picnoir/Systemd
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
36315 commits 2 branches 0 tags 166 MiB
Commit graph

22961 commits

Author SHA1 Message Date
Yu Watanabe 1ce7fecb4f udev-rules: make import_parent_into_properties() take sd_device 2018-11-07 13:35:03 +09:00
Yu Watanabe 13c7b75f2f udev-rules: make import_file_into_properties() take sd_device 2018-11-07 13:35:03 +09:00
Yu Watanabe 29b5eb5adf udev-rules: make import_property_from_string() take sd_device 
Also, this makes the function return negative errno, though its
return value is always ignored.
 2018-11-07 13:35:03 +09:00
Yu Watanabe f3d241feb2 udev: use sd_device for udev_event.dev_parent 2018-11-07 13:35:03 +09:00
Yu Watanabe 5ba7e79885 udev-rules: make match_attr() take sd_device instead of udev_device 2018-11-07 13:35:03 +09:00
Yu Watanabe 480ecb7d28 udev: use sd_device for udev_event.dev_db 
Also, this adds sd_device_unref for the object in udev_event_free()
for safety and readability of code.
 2018-11-07 13:35:03 +09:00
Yu Watanabe e0bb2ff94b udev-event: make udev_event_new() take sd_netlink and take a reference of that 2018-11-07 13:35:03 +09:00
Yu Watanabe 0f86dc900e udev-event: make udev_event_new() take exec_delay 2018-11-07 13:35:03 +09:00
Evgeny Vereshchagin 212bd73c78 networkd: make network_load_one "public" and add a fuzzer for it 2018-11-06 19:01:32 +01:00
Evgeny Vereshchagin e27aac11f2 networkd: make netdev_load_one "public" and add a fuzzer for it 2018-11-06 19:01:32 +01:00
Evgeny Vereshchagin 0621f03b13 networkd: remove a weird assertion from network_load_one 
The assertion was added in dbffab87f1 but I'm not exactly sure
why. Now it just crashes a fuzzer I wrote. I, of course, could fix the fuzzer
so that it would generate names ending in ".network" if anyone would tell
me why the assertion should be kept here.
 2018-11-06 19:01:32 +01:00
Evgeny Vereshchagin 3e180a2516 util: initialize _argtypes in VA_FORMAT_ADVANCE when systemd is built with MSan 
This gets around https://github.com/google/sanitizers/issues/992.
 2018-11-06 19:01:32 +01:00
Yu Watanabe 49a060aca8
Merge pull request #10654 from poettering/srand-rdrand 
random-util.c mini-fixes
 2018-11-07 02:03:09 +09:00
Marco Trevisan (Treviño) 6260d28b8a login: Don't mark framebuffer devices as master-of-seat devices 
Currently we consider any framebuffer device as enough to have a
valid graphical session, but this might lead to many false postives
like in the case of framebuffer devices that have a linked drm card
which is still in the process of being added, or for vesa fb, and
so it doesn't ensure us that we can have a proper graphical session.

Since these days we normally don't consider anything without a DRM
card able to provide a full graphical session, let's not set this
at this level.
Drivers which can provide a graphical session with the sole fb are
still free to mark any device as `master-of-seat`

Fixes #10435
 2018-11-06 20:02:26 +03:00
Lennart Poettering 39adc47495
Merge pull request #10658 from yuwata/udevd-worker-performance 
udev: improve performance
 2018-11-06 20:01:37 +03:00
Lennart Poettering 9edcbc4f44
Merge pull request #10602 from yuwata/sd-device-logs 
update log messages
 2018-11-06 19:33:51 +03:00
Lennart Poettering b992109b3e
Merge pull request #10633 from yuwata/sd-resolve-destroy 
Another solution to fix wireguard issues
 2018-11-06 19:30:59 +03:00
Yu Watanabe ed2e7967be time-util: make parse_sec() not accept "12.34.56" 
This also changes the rational number treatment.
So, the limitations introduced by 8079c90333
and f6a178e91d are relaxed.

