Lennart Poettering [Tue, 9 Mar 2021 21:03:00 +0000 (22:03 +0100)]
dissect-image: clean up meaning of DISSECT_IMAGE_MKDIR
Previously handling of DISSECT_IMAGE_MKDIR was pretty weird and broken:
it would control both if we create the top-level mount point when
mounting an image, and the inner mount points for images that consist of
multiple file systems. However, the latter is redundant, since
1f0f82f1311e4c52152b8e2b6f266258709c137d does this too, a few lines
further up – unconditionally!
Hence, let's make the meaning of DISSECT_IMAGE_MKDIR more strict: it
shall be only about the top-level mount point, not about the inner ones
(where we'll continue to create what is missing alwayway). Having a
separate flag for the top-level mount point is relevant, since the mount
point dir created by it will remain on the host fs – unlike the
directories we create inside the image, which will stay within the
image.
This slightly change of meaning is actually inline with what the flag is
actually used for and documented in systemd-dissect.
(cherry picked from commit
9842905ede9c7fdc541724ee5c6db7d46a47405d)
Michal Sekletar [Tue, 9 Mar 2021 16:22:32 +0000 (17:22 +0100)]
install: refactor find_symlinks() and don't search for symlinks recursively
After all we are only interested in symlinks either in top-level config
directory or in .wants and .requires sub-directories.
As a bonus this should speed up ListUnitFiles() roughly 3-4x on systems
with a lot of units that use drop-ins (e.g. SSH jump hosts with a lot of
user session scopes).
(cherry picked from commit
43b4e3058c106e663bbd5413e7bd106e55d6fd79)
Lennart Poettering [Wed, 10 Mar 2021 20:54:59 +0000 (21:54 +0100)]
coredumpctl: fflush() stdout before invoking gdb
Fixes: #18936
(cherry picked from commit
48f813c4aab307695b853cbd3ea3c85f22486898)
Frantisek Sumsal [Wed, 10 Mar 2021 15:41:35 +0000 (16:41 +0100)]
coredump: omit coredump info when -q is used with the `debug` verb
Skip printing the coredump info table when using the `debug` verb in
combination with the `-q/--quiet` option. Useful when trying to gather
coredump info non-interactively via scripted gdb commands.
Fixes: systemd/systemd#18935
(cherry picked from commit
a174da59c299f6197425707f23757dc9149e5fba)
Zbigniew Jędrzejewski-Szmek [Wed, 10 Mar 2021 09:17:23 +0000 (10:17 +0100)]
Revert "udev: do not execute hwdb builtin import twice or thrice"
This reverts commit
876c75fe870846b09b54423a6b719d80bc879b27.
The patch seems to cause usb devices to get some attributes set from the parent
PCI device. 'hwdb' builtin has support for breaking iteration upwards on usb
devices. But when '--subsystem=foo' is specified, iteration is continued. I'm
sure it *could* be figured out, but it seems hard to get all the combinations
correct. So let's revert to functional status quo ante, even if does the lookup
more than once unnecessarily.
Fixes #18125.
(cherry picked from commit
451ba55fecd8b494add2001b3ca3c1915c8fd655)
Lennart Poettering [Tue, 9 Mar 2021 19:43:02 +0000 (20:43 +0100)]
socket-util: refuse ifnames with embedded '%' as invalid
So Linux has this (insane — in my opinion) "feature" that if you name a
network interface "foo%d" then it will automatically look for the
interface starting with "foo…" with the lowest number that is not used
yet and allocates that.
We should never clash with this "magic" handling of ifnames, hence
refuse this, since otherwise we never know what the name is we end up
with.
We should probably switch things from a deny list to an allow list
sooner or later and be much stricter. Since the kernel directly enforces
only very few rules on the names, we'd need to do some research what is
safe and what is not first, though.
(cherry picked from commit
e5f8ce13bbaf0d8b9ff597692c67fba0e38b4200)
Anita Zhang [Mon, 8 Mar 2021 18:51:12 +0000 (10:51 -0800)]
oomd: add unit test to repro #18926
(cherry picked from commit
399d80ba8c604d57e7df6c5118ea3258cce026d9)
Anita Zhang [Mon, 8 Mar 2021 18:21:37 +0000 (10:21 -0800)]
oomd: wrap paths in oomd_insert_cgroup_context with empty_to_root
(cherry picked from commit
50c0578b619e7298375afdffec7a8b3a40a68c21)
Anita Zhang [Mon, 8 Mar 2021 18:35:31 +0000 (10:35 -0800)]
oomd: move TAKE_PTR to end of oomd_insert_cgroup_context()
Fixes #18926
(cherry picked from commit
45da27fa053898c1b94c175070a0dd63128875c9)
Zbigniew Jędrzejewski-Szmek [Mon, 8 Mar 2021 19:49:53 +0000 (20:49 +0100)]
timedated: fix skipping of comments in config file
Reading file '/usr/lib/systemd/ntp-units.d/80-systemd-timesync.list'
Failed to add NTP service "# This file is part of systemd.", ignoring: Invalid argument
Failed to add NTP service "# See systemd-timedated.service(8) for more information.", ignoring: Invalid argument
:(
(cherry picked from commit
03a81441b1a490f8fc2a19aeb9b23299657c380c)
Zbigniew Jędrzejewski-Szmek [Mon, 8 Mar 2021 10:37:17 +0000 (11:37 +0100)]
homed: disable event sources before unreffing them
C.f.
9793530228.
