[racket-dev] [plt] Push #24367: master branch updated

From: Ryan Culpepper (ryan at cs.utah.edu)
Date: Tue Feb 28 15:43:37 EST 2012

On my machine before the change, "raco setup -D" took 8m13s real, 13m52s 
user; after the change, it takes 4m0s real, 9m3s user.

It's useful to look at the progress output, particularly the place that 
a particular collection is scheduled on. Before, progress would stop on 
the "macro-debugger" collection for a long time. Now, you can still see 
individual places blocking for long periods of time by looking for long 
runs where a place number is absent.

On my machine, the "images" collection is scheduled on place 0, and it 
keeps 0 occupied for a long time---until around when setup hits 
"scribblings". Then place 2 is occupied by "gui-debugger", then 1 is 
occupied by "macro-debugger". CPUs utilization drops to around 75% when 
it hits "gui-debugger" and then to around 50% at "macro-debugger". Both 
of those collections depend on "images". It drops again, briefly, around 
"scribble", which is also around when "images" finishes---I'm not sure 
what that means. It also takes a while to climb back to around 100%, and 
I'm not sure why that happens either (possibly related to 
"scribble"/"scribblings"?).

It seems like it should help to issue more collections while those few 
block, but doing it naively (-j 5 or -j 6) doesn't work.

Ryan


On 02/28/2012 11:22 AM, Robby Findler wrote:
> On Tue, Feb 28, 2012 at 7:40 AM, Matthew Flatt<mflatt at cs.utah.edu>  wrote:
>> At Tue, 28 Feb 2012 07:21:24 -0600, Robby Findler wrote:
>>> Last I heard, Eli was saying that there was something seriously wrong
>>> with 'raco setup' on two cores. Did that ever get resolved?
>>
>> Commits 012ef60cd545ba and 534886dbe4b6ad (yesterday) were in response
>> and improved things on my machine, so it's probably worth checking again.
>>
>
> Here's what I get. Looks like we see improvements to 3 or so places
> and then don't anymore. But there also doesn't seem to be any bad
> behavior in there, at least for me (on dual quad-core 2.8 ghz xeon
> processors).
>
> time raco setup -Dj 1
> [...]
> real	17m26.143s
> user	14m19.715s
> sys	2m39.913s
>
>
> time raco setup -Dj 2
> [...]
> real	9m16.022s
> user	13m14.917s
> sys	3m49.053s
>
>
> time raco setup -Dj 3
> [...]
> real	7m57.370s
> user	13m50.130s
> sys	5m15.396s
>
> time raco setup -Dj 4
> [...]
> real	7m19.843s
> user	14m12.368s
> sys	6m45.303s
>
>
> time raco setup -Dj 5
> [...]
> real	7m12.761s
> user	14m43.506s
> sys	8m47.768s
>
>
> time raco setup -Dj 6
> [...]
> real	7m16.931s
> user	15m9.667s
> sys	10m46.150s
>
> time raco setup -Dj 7
> [...]
> real	7m29.394s
> user	15m30.409s
> sys	12m44.665s
>
> time raco setup -Dj 8
> [...]
> real	7m29.394s
> user	15m30.409s
> sys	12m44.665s
>
> Robby


Posted on the dev mailing list.