Re: [Qemu-devel] [PATCH for-2.3 0/5] aio: Support epoll by introducing qemu_poll abstraction

[Paolo Bonzini <pbonzini@redhat.com>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



On 25/11/2014 14:52, Stefan Hajnoczi wrote:
> Do you think it is okay to forget about <1 ms timeout precision?
> 
> If we go ahead with this, we'll need to rethink other timeouts in 
> QEMU. For example, is there a point in setting timer slack to 1 ns 
> if we cannot even specify ns wait times?
> 
> Perhaps timerfd is needed before we can use epoll.  Hopefully the 
> overall performance effect will be positive with epoll + timerfd, 
> compared to ppoll().

You can also use POLLIN with the epoll file descriptor, i.e. do ppoll
followed (if there's no timeout) by epoll_wait with zero timeout.

Paolo
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJUdIqEAAoJEL/70l94x66DdtsH/RIXbk6faPWdb3MXaHmAoH1E
z7q8cGuLPkI7XT54BYbiBFFn4MqS0XxLHLJ69CksFEC5u4wNl9vqsLLJrN/+uZe5
PznE7madjam32ZVtUbzfRwrBtO0KFgyXEiZfR9stAVXW+/KIUAaWU5rQ2IW6GqHg
skt2GGmKfrCbyvmxVhSt2oMDRZ7O2Tquox6eLYizQX6JJ3/5vDqpzXTKE/Ix+wnt
R3FA3IZkQuZMQPAFsMKj0AajN178RGiqXaB3UIR2YmPN1DWyjkfN05WmPgMpvZa/
eX70AhUdOjLzncfLU3bX9EAsml0s2Hsj5gKRYT6B5d8YK2b0ba3dCqgZRPV3+bo=
=8Awx
-----END PGP SIGNATURE-----