Hey Dave,
----- Original Message -----
> On 03/11/2015 09:38 PM, Nathan Scott wrote:
> > Hi guys,
> >
> > 828 and 833 are new failures from overnight commits - they look
> > likely to be just .out file issues, but I see 828 at least has
> > already been looked at (but still fails). 833 might be a lack
> > of valgrind installed on your test machine? - looks like its got
> > incorrect .out from not being run anyway; .bad files attached.
> >
> > Also, ISTR at one point Ken asking for additional performance
> > data before this change was made, but don't remember seeing any
> > go past - was/is everyone content with this change? Thanks.
> >
> Sorry, if this was premature. I had read Ken's responses to Frank's
> performance data as positive and I also read that Ken had pulled these
> changes and reviewed them favourably.
Yep, I don't think it was necessarily premature (except for those QA
failures :) - though I assume if Ken was finished with it, he'd have
merged it.
I mainly just want to make sure all bases were covered and since it
touches some of the most critical code paths in all PCP (potentially
regressing single-threaded performance, etc) that those questions were
fully addressed. e.g. is live mode affected positively / negatively?
how much of those reported times was I/O time - are we CPU bound yet
with warm cache? were the reported numbers warm/cold cached archives,
is there more work to do? ... that sort of thing (well, that's what I
would be looking for, maybe Ken had other ideas). More comprehensive
performance data I think is what Ken asked for, and I don't remember
seeing that so far, hence the follow-up - no big deal, I'm sure it'll
be forthcoming.
cheers.
--
Nathan
|