hv_{kvp,vss}_daemon used to communicate with the corresponding kernel
drivers over netlink, but now they use char devices. hv_fcopy_daemon
always used a char device. Rather than checking for Hyper-V
specifically, change all of the init scripts and systemd service
definitions to check for the appropriate device nodes.
Delete the check-hyperv program that we used to check for Hyper-V
in init scripts.
CONFIG_DEBUG_INFO and CONFIG_MODULE_SIG are added in gencontrol.py,
so be consistent with that.
This unfortunately requires some ugly escaping of quotes.
Checksumming the whole of debian/changelog when deciding whether to
run gencontrol.py results in (a) frequent changes to control.md5sum
and (b) the need to invoke various targets twice during development.
I originally made this change to address (a), which would be an
annoyance if and when we start using dgit. However, fixing (b) is a
nice benefit regardless of whether we do that.
The rt patch "of: allocate / free phandle cache outside of the
devtree_lock" drops the lock earlier in of_populate_phandle_cache()
and removes the "out" label which is no longer needed.
4.18.10 includes "of: fix phandle cache creation for DTs with no
phandles" which adds another "goto out" inside the locked section.
The previous textual conflict resolution between these changes left
the "goto out" in place, but it needs to be a "return".
Gencontrol.parse_changelog() makes a list of all Debian versions with
the same upstream version, but doesn't use it. This is left over from
the old multiple-series patch system which was removed in 2012.
SchemaItemBoolean and SchemaItemInteger attempt to raise an exception
of type Error when given invalid input, but this type has never been
defined. Use ValueError instead.
MakeFlags.__repr__ references a "flags" type (probably meant to be
MakeFlags). Gencontrol.write_config references the "file" type which
was removed in Python 3. Clearly neither of these methods are
actually used, so delete them.
- Add explicit imports for all needed modules, rather than indirectly
(accidentally!) importing them with "from ... import *"
- Replace all "from ... import *" statements, which inhibit static
checking, with explicit lists of names to import
- Delete the remaining unneeded imports reported by pyflakes