We'd crash when trying to access an already-deallocated object:
Thread no. 1 (7 frames)
#2 log_assert_failed_realm at ../src/basic/log.c:844
#3 event_inotify_data_drop at ../src/libsystemd/sd-event/sd-event.c:3035
#4 source_dispatch at ../src/libsystemd/sd-event/sd-event.c:3250
#5 sd_event_dispatch at ../src/libsystemd/sd-event/sd-event.c:3631
#6 sd_event_run at ../src/libsystemd/sd-event/sd-event.c:3689
#7 sd_event_loop at ../src/libsystemd/sd-event/sd-event.c:3711
#8 run at ../src/home/homed.c:47
The source in question is an inotify source, and the messages are:
systemd-homed[1340]: /home/ moved or renamed, recreating watch and rescanning.
systemd-homed[1340]: Assertion '*_head == _item' failed at src/libsystemd/sd-event/sd-event.c:3035, function event_inotify_data_drop(). Aborting.
on_home_inotify() got called, then manager_watch_home(), which unrefs the
existing inotify_event_source. I assume that the source gets dispatched again
because it was still in the pending queue.
I can't reproduce the issue (timing?), but this should
fix #17824, https://bugzilla.redhat.com/show_bug.cgi?id=
1899264.
(cherry picked from commit
cf5366387b24633284ee92285ea64a282270d591)
Zbigniew Jędrzejewski-Szmek [Mon, 8 Mar 2021 10:28:19 +0000 (11:28 +0100)]
homed: wrap some very long lines
(cherry picked from commit
23d24b76f3df765fa6dbe7fe815bd4f8e76c9bb5)
Yu Watanabe [Sat, 6 Mar 2021 15:05:03 +0000 (00:05 +0900)]
dissect: fix memleak
Fixes #18903.
(cherry picked from commit
f91861e49fea01c98ad0e1131b9141830c029f28)
Zbigniew Jędrzejewski-Szmek [Fri, 5 Mar 2021 14:16:18 +0000 (15:16 +0100)]
sd-bus: fix memleak in failure path in bus_match_parse()
(cherry picked from commit
8df3f44c90a3e479b717e8cc4bb26c95ebbc5cb0)
Zbigniew Jędrzejewski-Szmek [Thu, 4 Mar 2021 20:19:02 +0000 (21:19 +0100)]
sd-bus: fix memstream buffer extraction
I'm getting the following error under valgrind:
==305970== Invalid free() / delete / delete[] / realloc()
==305970== at 0x483E9F1: free (vg_replace_malloc.c:538)
==305970== by 0x4012CD: mfree (alloc-util.h:48)
==305970== by 0x4012EF: freep (alloc-util.h:83)
==305970== by 0x4017F4: LLVMFuzzerTestOneInput (fuzz-bus-match.c:58)
==305970== by 0x401A58: main (fuzz-main.c:39)
==305970== Address 0x59972f0 is 0 bytes inside a block of size 8,192 free'd
==305970== at 0x483FCE4: realloc (vg_replace_malloc.c:834)
==305970== by 0x4C986F7: _IO_mem_finish (in /usr/lib64/libc-2.33.so)
==305970== by 0x4C8F5E0: fclose@@GLIBC_2.2.5 (in /usr/lib64/libc-2.33.so)
==305970== by 0x49D2CDB: fclose_nointr (fd-util.c:108)
==305970== by 0x49D2D3D: safe_fclose (fd-util.c:124)
==305970== by 0x4A4BCCC: fclosep (fd-util.h:41)
==305970== by 0x4A4E00F: bus_match_to_string (bus-match.c:859)
==305970== by 0x4016C2: LLVMFuzzerTestOneInput (fuzz-bus-match.c:58)
==305970== by 0x401A58: main (fuzz-main.c:39)
==305970== Block was alloc'd at
==305970== at 0x483FAE5: calloc (vg_replace_malloc.c:760)
==305970== by 0x4C98787: open_memstream (in /usr/lib64/libc-2.33.so)
==305970== by 0x49D56D6: open_memstream_unlocked (fileio.c:97)
==305970== by 0x4A4DEC5: bus_match_to_string (bus-match.c:859)
==305970== by 0x4016C2: LLVMFuzzerTestOneInput (fuzz-bus-match.c:58)
==305970== by 0x401A58: main (fuzz-main.c:39)
==305970==
So the fclose() which is called from _cleanup_fclose_ clearly reallocates the
buffer (maybe to save memory?). open_memstream(3) says:
The locations referred to by these pointers are updated each time the
stream is flushed (fflush(3)) and when the stream is closed (fclose(3)).
This seems to mean that we should close the stream first before grabbing the
buffer pointer.
(cherry picked from commit
5963e6f43c4f33d5255ef0fb887cdf382bd51c9e)
Lennart Poettering [Thu, 4 Mar 2021 15:23:46 +0000 (16:23 +0100)]
run: tweak algorithm for generating unit name from dbus unique name
This reverts behaviour of systemd-run's unit name generation to the
status quo ante of #18871: we chop off the ":1." prefix if we can.
However, to address the issue that the unique name can overrun we then
do what #18871 did as fallback: only chop off the ":" prefix.
This way we should have pretty names that look like they always looked
in the common case, but in the case of a unique name overrun we still
will have names that work.
Follow-up for #18871
(cherry picked from commit
e6283cbf48a3821d03ec73252620fc1b04bd4588)
caoxia [Thu, 28 Jan 2021 12:50:08 +0000 (20:50 +0800)]
trans_time sec is int32,it will overflow if local system time is later than 2038.
(cherry picked from commit
370d3c31b4637d1b2faeec555da3283e49a6744f)
Anita Zhang [Thu, 4 Mar 2021 01:25:40 +0000 (17:25 -0800)]
run: update dbus unique names check
Some code in systemd-run checks that a bus's unique name must start with
`:1.`. However the dbus specification on unique connection names only specifies
that it must begin with a colon. And the freedesktop/dbus implementation allows
allows unique names to go up to `:INT_MAX.INT_MAX`. So update the
current check to only look for a colon at the beginning.
(cherry picked from commit
01584bf9e43db1dfb7ea4ba628f533b3066e81cc)
Zbigniew Jędrzejewski-Szmek [Wed, 3 Mar 2021 23:36:24 +0000 (00:36 +0100)]
core: fix mtime calculation of dropin files
Nominally, the bug was in unit_load_dropin(), which just took the last mtime
instead of calculating the maximum. But instead of adding code to wrap the
loop, this patch goes in the other direction.
