Am 10.03.2025 um 11:55 hat Stefan Hajnoczi geschrieben:
> On Fri, Mar 07, 2025 at 11:16:34PM +0100, Kevin Wolf wrote:
> > Adaptive polling has a big problem: It doesn't consider that an event
> > loop can wait for many different events that may have very different
> > typical latencies.
> >
> > Fo
On Fri, Mar 07, 2025 at 11:16:34PM +0100, Kevin Wolf wrote:
> Adaptive polling has a big problem: It doesn't consider that an event
> loop can wait for many different events that may have very different
> typical latencies.
>
> For example, think of a guest that tends to send a new I/O request soo
On Mon, Mar 10, 2025 at 12:11:44PM +0100, Kevin Wolf wrote:
> Am 10.03.2025 um 11:55 hat Stefan Hajnoczi geschrieben:
> > On Fri, Mar 07, 2025 at 11:16:34PM +0100, Kevin Wolf wrote:
> > > Adaptive polling has a big problem: It doesn't consider that an event
> > > loop can wait for many different ev
Adaptive polling has a big problem: It doesn't consider that an event
loop can wait for many different events that may have very different
typical latencies.
For example, think of a guest that tends to send a new I/O request soon
after the previous I/O request completes, but the storage on the hos