Fixes #10619.
 2018-11-06 19:24:47 +03:00
Yu Watanabe faae64fa3d udevd: increase default number of workers 
The uevent handling in udevd is not cpu hungry.
So, let's increase the default number of workers per cpu.
This decrease the number of queued uevents.

Without this commit (children_max is 16 on my laptop)
```
$ journalctl -b -u systemd-udevd.service | grep reached | wc -l
1544
```
With this commit (children_max is 30 on my laptop)
```
$ journalctl -b -u systemd-udevd.service | grep reached | wc -l
7
```
 2018-11-07 00:25:20 +09:00
Yu Watanabe eca195ec23 udevd: wait 3 seconds before killing worker processes 
During boot process, many worker processes are forked and killed.
To decrease cycles of forking and killing worker, let's wait
3 seconds before killing workers. If new uevent or inotify event
comes within the delay, the killing porcess will be cancelled.
 2018-11-07 00:25:13 +09:00
Yu Watanabe da14313418 udevd: drop redundant logic of receiving uevent 
If there exists pending uevents, then sd-event invokes uevent handler.
So, it is not necessary to receive the next uevent in inotify event.
 2018-11-06 23:53:55 +09:00
Yu Watanabe b6107f010a udevd: decrease indentation 2018-11-06 23:53:50 +09:00
Lennart Poettering a0ca258adf
Merge pull request #10597 from toanju/fix-networkd-l3-loss 
networkd: don't remove ip address or route
 2018-11-06 17:44:24 +03:00
Yu Watanabe 05e6d9c64b udevd: merge conditions to decrease indentation 2018-11-06 23:21:25 +09:00
Lennart Poettering ed63705975
Merge pull request #10650 from yuwata/udevadm-trigger-use-write-string-file 
udevadm: use write_string_file() helper function
 2018-11-06 16:46:25 +03:00
Tobias Jungel db688b7e55 networkd: cleanup for #10542 
fixes: 53b1f7d
 2018-11-06 16:42:12 +03:00
Lennart Poettering ce9d553d2e
Merge pull request #10652 from yuwata/lldp-ndisc-string-table 
lldp,ndisc: update debug logs
 2018-11-06 16:41:48 +03:00
Giuseppe Scrivano 875622c39e core, sysctl: skip ENOENT for /proc/sys/net/unix/max_dgram_qlen 
sysctl is disabled for /proc mounted from an user namespace thus entries like
/proc/sys/net/unix/max_dgram_qlen do not exist.  In this case, skip the error
and do not try to change the default for the AF_UNIX datagram queue length.
 2018-11-06 16:41:34 +03:00
Lennart Poettering 0e28c86f54 ask-password: fix minor memory leak on error path 
CID 1396557
 2018-11-06 16:41:01 +03:00
Tobias Jungel 7ecf0c3e17 networkd: don't remove route 
In case networkd is restarted this prevents a removal of an already existing
route that would be configured using networkd. With the proposed changes the
route will be kept on the interface without removing. This happens only on
physical hosts or VMs since networkd handles interface configuration slightly
different in containers.
 2018-11-06 13:28:12 +01:00
Tobias Jungel 30226d2718 networkd: don't remove ip address 
In case networkd is restarted this prevents a removal of an already existing IP
address that would be configured using networkd. With the proposed changes the
IP address will be kept on the interface without removing. This happens only on
physical hosts or VMs since networkd handles interface configuration slightly
different in containers.
 2018-11-06 13:26:37 +01:00
Yu Watanabe 7cadbe09e9 sd-device: use write_string_file() helper to write sysattr 2018-11-06 21:24:03 +09:00
