All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vivek Goyal <vgoyal@redhat.com>
To: JeffleXu <jefflexu@linux.alibaba.com>
Cc: miklos@szeredi.hu, "Dr. David Alan Gilbert" <dgilbert@redhat.com>,
	virtualization@lists.linux-foundation.org,
	joseph.qi@linux.alibaba.com, bo.liu@linux.alibaba.com,
	stefanha@redhat.com, linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH v2 0/4] virtiofs,fuse: support per-file DAX
Date: Wed, 21 Jul 2021 10:42:50 -0400	[thread overview]
Message-ID: <YPgyalU0avl9KI/U@redhat.com> (raw)
In-Reply-To: <YPgXuacFfJ/JVRjo@redhat.com>

On Wed, Jul 21, 2021 at 08:48:57AM -0400, Vivek Goyal wrote:
[..]
> > > So is "dax=inode" enough for your needs? What's your requirement,
> > > can you give little bit of more details.
> > 
> > In our use case, the backend fs is something like SquashFS on host. The
> > content of the file on host is downloaded *as needed*. When the file is
> > not completely ready (completely downloaded), the guest will follow the
> > normal IO routine, i.e., by FUSE_READ/FUSE_WRITE request. While the file
> > is completely ready, per-file DAX is enabled for this file. IOW the FUSE
> > server need to dynamically decide if per-file DAX shall be enabled,
> > depending on if the file is completely downloaded.
> 
> So you don't want to enable DAX yet because guest might fault on
> a section of file which has not been downloaded yet?
> 
> I am wondering if somehow user fault handling can help with this.
> If we could handle faults for this file in user space, then you
> should be able to download that particular page[s] and resolve
> the fault?

Stefan mentioned that can't we block when fuse mmap request comes
in and download corresponding section of file. Or do whatever you
are doing in FUSE_READ. 

IOW, even if you enable dax in your use case on all files,
FUSE_SETUPMAPPING request will give you control to make sure 
file section being mmaped has been downloaded.

Vivek

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

WARNING: multiple messages have this Message-ID (diff)
From: Vivek Goyal <vgoyal@redhat.com>
To: JeffleXu <jefflexu@linux.alibaba.com>
Cc: stefanha@redhat.com, miklos@szeredi.hu,
	linux-fsdevel@vger.kernel.org,
	virtualization@lists.linux-foundation.org,
	bo.liu@linux.alibaba.com, joseph.qi@linux.alibaba.com,
	"Dr. David Alan Gilbert" <dgilbert@redhat.com>
Subject: Re: [PATCH v2 0/4] virtiofs,fuse: support per-file DAX
Date: Wed, 21 Jul 2021 10:42:50 -0400	[thread overview]
Message-ID: <YPgyalU0avl9KI/U@redhat.com> (raw)
In-Reply-To: <YPgXuacFfJ/JVRjo@redhat.com>

On Wed, Jul 21, 2021 at 08:48:57AM -0400, Vivek Goyal wrote:
[..]
> > > So is "dax=inode" enough for your needs? What's your requirement,
> > > can you give little bit of more details.
> > 
> > In our use case, the backend fs is something like SquashFS on host. The
> > content of the file on host is downloaded *as needed*. When the file is
> > not completely ready (completely downloaded), the guest will follow the
> > normal IO routine, i.e., by FUSE_READ/FUSE_WRITE request. While the file
> > is completely ready, per-file DAX is enabled for this file. IOW the FUSE
> > server need to dynamically decide if per-file DAX shall be enabled,
> > depending on if the file is completely downloaded.
> 
> So you don't want to enable DAX yet because guest might fault on
> a section of file which has not been downloaded yet?
> 
> I am wondering if somehow user fault handling can help with this.
> If we could handle faults for this file in user space, then you
> should be able to download that particular page[s] and resolve
> the fault?

Stefan mentioned that can't we block when fuse mmap request comes
in and download corresponding section of file. Or do whatever you
are doing in FUSE_READ. 

IOW, even if you enable dax in your use case on all files,
FUSE_SETUPMAPPING request will give you control to make sure 
file section being mmaped has been downloaded.

Vivek


  reply	other threads:[~2021-07-21 14:43 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-16 10:47 [PATCH v2 0/4] virtiofs,fuse: support per-file DAX Jeffle Xu
2021-07-16 10:47 ` Jeffle Xu
2021-07-16 10:47 ` [PATCH v2 1/4] fuse: add fuse_should_enable_dax() helper Jeffle Xu
2021-07-16 10:47   ` Jeffle Xu
2021-07-16 10:47 ` [PATCH v2 2/4] fuse: Make DAX mount option a tri-state Jeffle Xu
2021-07-16 10:47   ` Jeffle Xu
2021-07-19 18:02   ` Vivek Goyal
2021-07-19 18:02     ` Vivek Goyal
2021-07-20  5:54     ` JeffleXu
2021-07-20  5:54       ` JeffleXu
2021-07-16 10:47 ` [PATCH v2 3/4] fuse: add per-file DAX flag Jeffle Xu
2021-07-16 10:47   ` Jeffle Xu
2021-07-19 18:41   ` Vivek Goyal
2021-07-19 18:41     ` Vivek Goyal
2021-07-20  7:19     ` JeffleXu
2021-07-20  7:19       ` JeffleXu
2021-07-20 19:40       ` Vivek Goyal
2021-07-20 19:40         ` Vivek Goyal
2021-07-21 12:35         ` JeffleXu
2021-07-21 12:35           ` JeffleXu
2021-07-19 19:44   ` Vivek Goyal
2021-07-19 19:44     ` Vivek Goyal
2021-07-20  6:51     ` JeffleXu
2021-07-20  6:51       ` JeffleXu
2021-07-20  9:22       ` JeffleXu
2021-07-20  9:22         ` JeffleXu
2021-07-20 19:27       ` Vivek Goyal
2021-07-20 19:27         ` Vivek Goyal
2021-07-21 14:14         ` JeffleXu
2021-07-21 14:14           ` JeffleXu
2021-07-21 14:40           ` Vivek Goyal
2021-07-21 14:40             ` Vivek Goyal
2021-07-16 10:47 ` [PATCH v2 4/4] fuse: support changing per-file DAX flag inside guest Jeffle Xu
2021-07-16 10:47   ` Jeffle Xu
2021-07-19 19:54   ` Vivek Goyal
2021-07-19 19:54     ` Vivek Goyal
2021-07-19 21:30 ` [PATCH v2 0/4] virtiofs,fuse: support per-file DAX Vivek Goyal
2021-07-19 21:30   ` Vivek Goyal
2021-07-20  5:25   ` JeffleXu
2021-07-20  5:25     ` JeffleXu
2021-07-20 19:18     ` Vivek Goyal
2021-07-20 19:18       ` Vivek Goyal
2021-07-21 12:32       ` JeffleXu
2021-07-21 12:32         ` JeffleXu
2021-07-21 12:48         ` Vivek Goyal
2021-07-21 12:48           ` Vivek Goyal
2021-07-21 14:42           ` Vivek Goyal [this message]
2021-07-21 14:42             ` Vivek Goyal
2021-08-04  6:51             ` JeffleXu
2021-08-04  6:51               ` JeffleXu

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=YPgyalU0avl9KI/U@redhat.com \
    --to=vgoyal@redhat.com \
    --cc=bo.liu@linux.alibaba.com \
    --cc=dgilbert@redhat.com \
    --cc=jefflexu@linux.alibaba.com \
    --cc=joseph.qi@linux.alibaba.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=stefanha@redhat.com \
    --cc=virtualization@lists.linux-foundation.org \
    /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.