I was able to replicate this on a simpler example that essentially does
nothing but hook into TS similar to a buffering null transform in both
directions. I will play with it a bit and once I have some colleges take a
quick look I did not mess anything up, I'll file a bug so that others might
take
On Wednesday, March 11, 2015, Nick Kew wrote:
> On Wed, 2015-03-11 at 03:10 -0500, Brian Rectanus wrote:
> > I certainly have some cycles if someone can help point me in the right
> > direction. Right now I am at a loss as to where to dig. Profiler just
> shows
> > a lot of nothing (epoll) happen
With regard to rescheduling and latency:
For ats_pagespeed, changing the fast_signal argument's default from false
to true in I_EThread::schedule helped improve latency when scheduling an
event from an external thread pool (up to 5 ms per scheduled operation).
This ticket [1] at least documents th
Latency problems with transforms could be missing TSVIOReenable() or
TSContCall() to wake up the other sides.
Look very carefully at some of the sample uses of Transforms (null-transform
append-transform) and try to match the pattern.
-William Bardwell
On Wed, 2015-03-11 at 03:10 -0500, Brian Rectanus wrote:
> I certainly have some cycles if someone can help point me in the right
> direction. Right now I am at a loss as to where to dig. Profiler just shows
> a lot of nothing (epoll) happening as you wrote.
Hmmm. Is this with or without asynchro
I certainly have some cycles if someone can help point me in the right
direction. Right now I am at a loss as to where to dig. Profiler just shows
a lot of nothing (epoll) happening as you wrote.
10sec @ 10ms polling under full load on CentOS 7 w/8x X5660 @ 2.80GHz
shows...
total
I've also observed unexplained latency when it comes to transformations, I
think it's time that we dig into this more. The reason we're observing an
increase in latency without a corresponding increase in CPU load is because
TS simply isn't doing anything, it appears that it's just rescheduling
tra
