ULE vs. 4BSD in RELENG_7
jroberson at chesapeake.net
Sun Nov 4 15:30:57 PST 2007
On Sun, 4 Nov 2007, Josh Carroll wrote:
>> Josh, I included one too many changes in the diff and it made the results
>> ambiguous. I've scaled it back slightly by removing the changes to
>> sched_pickcpu() and included the patch in this email again. Can you run
>> through your tests once more? I'd like to commit this part soon as it
>> helps in other cases. I'm most interested in the buildworld numbers for
> Sure, I can run through them. Should I change the slice_min value at
> all during testing to compare different values? Or is that part of the
> other patch that is not included in this (haven't fetched and grep'd
> it yet). :)
> I'll get the benchmark numbers as soon as I can.
Turns out the last patch I posted had a small compile error because I
edited it by hand to remove one section. Here's an updated patch that
fixes that and changes the min/max slice values to something more
reasonable. Slice min should be around 4 with a max of 12.
Also looks like 4BSD's slice handling got more precise recently as well.
jhb changed it from using a callout to an actual counter based on runtime.
> freebsd-performance at freebsd.org mailing list
> To unsubscribe, send any mail to "freebsd-performance-unsubscribe at freebsd.org"
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 19404 bytes
Url : http://lists.freebsd.org/pipermail/freebsd-performance/attachments/20071104/fafcdd5a/uleslice.bin
More information about the freebsd-performance