>On 24/07/2017 12:38, wang.yong...@zte.com.cn wrote: >> finally use g_main_loop_run to replace aio_poll in the
>> iothread_run function. > >That would make the performance of virtio-blk with iothreads worse, >unfortunately. aio_poll is much more optimized than g_main_loop_run. Hi Paolo, Any other good idea to achieve this? Thanks > >Paolo > >> After that IOThread >> runs the GMainContext event loop,chardev and IOThread can work together. >> >> How about it? If feasible, I will try to submit 原始邮件 发件人： <pbonz...@redhat.com> 收件人：王勇10170530 <stefa...@redhat.com> <jasow...@redhat.com> <marcandre.lur...@gmail.com> 抄送人： <zhang.zhanghaili...@huawei.com> <zhangchen.f...@cn.fujitsu.com> <lizhij...@cn.fujitsu.com> <qemu-devel@nongnu.org>王广10165992 日 期 ：2017年07月24日 19:59 主 题 ：Re: [PATCHv2 02/04] colo-compare: Processpactkets in the IOThreadofthe primary On 24/07/2017 12:38, wang.yong...@zte.com.cn wrote: > finally use g_main_loop_run to replace aio_poll in the > iothread_run function. That would make the performance of virtio-blk with iothreads worse, unfortunately. aio_poll is much more optimized than g_main_loop_run. Paolo > After that IOThread > runs the GMainContext event loop,chardev and IOThread can work together. > > How about it? If feasible, I will try to submit a patch.