All (correct) callers of config_parse() followed a very similar pattern to
calculate the maximum mtime. So let's simplify things by making config_parse()
assume that mtime is initialized and update it to the maximum. This makes all
the callers that care about mtime simpler and also fixes the issue in
unit_load_dropin().
config_parse_many_nulstr() and config_parse_many() are different, because it
makes sense to call them just once, and current ret_mtime behaviour make sense.
Fixes #17730, https://bugzilla.redhat.com/show_bug.cgi?id=
1933137.
(cherry picked from commit
da46a1bc3cd28ac36114002c216196dae004b05c)
Tom Shield [Tue, 2 Mar 2021 09:56:06 +0000 (03:56 -0600)]
udev: add i2c to 60-persistent-input.rules for by-path (#18808)
Add the i2c subsystem to those that create by-path links.
i2c devices may not have IDs so we can't rely on the by-id links
but they (or some of them) should at least have a path that we can use.
(cherry picked from commit
e595edf1a3d7318811325cf97663c4b5a3859c6e)
Martin Wilck [Thu, 25 Feb 2021 16:09:03 +0000 (17:09 +0100)]
sd-device: don't use BPF filtering for kernel monitors
BPF filtering accesses fields in the netlink header that are
only filled in by libudev, never by the kernel. Therefore adding
BPF filters for kernel monitors is pointless. Even false filtering
of kernel events might be possible; at least it's hard to prove that
it can't occur.
(cherry picked from commit
d8ce385fe3e5be91b6f414415e10f8897e41d942)
Lajos Veres [Sun, 28 Feb 2021 22:48:08 +0000 (22:48 +0000)]
Fix path typo in systemd.unit
/etc/systemd/systemd/ => /etc/systemd/system/
(cherry picked from commit
57733518817c1217b9f442ca790ff3f50705b1c8)
Lennart Poettering [Thu, 25 Feb 2021 14:04:25 +0000 (15:04 +0100)]
udev: when btrfs.ko is not available consider btrfs filesystems not ready
Let's add a special tweak to the btrfs builtin: if /dev/btrfs-control is
not there, let's consider all btrfs file systems as SYSTEMD_READY=0.
This is useful in initrds, where btrfs.ko might be missing. After the
initrd → host transition we can then retigger the device and undo the
SYSTEMD_READY=0 marking.
(cherry picked from commit
97e535c7248cc0457395e2d62b6e7d6c342a0bd2)
Lennart Poettering [Thu, 25 Feb 2021 13:36:18 +0000 (14:36 +0100)]
man: fix indentation of example
The man page otherwise looks very weirdly aligned.
(cherry picked from commit
8b596d51a9b347f6fe5d130cbb0d26693149ce25)
Zbigniew Jędrzejewski-Szmek [Thu, 25 Feb 2021 13:54:54 +0000 (14:54 +0100)]
man: advertise shared drop-ins more
systemd.unit(5) is a wall of text. And this particular feature can be very useful
in the context of resource control. Let's avertise this cool feature a bit more.
Fixes #17900.
(cherry picked from commit
a8136f1bc03d1bdf93b9071b4f82123b81a05c8e)
Zbigniew Jędrzejewski-Szmek [Thu, 25 Feb 2021 13:30:34 +0000 (14:30 +0100)]
man: fix two issues in udev(7)
Fixed #18050.
(cherry picked from commit
4fc8a70d9f062b7b51c6c808a24ca3fd22df668f)
Lennart Poettering [Wed, 24 Feb 2021 16:49:32 +0000 (17:49 +0100)]
cgroup: don't generate BPF firewall unsupported warning on wrong unit
Let's generate the warning on a unit that actually needs the BPF
firewall, and not confusingly already for a sibling of one.
(cherry picked from commit
a437c5e4da666d16f15649461ce45e0e6d735148)
Sergey Bugaev [Fri, 19 Feb 2021 12:29:02 +0000 (15:29 +0300)]
resolved: don't discard mDNS queries with bad flags
Even though RFC 6762 specifies these bits MUST be zero, it also says they MUST
be ignored on reception.
(cherry picked from commit
2aaf3765974c9f340c0c3d705e9f0609073cf83c)
Sergey Bugaev [Mon, 15 Feb 2021 16:03:27 +0000 (19:03 +0300)]
resolved: don't discard mDNS packets with ip6.arpa RRs
ip6.arpa is also a valid domain name to put in mDNS packets.
(cherry picked from commit
7675501540980eff80c0e055c219e5e1efde6b62)
Yu Watanabe [Wed, 24 Feb 2021 04:16:15 +0000 (13:16 +0900)]
timedate: do not ignore fix_system argument in SetLocalRTC method
Fixes #18391.
(cherry picked from commit
2be6c4758e3c27a3f502735881a355e5bfae97b0)
Yu Watanabe [Mon, 22 Feb 2021 13:39:31 +0000 (22:39 +0900)]
dhcp6: do not set T1 and T2 by dhcp6_option_append_pd() in client
Fixes #18090.
(cherry picked from commit
73b49d433c2c8e6304c8b82538bd4231d070fce4)
Yu Watanabe [Mon, 22 Feb 2021 12:54:32 +0000 (21:54 +0900)]
dhcp6: do not set T1 and T2 by dhcp6_option_append_ia() in client
(cherry picked from commit
e7613578b4747488bef9b558dc35e41d7075f18a)
Yu Watanabe [Tue, 23 Feb 2021 10:25:17 +0000 (19:25 +0900)]
dhcp6: make dhcp6_option_parse_{address,pdprefix}() return -EINVAL when received address or prefix is refused
And then the caller ignores the error.
Otherwise, `ret_liftime_valid` is not set even if they succeeds.
