Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 18 Sep 2007 01:28:27 -0700 (PDT)
From:      Jeff Roberson <jroberson@chesapeake.net>
To:        Julian Elischer <julian@elischer.org>
Cc:        arch@freebsd.org
Subject:   Re: Changing p_swtime and td_slptime to ticks
Message-ID:  <20070918012555.G558@10.0.0.1>
In-Reply-To: <46EF644E.9050207@elischer.org>
References:  <20070917165657.B558@10.0.0.1> <46EF644E.9050207@elischer.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 17 Sep 2007, Julian Elischer wrote:

> Jeff Roberson wrote:
>> Enclosed is a patch that fixes swapping with ULE.  ULE has never properly 
>> set p_swtime and td_slptime which are used by the swapout/swapin code to 
>> select the appropriate thread to swap.
>
> I have not looked at in the depth required, but 2 points that I was unable
> to check to my satisfaction before I got called away for work....
>
> 1/ the source of the ticks is a monotonically increasing count that never
> goes backwards or changes?

ticks is incremented each time hardclock() is called.  That's it.

>
> 2/ nothing that used to be accounted in seconds becomes accounted for in 
> ticks?

I scale back to seconds where it is required.  Really I think ticks would 
be the better metric in vm_glue.c but didn't want to make any drastic 
changes.

Thanks,
Jeff

>
>
>> 
>> In 4BSD these two variables are increment once per-second as schedcpu() 
>> iterates over all threads.  ULE does not have a once per-second loop 
>> iterating over all threads.  So I have changed p_swtime to p_swtick and 
>> td_slptime to td_slptick.  These record the value of 'ticks' when the 
>> thread slept or was last swapped in or out.
>> 
>> For backwards compatibility I leave the values in kinfo_proc with the 
>> legacy meaning by subtracting from ticks and dividing by hz.  I perform a 
>> similar transformation in the swapout code to convert to seconds.  This 
>> change does make it possible to use sub-second granular decisions in the 
>> swap code, however I'm not sure if that's really necessary.
>> 
>> So that I did not disturb the 4BSD mechanism I kept the original td_slptime 
>> in the td_sched area.  It should be possible to use td_slptick directly but 
>> especially this close to release I did not want to change 4BSD.
>> 
>> Feedback and testing welcome.
>> 
>> Thanks,
>> Jeff
>> 
>> 
>> ------------------------------------------------------------------------
>> 
>> _______________________________________________
>> freebsd-arch@freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-arch
>> To unsubscribe, send any mail to "freebsd-arch-unsubscribe@freebsd.org"
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070918012555.G558>