This adds support for the EFI stub to look for credential files and
sysext files next to the EFI kernel image being loaded, and pack them up
in an initrd cpio image, and pass them to the kernel.
Specifically, for a kernel image foo.efi it looks for
foo.efi.extra.d/*.cred and packs these files up in an initrd, placing it
inside a directory /.extra/credentials/. It then looks for
foo.efi.extra.d/*.raw and pack these files up in an initrd, placing them
inside a directory /.extra/sysexts/. It then concatenates any other
initrd with these two initrds, so they are combined.
Or in other words auxiliary files placed next to the kernel image are
picked up automatically by the EFI stub and be made available in the
initrd in the /.extra/ directory.
What's the usecase for this? This is supposed to be useful in context of
implementing fully trusted initrds, i.e. initrds that are not built
locally on the system and unsigned/unmeasured – as we do things
currently —, but instead are built by the vendor, and measured to TPM.
The idea is that a basic initrd is always linked into the kernel EFI
image anyway. This will already be sufficient for many cases. However,
in some cases it is necessary to parameterize initrds, or to extend the
basic initrds with additional subsystems (e.g. think complex storage, or
passing server info/certificates/… to initrds). The idea is that the
parameterization is done using the "credentials" logic we already have
in systemd, with these credential files (which can optionally be
encrypted+authenticated by TPM2) being placed in the ESP next to the
kernel image. And the initrd extension via the "sysext" logic we already
have in systemd too.
Note that the files read by this code are not verified immediately, they
are copied *as-is* and placed into /.extra/ in the initrd. In a trusted
environment they need to be validated later, but before first use. For
the credentials logic this should be done via the TPM2
encryption/authentication logic. For the sysext stuff the idea is that
this is done via signed images, as implemented by #20691.
"bootctl update" tries to add sd-boot to the EFI boot loader list if it
is not already there. To do so, it uses find_slot() which finds the
proper BootXXXX slot ID to use and also returns 1 if an existing sd-boot
entry was found at this ID or 0 if it is a new unused ID. In "update"
case install_variables() only writes the entry in case 0 (no existing
entry).
However, find_slot() erroneously returns 1 if it finds a gap in the Boot
IDs (i.e. when not resorting to max(ids) + 1). This causes
"bootctl update" to not add a missing systemd-boot boot entry if the
existing BootXXXX entry IDs are not consecutive.
Fix that by returning 0 in find_slot() when an empty gap ID is selected
to make it match the behavior when selecting an empty non-gap ID.
Compilation would fail because we could have HAVE_SMACK_RUN_LABEL without
HAVE_SMACK. This doesn't make much sense, so let's just make -Dsmack=false
completely disable smack.
Also, the logic in smack-setup.c seems dubious: '#ifdef SMACK_RUN_LABEL'
would evaluate to true even if -Dsmack-run-label='' is used. I think
this was introduced in the conversion to meson:
8b197c3a8a added
AC_ARG_WITH(smack-run-label,
AS_HELP_STRING([--with-smack-run-label=STRING],
[run systemd --system with a specific SMACK label]),
[AC_DEFINE_UNQUOTED(SMACK_RUN_LABEL, ["$withval"], [Run with a smack label])],
[])
i.e. it really was undefined if not specified. And it was same
still in 72cdb3e783 when configure.ac
was dropped.
So let's use the single conditional HAVE_SMACK_RUN_LABEL everywhere.
As far as I can see, we use this to get a list of ARPHRD_* defines (used in
particular for Type= in .link files). If we drop our copy, and build against
old kernel headers, the user will have a shorter list of types available. This
seems OK, and I don't think it's worth carrying our own version of this file
just to have newest possible entries.
7c5b9952c4 recently updated this file, but we'd
have to update it every time the kernel adds new entries. But if we look at
the failure carefully:
src/basic/arphrd-from-name.gperf:65:16: error: ‘ARPHRD_MCTP’ undeclared (first use in this function); did you mean ‘ARPHRD_FCPP’?
65 | MCTP, ARPHRD_MCTP
| ^~
| ARPHRD_FCPP
we see that the list we were generating was from the system headers, so it was
only as good as the system headers anyway, without the newer entries in our
bundled copy, if there were any. So let's make things simpler by always using
system headers.