(cherry picked from commit
1e84213a433f8a80c94dea34514744cc4937efd0)
Yu Watanabe [Mon, 22 Feb 2021 12:59:24 +0000 (21:59 +0900)]
dhcp6: do not use input value before checking
(cherry picked from commit
cf6c33bd6b6ca606f0756312bd73ae066c6aa9a8)
Lennart Poettering [Tue, 23 Feb 2021 14:02:33 +0000 (15:02 +0100)]
bootspec: assume that the root dir is at the top of its file system
Fixes: #17746
(cherry picked from commit
eceb61112c8ece03adfb4fcbc83e357875ca0ceb)
Zbigniew Jędrzejewski-Szmek [Tue, 23 Feb 2021 16:18:05 +0000 (17:18 +0100)]
man: add a description of how manager env block is set
(cherry picked from commit
d51586434677423e97decd0f8c02db0a266dd4e7)
Lennart Poettering [Tue, 23 Feb 2021 13:16:08 +0000 (14:16 +0100)]
efi stub: accept it if our loaded image has no FilePath field set
The firmware spec doesn't really say whether FilePath of the LoadedImage
protocol may be NULL or not. So far we assumed it to be non-NULL, but
apparently the FreeBSD UEFI chainloader sets it to NULL. Handle this
gracefully.
(Noticed and tracked down by Alexander Schreiber)
Fixes: #18733
(cherry picked from commit
685097b9cadf92d21dfa857c3a7ddc1ba115088e)
Lennart Poettering [Tue, 23 Feb 2021 13:07:46 +0000 (14:07 +0100)]
man: various improvements to systemd.generator(7)
In particular, make clear the .d/*.conf unit file drop-ins are OK to
generate from generators.
Inspired by: https://lists.freedesktop.org/archives/systemd-devel/2021-February/046148.html
(cherry picked from commit
3acf00a5a4ff656e2799f7f3e2544891b09bbc35)
Zbigniew Jędrzejewski-Szmek [Tue, 23 Feb 2021 00:31:15 +0000 (01:31 +0100)]
resolved: add another explanatory message to stub resolv.conf files
Apparently people do "cat /etc/resolv.conf" and not realize that they
are looking at a file in /run.
(cherry picked from commit
2b767e9222809103cb756e368f0d71e906623bca)
Lennart Poettering [Mon, 22 Feb 2021 16:38:02 +0000 (17:38 +0100)]
cryptsetup: unescape ID_PART_ENTRY_NAME udev property before using it
Fixes: #18729
(cherry picked from commit
fadd34dd5af9a26edf2906b237ac212169d39f0c)
Lennart Poettering [Mon, 22 Feb 2021 17:12:34 +0000 (18:12 +0100)]
man: try to improve documentation of conditions/asserts
Fixes: #18725
(cherry picked from commit
0a6aa7a238e1130ebe1b797a62da06645fb73ad4)
Lennart Poettering [Mon, 22 Feb 2021 16:52:20 +0000 (17:52 +0100)]
man: correct documentaiton of StandardInput='s defaults in regards to "data"
Fixes: #18710
(cherry picked from commit
c6e33c293ecb1cad5f975fe3cd12ec5aff3df221)
Tyler Hicks [Sun, 21 Feb 2021 18:53:22 +0000 (12:53 -0600)]
man: correct the unit file directory for attached images
Commit
83f72cd65fb8 ("man,docs: document the new unit file directory for
attached images") updated the docs and man page with the new unit file
directory for attached images but included a system.attached ->
systemd.attached typo in the man page portion of the change. Fix the
typo to document the correct path.
(cherry picked from commit
e4d54220a1ffa6629c0aad717a8b7601c0319657)
Zbigniew Jędrzejewski-Szmek [Fri, 19 Feb 2021 18:02:38 +0000 (19:02 +0100)]
sysctl.d: silence warning if net.core.default_qdisc cannot be set
Kernels can be compiled without the attribute. It's fine if this is not
set, so silence the warning.
(cherry picked from commit
fa98c99ea7f7c5bec3962fa52f4d3496a9777024)
Zbigniew Jędrzejewski-Szmek [Fri, 19 Feb 2021 08:50:09 +0000 (09:50 +0100)]
xdg-autostart-generator: ignore DBusActivatable=true
See https://wiki.gnome.org/HowDoI/DBusApplicationLaunching and
https://wiki.gnome.org/Initiatives/GnomeGoals/DBusActivatable for a description
of this key:
> Instead of the typical UNIX-style fork()/exec() approach to process creation,
> launching an application is done by sending a D-Bus message to the well-known
> name of that application, causing a D-Bus activation.
>
> Starting processes with D-Bus activation ensures that each application gets
> started in its own pristine environment, as a direct descendent of the
> session -- not in the environment of whatever its parent happened to be. This
> is important for ensuring the app ends up in the correct cgroup, for example.
So this motivation is not important for us: we launch stuff ourselves better.
This fixes warnings during boot:
systemd-xdg-autostart-generator[2274]: /etc/xdg/autostart/org.freedesktop.problems.applet.desktop:92: Unknown key name 'DBusActivatable' in section 'Desktop Entry', ignoring.
(cherry picked from commit
36290e0b97ba032cbd71723f2255917924eadc05)
Zbigniew Jędrzejewski-Szmek [Fri, 19 Feb 2021 08:47:32 +0000 (09:47 +0100)]
xdg-autostart-generator: reindent
(cherry picked from commit
9f11f565556410f9484db9e4991f6c04eeafffc7)
Zbigniew Jędrzejewski-Szmek [Thu, 18 Feb 2021 19:57:06 +0000 (20:57 +0100)]
Recommend drop-ins over modifications to the main config file
As discussed in https://github.com/systemd/systemd/pull/18347.
