All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jason Wang <jasowang@redhat.com>
To: Yongji Xie <xieyongji@bytedance.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	sgarzare@redhat.com, Parav Pandit <parav@nvidia.com>,
	akpm@linux-foundation.org, Randy Dunlap <rdunlap@infradead.org>,
	Matthew Wilcox <willy@infradead.org>,
	viro@zeniv.linux.org.uk, axboe@kernel.dk, bcrl@kvack.org,
	corbet@lwn.net, virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org, kvm@vger.kernel.org, linux-aio@kvack.org,
	linux-fsdevel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [RFC v2 09/13] vduse: Add support for processing vhost iotlb message
Date: Tue, 29 Dec 2020 04:11:08 -0500 (EST)	[thread overview]
Message-ID: <1356137727.40748805.1609233068675.JavaMail.zimbra@redhat.com> (raw)
In-Reply-To: <CACycT3soQoX5avZiFBLEGBuJpdni6-UxdhAPGpWHBWVf+dEySg@mail.gmail.com>



----- Original Message -----
> On Mon, Dec 28, 2020 at 4:43 PM Jason Wang <jasowang@redhat.com> wrote:
> >
> >
> > On 2020/12/28 下午4:14, Yongji Xie wrote:
> > >> I see. So all the above two questions are because VHOST_IOTLB_INVALIDATE
> > >> is expected to be synchronous. This need to be solved by tweaking the
> > >> current VDUSE API or we can re-visit to go with descriptors relaying
> > >> first.
> > >>
> > > Actually all vdpa related operations are synchronous in current
> > > implementation. The ops.set_map/dma_map/dma_unmap should not return
> > > until the VDUSE_UPDATE_IOTLB/VDUSE_INVALIDATE_IOTLB message is replied
> > > by userspace. Could it solve this problem?
> >
> >
> >   I was thinking whether or not we need to generate IOTLB_INVALIDATE
> > message to VDUSE during dma_unmap (vduse_dev_unmap_page).
> >
> > If we don't, we're probably fine.
> >
> 
> It seems not feasible. This message will be also used in the
> virtio-vdpa case to notify userspace to unmap some pages during
> consistent dma unmapping. Maybe we can document it to make sure the
> users can handle the message correctly.

Just to make sure I understand your point.

Do you mean you plan to notify the unmap of 1) streaming DMA or 2)
coherent DMA?

For 1) you probably need a workqueue to do that since dma unmap can
be done in irq or bh context. And if usrspace does't do the unmap, it
can still access the bounce buffer (if you don't zap pte)?

Thanks


> 
> Thanks,
> Yongji
> 
> 


WARNING: multiple messages have this Message-ID (diff)
From: Jason Wang <jasowang@redhat.com>
To: Yongji Xie <xieyongji@bytedance.com>
Cc: axboe@kernel.dk, corbet@lwn.net, kvm@vger.kernel.org,
	"Michael S. Tsirkin" <mst@redhat.com>,
	linux-aio@kvack.org, netdev@vger.kernel.org,
	Randy Dunlap <rdunlap@infradead.org>,
	Matthew Wilcox <willy@infradead.org>,
	virtualization@lists.linux-foundation.org, linux-mm@kvack.org,
	bcrl@kvack.org, viro@zeniv.linux.org.uk,
	Stefan Hajnoczi <stefanha@redhat.com>,
	linux-fsdevel@vger.kernel.org, akpm@linux-foundation.org
Subject: Re: [RFC v2 09/13] vduse: Add support for processing vhost iotlb message
Date: Tue, 29 Dec 2020 04:11:08 -0500 (EST)	[thread overview]
Message-ID: <1356137727.40748805.1609233068675.JavaMail.zimbra@redhat.com> (raw)
In-Reply-To: <CACycT3soQoX5avZiFBLEGBuJpdni6-UxdhAPGpWHBWVf+dEySg@mail.gmail.com>



