.
Reported-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.vnet.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.vnet.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Fix the following undesired behavior of lsluns to scan all resources if the
user provided filters did not match anything:
No valid combination found for adapter '5080'. Removing from resource
list.
No valid parameters left, using all available resources in system.
Scanning for LUNs on adapter 0.0.5080
...
Scanning can be resource consumptive. So if a user already wants to filter,
possibly to reduce resource consumption, he does not want to happen to scan
everything and thus consume the worst case of resources.
Instead print a message to inform the user why nothing was scanned.
Reported-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.vnet.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.vnet.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Fix a possible buffer overflow.
The buffer overflow is only theoretical since the device name is max
8 characters in length.
This fixes following gcc 7 warning:
dasdinfo.c: In function 'main':
dasdinfo.c:576:33: warning: '%s' directive writing up to 255 bytes into a
region of size 69 [-Wformat-overflow=]
sprintf(*uidfile,"/sys/block/%s/device/uid",
^~
In file included from /usr/include/stdio.h:862:0,
from dasdinfo.c:15:
/usr/include/bits/stdio2.h:33:10: note: '__builtin___sprintf_chk' output
between 23 and 278 bytes into a destination of size 80
return __builtin___sprintf_chk (__s, __USE_FORTIFY_LEVEL - 1,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
__bos (__s), __fmt, __va_arg_pack ());
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The information shown with the --help option did not mention the -v or
--version option which can be used to retrieve the version information.
This patch fixes it.
Fixes: 7179f5de2a1582 ("cpi: Introduce new udev rule and integrate with systemd unit")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.vnet.ibm.com>
Reference-ID: VS1706
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
This patch fixes a segfault in the qeth get_layer function that occurs
when an unknown attribute is used in combination with the force option.
Signed-off-by: Hans Wippel <hwippel@linux.vnet.ibm.com>
Reviewed-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
This patch adds support for the qeth attributes that are used to
configure VNIC Characteristics.
VNIC Characteristics and BridgePort are mutually exclusive. Thus, this
patch also adds a check function for conflicting settings to the qeth
code. Also, the output when VNIC Characteristics is active is added to
the BridgePort attribute descriptions.
Signed-off-by: Hans Wippel <hwippel@linux.vnet.ibm.com>
Acked-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Re-add a missing line that was removed by accident when creating the
new s390-tools-2.0.0 release from s390-tools-1.39.0.
Fixes: b627b8d8e1 ("Initial s390-tools-2.0.0 import")
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
With GCC 7 we get the following warning as a potential overflow might
happen, if d_name gets too big:
dasdinfo.c: In function 'main':
dasdinfo.c:611:37: warning: '%s' directive writing up to 255 bytes into
a region of size 69 [-Wformat-overflow=]
sprintf(sys_dev_path, "/sys/block/%s/dev", dir_entry->d_name);
^~
dasdinfo.c:611:3: note: 'sprintf' output between 16 and 271 bytes into a
destination of size 80
sprintf(sys_dev_path, "/sys/block/%s/dev", dir_entry->d_name);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This could be fixed by simply increasing the buffer size. However, there
is a little bit more to it and the way files are currently read can be
simplified.
Do this by using the libutil functions to read files and clean up a
little along the way.
Reviewed-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Before writing data to the disk, fdasd always checks whether a partition
is specified to prevent writing partition tables to a partition.
However, this also prevents the user from writing partition tables to
loop devices > loop0, even when the force option is specified.
Technically, those are not partitions.
Therefore, allow writing partition tables to partitions when the force
option is specified.
Fixes: #4
Reviewed-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
This patch adds the qeth attributes that are used to configure
VNIC Characteristics to lsqeth.
The output of lsqeth comprises two columns: one for attribute names, one
for attribute values. Not all VNIC Characteristics attribute names fit
in the corresponding column. Thus, this patch also increases the width
of the attribute names column by 2 characters.
Signed-off-by: Hans Wippel <hwippel@linux.vnet.ibm.com>
Reviewed-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Currently, the qeth attributes that are used to configure IPv6 NDP proxy
are described as ARP proxy attributes. This patch changes the titles and
descriptions from ARP to NDP.
Signed-off-by: Hans Wippel <hwippel@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Add a new udev rule that is triggered when a KVM VM is started
on the system the very first time. This rule reports over the
sysfs CPI kernel interface that the system is a KVM host.
