Re: [benchmark] 1% performance overhead of paravirt_ops on nativekernels

[Sander Eikelenboom]

The argument that the 1% perfomance hit for native kernels requires extra work for
package maintainers because they have to make a seperate package for native
(without the performance hit) and a xen version (with paravirt_ops) seems
bogus.

At the moment they also have to provide the 2 packages, 1 native recent
kernel, and 1 older back or forward ported xen kernel (2.6.18 or 2.6.26 for
debian)

So this probably wouldn't change much, although it would be nice for
paravirt_ops to perform like native, i think it shouldn't have to be a show
stopper for inclusion of Xen in mainline.

--
        Sander

--
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/