Luca Boccassi [Sun, 5 Dec 2021 13:42:38 +0000 (13:42 +0000)]
meson: remove openssl dependency from repart
No longer needed since
ade99252e2cdd9eeff78566789008996d27e4dc0
Frantisek Sumsal [Sat, 4 Dec 2021 11:11:31 +0000 (12:11 +0100)]
tree-wide: check for NULLs in more places
Fixes issues pointed out by the `cpp/inconsistent-null-check` LGTM
query.
Luca Boccassi [Sun, 5 Dec 2021 14:10:09 +0000 (14:10 +0000)]
NEWS: add more entries for v250
Yu Watanabe [Sat, 4 Dec 2021 20:24:01 +0000 (05:24 +0900)]
Merge pull request #21563 from yuwata/network-IPoIB-support
network: IPoIB support
Yu Watanabe [Sat, 4 Dec 2021 06:27:53 +0000 (15:27 +0900)]
sd-netlink: fix implicit cast to boolean
Yu Watanabe [Mon, 8 Nov 2021 03:19:56 +0000 (12:19 +0900)]
network: set MTU after IPoIB configs are applied
MTU is updated when IB mode is changed.
Yu Watanabe [Mon, 8 Nov 2021 00:41:51 +0000 (09:41 +0900)]
network: add support to configure IPoIB interfaces
Yu Watanabe [Sun, 7 Nov 2021 21:34:43 +0000 (06:34 +0900)]
network/netdev: add support to create IPoIB subinterface
Yu Watanabe [Sun, 7 Nov 2021 21:43:17 +0000 (06:43 +0900)]
sd-netlink: add support for IPoIB
Yu Watanabe [Tue, 9 Nov 2021 06:51:58 +0000 (15:51 +0900)]
network/netdev: generate persistent MAC address when creating netdev interface
Preparation for later commits.
Frantisek Sumsal [Fri, 3 Dec 2021 16:07:34 +0000 (17:07 +0100)]
ci: consider `cryptolib` in the group identifier
otherwise we end up with more than one job with the same identifier in
one run, causing some of them to get cancelled unexpectedly.
A quick follow-up to
85bd394df57fe45c2873605e2c1d1d79e83e853d.
Luca Boccassi [Fri, 3 Dec 2021 18:37:33 +0000 (18:37 +0000)]
Merge pull request #21607 from mrc0mmand/ci-install-libbpf
ci: run build test with BPF-related stuff as well
Luca Boccassi [Fri, 3 Dec 2021 18:25:19 +0000 (18:25 +0000)]
Merge pull request #21582 from mrc0mmand/lgtm-uninitialized
lgtm: enable more queries
Frantisek Sumsal [Fri, 3 Dec 2021 15:30:56 +0000 (16:30 +0100)]
ci: install libbpf
Frantisek Sumsal [Fri, 3 Dec 2021 06:37:32 +0000 (07:37 +0100)]
analyze: fix build with -Db_ndebug=true
Frantisek Sumsal [Thu, 2 Dec 2021 20:49:02 +0000 (21:49 +0100)]
meson: support versioned llvm binaries in BPF detection
Franck Bui [Fri, 3 Dec 2021 10:23:36 +0000 (11:23 +0100)]
Bump the max number of inodes for /dev to 128k
Follow-up for
7d85383edbab73274dc81cc888d884bb01070bc2.
Apparently the previous limit set on the max number of inodes for /dev was too
small as a system with 4096 LUNs attached can consume up to 95k inodes for
symlinks:
# /bin/df -i
Filesystem Inodes IUsed IFree IUse% Mounted on
devtmpfs
49274377 95075
49179302 1% /dev
Hence this patch bumps the limit from 64k to 128k although the new limit is
still pretty arbitrary (that said, not sure if it really makes sense to put
such absolute limit number).
Zbigniew Jędrzejewski-Szmek [Fri, 3 Dec 2021 12:44:53 +0000 (13:44 +0100)]
Merge pull request #21170 from keszybz/delibgcryptify
Allow systemd-resolved and systemd-importd to use libgcrypt or libopenssl
Gibeom Gwon [Fri, 3 Dec 2021 06:10:50 +0000 (15:10 +0900)]
cryptenroll: fix wrong error messages
PKCS#11 -> FIDO2 in cryptenroll-fido2.c
Zbigniew Jędrzejewski-Szmek [Fri, 3 Dec 2021 07:11:31 +0000 (08:11 +0100)]
Merge pull request #21599 from loongarch64/dev-syscalls
Add LoongArch 64bit syscalls
Frantisek Sumsal [Thu, 2 Dec 2021 17:42:31 +0000 (18:42 +0100)]
meson: correctly display enabled features
In
9cf75222f20 the conf.get() statements for `bpf-framework` and
`valgrind` were dropped, which causes the respective features to always
show as disabled (since they don't follow the "standard" naming scheme
with HAVE_/ENABLE_ prefixes).
