On Mon, 2007-05-28 at 14:53 +0200, Tristan Gingold wrote:
On Thu, May 10, 2007 at 02:15:34PM -0600, Alex Williamson wrote:
On Wed, 2007-05-09 at 06:10 +0200, Tristan Gingold wrote:
Hi,
a stripped-down version of a previous patch: reimplement
vcpu_get_psr.
Should be a noop, I don't think performance should be affected.
Unfortunately...
with patch:
real 6m0.483s
user 21m10.680s
sys 1m18.470s
without patch:
real 4m47.767s
user 16m55.060s
sys 0m53.240s
Hi,
I can now reproduce this slow-down.
However I don't understand your figures: how can real time be less than user
time ? Isn't real time the elapsed time ?
Parallel build, -j4. The real time is the elapsed time, user is
cumulative across all 4 CPUs. Thanks,
Alex