On Fri, Feb 12, 2021 at 3:46 PM Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
>
> On Fri, Feb 12, 2021 at 12:44:00PM +0800, Nicolas Boichat wrote:
> > Filesystems such as procfs and sysfs generate their content at
> > runtime. This implies the file sizes do not usually match the
> > amount of data that can be read from the file, and that seeking
> > may not work as intended.
> >
> > This will be useful to disallow copy_file_range with input files
> > from such filesystems.
> >
> > Signed-off-by: Nicolas Boichat <drinkcat@xxxxxxxxxxxx>
> > ---
> > I first thought of adding a new field to struct file_operations,
> > but that doesn't quite scale as every single file creation
> > operation would need to be modified.
>
> Even so, you missed a load of filesystems in the kernel with this patch
> series, what makes the ones you did mark here different from the
> "internal" filesystems that you did not?
Which ones did I miss? (apart from configfs, see the conversation on
patch 6/6). Anyway, hopefully auditing all filesystems is an order of
magnitude easier task, and easier to maintain in the longer run ,-)
> This feels wrong, why is userspace suddenly breaking? What changed in
> the kernel that caused this? Procfs has been around for a _very_ long
> time :)
Some of this is covered in the cover letter 0/6. To expand a bit:
copy_file_range has only supported cross-filesystem copies since 5.3
[1], before that the kernel would return EXDEV and userspace
application would fall back to a read/write based copy. After 5.3,
copy_file_range copies no data as it thinks the input file is empty.
[1] I think it makes little sense to try to use copy_file_range
between 2 files in /proc, but technically, I think that has been
broken since copy_file_range fallback to do_splice_direct was
introduced (eac70053a141 ("vfs: Add vfs_copy_file_range() support for
pagecache copies", ~4.4).
>
> thanks,
>
> greg k-h
