On Wed, Jul 11, 2018 at 05:15:40PM +0100, Al Viro wrote:

> Actually, looking at the entire thing, I'm rather tempted to go for
>       alloc_empty_file(f_flags, cred)
> setting both f_flags and f_flags-derived part of f_mode, making
>       alloc_file_pseudo(inode, mnt, name, f_flags, ops)
>       alloc_file_clone(base, f_flags, ops)
> do the same automatically as they call alloc_empty_file().

See vfs.git#work.open3; one commit added just before "pass creds
to get_empty_filp(), make sure dentry_open() passes the right creds",
one just after.

The first one ("alloc_file(): switch to passing O_... flags instead of
 FMODE_... mode") pulls ->f_flags assignments into preceding alloc_file(),
the second ("pass ->f_flags value to alloc_empty_file()") makes
alloc_empty_file() set ->f_flags and ->f_flags-derived part of ->f_mode,
with do_dentry_open() leaving these parts of ->f_mode alone.

The rest is pretty much identical to the previous iteration of the
series, except that alloc_file wrappers inherit the switch to passing

A question regarding the customs in such situations - are previous
Reviewed-by/Acked-by normally kept across rebases like that?

Reply via email to