Re:63 Cores Blocked by Seven Instructions
by brucedawson on Wednesday October 23, 2019 @03:07AM
(#21331928)
I have seen no signs of samples hitting the most expensive instruction in ETW trace. Across other cases I have looked at the samples tend to cluster after expensive instructions, not on them.
I guess it's possible that all of the cases I looked at were distorted by macro fusion but I don't think so.
Re:63 Cores Blocked by Seven Instructions
by BeeOnRope on Wednesday October 23, 2019 @11:41AM
(#21335756)
Interesting, so I guess ETW is using "normal" interrupts which have skid, and the CPU just having enough retire bandwidth to retire everything in one cycle.
Re:63 Cores Blocked by Seven Instructions
by Const-me on Tuesday October 22, 2019 @09:37AM
(#21323426)
> macro fuse on this CPU, so you actually will never get any samples on the first of the two fused instructions
Haven’t thought about µ-ops fusion in this case. Yes, that explanation is very plausible.
I believe cmp/jne macro fuse on this CPU, so you actually will never get any samples on the first of the two fused instructions: rather they all show on the second one. You see this same effect on Linux when sampling with the cycles:ppp event.