(cherry picked from commit
09db71a4a3fb9cee19f89cb75ad453caa46209b8)
Zbigniew Jędrzejewski-Szmek [Thu, 18 Feb 2021 09:48:08 +0000 (10:48 +0100)]
rfkill: use short writes and accept long reads
I'm seeing the following with kernel-core-5.10.16-200.fc33.x86_64:
$ sudo SYSTEMD_LOG_LEVEL=debug build/systemd-rfkill
Reading struct rfkill_event: got 8 bytes.
A new rfkill device has been added with index 0 and type bluetooth.
Found cgroup2 on /sys/fs/cgroup/, full unified hierarchy
Found container virtualization none.
rfkill0: Operating on rfkill device 'tpacpi_bluetooth_sw'.
Writing struct rfkill_event successful (8 of 9 bytes).
Loaded state '0' from /var/lib/systemd/rfkill/platform-thinkpad_acpi:bluetooth.
Reading struct rfkill_event: got 8 bytes.
A new rfkill device has been added with index 1 and type wwan.
rfkill1: Operating on rfkill device 'tpacpi_wwan_sw'.
Writing struct rfkill_event successful (8 of 9 bytes).
Loaded state '0' from /var/lib/systemd/rfkill/platform-thinkpad_acpi:wwan.
Reading struct rfkill_event: got 8 bytes.
A new rfkill device has been added with index 2 and type bluetooth.
rfkill2: Operating on rfkill device 'hci0'.
Writing struct rfkill_event successful (8 of 9 bytes).
Loaded state '0' from /var/lib/systemd/rfkill/pci-0000:00:14.0-usb-0:7:1.0:bluetooth.
Reading struct rfkill_event: got 8 bytes.
A new rfkill device has been added with index 3 and type wlan.
rfkill3: Operating on rfkill device 'phy0'.
Writing struct rfkill_event successful (8 of 9 bytes).
Loaded state '0' from /var/lib/systemd/rfkill/pci-0000:04:00.0:wlan.
All events read and idle, exiting.
We were expecting a read of exactly RFKILL_EVENT_SIZE_V1==8 bytes. But the
structure has 9 after [1].
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=
14486c82612a177cb910980c70ba900827ca0894
For some reason the kernel does not accept the full structure size, but cuts
the write short after 8 bytes:
static ssize_t rfkill_fop_write(struct file *file, const char __user *buf,
size_t count, loff_t *pos)
{
struct rfkill_event ev;
/* we don't need the 'hard' variable but accept it */
if (count < RFKILL_EVENT_SIZE_V1 - 1)
return -EINVAL;
/*
* Copy as much data as we can accept into our 'ev' buffer,
* but tell userspace how much we've copied so it can determine
* our API version even in a write() call, if it cares.
*/
count = min(count, sizeof(ev));
if (copy_from_user(&ev, buf, count))
return -EFAULT;
... so it should accept the full size. I'm not sure what is going on here.
But we don't care about the extra fields, so let's accept a write as long as
it's at least RFKILL_EVENT_SIZE_V1.
Fixes #18677.
(cherry picked from commit
a71c09685021cbcecb7566a00342421f635cc002)
Zbigniew Jędrzejewski-Szmek [Thu, 18 Feb 2021 09:40:16 +0000 (10:40 +0100)]
rfkill: improve error logging
If we get something of unexpected size, log the sizes. Also, don't log twice.
(cherry picked from commit
6c7afdeab010025c80508effd2b0039bc5181b82)
Lennart Poettering [Wed, 17 Feb 2021 22:11:42 +0000 (23:11 +0100)]
journald: when we fail to add a new entry to a journal, return the seqno
Fixes: #18005
(cherry picked from commit
0eaee8281d4699903b8b2cce18c836d4a144aee3)
Richard Laager [Thu, 18 Feb 2021 07:11:53 +0000 (01:11 -0600)]
Remove outdated disable_ipv6 docs
This was changed in commit
482efedc081b0c4bf2e77a3dee6b979d9c9a5765,
which was released in v243, to only enable and never disable IPv6.
Signed-off-by: Richard Laager <rlaager@wiktel.com>
(cherry picked from commit
f542f3b2ed3cf3e71692d1736f3fdd0ebdc226ef)
Vito Caputo [Sun, 29 Nov 2020 00:28:08 +0000 (16:28 -0800)]
journal-file: fix archiving offline journals
The existing set_offline() short-circuit erroneously included
when f->archive was true and header->state was STATE_OFFLINE.
This commit makes the short-circuit f->archive aware, so it will
only catch scenarios where there's not an offlining in progress
and the header state matches the target state of either archived
or offline.
Fixes https://github.com/systemd/systemd/issues/17770
(cherry picked from commit
bb1296b55a222cbab24d533006e730529009258c)
Luca Boccassi [Mon, 18 Jan 2021 20:15:03 +0000 (20:15 +0000)]
stat-util: fix dir_is_empty_at without path
Use the right FD, and do a fd_reopen instead of a dup, since the
latter will still share the internal pointer which then gets
moved by FOREACH_DIRENT, affecting the caller's FD.
(cherry picked from commit
b8f762f2fe59c94323c95d2aadea68612dca2b04)
Lennart Poettering [Mon, 15 Feb 2021 22:31:54 +0000 (23:31 +0100)]
journalctl: rotation is not a reason to warn, but certainly noteworthy
Downgrade the phrasing, since it is a bit misleading.
Fixes: #18465
(cherry picked from commit
71311efe2377d7e37a6af76f110882eed2332f4e)
Alan Perry [Sat, 13 Feb 2021 03:08:28 +0000 (19:08 -0800)]
no blkid for mmcblk[0-9]boot[0-9]
Don't run blkid on mmcblk[0-9]boot[0-9] devices because they contain
bootloaders or boot parameters, and not partitions or file systems.
