Re: [PATCH 08/15] xfs: don't modify file and inode flags for shmem files

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Jan 03, 2024 at 04:01:45PM -0800, Darrick J. Wong wrote:
> I actually want S_PRIVATE here to avoid interference from all the
> security hooks and whatnot when scrub is using an xfile to stash a
> large amount of data.  Shouldn't this patch change xfile_create to call
> shmem_kernel_file_setup instead?

Yes, and it used to do that before I reshuffled it..

> > -	inode->i_mode &= ~0177;
> > -	inode->i_uid = GLOBAL_ROOT_UID;
> > -	inode->i_gid = GLOBAL_ROOT_GID;
> 
> Also, I don't know if it matters that the default uid/gid are now going
> to be whatever the defaults would be for a new file instead of root
> only.  That seems like it could invite problems, but otoh xfiles are
> never installed in the fd table so userspace should never get access
> anyway.

In-kernel shm files are created on shm_mnt, which is owned by the global
root, so this will do the right thing.




[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux