Since mdevctl commit acf78c1ff6c9 it is now possible for the
get-attributes event to occur between a pre-define and post-define.
This is done in order to obtain the active attributes for the device
before writing them to the config file, and implies that the
get-attributes cannot re-obtain the file lock. For other cases
where mdevctl calls get-attributes, the file lock is not already
held and must be obtained by ap-check before reading attributes from
active devices.
To solve this, let's use the knowledge that mdevctl is a single-threaded
tool and add a test to detect this scenario. If the file lock is
already held by the parent during a get-attributes, don't attempt to
re-acquire it.
Reported-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
Move DF_S390_DUMPER_SIZE_* constants from zdump/df_s390.h to
boot/loaders_layout.h since ccw dumper size depends on the zipl boot
loader layout (to keep it all in one place).
Rename DF_S390_DUMPER_SIZE_* constants to STAGE2_DUMPER_SIZE_*
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>
Add --no-compress option to explicitly omit compression for single-volume
DASD dumper. Used primarily for test purposes.
Since only the lowest byte of mvdump_force field (struct
stage2dump_parm_tail) has been used, split it in two byte fields and use
one for the new no_compress attribute.
Update zipl help and zipl man page with the new parameter info.
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>
Move struct stage2dump_parm_tail from stage2dump.h in to
include/dump/s390_dump.h
Pass the entire stage2dump_parm_tail structure to install_dump_* functions
instead of individual parameters.
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>
Use a separate linker script eckd2dump_sv.lds for single volume dumper with
compression support.
The new dump tool with zlib compression support now has a size of 8 pages.
Since DASD stand alone dump requires a block size of 4K, we are not
affected by the stage 2 size limitations and can load the dumper to
stage 2 as before. We just need to move the HEAP section for ECKD dumper
in the layout definitions up to 0xb000 address. Also expand the stack by
unused 0x400 bytes.
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>
Integrate zlib DFLTCC deflate compression to single volume dasd dumper
using the existing s390 extended dump format. Compression takes place
only if DFLTCC facility is available, otherwise dump is written
uncompressed as before.
First megabyte of memory is always written uncompressed and afterwards
this area is used for zlib workspace and for the compression output buffer.
The compression takes place in chunks of data of equal size (currently 1MB)
and the offset of each compressed chunk is stored in the dump segment
header. Since existing dump segment headers of 1 page size are used, we
need to limit the maximum size of compressed dump segments.
Chunk is written uncompressed in case of compression error or if
deflate compression only makes it bigger.
Thus every chunk of data is compressed separately and can be decompressed
independently. The main reason for that is to enable zgetdump to make fast
read seeks. Otherwise, zgetdump would need to decompress a big dump segment
in the worst case to extract a single piece of data.
Put compression related functions and structures to eckd2dump_zlib.c
and eckd2dump_zlib.h
Update zipl man page with the general information of zlib compression
support.
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>
Instead of using its own DF_S390_ constants and df_s390_ structs
in df_s390.h, include those from "dump/s390_dump.h" in order to minimize
duplicates. Adjust the code, where required, to use <stdint> types
instead of those defined in zt_common.h (e.g. use uint64_t instead of u64).
Adjust zdump include statements.
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>
Move common dump related structures and constants to the global header
"dump/s390_dump.h" in order to get rid of many duplicates in zgetdump code.
Adjust zipl include statements and update Copyright statements.
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>
- Instead of always clearing the memory on reipl after the ccw dump has
been taken, check for the special OS_INFO_FLAG_REIPL_CLEAR flag in
os_info flags entry (indicates if sysfs 'clear' attribute has been set
on the panicked system) and trigger diag308 with a proper subcode.
- Get rid of superfluous ipib_info structure in stage2dump.c.
- Collect ipl_pbt constants in boot/ipl.h header.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Acked-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add DIAG308_LOAD_NORMAL diag308 subcode (used by FCP/NVMe normal ipl).
Rename other diag308_subcode and diag308_rc constants to be in sync with
kernel naming.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Introduce new os_info flags entry pointing to the field with bit flags.
The flag OS_INFO_FLAG_REIPL_CLEAR indicates that 'clear' sysfs attribute
has been set on a panicked system.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
- Add os_info_check() function to verify os_info address, magic and
checksum.
- Add os_info_entry_is_valid() function to check whether requested entry
is present and valid.
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Acked-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
- Move zipl/boot/error.h to include/boot
- Adjust include statements in zipb/boot and genprotimg/boot
- Remove error.h from tunedasd/src/tunedasd.c as not needed
- Fix tunedasd/src/Makefile
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
- Move struct os_info from kdump.h to the new header os_info.h.
- Place os_info.h to include/boot in order to use it in zgetdump
code as well.
- Replace hardcoded value of OS_INFO_CSUM_SIZE with a properly
calculated one.
- Rename os_info_check() of kdump.h to kdump_os_info_check().
Signed-off-by: Mikhail Zaslonko <zaslonko@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
`load_wait_psw` is defined in include/boot/s390.h but is implemented in
zipl/boot/libc.c. This results in a dangling definition for s390.h users
other than zipl, due to the missing implementation. Since the function
is only relevant for zipl, move the definition to the corresponding
header file zipl/boot/libc.h
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
include/boot/s390.h relies on s390 specific kernel headers. However,
some tools used this header for non-s390 code (outside of boot-loaders).
To solve this the PSW and PAGE_SIZE definitions are now in separate
header files. All includes for s390.h which are not in boot-loader code
are replaced with one of psw.h or page.h.
This fixes the compilation failure on ppc64le due to conflicting types
for `__vector128`.
Fixes: https://github.com/ibm-s390-linux/s390-tools/issues/151
Reviewed-by: Jan Höppner <hoeppner@linux.ibm.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
Make the listed changes which are needed to re-use some definitions
by the new zipl-debug(8) tool introduced by the next patch in the
series:
. Add definitions to boot_defs.h:
. a named union disk_bloclkptr (instead of the anonymous one);
. a named structure disk_program_table;
. Remove a 'typedef union disk_blockptr_t' from boot/stage2.h
(cleanup), use the named union in boot_defs.h instead;
. Move the definition of scsi_layout types and the function
get_scsi_layout() from install.c to install.h;
. Move definition of PROGRAM_TABLE_BLOCK_SIZE from bootmap.c to
bootmap.h
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>
In preparation of decoupling the program table position from
the block pointer format, rename:
PROGRAM_TABLE_0 to LEGACY_BLKPTR_FORMAT_ID
PRAGRAM_TABLE_1 to BLKPTR_FORMAT_ID
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Acked-by: Stefan Haberland <sth@linux.ibm.com>
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
Use linker script for all bootloaders instead of having the load
addresses of the bootloaders hard-coded in the Makefile. This allows us
to simplify the Makefile and it fixes the dependency tracking for the
bootloaders *0.bin, *1.bin, and *1b.bin.
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 macro in the next patch. In order to match
the naming scheme rename `STAGE1B_LOAD_ADDR` to `STAGE1B_LOAD_ADDRESS`.
Acked-by: Eduard Shishkin <edward6@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>
Enforce that the first glib.h include is done via glib-helper.h for libpv
so that glib version checks are in place.
Change zdump and pvattest such that they never include glibstuff before
libpv/glib-helper.h
Reviewed-by: Jan Hoeppner <hoeppner@linux.ibm.com>
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
The struct scsi_dump_sb is defined with the __packed attribute and
produces the following compile warning in zfcpdump:
zfcpdump_part.c: In function ‘csum_update’:
zfcpdump_part.c:125:41: warning: taking address of packed member of ‘struct scsi_dump_sb’ may result in an unaligned pointer value [-Waddress-of-packed-member]
125 | dump_sb.csum_size, &dump_sb.csum)) {
| ^~~~~~~~~~~~~
However, all members are of the same size and the __packed attribute is
unnecessary. Remove the attribute to get rid of the warning.
Add assertion to ensure correct size of the struct during compile time.
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
Make zipl tool prepare ECKD DASD for booting by different IPL
programs (with the same boot record installed).
Besides standard CCW-type IPL, user gets an ability to trigger
(with the same boot record installed!) List-Directed IPL. This
allows to use the feature of secure boot from ECKD DASD (which
is not available for CCW-type IPL).
When using the old boot interfaces, the usual CCW-type IPL is
triggered for DASD. Also for compatibility reasons zipl(8) tool
is modified to create and install one, or two "similar" program
tables per boot partition, depending on job and disk type. The
"similar" program tables differ only in block pointers format.
The old IPL programs (CCW-type IPL) use program table based on the
old format.
All program tables are packed to the same bootmap file. Their
order and logical offsets in the file are not significant (not
used by anyone).
The picture below shows which program table is used for IPL of
specified type from disk of specified type. Here "0" and "1" are
identifiers of program tables based on the old and new block
pointers format respectively. E.g. program table "0" is used for
CCW-type IPL from ECKD DASD. LD-IPL from DASD FBA is unsupported
(respectively, only one program table "0" is used), etc.
CCW-IPL LD-IPL
SCSI X 0
DASD FBA 0 X
ECKD DASD LDL 0 X
ECKD DASD CDL 0 1
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>
Add an option to tunedasd to trigger a copy pair swap using the appropriate
ioctl for DASD devices.
-s, --copy-pair-swap COPY_PAIR
This command requires a comma separated pair of primary,secondary to be
specified. In case of success the old secondary will become the new primary
device and the old primary will become a secondary device.
Example:
tunedasd /dev/dasda -s 0.0.9700,0.0.9740
This will set the old secondary device 0.0.9740 as the new primary.
The old primary device 0.0.9700 will automatically become a secondary
device.
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>
Implement simple file-locking routines that use process PIDs for stale
lock detection. The implementation is meant to be a simplified subset of
what liblockfile was previously being used for by libap, allowing the
external dependency to be removed.
This initial implementation provides a series of functions that allow for
creating/release file locks using either the current process PID or the
PID of the current process parent. When creating a file lock, first a
temporary file is created and the appropriate PID (either this process
PID or the parent process PID) is placed in the file to specify the owner
of the lock. Then an attempt is made to link that file to the desired
file location; if this succeeds, the lock is now held on behalf of the
specified PID. If it fails, this implies the file already exists
(meaning the lock is already held). In this case, stale lock detection
is performed by reading the PID from the file and ensuring that the
associated process still exists -- if it does not, then the lock is
presumed stale and destroyed. If the process still exists, then either
the lock request fails or the caller will sleep and retry, depending on
an optional retry setting.
A lock remains valid until either 1) it is released via the corresponding
util_lockfile function, which will delete the corresponding file 2) the
associated PID no longer exists, which leaves the file in-place but will
cause it to be destroyed the next time a different process attempts to
lock that file or 3) the file is directly removed (e.g. rm).
A typical usecase for such support would be to provide a means for
multiple invocations of the same (or different) tools to ensure that they
do not access the same shared resource simultaneously. For example,
ap-check, chzdev and lszdev all have a need to view and/or modify the AP
and vfio-ap configuration files; util_lockfile can be used to ensure that
only one instance of any of these utilities do that at a time by ensuring
they all use the same lockfile.
Additionally, providing the ability to specify the parent PID rather than
the current PID allows for a general purpose tool (like mdevctl) to
invoke a sub-program (ap-check) to acquire and release a lockfile as
necssary while allowing stale lock detection to be controlled by that
parent PID, allowing the lock to remain held over multiple sub-program
invocations.
Note that this implementation is sufficient for our current usage (e.g.
lockfiles placed in tmpfs) but does not take into consideration things
like NFS, which a more complete lockfile solution like liblockfile does.
GitHub-ID: https://github.com/ibm-s390-linux/s390-tools/issues/142
Suggested-by: Luca BRUNO <luca.bruno@coreos.com>
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@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 have
introduced Protected Virtualization (PV) dump support - also named
confidential dump support.
The confidential dump support works by integrating the firmware into the
dump process. New Ultravisor calls (UVC) 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 (CCK) for the
dumped guest. The output format is very similar the normal s390 vmcore
ELF format, it's only enriched by new sections where the returned data
from the UVC "Complete Configuration Dump" and the UVC "Dump
Configuration Storage State" is stored. The encrypted CPU data is stored
in a new note type `NT_S390_PV_CPU_DATA`. The old note types do still
exists but without any confidential data stored. The memory data is
stored in the LOAD segment as usual but for PV dumps it's fully AES-XTS
encrypted.
This commit adds support for reading/decrypting PV guest dumps to
zgetdump by introducing a new DFI input module (`dfi_pv_elf.c`). For
specifying the customer communication key a new command line option
`--key` is added.
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>
It's often useful to know how much data was actually copied, therefore let's
introduce an nullable parameter `@copied` to `pv_gbytes_memcpy`.
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>
Use `PV_CERT_ERROR` error domain whenever useful. Currently, it's only used
internally in libpv and therefore there is no problem in changing it.
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>
The kernel already provides the same macro as does the zt_common.h file,
thus check for its availability first.
CC zconf/qeth/lsqeth.o
In file included from ../../include/lib/util_base.h:17,
from lsqeth.c:30:
../../include/lib/zt_common.h:27: warning: "_AC" redefined
27 | #define _AC(X, TYPE) X##TYPE
|
In file included from /usr/include/linux/ethtool.h:17,
from lsqeth.c:12:
/usr/include/linux/const.h:21: note: this is the location of the previous definition
21 | #define _AC(X,Y) __AC(X,Y)
|
GitHub-ID: https://github.com/ibm-s390-linux/s390-tools/pull/137
Signed-off-by: Dan Horák <dan@danny.cz>
Signed-off-by: Jan Höppner <hoeppner@linux.ibm.com>
Add a check to detect Processor Activity Instrumentation (PAI)
Extension 1 facility unit.
Signed-off-by: Thomas Richter <tmricht@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>
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>
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>