Yu Watanabe 835d18ba20 fileio: introduce WRITE_STRING_FILE_NOFOLLOW flag for write_string_file() and friends 2018-11-06 21:24:03 +09:00
Yu Watanabe 57512c893e tree-wide: set WRITE_STRING_FILE_DISABLE_BUFFER flag when we write files under /proc or /sys 2018-11-06 21:24:03 +09:00
Yu Watanabe 76cdddfb9d udevd: use set_oom_score_adjust() to set OOM score 2018-11-06 21:24:03 +09:00
Yu Watanabe 92c40e1dc8 udevadm: use write_string_file() helper function 2018-11-06 21:24:03 +09:00
Yu Watanabe 90208b8217 test: add tests for string tables for lldp and ndisc events 2018-11-06 20:15:09 +09:00
Yu Watanabe a2dcda328a ndisc: improve debug log message 2018-11-06 20:15:09 +09:00
Yu Watanabe 4f0e4d29b0 lldp: improve debug log message 2018-11-06 20:15:09 +09:00
Lennart Poettering 92025e8f52 random-util: initialize srand() from RDRAND 
It's cheap to get RDRAND and given that srand() is anyway not really
useful for trusted randomness let's use RDRAND for it, after all we have
all the hard work for that already in place.
 2018-11-06 12:12:18 +01:00
Lennart Poettering 54bf23151f random-util: we don't intend to write to auxv, hence make it const 2018-11-06 12:12:18 +01:00
Yu Watanabe 6bee206591 udev-node: add one more debug log when failed to create symlink 2018-11-06 15:45:22 +09:00
Yu Watanabe a126a38a9f udev-node: drop redundant log message 
As node_symlink() logs almost same message.
 2018-11-06 15:45:22 +09:00
Yu Watanabe 6174a243f9 udev: setting the same log level to SYSTEMD realm 
Otherwise, many debug logs from libsystemd are dropped.
 2018-11-06 15:45:22 +09:00
Yu Watanabe a5db6debcd test: check returned value of sd_device_get_usec_since_initialized() 2018-11-06 15:45:22 +09:00
Yu Watanabe 754d6c26b4 test: show is_initialized flag in log message of test-sd-device 
And drop devpath, as it is redundant for most cases.
 2018-11-06 15:45:22 +09:00
Yu Watanabe ecbe9873a9 test: show name of testing functions in test-sd-device 2018-11-06 15:45:22 +09:00
Yu Watanabe c7d54daef9 sd-device: normalize debug messages 2018-11-06 15:45:22 +09:00
Yu Watanabe 38d1555d7c sd-device: drop subsystem from log_device_*() 
The sd_device object always has syspath and sysname, but subsytem may not.
Also, it may take some costs to get subsystem.
So, let's drop subsystem from logs.
 2018-11-06 15:45:22 +09:00
Yu Watanabe cc0bf5e148 udev-watch: adjust level of log messages 2018-11-06 11:15:39 +09:00
Yu Watanabe 7fe3324c5e udev-watch: make udev_watch_lookup() return 1 when device found 2018-11-06 11:15:39 +09:00
Yu Watanabe b7759e0403 udev: drop redundant log message and fix returned error code 2018-11-06 11:15:39 +09:00
Lennart Poettering 8912a99cea
Merge pull request #10606 from yuwata/udev-tests 
test: mask hwdb-update.services in most tests
 2018-11-05 23:14:42 +03:00
Lennart Poettering a54e373163
Merge pull request #10618 from yuwata/fix-10615 
network: fix several issues in config parser
 2018-11-05 17:37:25 +03:00
Yu Watanabe 3d8d1f13d5 test: add tests for config_parse_address() 2018-11-05 16:46:26 +09:00
Yu Watanabe b7cb445235 network: use in_addr_default_prefix_from_string_auto() in config_parse_address() 2018-11-05 16:46:26 +09:00
Yu Watanabe c1d58204b2 test: add tests for in_addr_default_prefix_from_string() or friend 2018-11-05 16:46:26 +09:00
Yu Watanabe a4798d4e6d util: introduce in_addr_default_prefix_from_string() and friends 
It is similar to in_addr_prefix_from_string() but it determines