Frantisek Sumsal [Tue, 30 Nov 2021 22:40:28 +0000 (23:40 +0100)]
lgtm: enable more (and potentially useful) queries
Not all available queries on LGTM are enabled by default, but some of
the excluded ones might come in handy, hence let's enable them
explicitly.
Frantisek Sumsal [Thu, 2 Dec 2021 15:55:17 +0000 (16:55 +0100)]
lgtm: don't treat the custom note as a list of tags
Just a cosmetic change.
Frantisek Sumsal [Wed, 1 Dec 2021 20:31:43 +0000 (21:31 +0100)]
lgtm: detect uninitialized variables using the __cleanup__ attribute
This is a slightly modified version of the original
`cpp/uninitialized-local` CodeQL query which focuses only on variables
using the cleanup macros. Since this has proven to cause issues in the
past, let's panic on every uninitialized variable using any of the
cleanup macros (as long as they're written using the __cleanup__
attribute).
Some test results from a test I used when writing the query:
```
#define _cleanup_foo_ __attribute__((__cleanup__(foo)))
#define _cleanup_(x) __attribute__((__cleanup__(x)))
static inline void freep(void *p) {
*(void**)p = mfree(*(void**) p);
}
#define _cleanup_free_ _cleanup_(freep)
static inline void foo(char **p) {
if (*p)
*p = free(*p);
}
int main(void) {
__attribute__((__cleanup__(foo))) char *a;
char *b;
_cleanup_foo_ char *c;
char **d;
_cleanup_free_ char *e;
int r;
r = fun(&e);
if (r < 0)
return 1;
puts(a);
puts(b);
puts(c);
puts(*d);
puts(e);
return 0;
}
```
```
+| test.c:23:14:23:14 | e | The variable $@ may not be initialized here, but has a cleanup handler. | test.c:20:26:20:26 | e | e |
+| test.c:27:10:27:10 | a | The variable $@ may not be initialized here, but has a cleanup handler. | test.c:16:45:16:45 | a | a |
+| test.c:29:10:29:10 | c | The variable $@ may not be initialized here, but has a cleanup handler. | test.c:18:25:18:25 | c | c |
```
Luca Boccassi [Wed, 1 Dec 2021 14:16:24 +0000 (14:16 +0000)]
core: support user manager with Condition[Memory/CPU/IO]Pressure
Get the cgroup root path from the current PID, so that when
ran by the user manager we can get to the right path.
Eg: foo.slice:10% will check under:
/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/foo.slice/cpu.pressure
Follow-up for
81513b382b24a7f3602987f71042d075ca27d1a5
Zbigniew Jędrzejewski-Szmek [Thu, 2 Dec 2021 10:29:45 +0000 (11:29 +0100)]
meson: disallow the combination of cryptolib=openssl and dns-over-tls=gnutls
It could work, but it doesn't make much sense. If we already have openssl as
the cryptolib that provides the necessary support, let's not bring in another
library. Disallowing this simplifies things and reduces our support matrix.
Zbigniew Jędrzejewski-Szmek [Wed, 1 Dec 2021 11:50:01 +0000 (12:50 +0100)]
ci: expand the test framework to cover openssl
Yu Watanabe [Thu, 2 Dec 2021 06:53:57 +0000 (15:53 +0900)]
Merge pull request #21584 from yuwata/network-wireguard-cleanups
network: cleanups for wireguard
Yu Watanabe [Thu, 2 Dec 2021 00:53:49 +0000 (09:53 +0900)]
Revert "network: address: drop deprecated temporary address"
This reverts commit
528da64a0c734bd45e953b5c57d82e2a727435ef.
The commit is a bad way to fix #19838, and introduces #21593.
Fixes #21593.
