Re: Inconsistent timing results of multithreaded program on an SMP machine.
From: Con Kolivas
Date: Sun Nov 20 2005 - 05:27:58 EST
On Sun, 20 Nov 2005 21:18, Marcel Zalmanovici wrote:
> Hi Con,
> Thanks for the quick reply.
Please reply below what you are replying to so we can keep track of email
> <kernel@xxxxxxxxx To: Marcel
> Zalmanovici/Haifa/IBM@IBMIL rg> cc:
> linux-kernel@xxxxxxxxxxxxxxx Subject: Re: Inconsistent timing results of
> multithreaded program on an SMP 20/11/2005 11:54 machine.
> On Sun, 20 Nov 2005 20:27, Marcel Zalmanovici wrote:
> > Hi,
> > I am trying, as part of my thesis, to make some improvement to the linux
> > scheduler.
> > Therefore I've written a small multithreaded application and tested how
> > long on average it takes for it to complete.
> > The results were very surprising. I expected to see the completion time
> > vary 1 to 2 seconds at most.
> > Instead what I've got was an oscillation where the maximum time was twice
> > and more than the minimum!! For a short test results ranged ~7sec to ~16
> > sec, and the same happened to longer tests where the min was ~1min and
> > max ~2:30min.
> > Does anyone have any clue as to what might happen ?
> > Is there anything I can do to get stable results ?
> > Here is a small test case program:
> > (See attached file: sched_test.c)
> > The test was always done on a pretty much empty machine. I've tried both
> > kernel 2.6.4-52 and 22.214.171.124 but the results were the same.
> > I'm working on a Xeon Intel machine, dual processor, hyperthreaded.
> I suspect what you're seeing is the random nature of threads to bind either
> a hyperthread sibling or a real core. If all your threads land on only one
> physical cpu and both hyperthread siblings it will run much slower than if
> half land on one physical cpu and half on the other physical cpu. To
> this, try setting cpu affinity just for one logical core of each phyiscal
> (like affinity for cpu 1 and 3 only for example) or disabling hyperthread
> the bios.
Ok I've had a look at the actual program now ;) Are you timing the time it
takes to completion of everything?
This part of your program:
for (i= 0; i<8; i++)
Cares about the order the threads finish. Do you think this might be affecting
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/