(cherry picked from commit
8db704b28b4fd4d13e376cf3495005010443937e)
Lennart Poettering [Fri, 12 Feb 2021 20:16:27 +0000 (21:16 +0100)]
udev: make net_setup_link builtin quiet when link vanishes while we operate on it
Fixes: #16175
(cherry picked from commit
3ec2f7f2e3fc04699d8de27d3cc2e5eecf9a3698)
Yu Watanabe [Wed, 3 Feb 2021 15:52:28 +0000 (00:52 +0900)]
network: dhcp6: fix condtion check
(cherry picked from commit
b54aed29c61b9b86dc9cfa214c946586ea1f4972)
Zbigniew Jędrzejewski-Szmek [Thu, 11 Feb 2021 11:00:09 +0000 (12:00 +0100)]
homework: fix unitialized variable
Coverity CID#
1444703.
(cherry picked from commit
a4aa5742f75f0753c4dfa812059af608c9bab7b6)
Zbigniew Jędrzejewski-Szmek [Thu, 11 Feb 2021 08:50:49 +0000 (09:50 +0100)]
fsck: make sure we don't read an unitialized variable
This use on %n was completely unnecessary: fprintf returns the number of
characters written. And the issue was that if fprintf failed for whatever
reason, it would not process the %n and m would be unitialized. Rework the
code a bit to simplify it.
Coverity CID#
1444708.
(cherry picked from commit
333ab199a12c7b060d3a3f4d50a8f73ee4fd5ebd)
Zbigniew Jędrzejewski-Szmek [Thu, 11 Feb 2021 08:35:52 +0000 (09:35 +0100)]
sd-journal: add forgotten unmap in error path
Bug introduced in
4b5bc5396c090ee41c45cab9052372d296c4a2f4 :(
Coverity CID#
1444709.
(cherry picked from commit
363729c4704b9838f351eaba3acb8b1ed16fae0e)
Lennart Poettering [Wed, 10 Feb 2021 09:50:23 +0000 (10:50 +0100)]
analyze: slightly reword PrivatTmp= message
Apparently there way confusion about "does not apply". Let's say "is not
appropriate".
Fixes: #13095
(cherry picked from commit
77552b9520ba0d47cbf33cdbe1ddedb9ce9b5bf3)
Lennart Poettering [Tue, 9 Feb 2021 14:03:46 +0000 (15:03 +0100)]
inhibit: cut off overly long "who" fields
systemd-inhibit when invoked with a command line will put the whole
command line in the "who" field of the inhibitor lock. This can get
extremely long for shell expressions, making the table "systemd-inhibit
--list" shows ridiculously weirdly formatted. Let's put a limit on the
column width: half of the screen, not more.
(cherry picked from commit
619b0c077cbb12b85c9d752779095b99cfd2c66c)
Lennart Poettering [Tue, 9 Feb 2021 14:02:21 +0000 (15:02 +0100)]
format-table: don't hit assert if column got less width than it asked for
If one field in a specific column has a maximum size limit, other fields
in the same column might affected by it and get less than they asked
for. Let's make sure we can handle this, and don't assert on this
because surprisingly we got less than what we asked for.
(cherry picked from commit
2cd9f773e29873df1b9f9b09c6c4f78853b16215)
Antonius Frie [Mon, 8 Feb 2021 08:15:15 +0000 (09:15 +0100)]
Use correct config parser for MountAPIVFS (#18501)
As far as I can see, at some point the parser function for MountAPIVFS
was changed from the generic bool parser to a custom implementation, to
allow the context to keep track of whether MountAPIVFS had been set
explicitly. If not, exec_context_get_effective_mount_apivfs would fall
back to a default value. However, the corresponding entry in the big
parser table wasn't updated, meaning that the old bool parser was still
used, meaning that context->mount_apivfs_set remained at its default
value of false, meaning that the default value was always used and the
config option was effectively ignored.
Fix for
5e98086d1629f5c5b73645ba2568de4b09b7d958.
(cherry picked from commit
4d0052d52f1359a054e83037c733bb3330ba4124)
Julia Cartwright [Fri, 5 Feb 2021 21:16:18 +0000 (15:16 -0600)]
man: fix misspellings of "NameSeviceSwitch"
Fixup NameSeviceSwitch -> NameServiceSwitch spelling error in
documentation.
Found through inspection of documentation.
(cherry picked from commit
afe6a9c48e956c8ddfaa0f201cf371b01a22f970)
Lennart Poettering [Fri, 5 Feb 2021 17:43:55 +0000 (18:43 +0100)]
resolved: suppress ifindex info in varlink JSON responses if zero
If we don't have ifindex info, don't set the field for it.
We already do that for parsed IP address replies, let's do it for all
cases: it's a bit nicer to suppress the ifindex prop if it doesn't apply
than to pass it invalid.
This is the other side of #18482, i.e. fixes things so that the parser
doesn't get tripped up by this.
(This too makes a problem go away we should track down properly, i.e.
figure out how the ifindex got lost in
https://github.com/systemd/systemd/pull/17823#issuecomment-
742439422 )
(cherry picked from commit
f8f5b8d864d3d034dbbfb84fdad4f2cbeb7790f9)
Lennart Poettering [Fri, 5 Feb 2021 17:46:18 +0000 (18:46 +0100)]
nss-resolve: shortcut fixing of ifindex if it's zero anyway
(cherry picked from commit
425ed13991153bd274a86009459b00ec8a709f30)
Lennart Poettering [Fri, 5 Feb 2021 17:44:22 +0000 (18:44 +0100)]
nss-resolve: accept zero ifindex when parsing resolved reply
Sometimes a reply isn't associated to any specific interface, it might
be a general truth (for example served from /etc/hosts or so). In this
case the server might pass ifindex == 0. Accept that.
https://github.com/systemd/systemd/pull/17823#issuecomment-
742439422
(cherry picked from commit
6da52ca861a1cae3f453d50dbb6fac6666a73f4b)
Ali Abdallah [Thu, 21 Jan 2021 06:37:21 +0000 (07:37 +0100)]
PATCH] Always free deserialized_subscribed on reload
Otherwise, it will keep consuming memory on systemctl daemon-reload.
(cherry picked from commit
3deed59afdc2c18ecb76fe90b9bba0cd66045dfa)
Zbigniew Jędrzejewski-Szmek [Thu, 4 Feb 2021 08:48:19 +0000 (09:48 +0100)]
core: map io.bfq.weight to 1..1000
Aaargh. See the comment in the code.
Apparently the range is like that:
$ sudo bash -c 'echo "default 1001" >/sys/fs/cgroup/user.slice/io.bfq.weight'
bash: line 0: echo: write error: Numerical result out of range
$ uname -r
5.11.0-0.rc4.129.fc34.x86_64
(cherry picked from commit
29eb0eefd14afc9a2424781a28b376db47c3c570)
Zbigniew Jędrzejewski-Szmek [Thu, 4 Feb 2021 10:05:39 +0000 (11:05 +0100)]
man: reword the description of "main conf file"
While not really "wrong", the text we had could use a little editing.
Fixes #18458.
(cherry picked from commit
a7a1887869695f16a1114c355c13d731ed7be109)
Yu Watanabe [Tue, 29 Dec 2020 18:32:07 +0000 (03:32 +0900)]
test-network: add tests for fib rules with Family=both vs networkctl reload or reconfigure
[#18109,7/7] cherry-picked from
49ff3f34d549ad214924548f46a71315b5cd23c6
Yu Watanabe [Tue, 29 Dec 2020 18:15:44 +0000 (03:15 +0900)]
network: drop fib rules configured with Family=both
[#18109,6/7] cherry-picked from
a75466ed198fad0f50054b4715cfc55c17ffba09
Yu Watanabe [Tue, 29 Dec 2020 18:07:33 +0000 (03:07 +0900)]
network: set RoutingPolicyRule::family based on Family= setting
[#18109,5/7] cherry-picked from
49de8d5cedc9f381a8b61cc9dcd40781cacf7e68
Yu Watanabe [Tue, 29 Dec 2020 17:45:25 +0000 (02:45 +0900)]
network: always re-configure rules even if already exist
routing_policy_rule_get() in link_set_routing_policy_rules() does not
work when [RoutingPolicyRule] section does not have From= or To=.
[#18109,4/7] cherry-picked from
c1934a8f2f37b8e64d4b65f465d823c2a96abbc7
Yu Watanabe [Tue, 29 Dec 2020 18:17:35 +0000 (03:17 +0900)]
network: drop unnecessary checks
By the previous commit, the checks are not necessary any more.
[#18109,3/7] cherry-picked from
57fe5a42f0975ca12ff856ca221c081a1ad3e87e
Yu Watanabe [Tue, 29 Dec 2020 17:03:48 +0000 (02:03 +0900)]
network: constify several arguments
[#18109,1/7] cherry-picked from
3141c8173632c100654144faa3d78b84bdffd72f
Yu Watanabe [Wed, 28 Oct 2020 08:02:41 +0000 (17:02 +0900)]
network: do not serialize/deserialize routing policy rules
We already handle foreign routing policy rules correctly by the previous
commit. So, the serialization/deserialization of rules are not necessary
anymore.
[#17477,5/5] cherry-picked from
a9d240f4bfefcd85de22381f6250e1e1ab58ea55
Yu Watanabe [Wed, 28 Oct 2020 11:27:23 +0000 (20:27 +0900)]
network: drop unnecessary routing policy rules
networkd already drop foreign address, routes, and nexthops on startup,
except those created by kernel. However, previously, routing policy
rules were not. The logic of serialization/deserialization of rules only
works for rules created by previous invocation of networkd, and does not
work for one created by other tools like `ip rule`.
This makes networkd drop foreign routing policy rules except created by
kernel on startup. Also, remove rules created by networkd when the
corresponding links are dropped or networkd is stopping.
[#17477,4/5] cherry-picked from
0b81225e5791f660506f7db0ab88078cf296b771
Yu Watanabe [Thu, 29 Oct 2020 02:55:12 +0000 (11:55 +0900)]
network: treat rule which has l3mdev flag as created by kernel
[#17477,3/5] cherry-picked from
e737dce5445ac2359f4f260aa462056b02d68d54
Yu Watanabe [Thu, 29 Oct 2020 02:34:36 +0000 (11:34 +0900)]
network: adjust protocol of rules sent from kernel when kernel does not support FRA_PROTOCOL
Otherwise, each configured rule is treated as foreign.
[#17477,2/5] cherry-picked from
569eeb0c059047601f16b9f6df2bbbc1582793a1
Yu Watanabe [Wed, 28 Oct 2020 07:54:51 +0000 (16:54 +0900)]
network: set FRA_PROTOCOL to RTPROT_STATIC by default
[#17477,1/5] cherry-picked from
1e5fd3216a8c73b636ad61e6ebee0c161b6c2063
Yu Watanabe [Wed, 28 Oct 2020 08:28:36 +0000 (17:28 +0900)]
network: use netlink_message_read_in_addr_union() where applicable
[#17960,4/4] cherry-picked from
c18c53c36e39f873a7a83e1c0fad1a12c3a650d3
Yu Watanabe [Wed, 28 Oct 2020 08:22:58 +0000 (17:22 +0900)]
network: introduce log_routing_policy_rule_debug()
[#17960,3/4] cherry-picked from
ea81208f03afdc03e7e6efd1b5d9515c1ebd4e72
Yu Watanabe [Thu, 29 Oct 2020 02:41:01 +0000 (11:41 +0900)]
network: introduce routing_policy_rule_equal()
[#17960,2/4] cherry-picked from
40424f1ad97db59958e40a395986a5e248da6f05
Yu Watanabe [Wed, 28 Oct 2020 07:16:58 +0000 (16:16 +0900)]
network: make routing_policy_rule_remove() take Manager instead of Link
As routing policy rules are managed by Manager.
