g_memdup() is subject to an integer overflow on 64-bit machines if the
object being copied is larger than UINT_MAX bytes. I suspect none of
these objects can actually be that large in practice, but it's easier
to replace all the calls than it is to assess whether we need to
replace them.
A backport in libglnx is used on systems where GLib is older than 2.68.x.
Signed-off-by: Simon McVittie <smcv@collabora.com>
It seems likely that we're allocating more RAM here. What we really
need to do is for cosa to express the necessary RAM
requirements declaratively, then we compute pod requirements
from that.
This adds a tag filter to the logic which emits version labels for
COPR build, so that it avoids mistakenly picking up tags belonging
to the Rust bindings.
Basically, it should not be necessary - the policy should be
up-to-date. We don't want to force on continual policy rebuilds.
Even trying to run bwrap when we're *not* in a booted
root can cause failures in nested containerization scenarios.
Closes: https://github.com/ostreedev/ostree/issues/2758
Signed-off-by: Ricardo Noriega <rnoriega@redhat.com>
Use G_IO_ERROR_BUSY to represent 500 error
Signed-off-by: Ricardo Noriega <rnoriega@redhat.com>
Make G_IO_ERROR_BUSY independent from glib version
Signed-off-by: Ricardo Noriega <rnoriega@redhat.com>
This was only there to reproduce a bug we hit long ago
with bootloader file ordering. We're extremely unlikely
to reintroduce such a bug, and it's not worth carrying around
this code.
As noted in readdir(3), in the presence of long filenames it is
possible for a directory entry to be larger than `sizeof (struct dirent)`.
Copy the full length instead.
Signed-off-by: Simon McVittie <smcv@collabora.com>
This converts several invariant checks to asserts. Most of the
functions in this file were already using assertions, so this aligns
the remaining few outliers to the rest.
We're hitting memory limits when running the reprovisioning tests. We
should investigate why we need 1Gi of overhead, but for now to get
unblocked let's just bump it.
This enhances the logic handling GVariant parameters within
`ostree_repo_static_delta_generate()`.
Several of those entries are expected to be zero-terminated values,
and this implicit assumption has been observed to be an hidden trap
in languages where strings and arrays may not carry a terminator value
(e.g. Rust).
In order to improve the situation, this makes the documentation more
explicit and actively tries to catch invalid input parameters.
More fallout from https://github.com/coreos/coreos-ci-lib/pull/116. We
need to be more honest about how much memory we need.
While we're here, switch the `kola testiso` invocation to use
`fcosKolaTestIso` instead. This will add coverage for UEFI and UEFI
SecureBoot testing.
Simon McVittie
Colin Walters
Luca BRUNO
Ricardo Noriega
dependabot[bot]
Jonathan Lebon
Manuel Stühn
Dan Nicholson