[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CACycT3u5x1wJi8Q1Pm88ckivEUHV11jrzrsaU7JRmVP73M9+DA@mail.gmail.com>
Date: Mon, 19 Dec 2022 14:39:09 +0800
From: Yongji Xie <xieyongji@...edance.com>
To: Jason Wang <jasowang@...hat.com>
Cc: "Michael S. Tsirkin" <mst@...hat.com>,
Thomas Gleixner <tglx@...utronix.de>,
Christoph Hellwig <hch@....de>,
virtualization <virtualization@...ts.linux-foundation.org>,
linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 09/11] vduse: Add enable_irq_wq sysfs interface for virtqueues
On Fri, Dec 16, 2022 at 1:43 PM Jason Wang <jasowang@...hat.com> wrote:
>
> On Mon, Dec 5, 2022 at 5:03 PM Xie Yongji <xieyongji@...edance.com> wrote:
> >
> > Add enable_irq_wq sysfs interface to control whether
> > use workqueue to inject irq or not. The vhost-vdpa case
> > can benefit from it.
>
> Do we have a benchmark result for this?
>
It can reduce 2~3 us latency in our sync I/O test.
> Or I wonder if we can extend set_vq_cb() by associating an eventfd
> then VDUSE can signal that eventfd directly?
>
It looks good to me. I can try this way in v3.
Thanks,
Yongji
Powered by blists - more mailing lists