On Thu, 20 Mar 2003 11:58, Nick Piggin wrote:
> Con Kolivas wrote:
> >Contest results for mm2:
>
> Contest is starting to look good. Especially in
> loads and lcpu.
Very good point. Some of the loads are exhibiting better overall cpu usage and
dropping the compile time (tar loads are a good example)
> >no_load:
> >Kernel [runs] Time CPU% Loads LCPU% Ratio
> >2.5.65 3 80 95.0 0.0 0.0 1.00
> >2.5.65-mm1 3 79 94.9 0.0 0.0 1.00
> >2.5.65-mm2 3 79 94.9 0.0 0.0 1.00
>
> AS is now on par with deadline here which is nice.
Excellent.
> [snip]
>
> >dbench_load:
> >Kernel [runs] Time CPU% Loads LCPU% Ratio
> >2.5.65 3 542 14.2 9.0 62.5 6.78
> >2.5.65-mm1 3 361 21.1 6.3 55.4 4.57
> >2.5.65-mm2 3 437 17.4 7.7 60.6 5.53
>
> I don't know if this is a good balance shift or not. Maybe not
> related to AS but I'll investigate.
I dont think there are other tweaks in mm2 that could be responsible. It's
extremely hard to know what exactly is the best balance in this load. If
dbench 16 should get 80% of the total cpu usage (since the kernel compile has
4 processes) then mm2 is very close at 78%. However I didnt think dbench was
supposed to be a particularly cpu bound task. The overall cpu usage is higher
which is good though.
> Thanks Con
A pleasure,
Con
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Mar 23 2003 - 22:00:29 EST