the prefix length from address if it is not specified.
 2018-11-05 16:46:26 +09:00
Yu Watanabe 67944f5c26 util: use strndup() instead of strndupa() 
As the input string may be comes from command line or config files.
 2018-11-05 16:46:00 +09:00
Yu Watanabe a7cc45caad network: sort headers in wireguard.h 2018-11-05 13:19:02 +09:00
Yu Watanabe fc72155321 network: use structured initializers in wireguard.c 2018-11-05 13:19:02 +09:00
Yu Watanabe c195364da5 network: fix crash in wireguard_done() 
Fixes another issue in #10629.
 2018-11-05 13:19:02 +09:00
Yu Watanabe 56ba90c2df network: use destroy callback to unref netdev attached to event source 2018-11-05 13:19:02 +09:00
Yu Watanabe 8173d1d0ec network: use destroy callback to clear resolved wireguard endpoints 2018-11-05 13:19:02 +09:00
Yu Watanabe c4397d94c3 network: link_drop() and netdev_drop() remove reference from manager 2018-11-05 13:19:02 +09:00
Yu Watanabe 2382c9367b sd-event: add sd_event_source_{get,set}_floating() 2018-11-05 13:19:02 +09:00
Yu Watanabe b3ae7237c6 sd-resolve: introduce sd_resolve_query_{get,set}_floating() 2018-11-05 13:19:02 +09:00
Yu Watanabe a8319dea1d sd-resolve: introduce sd_resolve_query_{get,set}_destroy_callback() 2018-11-05 13:19:02 +09:00
Yu Watanabe 383bb2bc1a vlan-util: add assertions to parse_vlanid() 2018-11-04 00:31:46 +09:00
Yu Watanabe 2e5da68b7d network: add missing .sections element for vcan and vrf 
Fixes #10629.
 2018-11-04 00:31:46 +09:00
Yu Watanabe de4224aa22 network: fix memleak abot Address.label 
Also fix possible memleak about Address.section.
Fixes #10628.
 2018-11-04 00:31:46 +09:00
Yu Watanabe 7946d57f18 network: drop error cause in log message 
Fixes for 21486d9e99.
 2018-11-04 00:31:46 +09:00
Yu Watanabe ecee0abe77 network: fix memleak in static prefix 
The fix by 744faf5e4e is not perfect,
as the section leaks if `hashmap_put()` in `prefix_new_static()` fails.
 2018-11-04 00:31:46 +09:00
Yu Watanabe 6f1a96ded4 netdev: also support GENEVE.UDP6ZeroChecksum{Tx,Rx}= 
After 53c06862c1, we prefer Checksum
instead of CheckSum. Let's do so for GENEVE netdev.
 2018-11-04 00:31:46 +09:00
Yu Watanabe fd3005f9b8 network: fix worng offset for .network parser 2018-11-04 00:31:46 +09:00
Yu Watanabe 899f0d259d network: fix memleak in config_parse_hwaddr() 
Fixes #10615.
 2018-11-04 00:31:46 +09:00
Yu Watanabe 44386b449b network: improve readability of config_parse_ifalias() 2018-11-04 00:31:46 +09:00
Yu Watanabe 25ed70f76f network: fix possible memleak caused by multiple setting of Bridge=, Bond= or VRF= 2018-11-04 00:31:46 +09:00
Yu Watanabe 5f74e49e85 test: add test for radv_prefix_delegation_{from,to}_string() 2018-11-04 00:31:46 +09:00
Yu Watanabe 6b1dec6696 network: introduce radv_prefix_delegation_{from,to}_string() 2018-11-04 00:31:46 +09:00
Yu Watanabe 8a4871c7d1 network: fix indentation 2018-11-04 00:31:46 +09:00
Yu Watanabe 8627d1120a network: invert the order in comparison 2018-11-04 00:31:46 +09:00
Yu Watanabe 1176b054e1 network: shorten code a bit 2018-11-04 00:31:06 +09:00
Yu Watanabe 76281c6408 test: also run compile tests for libudev.h 2018-11-03 18:00:47 +09:00
Franck Bui 1d88bce2a0 coredump: only install coredump.conf when ENABLED_COREDUMP=true 2018-11-02 14:31:12 +01:00
Tobias Jungel 99f68ef02d networkd: add missing bonding options (#10542) 
Add support for bonding options system prio, port key and actor system mac.

