On 27.06.24 18:03, David Woodhouse wrote:

I've updated the tree at
https://git.infradead.org/users/dwmw2/linux.git/shortlog/refs/heads/vmclock
(but not yet the qemu one).

I think I've taken into account all your comments apart from the one
about non-64-bit counters wrapping. I reduced the seq_count to 32 bit
to make room for a 32-bit flags field, added the time type
(UTC/TAI/MONOTONIC) and a smearing hint, with some straw man
definitions for smearing algorithms for which I could actually find
definitions.

The structure now looks like this:


struct vmclock_abi {

[...]

	/*
	 * What time is exposed in the time_sec/time_frac_sec fields?
	 */
	uint8_t time_type;
#define VMCLOCK_TIME_UNKNOWN		0	/* Invalid / no time exposed */
#define VMCLOCK_TIME_UTC		1	/* Since 1970-01-01 00:00:00z */
#define VMCLOCK_TIME_TAI		2	/* Since 1970-01-01 00:00:00z */
#define VMCLOCK_TIME_MONOTONIC		3	/* Since undefined epoch */

	/* Bit shift for counter_period_frac_sec and its error rate */
	uint8_t counter_period_shift;

	/*
	 * Unlike in NTP, this can indicate a leap second in the past. This
	 * is needed to allow guests to derive an imprecise clock with
	 * smeared leap seconds for themselves, as some modes of smearing
	 * need the adjustments to continue even after the moment at which
	 * the leap second should have occurred.
	 */
	int8_t leapsecond_direction;
	uint64_t leapsecond_tai_sec; /* Since 1970-01-01 00:00:00z */

	/*
	 * Paired values of counter and UTC at a given point in time.
	 */
	uint64_t counter_value;
	uint64_t time_sec; /* Since 1970-01-01 00:00:00z */

Nitpick: The comment is not valid any more for TIME_MONOTONIC.