----- Original Message -----
> On Mon, Dec 28, 2020 at 4:43 PM Jason Wang <jasowang@redhat.com> wrote:
> >
> >
> > On 2020/12/28 下午4:14, Yongji Xie wrote:
> > >> I see. So all the above two questions are because VHOST_IOTLB_INVALIDATE
> > >> is expected to be synchronous. This need to be solved by tweaking the
> > >> current VDUSE API or we can re-visit to go with descriptors relaying
> > >> first.
> > >>
> > > Actually all vdpa related operations are synchronous in current
> > > implementation. The ops.set_map/dma_map/dma_unmap should not return
> > > until the VDUSE_UPDATE_IOTLB/VDUSE_INVALIDATE_IOTLB message is replied
> > > by userspace. Could it solve this problem?
> >
> >
> >   I was thinking whether or not we need to generate IOTLB_INVALIDATE
> > message to VDUSE during dma_unmap (vduse_dev_unmap_page).
> >
> > If we don't, we're probably fine.
> >
> 
> It seems not feasible. This message will be also used in the
> virtio-vdpa case to notify userspace to unmap some pages during
> consistent dma unmapping. Maybe we can document it to make sure the
> users can handle the message correctly.

Just to make sure I understand your point.

Do you mean you plan to notify the unmap of 1) streaming DMA or 2)
coherent DMA?

For 1) you probably need a workqueue to do that since dma unmap can
be done in irq or bh context. And if usrspace does't do the unmap, it
can still access the bounce buffer (if you don't zap pte)?

Thanks


> 
> Thanks,
> Yongji
> 
> 

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

  reply	other threads:[~2020-12-29  9:13 UTC|newest]

