fuzz: limit the maximum size of test inputs for a few parsers
We have a few cases or reported issues which are about a timeout to parse
the input in 25 s. In all cases, the input is a few hundred kb. We don't really
care if the config parsers are super efficent, so let's set a limit on the input
size to avoid triggering such issues. The parsers often contain quadratic
algorithms. This is OK, because the numbers of elements are almost always very
small in real use. Rewriting the code to use more complicated data structures
to speed this up would not only complicate the code, but also pessimize behaviour
for the overwhelmingly common case of small samples. Note that in all those
cases, the input data is trusted. We care about memory correctness, and not
not so much about efficiency.
The size checks are done twice: using options for libfuzzer, and using an
internal check for afl. Those should be changed together. I didn't use a define,
because there is no easy mechanism to share the define between the two files.
This is almost the same as 0e636bf51a. I looked through the code,
and I don't see any more instances of this pattern, so hopefully this
will be the last one.
efivars: make sure efi_loader_entry_name_valid() is always compiled in
The string validation should always be done correctly, and not fail just
because EFI is turned off. After all an EFI loader entry name string
remains properly formatted regardless if EFI is on or off...
../src/login/logind-dbus.c: In function ‘property_get_reboot_to_boot_loader_entry’:
../src/login/logind-dbus.c:2772:29: error: implicit declaration of function ‘efi_loader_entry_name_valid’; did you mean ‘efi_loader_get_features’? [-Werror=implicit-function-declaration]
} else if (!efi_loader_entry_name_valid(v)) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~
efi_loader_get_features
../src/login/logind-dbus.c:2772:29: warning: nested extern declaration of ‘efi_loader_entry_name_valid’ [-Wnested-externs]
This adds stub for efi_loader_entry_name_valid() to enable compilation.
Lubomir Rintel [Sun, 10 Mar 2019 23:00:03 +0000 (00:00 +0100)]
hwdb: label the trackpoint on IBM UltraNav keyboard as a pointing stick (#11934)
This is the TrackPoint on an older IBM-branded ThinkPad-y USB keyboard.
It needs ID_INPUT_POINTINGSTICK=1 for TrackPoint scrolling to work, and
also for the AttrTrackpointMultiplier quirk to apply in libinput
(otherwise the TrackPoint is too slooooow).
If the QEMU_SMP value has not been explicitly set, try to determine it
from the number of online CPUs using the nproc utility. If this approach
fails, fall back to the default value QEMU_SMP=1.
This change should significantly help when running integration tests
under QEMU on multicore systems.
units: turn of ProtectHostname= again for services hat need to know about system hostname changes
ProtectHostname= turns off hostname change propagation from host to
service. This means for services that care about the hostname and need
to be able to notice changes to it it's not suitable (though it is
useful for most other cases still).
Let's turn it off hence for journald (which logs the current hostname)
for networkd (which optionally sends the current hostname to dhcp
servers) and resolved (which announces the current hostname via
llmnr/mdns).
core: support netns joining also for sockets created by .socket unit
Similar to the cgroup magic we nowadays do when listening to sockets, to
assign them the right bpf programs, let's also do the same and join the
specified netns in the child process.
This allows people to listen in sockets in specific namespaces, or join
multiple services and socket units together to live in the same
namespace.
The new call allows us to open a netns from the file system, and store
it in a "storage fd pair". It's supposed to work with setup_netns() and
allows pre-population of the netns used with one opened from the file
system.
random-seed: don't use "label" version of mkdir_parents()
We don't load the selinux label database anyway, hence this is not going
to do anything at all. Let's simplify this hence and drop the explicit
request.
Also note that today SELinux can take the filename into account when
automatically deducing the label for a new file, hence even if this code
actually would have done something it is redundant today.
nspawn: move payload to sub-cgroup first, then sync cgroup trees
if we sync the legacy and unified trees before moving to the right
subcgroup then ultimately the cgroup paths in the hierarchies will be
out-of-sync... Hence, let's move the payload first, and sync then.
remove "." path components from required mount paths
unit_require_mounts_for may be passed path arguments that contain "."
components like for user's home directories where "." is sometimes used
to specify some form of anchor point.
This change stops considering such path as an error and removes the "."
components instead.
shared/bootspec: do not fail on errors when reading EFI vars
It seems that my EFI storage is corrupted. The kernel reports the file as present, but returns
an error when reading. Nevertheless, this shouldn't prevent me from reading the entry
list.
meson: use a convenience library for the sources shared between core/ and the outside
This avoids double compilation. Those files are tiny, so it doesn't save time,
but we avoid repeated warnings and errors, and it's generally cleaner to it
this way.
The number of commands in 'ninja -C build clean && ninja -C build' drops from
1462 to 1455 for me.
pkgconfig: avoid double slash with split-usr configuration
By defining rootprefix= we avoid a double slash in $systemdsystemunitdir and
other variables. This fixes a regression introduced in 1c2c7c6cb3d92315624f3711114e86b0acfbce63 where the variables using rootprefix=/
would start with a double slash. This should be interpreted the same, but is
certainly ugly.
The rootprefix variable was added to systemd.pc in 1c2c7c6cb3d92315624f3711114e86b0acfbce63, so there is no question of backwards
compatiblity. If people try to "override" the prefix and specify
--define-variable=rootprefix=/, they will get a double slash, which should be
OK, and is the same as --define-variable=rootprefix=/something/, which also
results in a double slash somewhere in the strings.
systemctl: add support for booting into boot menu/entry
(This also removes support for booting into the EFI firmware setup
without logind. That's because otherwise the non-EFI fallback logind
implements can't work.)
logind: add support for booting into the boot menu or a specific boot menu entry
This behaves similar to the "boot into firmware" logic, and also allows
either direct EFI operation (which sd-boot supports and others might
support eventually too) or override through env var.
bootspec: get rid of find_default_boot_entry() entirely
Now only two operations are left. Let's just move this into the caller,
since it should make things simpler, clearer and shorter, in particular
as there's only a single user for this.
projects / thirdparty / systemd.git / log
