| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
|
|
|
|
|
|
| |
This isn't quite ready for other buildroots yet, just mark it as Fedora
only.
Signed-off-by: Laura Abbott <labbott@redhat.com>
|
|
|
|
|
|
|
|
| |
We've been doing a workaround of using a uuid for the build-id for
userspace binaries to let us get parallel debuginfo (the trick we
use for the kernel doesn't extend to userspace binaries). This
works but isn't ideal for reproducibility. Given we don't actually
care about this debuginfo, just remove it from the buildroot completely.
|
|
|
|
| |
Fix up some whitespace
|
|
|
|
|
|
| |
Now that we have the rest of the flags hooked up, go ahead
and build with the kpatch flags. This should have a minimal
effect at runtime.
|
|
|
|
|
|
| |
Borrowed from downstream, if we're not doing an arch build and
only building a sub-package don't bother building the kernel
meta package.
|
|
|
|
|
| |
We don't need to do some config work if we're not doing
an arch build.
|
|
|
|
|
| |
Borrowed from downstream, we can be a little bit smarter
on how we compress modules.
|
| |
|
|
|
|
|
| |
This was a workaround for some bad diffs that came in. Upstream
has mostly stopped doing that so we can drop this workaround.
|
|
|
|
| |
The el7 buildroot now has pathfix so we can drop these checks.
|
| |
|
| |
|
|
|
|
|
|
| |
This reverts commit 2147ca93975deaf220619e9096e0b84d879febc9.
Upstream removed the driver, hopefully it'll make it into 5.5.
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The "regular" definition of the with_kabichk macro does not need to
be commented out while it is temporarily (re-)defined as 0 on the
next line. This avoids a "Macro expanded in comment" warning.
Remove comments on the same line as %endif directives. Instead, make
a couple of these conditionals easier to read (and consistent with
the surrounding ones) by deleting a few blank lines.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The test was failing (after hanging for several hours) because of a
hard-coded assumption about which file descriptor would be returned
by memfd_create(). The actual file descriptor gets printed by the
test program though, so read this and use it instead. (Adjust the
test program so that this is not printed until the file is ready.)
Resolve compiler warnings about implicit function declarations, by
using the #include statements listed in the kernel.org man-pages.
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This reverts commit 89c7d2ae30a43ba2960ce14cddd50f8d8a7da92f, which
was causing kernel modules to be installed in a location other than
/lib/modules/$(uname -r), preventing modprobe from locating them.
CI builds no longer contain the Pagure UID in the package release:
https://github.com/CentOS-PaaS-SIG/upstream-fedora-pipeline/pull/199
|
| |
| |
| |
| |
| | |
This is a staging driver that implements the Linux-guest side of folder
sharing.
|
| | |
|
|/ |
|
| |
|
| |
|
|
|
|
|
| |
This enables support for Macbook8,1 and later, along with the
MacbookPro13,* and MacbookPro14,*.
|
| |
|
| |
|
|\ |
|
| |
| |
| |
| |
| | |
This aligns the setting with upstream defconfigs along with the ARK
configuration.
|
|/ |
|
| |
|
| |
|
|\
| |
| |
| | |
1750194)`
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| | |
We don't actually need to be packaging most of these files so
switch to a smaller subset.
|
| |
| |
| |
| |
| | |
These are not used directly in Fedora but for easy of syncing
let's keep them around.
|
| | |
|
| | |
|
| |
| |
| |
| |
| | |
This is a package for dumping on s390x systems. We don't build
this for Fedora but bring it in for now.
|
|/ |
|
| |
|
|
|
|
|
|
|
| |
This reverts commit 8044841427b22ab04ec6dd8781fbe4be2be411df.
There's an issue with the weak-updates script that needs to
be fixed. Revert this for now.
|
| |
|
| |
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
OSTree is a far, far more sophisticated wrapper around the `link()`
system call than the `hardlink` package - it supports using
as a mechanism for transactional offline updates, fetching over
HTTP with GPG signatures and deltas, etc. rpm-ostree uses it
for everything.
Having the `kernel-devel` package run `hardlink` just adds
latency to `rpm-ostree compose tree` unnecessarily.
|
| | |
|