Thread overview: 97+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-22 14:52 [RFC v2 00/13] Introduce VDUSE - vDPA Device in Userspace Xie Yongji
2020-12-22 14:52 ` [RFC v2 01/13] mm: export zap_page_range() for driver use Xie Yongji
2020-12-22 15:44   ` Christoph Hellwig
2020-12-22 15:44     ` Christoph Hellwig
2020-12-22 14:52 ` [RFC v2 02/13] eventfd: track eventfd_signal() recursion depth separately in different cases Xie Yongji
2020-12-22 14:52 ` [RFC v2 03/13] eventfd: Increase the recursion depth of eventfd_signal() Xie Yongji
2020-12-22 14:52 ` [RFC v2 04/13] vdpa: Remove the restriction that only supports virtio-net devices Xie Yongji
2020-12-22 14:52 ` [RFC v2 05/13] vdpa: Pass the netlink attributes to ops.dev_add() Xie Yongji
2020-12-22 14:52 ` [RFC v2 06/13] vduse: Introduce VDUSE - vDPA Device in Userspace Xie Yongji
2020-12-23  8:08   ` Jason Wang
2020-12-23  8:08     ` Jason Wang
2020-12-23 14:17     ` Yongji Xie
2020-12-23 14:17       ` Yongji Xie
2020-12-24  3:01       ` Jason Wang
2020-12-24  3:01         ` Jason Wang
2020-12-24  8:34         ` Yongji Xie
2020-12-24  8:34           ` Yongji Xie
2020-12-25  6:59           ` Jason Wang
2020-12-25  6:59             ` Jason Wang
2021-01-08 13:32   ` Bob Liu
2021-01-08 13:32     ` Bob Liu
2021-01-10 10:03     ` Yongji Xie
2021-01-10 10:03       ` Yongji Xie
2020-12-22 14:52 ` [RFC v2 07/13] vduse: support get/set virtqueue state Xie Yongji
2020-12-22 14:52 ` [RFC v2 08/13] vdpa: Introduce process_iotlb_msg() in vdpa_config_ops Xie Yongji
2020-12-23  8:36   ` Jason Wang
2020-12-23  8:36     ` Jason Wang
2020-12-23 11:06     ` Yongji Xie
2020-12-23 11:06       ` Yongji Xie
2020-12-24  2:36       ` Jason Wang
2020-12-24  2:36         ` Jason Wang
2020-12-24  7:24         ` Yongji Xie
2020-12-24  7:24           ` Yongji Xie
2020-12-22 14:52 ` [RFC v2 09/13] vduse: Add support for processing vhost iotlb message Xie Yongji
2020-12-23  9:05   ` Jason Wang
2020-12-23  9:05     ` Jason Wang
2020-12-23 12:14     ` [External] " Yongji Xie
2020-12-23 12:14       ` Yongji Xie
2020-12-24  2:41       ` Jason Wang
2020-12-24  2:41         ` Jason Wang
2020-12-24  7:37         ` Yongji Xie
2020-12-24  7:37           ` Yongji Xie
2020-12-25  2:37           ` Yongji Xie
2020-12-25  2:37             ` Yongji Xie
2020-12-25  7:02             ` Jason Wang
2020-12-25  7:02               ` Jason Wang
2020-12-25 11:36               ` Yongji Xie
2020-12-25 11:36                 ` Yongji Xie
2020-12-25  6:57           ` Jason Wang
2020-12-25  6:57             ` Jason Wang
2020-12-25 10:31             ` Yongji Xie
2020-12-25 10:31               ` Yongji Xie
2020-12-28  7:43               ` Jason Wang
2020-12-28  7:43                 ` Jason Wang
2020-12-28  8:14                 ` Yongji Xie
2020-12-28  8:14                   ` Yongji Xie
2020-12-28  8:43                   ` Jason Wang
2020-12-28  8:43                     ` Jason Wang
2020-12-28  9:12                     ` Yongji Xie
2020-12-28  9:12                       ` Yongji Xie
2020-12-29  9:11                       ` Jason Wang [this message]
2020-12-29  9:11                         ` Jason Wang
2020-12-29  9:11                         ` Jason Wang
2020-12-29 10:26                         ` Yongji Xie
2020-12-29 10:26                           ` Yongji Xie
2020-12-30  6:10                           ` Jason Wang
2020-12-30  6:10                             ` Jason Wang
2020-12-30  7:09                             ` Yongji Xie
2020-12-30  7:09                               ` Yongji Xie
2020-12-30  8:41                               ` Jason Wang
2020-12-30  8:41                                 ` Jason Wang
2020-12-30 10:12                                 ` Yongji Xie
2020-12-30 10:12                                   ` Yongji Xie
2020-12-31  2:49                                   ` Jason Wang
2020-12-31  2:49                                     ` Jason Wang
2020-12-31  5:15                                     ` Yongji Xie
2020-12-31  5:15                                       ` Yongji Xie
2020-12-31  5:49                                       ` Jason Wang
2020-12-31  5:49                                         ` Jason Wang
2020-12-31  6:52                                         ` Yongji Xie
2020-12-31  6:52                                           ` Yongji Xie
2020-12-31  7:11                                           ` Jason Wang
2020-12-31  7:11                                             ` Jason Wang
2020-12-31  8:00                                             ` Yongji Xie
2020-12-31  8:00                                               ` Yongji Xie
2020-12-22 14:52 ` [RFC v2 10/13] vduse: grab the module's references until there is no vduse device Xie Yongji
2020-12-22 14:52 ` [RFC v2 11/13] vduse/iova_domain: Support reclaiming bounce pages Xie Yongji
2020-12-22 14:52 ` [RFC v2 12/13] vduse: Add memory shrinker to reclaim " Xie Yongji
2020-12-22 14:52 ` [RFC v2 13/13] vduse: Introduce a workqueue for irq injection Xie Yongji
2020-12-23  6:38 ` [RFC v2 00/13] Introduce VDUSE - vDPA Device in Userspace Jason Wang
2020-12-23  6:38   ` Jason Wang
2020-12-23  8:14   ` Jason Wang
2020-12-23  8:14     ` Jason Wang
2020-12-23 10:59   ` Yongji Xie
2020-12-23 10:59     ` Yongji Xie
2020-12-24  2:24     ` Jason Wang
2020-12-24  2:24       ` Jason Wang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1356137727.40748805.1609233068675.JavaMail.zimbra@redhat.com \
    --to=jasowang@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=axboe@kernel.dk \
    --cc=bcrl@kvack.org \
    --cc=corbet@lwn.net \
    --cc=kvm@vger.kernel.org \
    --cc=linux-aio@kvack.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mst@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=parav@nvidia.com \
    --cc=rdunlap@infradead.org \
    --cc=sgarzare@redhat.com \
    --cc=stefanha@redhat.com \
    --cc=viro@zeniv.linux.org.uk \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=willy@infradead.org \
    --cc=xieyongji@bytedance.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.