These options exist in the linux kernel since 4.2
(torvalds/linux@171a42c38c)

Details:
https://www.kernel.org/doc/Documentation/networking/bonding.txt
 2018-11-02 10:31:20 +09:00
Yu Watanabe cb700a11d9 network: drop white space in log message 2018-11-01 23:41:29 +09:00
Yu Watanabe 0d7febd002 network: fix hash function for routing policy rule 2018-11-01 23:40:56 +09:00
Yu Watanabe e6b65ab760 network: fix return value of routing_policy_rule_get() 
To distinguish source of rule.
 2018-11-01 23:38:11 +09:00
Yu Watanabe a6276f0f3f network: fix memleak in routing policy 
As set_put() may fail if input is duplicated.
 2018-11-01 23:24:20 +09:00
Yu Watanabe 744faf5e4e network: fix memleak in static prefix 
Fixes #10608.
 2018-11-01 21:46:38 +09:00
Yu Watanabe f19ee68140 network: fix segfault in manager_free() 
Fixes #10605.
 2018-11-01 20:26:02 +09:00
Lennart Poettering 6619ad889d nspawn: beef up netns checking a bit, for compat with old kernels 
Fixes: #10544
 2018-10-31 21:42:45 +03:00
Xiang Fan c7b7d74e81 ask-password: check keyring in ask_password_tty and ask_password_agent 
A race condition happens when calling ask_password_auto() multiple times
to unlock several disks on boot and effectively no password caching is
utilized. This patch fixes it by polling the cache when waiting for
the password.
 2018-10-31 18:26:58 +01:00
Lennart Poettering 45313bd921
Merge pull request #10010 from msekletar/cryptsetup-generator-keydev-followups 
cryptsetup-generator: keydev support - followups
 2018-10-31 18:26:25 +01:00
Lennart Poettering e2d39e549f nspawn: add proper error message if setns() on network namespace fd fails 
Addresses: https://github.com/systemd/systemd/pull/10589#issuecomment-434670595
 2018-10-31 18:07:30 +01:00
Lennart Poettering 1ad6e8b302 core: split environment block mantained by PID 1's Manager object in two 
This splits the "environment" field of Manager into two:
transient_environment and client_environment. The former is generated
from configuration file, kernel cmdline, environment generators. The
latter is the one the user can control with "systemctl set-environment"
and similar.

Both sets are merged transparently whenever needed. Separating the two
sets has the benefit that we can safely flush out the former while
keeping the latter during daemon reload cycles, so that env var settings
from env generators or configuration files do not accumulate, but
dynamic API changes are kept around.

Note that this change is not entirely transparent to users: if the user
first uses "set-environment" to override a transient variable, and then
uses "unset-environment" to unset it again things will revert to the
original transient variable now, while previously the variable was fully
removed. This change in behaviour should not matter too much though I
figure.

Fixes: #9972
 2018-10-31 18:00:53 +01:00
Lennart Poettering bea1a01310 strv: wrap strv_new() in a macro so that NULL sentinel is implicit 2018-10-31 18:00:52 +01:00
Lennart Poettering fc03c53653
Merge pull request #10573 from faheel/master 
Convert remaining docs to Markdown
 2018-10-31 15:15:41 +01:00
Lennart Poettering 0aeff581cc
Merge pull request #10556 from yuwata/udev-event-cleanup-2 
udev-event: several fixes for previous PR
 2018-10-31 12:10:12 +01:00