Thanks all for the comments.
I have a patchset ready that hopefully wraps together the extendability
suggested by Nikolaus, that I agree is a good idea.
The way I tried to make it more flexible is first of all transitioning to a
ioctl(), as suggested by both Jann and Miklos, and by using a data
Thanks Jann,
Sorry for the late reply, I wanted to better understand the problems you
mentioned and explore the ioctl solution before coming back to discussion.
I have a new patch ready which addresses the feedbacks received, and that
has been converted to using a new ioctl. And I have to say I'm
On Wed, Aug 19, 2020 at 11:25 AM Amir Goldstein wrote:
> > What I have in mind is things like not coupling the setup of the
> > passthrough fds to open(), but having a separate notification message for
> > this (like what we use for invalidation of cache), and adding not just
> > an "fd" field
On Tue, Aug 18, 2020 at 3:53 PM Alessio Balsini wrote:
> Thank you both for the important feedback,
>
> I tried to consolidate all your suggestions in the new version of the
> patch, shared below.
>
> As you both recommended, that tricky ki_filp swapping has been removed,
> taking overlayfs as a
On Wed, Aug 19, 2020 at 9:03 AM Nikolaus Rath wrote:
>
> Hi Alessio,
>
> Thank you for working on this, I'm excited to see things moving again on
> this front!
>
> What I would really like to see in the long-term is the ability for FUSE
> to support passthrough for specific areas of a file, i.e.
Hi Alessio,
Thank you for working on this, I'm excited to see things moving again on
this front!
What I would really like to see in the long-term is the ability for FUSE
to support passthrough for specific areas of a file, i.e. the ability to
specify different passthrough fds for different
Thank you both for the important feedback,
I tried to consolidate all your suggestions in the new version of the
patch, shared below.
As you both recommended, that tricky ki_filp swapping has been removed,
taking overlayfs as a reference for the management of asynchronous
requests.
The V7 below
On 8/12/20 12:29 PM, Jann Horn wrote:
>> + passthrough_inode = file_inode(passthrough_filp);
>> +
>> + iocb->ki_filp = passthrough_filp;
>
> Hmm... so we're temporarily switching out the iocb's ->ki_filp here? I
> wonder whether it is possible for some other code to look at ->ki_filp
On Thu, Aug 13, 2020 at 3:28 PM Alessio Balsini wrote:
> On Wed, Aug 12, 2020 at 08:29:58PM +0200, 'Jann Horn' via kernel-team wrote:
[...]
> > On Wed, Aug 12, 2020 at 6:15 PM Alessio Balsini wrote:
> > > Add support for filesystem passthrough read/write of files when enabled in
> > > userspace
Hi Jann,
Thank you for looking into this.
On Wed, Aug 12, 2020 at 08:29:58PM +0200, 'Jann Horn' via kernel-team wrote:
> [+Jens: can you have a look at that ->ki_filp switcheroo in
> fuse_passthrough_read_write_iter() and help figure out whether that's
> fine? This seems like your area of
[+Jens: can you have a look at that ->ki_filp switcheroo in
fuse_passthrough_read_write_iter() and help figure out whether that's
fine? This seems like your area of expertise.]
On Wed, Aug 12, 2020 at 6:15 PM Alessio Balsini wrote:
> Add support for filesystem passthrough read/write of files
Add support for filesystem passthrough read/write of files when enabled in
userspace through the option FUSE_PASSTHROUGH.
There are filesystems based on FUSE that are intended to enforce special
policies or trigger complicate decision makings at the file operations
level. Android, for example,
12 matches
Mail list logo