[RFC PATCH 0/4] treewide: Use clocksource id for get_device_system_crosststamp()

[Peter Hilber]

This patch series changes struct system_counterval_t to identify the
clocksource through enum clocksource_ids, rather than through struct
clocksource *. The net effect of the patch series is that
get_device_system_crosststamp() callers can supply clocksource ids instead
of clocksource pointers, which can be problematic to get hold of.

For this, modify some code which is relevant to
get_device_system_crosststamp(), in timekeeping, ptp/kvm, x86/kvm, and
x86/tsc.

The series does the following: First, introduce clocksource ids for x86 TSC
and kvm-clock. Then, refactor the x86 TSC a tiny bit to keep changes in the
last, "treewide" patch to a minimum. In the treewide patch, replace
system_counterval_t.cs by .cs_id.

This series should not alter any behavior. Out of the existing
get_device_system_crosststamp() users, only ptp_kvm has been tested (on
x86-64 and arm64). This series is a prerequisite for the virtio_rtc driver
(RFC v2 to be posted). Through this series, virtio_rtc can work without
modifying arm_arch_timer.


Peter Hilber (4):
  x86/tsc: Add clocksource ids for TSC and early TSC
  x86/kvm: Add clocksource id for kvm-clock
  x86/tsc: Use bool, not pointer, for ART availability
  treewide: Use clocksource id for struct system_counterval_t

 arch/x86/kernel/kvmclock.c      |  2 ++
 arch/x86/kernel/tsc.c           | 23 +++++++++++++++--------
 drivers/ptp/ptp_kvm_common.c    |  3 ++-
 include/linux/clocksource_ids.h |  3 +++
 include/linux/timekeeping.h     |  4 ++--
 kernel/time/timekeeping.c       |  3 ++-
 6 files changed, 26 insertions(+), 12 deletions(-)


base-commit: 47762f08697484cf0c2f2904b8c52375ed26c8cb
-- 
2.39.2