[#17960,1/4] cherry-picked from
18f2ee33102d1ff4f658829f971697895d813cfd
Luca Boccassi [Mon, 22 Feb 2021 21:31:44 +0000 (21:31 +0000)]
MountImages: parse unit specifiers after permissive hyphen
The permissive bit it not something a specifier might synthetise
(cherry picked from commit
6c3f7ca036829825f344ffb364354faee5db174e)
Luca Boccassi [Mon, 22 Feb 2021 10:12:34 +0000 (10:12 +0000)]
MountImages: fix exec_context_dump printf
Use strempty as options might not be set, and add the separator
for each option tuple
(cherry picked from commit
79e20ceb59157662eb32fdf3a7eb1e4d9d8bf00b)
Luca Boccassi [Thu, 28 Jan 2021 13:25:41 +0000 (13:25 +0000)]
MountImages: actually support optional paths
ENOENT did not cause an image mount to be skipped, fix it
(cherry picked from commit
b850a9b29f2bc14836415f0c85d42e7d4990285d)
Conflicts:
src/core/namespace.c
Luca Boccassi [Fri, 5 Mar 2021 14:53:20 +0000 (14:53 +0000)]
portablectl: start/stop path/timer units too
Partial backport from:
https://github.com/systemd/systemd/commit/
e26fe5f91158a0d9133e61347650dfb73aadd6bb
Lennart Poettering [Wed, 10 Feb 2021 22:05:51 +0000 (23:05 +0100)]
sd-netlink: spread out sequence numbers a bit
An (imperfect) fix for #14760.
This makes collisions unlikely, but still theoretically possible.
Fixes: #14760
(cherry picked from commit
ac3bc1b819b66f925145629a5c2ccb0a098f0446)
Lennart Poettering [Wed, 10 Feb 2021 17:12:58 +0000 (18:12 +0100)]
sd-netlink: reduce indentation levels a bit
(cherry picked from commit
baf78f1a51793c3dbbeb2b420783573d627f85a8)
Lennart Poettering [Wed, 10 Feb 2021 17:11:20 +0000 (18:11 +0100)]
sd-netlink: use getsockopt_int() where appropriate
(cherry picked from commit
13ec9f103b53ecc555a16048c3901d3320b26ef3)
Lennart Poettering [Wed, 10 Feb 2021 16:59:46 +0000 (17:59 +0100)]
sd-netlink: revamp message serial handling
Let's use uint32_t everywhere to maintain the seqno, since that's what
the kernel does. Prviously in the reply_callback logic we used 64bit,
for no apparent reason.
Using 32bit also provides us with the benefit that we can avoid using
uint64_hash_ops, and can use trivial_hash_ops instead for the reply
hashmap, so that we can store the seqno in the key pointer directly.
While we are at it, let's make sure we never run into serial collisions
internally (32bit is a lot, but not that much), and let's put a limit on
outstanding serials, to catch programming errors.
(cherry picked from commit
b522c4b92a1a8999e008002f0a30acbaf58b55e4)
Lennart Poettering [Mon, 16 Nov 2020 11:41:39 +0000 (12:41 +0100)]
socket-util: add getsockopt_int() helper
(cherry picked from commit
4e25d4cf0f78ef7c960f2b77c941e3d8751535df)
Lennart Poettering [Tue, 10 Nov 2020 13:48:13 +0000 (14:48 +0100)]
resolved: gracefully handle with packets with too large RR count
Apparently, there are plenty routers in place that report an incorrect
RR count in the packets: they declare more RRs than are actually
included.
Let's accept these responses, but let's downgrade them to baseline, i.e.
let's suppress OPT in this case: if they don't even get the RR count
right, let's operate on the absolute baseline, and not bother with
anything fancier such as EDNS.
Prompted-by: https://github.com/systemd/systemd/issues/12841#issuecomment-724063973
Fixes: #3980
Most likely fixes: #12841
(cherry picked from commit
18674159ebbf622a9e6e5a45cc36b38f74dae315)
Lennart Poettering [Thu, 17 Dec 2020 15:19:09 +0000 (16:19 +0100)]
test: fix fd_is_mount_point() check
So the currentl and only fd_is_mount_point() check is actually entirely
bogus: it passes "/" as filename argument, but that's not actually a
a valid filename, but an absolute path.
fd_is_mount_point() is written in a way tha the fd refers to a directory
and the specified path is a file directly below it that shall be
checked. The test call actually violated that rule, but still expected
success.
Let's fix this, and check for this explicitly, and refuse it.
Let's extend the test and move it to test-mountpoint-util.c where the
rest of the tests for related calls are placed.
Replaces: #18004
Fixes: #17950
(cherry picked from commit
95231c7215c3ff14c491eb1d2a93312a8fe0c4f6)
Yu Watanabe [Thu, 10 Dec 2020 23:34:13 +0000 (08:34 +0900)]
sd-device: make TAGS= property prefixed and suffixed with ":"
The commit
6f3ac0d51766b0b9101676cefe5c4ba81feba436 drops the prefix and
suffix in TAGS= property. But there exists several rules that have like
`TAGS=="*:tag:*"`. So, the property must be always prefixed and suffixed
with ":".
Fixes #17930.
(cherry picked from commit
8d80f275303385eb9e2452f97f9c11be5c196862)
Yu Watanabe [Sun, 6 Dec 2020 12:10:34 +0000 (21:10 +0900)]
sd-device: keep escaped strings in DEVLINK= property
This fixes a bug introduced by
87a4d416e5126b6fb2528ae192a6a6a8033539ce.
Fixes #17772.
(cherry picked from commit
31063db0b4842a452c6eadde5b969aaee2a0516a)
Yu Watanabe [Sun, 6 Dec 2020 11:11:37 +0000 (20:11 +0900)]
sd-device: use set_strjoin()
This slightly changes TAGS= and CURRENT_TAGS= properties:
Before
E: TAGS=:aaa:bbb:
After
E: TAGS=aaa:bbb
(cherry picked from commit
6f3ac0d51766b0b9101676cefe5c4ba81feba436)