Re: [RFC][PATCH 2/4] time: Fix CLOCK_MONOTONIC_RAW sub-nanosecond accounting

2017-05-30 Thread Daniel Mentz
On Sat, May 27, 2017 at 12:36 AM, Ingo Molnar wrote: > > * John Stultz wrote: > >> + u64 nsecps; > > What does the 'ps' postfix stand for? It's not obvious (to me). > I guess that nsecps stands for "nanoseconds per second", although the code appears

Re: [RFC][PATCH 2/4] time: Fix CLOCK_MONOTONIC_RAW sub-nanosecond accounting

2017-05-30 Thread Daniel Mentz
On Sat, May 27, 2017 at 12:36 AM, Ingo Molnar wrote: > > * John Stultz wrote: > >> + u64 nsecps; > > What does the 'ps' postfix stand for? It's not obvious (to me). > I guess that nsecps stands for "nanoseconds per second", although the code appears to be storing that value left shifted by

Re: [RFC][PATCH 2/4] time: Fix CLOCK_MONOTONIC_RAW sub-nanosecond accounting

2017-05-27 Thread Ingo Molnar
* John Stultz wrote: > Due to how the MONOTONIC_RAW accumulation logic was handled, > there is the potential for a 1ns discontinuity when we do > accumulations. This small discontinuity has for the most part > gone un-noticed, but since ARM64 enabled CLOCK_MONOTONIC_RAW

Re: [RFC][PATCH 2/4] time: Fix CLOCK_MONOTONIC_RAW sub-nanosecond accounting

2017-05-27 Thread Ingo Molnar
* John Stultz wrote: > Due to how the MONOTONIC_RAW accumulation logic was handled, > there is the potential for a 1ns discontinuity when we do > accumulations. This small discontinuity has for the most part > gone un-noticed, but since ARM64 enabled CLOCK_MONOTONIC_RAW > in their vDSO

[RFC][PATCH 2/4] time: Fix CLOCK_MONOTONIC_RAW sub-nanosecond accounting

2017-05-26 Thread John Stultz
Due to how the MONOTONIC_RAW accumulation logic was handled, there is the potential for a 1ns discontinuity when we do accumulations. This small discontinuity has for the most part gone un-noticed, but since ARM64 enabled CLOCK_MONOTONIC_RAW in their vDSO clock_gettime implementation, we've seen

[RFC][PATCH 2/4] time: Fix CLOCK_MONOTONIC_RAW sub-nanosecond accounting

2017-05-26 Thread John Stultz
Due to how the MONOTONIC_RAW accumulation logic was handled, there is the potential for a 1ns discontinuity when we do accumulations. This small discontinuity has for the most part gone un-noticed, but since ARM64 enabled CLOCK_MONOTONIC_RAW in their vDSO clock_gettime implementation, we've seen