UBUNTU: SAUCE: x86/cpu: Add support for modern x86-64-v* march
Add support for setting march to x86-64-v2, x86-64-v3, x86-64-v4 with
tuning set to an early family of CPUs that support such instruction
levels. By default gcc sets generic tuning for x86-64-v*, which is
suboptimal for all brands of CPUs with such instruction set support.
Signed-off-by: Dimitri John Ledkov <email address hidden>
dkms upstream, debian, lunar-proposed is changing return code for skip
from 9 to 77. To support either editions of the dkms, update
build_dkms invocation to support either return codes.
lunar kernel started to export MAKEFLAGS to speed up and optimize our
packaging builds. However, this can cause build_dkms to FTBFS if a
given dkms module uses submake and relies on built-in Make variables
to be set. Address this by unsetting MAKEFLAGS when invoking
build-dkms.
Signed-off-by: Dimitri John Ledkov <email address hidden>
Signed-off-by: Andrea Righi <email address hidden>
CONFIG_EXPOLINE_EXTERN has been enabled for s390x [LP:#2013209]. While
this works as expected on the base kernels, it might not work on some
derivatives: for example focal:hwe-5.15 (derived from Jammy). On Focal,
this config can't be enabled due to the GCC version it comes with.
CONFIG_EXPOLINE_EXTERN requires >= 110200 while Focal comes with 90400.
This change adds a check on the existence of the expoline object file(s)
that are packaged later. This also makes the code more robust generally.
Signed-off-by: Andrei Gherzan <email address hidden>
Signed-off-by: Andrea Righi <email address hidden>
Commit 65b32f801bfb ("uapi: move IPPROTO_L2TP to in.h") moved the
definition of IPPROTO_L2TP from a define to an enum, but since
__stringify doesn't work properly with enums, we ended up breaking the
modalias strings for the l2tp modules: