On 04/08/14 11:16, Nathan Scott wrote:
Hi Ken,
I was having a (very!) detailed discussion with someone after
the PyCon.AU talk yesterday, who was quite interested in the
event tracing support in PCP. One issue he raised was our use
of microsecond-resolution timestamps (over nanoseconds, which
he'd found valuable in the past with other tools).
From a quick audit, this time-resolution is fairly well baked
into the protocol and on-disk format now - any thoughts on how
one might best go about increasing this resolution? (I'm only
thinking events here, and not the top level pmResult timestamp
- its not clear the latter would realistically benefit from a
finer granularity in practice).
The best I've come up with so far would involve a __pmTimespec,
a pmHighResEventRecord, and a new PM_TYPE_HIGHRES_EVENT (PMDAs
could then use clock_gettime(2) and/or preserve the resolution
of a nanosec timestamp delivered to them) ... any alternatives,
or better ideas that spring to mind?
Not much choice other than what you've proposed I think.
For me ...
kenj@bozo:/tmp$ a.out
sizeof: __pmTimeval 8 timeval 16 timespec 16
|