Currently the cpi.service systemd unit already works with
the CPI sysfs interface. Introduce a new internal utility
script "cpictl" that is used by both the udev rule and
the systemd unit to guarantee coordinated access to the
kernel CPU interface.
Signed-off-by: Claudio Imbrenda <imbrenda@linux.vnet.ibm.com>
Reviewed-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Reviewed-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
With gcc 7 we get warnings for code like the following:
44 #define MAX_BUFFER_SIZE_FOR_SUBCHANNEL_ATTRIBUTES 37
406 char buf[MAX_BUFFER_SIZE_FOR_SUBCHANNEL_ATTRIBUTES];
416 device = de_vec[0]->d_name;
417 snprintf(buf, sizeof(buf), "%s", device);
$ grep -r 'd_name\[' /usr/include/
/usr/include/bits/dirent.h: char d_name[256];
The compiler assumes that d_name can be up to 255 characters. Therefore
it produces the following warning:
CC zconf/css/lscss.o
lscss.c: In function 'print_sch_io':
lscss.c:417:31: warning: '%s' directive output may be truncated writing
up to 255 bytes into a region of size 37 [-Wformat-truncation=]
snprintf(buf, sizeof(buf), "%s", device);
^~
In file included from /usr/include/stdio.h:936:0,
from lscss.c:10:
/usr/include/bits/stdio2.h:64:10: note: '__builtin_snprintf' output
between 1 and 256 bytes into a destination of size 37
return __builtin___snprintf_chk (__s, __n, __USE_FORTIFY_LEVEL - 1,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
__bos (__s), __fmt, __va_arg_pack ());
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fix this and use larger buffer sizes to keep gcc quiet.
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
With gcc 7 we get warnings like the following:
fdasd.c: In function 'main':
fdasd.c:3055:4: warning: this statement may fall through
[-Wimplicit-fallthrough=]
fdasd_exit(&anchor, 0);
^~~~~~~~~~~~~~~~~~~~~~
fdasd.c:3056:3: note: here
default:
^~~~~~~
Fix this by marking functions with "__noreturn" to help gcc.
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
With gcc 7 the compiler checks for sprintf() the maxium possible resulting
string based on the used datatypes.
Example:
struct vol {
...
char bus_id[9];
...
}
sprintf(vol->bus_id, "0.%x.%04x", ssid, vol_parm->devno);
The compiler can prove that "ssid" comes from "u8" and "vol_parm->devno"
from "u16". Therefore the resulting maximum string can be 0.ff.ffff which
requires 10 bytes.
This leads to the following warning:
dfi_s390mv.c: In function 'volumes_init':
dfi_s390mv.c:243:45: warning: '__builtin___snprintf_chk' output may be
truncated before the last format character [-Wformat-truncation=]
snprintf(vol->bus_id, sizeof(vol->bus_id), "0.%x.%04x", ssid,
^~~~~~~~~~~
In file included from /usr/include/stdio.h:936:0,
from dfi_s390mv.c:15:
/usr/include/bits/stdio2.h:64:10: note: '__builtin___snprintf_chk' output
between 9 and 10 bytes into a destination of size 9
return __builtin___snprintf_chk (__s, __n, __USE_FORTIFY_LEVEL - 1,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
__bos (__s), __fmt, __va_arg_pack ());
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To get rid of the warning use 10 instead of 9 bytes for the bus_id.
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The newer perf_event_open syscall needs a struct
perf_event_attr with the size field set correctly. Older
versions did not even provide this field. However, this
field is now set to the correct size value.
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
There was a too tight buffer allocated on the stack
which was overwritten by just one byte caused to
dump the application on function exit's stack check.
Reviewed all the buffer sizes and adapted them
to the need of the functions.
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
If called without parameters it updates the authors file with new authors
since last release tag. With the -t option a specific tag can be specified.
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Reviewed-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
A set of scripts and a short documentation describing how to build
a network boot image simulating a PXELINUX-style boot process.
Starting with QEMU 2.10 it is possible to boot a KVM guest over a
network interface using DHCP/BOOTP. The boot process is triggered by
the network boot firmware that is part of QEMU and follows the usual
network boot pattern: a DHCP request is issued by the client and
answered by a DHCP/BOOTP server. The DHCP reply will contain a
TFPT server identification and a bootfile name. The client will
retrieve the bootfile from the TFTP server, load it into memory
and IPL it.
A very common way of setting up a boot server has been defined
by PXELINUX, an open source implementation of PXE. With PXELINUX
the bootfile is a small network boot loader that will retrieve
a potentially client-specific configuration file containing
further instructions for the final boot process (kernel, ramdisk,...).
The set of sample scripts contained in the netboot directory provide
directions for a Linux distributor or a boot server administrator
on how to build a network boot image usable for a simplified
PXELINUX-style network boot setup for s390.
Note that the sample scripts are implementing only a subset of PXELINUX
functionality, specifically the config file parsing. In order to
get full functionality, a more specialized boot loader program
like petitboot or pxe-kexec must be used in the ramdisk.
Further, a sample Dockerfile is provided along with instructions on how
to build the network boot image in a Docker container.
Signed-off-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The -d and -e options did not show any failure message when
executed with insufficient permissions (e.g. not root user).
Now the behaviour is equal to the other options: A message
'chzcrypt: Error - can't write to xxx. Wrong permissions or
wrong tools version' is printed.
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The -c option was listed with <device-id> and the text
explained that device id may be a card or queue device id.
The --capability option only allows a card as argument so
the man page has been corrected to show this now correctly.
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The sysfs files ap_control_domain_mask and ap_usage_domain_mask
may hold just a string "not supported" if there's no crypto
configuration available. However, lszcrypt always processed
the content of these files as hex number and so produced funny
output if there's no configuration data available.
Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
During processing, the zdev-provided dracut module creates a temporary
file using the mktemp tool. Due to a missing path specification
in the filename template, mktemp tries to create the temporary file
in the current working directory which fails if that directory is not
writable:
# cd /sys
# dracut -Nf
mktemp: failed to create file via template 'dracut-zdev.XXXX': No such
file or directory
chzdev: Could not write to file : No such file or directory
sed: can't read : No such file or directory
chzdev: Could not open file : No such file or directory
Fix this by specifying option --tmpdir when calling mktemp.
Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Abbreviated commands are not recognized and zkey issues an invalid
command error.
In is_command(), the abbreviated command string is copied into the
command_str variable. Because this variable is not initialized and,
thus, might contain arbitrary data, a NUL-terminated is not guaranteed.
The following string comparison is very likely to fail. Correct this
problem by comparing up to the length of the command string only.
Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Few options are displayed with their argument concatenated. This
is due to the nature of the .BR roff macro. Ensure to separate them
with a blank (using "\~" which is an unbreakable space that stretches
like a normal inter-word space when a line is adjusted).
Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
When validating an XTS key, the 2 key parts are validated separately.
This can happen on different crypto cards, when more than one crypto
card is available to the system. Each key part can be validated on
a different card, however, the 2 cards need to have the same master
key, thus the master key validation pattern contained in the 2 key
parts are checked.
Signed-off-by: Ingo Franzki <ifranzki@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Set the umask for the zkey process to clear permissions for genernated
secure key files for group and others.
Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
The LDFLAGS content is added before the object files causing unresolved
symbol dereferences during the build. This actually happens so far on
Debian/Ubuntu instances.
Hence, use LDLIBS instead.
Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
log_alert "Changing the re-IPL device failed. The current re-IPL settings might be inconsistent. Check and correct the settings (see the README.md of chreipl-fcp-mpath) to make sure that the current re-IPL device is valid."
return 14
fi
firmware_unlock_exclusive
if [ "${sdev_busid}" != "${ipl_busid}" ] \
|| [ "${sdev_wwpn}" != "${ipl_wwpn}" ] \
|| [ "${sdev_lun}" != "${ipl_lun}" ]; then
log_note "Changed re-IPL path to: ${sdev_busid}:${sdev_wwpn}:${sdev_lun}."
fi
# Try to update the information in the ID file if we have gotten the
if '@DEBUG@'; then hexdump -vC "${ID_FILE}" 1>&2; fi
fi
return 0
}
declare -g SDEV=""
if [[ "${DM_UUID}" == mpath-* ]]; then
# Assume Multipath Device Mapper Device;
# e.g.: DEVPATH = /devices/virtual/block/dm-0
for sdev in /sys/"${DEVPATH}"/slaves/sd*/device; do
if sdev_test_path_state "${sdev}"; then
SDEV="${sdev}"
break
fi
done
# No path of the multipath-device that represents the IPL volume is
# online.
if [ "${SDEV}" = "" ]; then
log_crit "The re-IPL device cannot be changed because no operational path to the re-IPL volume remains. The next re-IPL might fail unless you re-attach or enable at least one valid path to the re-IPL volume."
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
