Re: [PATCH v3 2/3] x86/hyperv: move TSC reading method to asm/mshyperv.h

[Thomas Gleixner]

On Fri, 3 Mar 2017, Stephen Hemminger wrote:
> static inline u64 hv_read_tsc_page(const struct ms_hyperv_tsc_page *tsc_pg)
> {
> 	u64 scale, offset, cur_tsc;
> 	u32 start;
> 
> 	/*
> 	 * The protocol for reading Hyper-V TSC page is specified in Hypervisor
> 	 * Top-Level Functional Specification ver. 3.0 and above. To get the
> 	 * reference time we must do the following:
> 	 * - READ ReferenceTscSequence
> 	 *   A special '0' value indicates the time source is unreliable and we
> 	 *   need to use something else. The currently published specification
> 	 *   versions (up to 4.0b) contain a mistake and wrongly claim '-1'
> 	 *   instead of '0' as the special value, see commit c35b82ef0294.
> 	 * - ReferenceTime =
> 	 *        ((RDTSC() * ReferenceTscScale) >> 64) + ReferenceTscOffset
> 	 * - READ ReferenceTscSequence again. In case its value has changed
> 	 *   since our first reading we need to discard ReferenceTime and repeat
> 	 *   the whole sequence as the hypervisor was updating the page in
> 	 *   between.
> 	 */
> 	do {
> 		start = READ_ONCE(tsc_pg->tsc_sequence);
> 		smp_rmb();
> 
> 		if (unlikely(!start))
> 			return U64_MAX;
> 
> 		scale = tsc_pg->tsc_scale;
> 		offset = tsc_pg->tsc_offset;
> 
> 		/*
> 		 * Make sure we read sequence after we read all other values
> 		 * from TSC page.
> 		 */
> 		smp_rmb();
> 	} while (unlikely(READ_ONCE(tsc_pg->tsc_sequence != start)));
> 
> 	cur_tsc = rdtsc_ordered();

That's wrong. You need to read the TSC value together with the scale and
offset. That's needs to be "atomic". You can only do the mult/shift
outside.

> 	return mul_u64_u64_shr(cur_tsc, scale, 64) + offset;
> }

Thanks,

	tglx