Yu Watanabe [Mon, 29 Nov 2021 15:20:03 +0000 (00:20 +0900)]
network/wireguard: search valid address of the endpoint from all struct addrinfo entries
Yu Watanabe [Mon, 29 Nov 2021 12:21:21 +0000 (21:21 +0900)]
network/wireguard: cleanups for resolving endpoints
This makes
- drop peers_with_unresolved_endpoint and peers_with_failed_endpoint,
- drop destroy handler for sd_resolve_query, and manage each query by peer,
- add random fluctuation to the timeout for retry handler,
- retry timer event source is now managed by peer,
- use sd_event_source_disable_unref().
Yu Watanabe [Mon, 29 Nov 2021 13:07:29 +0000 (22:07 +0900)]
network/wireguard: do not resolve Endpoint= if an IP address is specified
Also verify the domain name and port.
Yu Watanabe [Wed, 1 Dec 2021 23:31:55 +0000 (08:31 +0900)]
parse-util: refuse leading white space in port number
When parse_ip_port() is directly used in a conf parser, then that's
fine, as the rvalue is already truncated.
When parse_ip_port() is used when e.g. parsing IP address with port,
then we should really refuse white space after colon.
Yu Watanabe [Wed, 1 Dec 2021 23:16:23 +0000 (08:16 +0900)]
Merge pull request #21585 from yuwata/network-radv-uplink-interface-auto-with-dhcp6-pd
network: cleanups for uplink interface handling for RADV and DHCP6-PD
Yu Watanabe [Wed, 1 Dec 2021 22:47:17 +0000 (07:47 +0900)]
Merge pull request #21583 from bluca/bpf_assert
cgroup: don't emit BPF firewall warning when manager is in test mode
Luca Boccassi [Wed, 1 Dec 2021 21:20:30 +0000 (21:20 +0000)]
Merge pull request #21591 from yuwata/core-bpf-firewall-unsupported-reason
core/bpf-firewall: make bpf_firewall_supported() always set unsupport…
Luca Boccassi [Wed, 1 Dec 2021 01:04:54 +0000 (01:04 +0000)]
elf-util: do not ignore prctl() errors
We want to avoid loops, so fail and return if we can't disable
core dumping
CID#
1467004
Xiaotian Wu [Wed, 1 Dec 2021 15:53:17 +0000 (23:53 +0800)]
syscalls: run ninja update-syscall-tables
Yu Watanabe [Wed, 1 Dec 2021 15:07:21 +0000 (00:07 +0900)]
tree-wide: fix typo
Xiaotian Wu [Tue, 8 Dec 2020 12:02:13 +0000 (20:02 +0800)]
syscalls: add LoongArch 64bit syscalls
KennthStailey [Wed, 1 Dec 2021 14:43:40 +0000 (09:43 -0500)]
Fixed typo
`ip set dev eth0` should be `ip link set dev eth0`
Yu Watanabe [Wed, 1 Dec 2021 12:07:22 +0000 (21:07 +0900)]
core/bpf-firewall: make bpf_firewall_supported() always set unsupported reason when BPF_FIREWALL_UNSUPPORTED is returned
Otherwise, log_unit_full_errno() in emit_bpf_firewall_warning() will
trigger an assertion.
Yu Watanabe [Wed, 1 Dec 2021 12:37:17 +0000 (21:37 +0900)]
core/cgroup: propagate errors on detecting supported features
Yu Watanabe [Wed, 1 Dec 2021 12:34:37 +0000 (21:34 +0900)]
core/restrict-netif: make restrict_network_interfaces_supported() return negative errno only when critical error
Other errors are handled as the functionality is not supported.
This also drops unnecessary SYNTHETIC_ERRNO().
Luca Boccassi [Tue, 30 Nov 2021 23:49:58 +0000 (23:49 +0000)]
test: run commands with debug level logs in TEST-65-ANALYZE
Luca Boccassi [Tue, 30 Nov 2021 23:49:33 +0000 (23:49 +0000)]
cgroup: don't emit BPF firewall warning when manager is in test mode
Support for BPF might not have been checked, since it's not necessary
in test mode (eg: running offline analysis of units). This causes an
assert:
Assertion '(_error) != 0' failed at src/core/bpf-firewall.c:914, function emit_bpf_firewall_warning(). Aborting.
Export SYSTEMD_LOG_LEVEl=debug in TEST-65-ANALYZE is enough to trigger
this assert while doing an offline analysis of a unit that has some
firewall/network restrictions set.
Skip the warning if the manager is in test mode.