And if somebody wants to fix things so that we always have the newest list,
then we should just generate and store the converted list, not the full header.
In a follow-up patch we are going to modify the initrd, hence prepare
the pointers/"physical addresses" to it, first, so that we can do so
easily.
Also, do some other tweaks and cleanups to physical address/pointer
conversion.
Units are copied out via sendmsg datafd from images, but that means
the SELinux labels get lost in transit. Extract them and copy them over.
Given recvmsg cannot use multiple IOV transparently when the sizes are
variable, use a '\0' as a separator between the filename and the label.
Compared to PID1 where systemd-oomd has to be the client to PID1
because PID1 is a more privileged process than systemd-oomd, systemd-oomd
is the more privileged process compared to a user manager so we have
user managers be the client whereas systemd-oomd is now the server.
The same varlink protocol is used between user managers and systemd-oomd
to deliver ManagedOOM property updates. systemd-oomd now sets up a varlink
server that user managers connect to to send ManagedOOM property updates.
We also add extra validation to make sure that non-root senders don't
send updates for cgroups they don't own.
The integration test was extended to repeat the chill/bloat test using
a user manager instead of PID1.
Gets rid of a few gotos, allows removing the extra ret variable and
will also be used in a future commit by the codepath that receives
cgroups from user instances of systemd.
i.e. let's pick some files we know are too large, or where struct stat's
.st_size is zero even though non-empty, and test read_virtual_file()
with that, to ensure things are handled sensibly. Goal is to ensure all
three major codepaths in read_virtual_file() are tested.
Prompted-by: #20743
We mishandled the case where the size we read from the file actually
matched the maximum size fully. In that case we cannot really make a
determination whether the file was fully read or only partially. In that
case let's do another loop, so that we operate with a buffer, and
we can detect the EOF (which will be signalled to us via a short read).
There's a very gradual increase of anonymous memory in systemd-journald that
blames to 2ac67221bb.
systemd-journald makes many calls to read /proc/PID/cmdline and
/proc/PID/status, both of which tend to be well under 4K. However the
combination of allocating 4M read buffers, then using `realloc()` to
shrink the buffer in `read_virtual_file()` appears to be creating
fragmentation in the heap (when combined with the other allocations
systemd-journald is doing).
To help mitigate this, try reading /proc with a 4K buffer as
`read_virtual_file()` did before 2ac67221bb.
If it isn't big enough then try again with the larger buffers.
According to the documentation, Setting the data threshold to zero disables the
data threshold alltogether. Let's make sure we actually implement this behaviour
in sd_journal_enumerate_fields() by only applying the data threshold if it exceeds
zero.
Device-tree based devices can't get the chassis type from DMI or ACPI,
and so far need a custom `/etc/machine-info` to set this property right.
A new 'chassis-type' toplevel device tree property has recently been
approved into the DT specification, making it possible to automate
chassis type detection on such devices.
This patch therefore falls back to reading this device-tree property if
nothing is available through both DMI and ACPI.
Signed-off-by: Arnaud Ferraris <arnaud.ferraris@collabora.com>
Currently `nulstr_contains` returns a boolean, making it difficult to
identify which of the input strings matches the "needle".
Adding a new `nulstr_get()` function, returning a const pointer to the
matching string, eases this process and allows us to directly re-use the
result of a call to this function without additional processing or
memory allocation.
Signed-off-by: Arnaud Ferraris <arnaud.ferraris@collabora.com>
let's do what we did for sysctl_write()/sysctl_write_ip_property() also
for the read paths: i.e. make one a wrapper of the other, and add more
careful input validation.
Let's add similar path validation to sysctl_read() as we already have in
sysctl_write().
Let's also drop the trailing newline from the returned string, like
sysctl_read_ip_property() already does it.
(I checked all users of this, they don't care)
It does the same stuff, let's use the same codepaths as much as we can.
And while we are at it, let's generate good error codes in case we are
called with unsupported parameters/let's validate stuff more that might
originate from user input.
Lennart Poettering
Anssi Hannula
Dan Streetman
Yu Watanabe
Zbigniew Jędrzejewski-Szmek
Luca Boccassi
Daan De Meyer
Anita Zhang
Arnaud Ferraris
Jan Janssen