Re: linux-next: manual merge of the pidfd tree with the f2fs tree

From: Stephen Rothwell
Date: Sun Feb 21 2021 - 17:51:02 EST


Hi all,

On Mon, 15 Feb 2021 08:01:44 +1100 Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx> wrote:
>
> On Mon, 25 Jan 2021 16:48:06 +1100 Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx> wrote:
> >
> > Today's linux-next merge of the pidfd tree got a conflict in:
> >
> > fs/f2fs/file.c
> >
> > between commits:
> >
> > 19bb7075d1a9 ("f2fs: fix out-of-repair __setattr_copy()")
> > 7cf2e6173b2d ("f2fs: enhance to update i_mode and acl atomically in f2fs_setattr()")
> >
> > from the f2fs tree and commit:
> >
> > 2f221d6f7b88 ("attr: handle idmapped mounts")
> > e65ce2a50cf6 ("acl: handle idmapped mounts")
> >
> > from the pidfd tree.
> >
> > I fixed it up (see below) and can carry the fix as necessary. This
> > is now fixed as far as linux-next is concerned, but any non trivial
> > conflicts should be mentioned to your upstream maintainer when your tree
> > is submitted for merging. You may also want to consider cooperating
> > with the maintainer of the conflicting tree to minimise any particularly
> > complex conflicts.
> >
> > diff --cc fs/f2fs/file.c
> > index 7db27c81d034,8f1e97e7d242..000000000000
> > --- a/fs/f2fs/file.c
> > +++ b/fs/f2fs/file.c
> > @@@ -851,9 -848,9 +852,10 @@@ static void __setattr_copy(struct user_
> > inode->i_ctime = attr->ia_ctime;
> > if (ia_valid & ATTR_MODE) {
> > umode_t mode = attr->ia_mode;
> > + kgid_t kgid = i_gid_into_mnt(mnt_userns, inode);
> >
> > - if (!in_group_p(inode->i_gid) &&
> > - !capable_wrt_inode_uidgid(inode, CAP_FSETID))
> > - if (!in_group_p(kgid) && !capable(CAP_FSETID))
> > ++ if (!in_group_p(kgid) &&
> > ++ !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FSETID))
> > mode &= ~S_ISGID;
> > set_acl_inode(inode, mode);
> > }
> > @@@ -958,14 -948,13 +961,14 @@@ int f2fs_setattr(struct user_namespace
> > spin_unlock(&F2FS_I(inode)->i_size_lock);
> > }
> >
> > - __setattr_copy(inode, attr);
> > + __setattr_copy(&init_user_ns, inode, attr);
> >
> > if (attr->ia_valid & ATTR_MODE) {
> > - err = posix_acl_chmod(inode, f2fs_get_inode_mode(inode));
> > -
> > + err = posix_acl_chmod(&init_user_ns, inode,
> > + f2fs_get_inode_mode(inode));
> > - if (err || is_inode_flag_set(inode, FI_ACL_MODE)) {
> > - inode->i_mode = F2FS_I(inode)->i_acl_mode;
> > + if (is_inode_flag_set(inode, FI_ACL_MODE)) {
> > + if (!err)
> > + inode->i_mode = F2FS_I(inode)->i_acl_mode;
> > clear_inode_flag(inode, FI_ACL_MODE);
> > }
> > }
>
> With the merge window about to open, this is a reminder that this
> conflict still exists.
>
> The fsf2 tree commits are now
>
> 17232e830afb ("f2fs: enhance to update i_mode and acl atomically in f2fs_setattr()")
> 2562515f0ad7 ("f2fs: fix out-of-repair __setattr_copy()")

This is now a conflict between the pidfd tree and Linus' tree.

--
Cheers,
Stephen Rothwell

Attachment: pgpg4HZ6mUOUL.pgp
Description: OpenPGP digital signature