Zbigniew Jędrzejewski-Szmek [Tue, 2 Nov 2021 13:37:19 +0000 (14:37 +0100)]
ci: temporarily set -Wno-deprecated-declarations in Packit
to suppress OpenSSL 3.0 deprecation warnings (until a proper solution is
deployed): RSA_free, EC_KEY_free, RSA_set0_key, RSA_size, EVP_PKEY_assign,
EC_KEY_set_group, and others are deprecated.
Zbigniew Jędrzejewski-Szmek [Tue, 2 Nov 2021 09:45:20 +0000 (10:45 +0100)]
resolved: do not use BN_dup() unnecessarilly
Suggested in https://github.com/systemd/systemd/pull/21170#discussion_r738696794
Zbigniew Jędrzejewski-Szmek [Wed, 27 Oct 2021 13:39:48 +0000 (15:39 +0200)]
port string_hashsum from libgcrypt to openssl^gcrypt
This allows resolved and importd to be built without libgcrypt.
Note that we now say either 'cryptographic library' or 'cryptolib'.
Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
Kevin Kuehler [Fri, 11 Dec 2020 00:08:11 +0000 (16:08 -0800)]
basic/openssl-util: Add sha256 hash wrapper
Kevin Kuehler [Fri, 11 Dec 2020 00:08:30 +0000 (16:08 -0800)]
resolve: Port nsec3 code to openssl^gcrypt
Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
Yu Watanabe [Wed, 1 Dec 2021 11:34:48 +0000 (20:34 +0900)]
test-network: re-arrange DHCP6-PD subnet IDs to test searching free subnet ID
Yu Watanabe [Wed, 1 Dec 2021 06:08:00 +0000 (15:08 +0900)]
test-network: add testcases for uplink interface detection for RADV with DHCPv6-PD
Yu Watanabe [Wed, 1 Dec 2021 10:43:40 +0000 (19:43 +0900)]
network: dhcp6-pd: exclude all explicitly specified subnet IDs when searching free IDs
When the upstream link gained a lease, then several downstream links may
not appear yet. Previously, the explicitly specified subnet ID for a
downstream link which appears later may be already assigned to an
interface which does not request specific subnet ID.
To avoid such situation, this makes all specified IDs are excluded when
searching free IDs.
As a side effect, we can avoid the second call of
dhcp6_pd_distribute_prefix().
Yu Watanabe [Wed, 1 Dec 2021 10:24:22 +0000 (19:24 +0900)]
network: dhcp6-pd: introduce dhcp6_pd_is_uplink()
This also merges dhcp6_pd_resolve_uplink() with dhcp6_pd_find_uplink()
Yu Watanabe [Wed, 1 Dec 2021 05:48:51 +0000 (14:48 +0900)]
network: dhcp6: make UplinkInterface=:self take effect only when DHCPv6PrefixDelegation= is enabled
Yu Watanabe [Wed, 1 Dec 2021 05:37:26 +0000 (14:37 +0900)]
network: radv: use the uplink interface used in DHCPv6-PD
Arvid E. Picciani [Wed, 1 Dec 2021 08:21:04 +0000 (09:21 +0100)]
docs: document that systemd closes console during normal runtime
fixes #21451
Signed-off-by: Arvid E. Picciani <aep@exys.org>
Luca Boccassi [Wed, 17 Nov 2021 10:00:12 +0000 (10:00 +0000)]
core: add Condition[Memory/CPU/IO]Pressure
By default checks PSI on /proc/pressure, and causes a unit to be skipped
if the threshold is above the given configuration for the avg300
measurement.
Also allow to pass a custom timespan, and a particular slice unit to
check under.
Fixes #20139
Yu Watanabe [Wed, 1 Dec 2021 03:54:00 +0000 (12:54 +0900)]
Merge pull request #21581 from keszybz/really-random-fixlets
Really random fixlets
Yu Watanabe [Wed, 1 Dec 2021 03:52:27 +0000 (12:52 +0900)]
Merge pull request #21564 from yuwata/errno-eagain-eintr
tree-wide: introduce ERRNO_IS_TRANSIENT() and use it
Yu Watanabe [Wed, 1 Dec 2021 03:52:10 +0000 (12:52 +0900)]
Merge pull request #21536 from medhefgo/test
test: Introduce TEST_RET macro
Yu Watanabe [Wed, 1 Dec 2021 03:51:48 +0000 (12:51 +0900)]
Merge pull request #21288 from loongarch64/la64/main
Add support for the LoongArch architecture.
Luca Boccassi [Tue, 30 Nov 2021 23:30:31 +0000 (23:30 +0000)]
Merge pull request #21553 from yuwata/network-wireguard-add-routes-to-allowed-ips
network: wireguard: add routes to AllowedIPs=
Kevin Kuehler [Fri, 11 Dec 2020 00:08:28 +0000 (16:08 -0800)]
resolve: Port dnskey verification by ds to openssl^gcrypt
Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
Kevin Kuehler [Fri, 11 Dec 2020 00:08:26 +0000 (16:08 -0800)]
resolve: Port dnssec verify from gcrypt to openssl^gcrypt
Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
Zbigniew Jędrzejewski-Szmek [Thu, 21 Oct 2021 14:00:08 +0000 (16:00 +0200)]
resolved: split out function to hash signature
dnssec_verify_rrset() is just too long.
Zbigniew Jędrzejewski-Szmek [Thu, 28 Oct 2021 12:03:44 +0000 (14:03 +0200)]
resolved: split out function to generate signature
dnssec_verify_rrset() is just too long.
Kevin Kuehler [Fri, 11 Dec 2020 00:08:23 +0000 (16:08 -0800)]
resolve: Add coverage for dnssec ecdsa (rfc6605)
Zbigniew Jędrzejewski-Szmek [Tue, 2 Nov 2021 08:58:04 +0000 (09:58 +0100)]
import: port importd from libgcrypt to openssl^gcrypt
This is heavily based on Kevin Kuehler's work, but the logic is also
significantly changed: instead of a straighforward port to openssl, both
versions of the code are kept, and at compile time we pick one or the other.
The code is purposefully kept "dumb" — the idea is that the libgcrypt codepaths
are only temporary and will be removed after everybody upgrades to openssl 3.
Thus, a separate abstraction layer is not introduced. Instead, very simple
ifdefs are used to select one or the other. If we added an abstraction layer,
we'd have to remove it again afterwards, and it don't think it makes sense to
do that for a temporary solution.
Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
# Conflicts:
# meson.build
Zbigniew Jędrzejewski-Szmek [Tue, 2 Nov 2021 08:44:12 +0000 (09:44 +0100)]
meson: add config setting to select between openssl and gcrypt
This is not pretty, but it is supposed to be only a temporary measure.
Kevin Kuehler [Wed, 20 Oct 2021 10:21:18 +0000 (12:21 +0200)]
repart: port to our home-grown hmac_sha256
This reduces dependencies. The speed of the code here is uimportant, because we
hash only a tiny amount of input data.
Debian and Ubuntu currently build without repart, see
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=976959
> repart requires openssl and so far I tried to avoid linking against
> both gnutls and openssl.
Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
Zbigniew Jędrzejewski-Szmek [Tue, 23 Nov 2021 14:05:58 +0000 (15:05 +0100)]
tmpfiles: 'st' may have been used uninitialized
Zbigniew Jędrzejewski-Szmek [Sun, 28 Nov 2021 11:30:52 +0000 (12:30 +0100)]
man: prettify line number handling in python example
This way the 'line_number' variable contains the actual line number as we think
of it, instead of adjusting it on output.
Zbigniew Jędrzejewski-Szmek [Fri, 26 Nov 2021 09:28:18 +0000 (10:28 +0100)]
test-sizeof: show stack and heap randomization
It's useful to reassure yourself those those things actually work ;)
Zbigniew Jędrzejewski-Szmek [Tue, 30 Nov 2021 20:49:53 +0000 (21:49 +0100)]
Merge pull request #21580 from bluca/core_fork
Switch coredump ELF parsing to fork+dlopen
Yu Watanabe [Mon, 29 Nov 2021 07:49:24 +0000 (16:49 +0900)]
test-network: add tests for creating routes to wireguard's AllowedIPs=
Yu Watanabe [Mon, 29 Nov 2021 07:36:42 +0000 (16:36 +0900)]
network: wireguard: automatically configure routes to addresses specified in AllowedIPs=
Closes #14176.
Yu Watanabe [Mon, 29 Nov 2021 07:35:46 +0000 (16:35 +0900)]
network: route: expose route_hash_ops
Yu Watanabe [Mon, 29 Nov 2021 06:13:03 +0000 (15:13 +0900)]
network/wireguard: drop unnecessary .in6 specifier
This should not change any behavior, as the size of the IPv4 address is
smaller than IPv6.
Yu Watanabe [Mon, 29 Nov 2021 10:38:24 +0000 (19:38 +0900)]
network: route: read RTA_TABLE attribute to get route table
If the table of a route is larger than 255, then the value is stored in
RTA_TABLE attribute. But the attribute is not supported by old kernels.
So, first try to read the value from RTA_TABLE attribute, then fallback
to the value in the message header.
Yu Watanabe [Mon, 29 Nov 2021 10:38:03 +0000 (19:38 +0900)]
network: route: use typesafe function
Zbigniew Jędrzejewski-Szmek [Tue, 30 Nov 2021 17:42:32 +0000 (18:42 +0100)]
oomd: use type suffix instead of casting
The end result is the same.
Yu Watanabe [Tue, 30 Nov 2021 06:52:54 +0000 (15:52 +0900)]
network: address: drop deprecated temporary address
Fixes #19838.
Luca Boccassi [Wed, 24 Nov 2021 20:26:19 +0000 (20:26 +0000)]
TODO: update dlopen todo list
Luca Boccassi [Mon, 22 Nov 2021 11:11:21 +0000 (11:11 +0000)]
elf-util: switch libelf/libdw to dlopen()
In order to avoid inflating the dependency list for the core
library, use dlopen when inspecting elfs, since it's only
used in two non-core executables.
Luca Boccassi [Wed, 24 Nov 2021 20:23:02 +0000 (20:23 +0000)]
meson: remove libdw dependency from pstore
systemd-pstore does not use any symbol from libdw, and never did,
but the dependency was listed since the beginning
Luca Boccassi [Sun, 21 Nov 2021 17:31:20 +0000 (17:31 +0000)]
elf-util: add stub for builds without libdw
Luca Boccassi [Thu, 18 Nov 2021 00:51:41 +0000 (00:51 +0000)]
elf-util: split out helper to parse build-id
Luca Boccassi [Wed, 17 Nov 2021 01:44:33 +0000 (01:44 +0000)]
coredump: move elf parsing utilities to src/shared/elf-util
Luca Boccassi [Wed, 24 Nov 2021 20:13:44 +0000 (20:13 +0000)]
coredump: always log on failure in parse_core()
Luca Boccassi [Wed, 17 Nov 2021 01:27:57 +0000 (01:27 +0000)]
coredump: make the buffer optional when analyzing a core file
Allow later usage when we only want to fetch the JSON packaging metadata
Luca Boccassi [Sun, 21 Nov 2021 17:05:28 +0000 (17:05 +0000)]
coredump: analyze object with libdwelf in forked process
Parsing objects is risky as data could be malformed or malicious,
so avoid doing that from the main systemd-coredump process and
instead fork another process, and set it to avoid generating
core files itself.
Luca Boccassi [Mon, 29 Nov 2021 11:31:00 +0000 (11:31 +0000)]
coredump: use _cleanup_ pattern
Note that c.f needs to be closed _before_ taking or freeing
the buf pointer, as it might be invalidated
Luca Boccassi [Mon, 29 Nov 2021 10:23:49 +0000 (10:23 +0000)]
coredump: flush accumulating buffer after parsing core file
Luca Boccassi [Tue, 30 Nov 2021 02:01:53 +0000 (02:01 +0000)]
coredump: check return of dwarf_getscopes
Zbigniew Jędrzejewski-Szmek [Tue, 30 Nov 2021 15:36:36 +0000 (16:36 +0100)]
Merge pull request #21578 from bluca/json_spec
docs: clarify JSON terms in COREDUMP_PACKAGE_METADATA
Xiaotian Wu [Mon, 19 Apr 2021 13:03:32 +0000 (21:03 +0800)]
gpt: update the gpt test case for LoongArch
Wu Xiaotian [Tue, 1 Sep 2020 04:13:18 +0000 (12:13 +0800)]
architecture: Add support for the LoongArch architecture
Rasmus Villemoes [Tue, 30 Nov 2021 11:50:19 +0000 (12:50 +0100)]
sysusers: avoid creating spurious "nobody" group
On distros using Debian's base-passwd, the name of the group with gid 65534 is
nogroup. Currently, systemd-sysusers creates a spurious "nobody" group
systemd-sysusers[243]: Creating group nobody with gid 996
That's both confusing and redundant, as the nobody user still has primary group
65534 aka nogroup, and the nobody group simply goes completely unused.
So explicitly specify the primary group of the nobody user, and add a line
ensuring that that group exists.
This is not a problem for Debian (or Ubuntu) itself, as they add their own
version of basic.conf in their systemd build logic. But it appears on for
example Yocto/OpenEmbedded.