Jeff Layton <jlayton@xxxxxxxxxx> wrote: > > + /* check parameters */ > > + ret = -EOPNOTSUPP; > > + if (d_is_negative(root) || > > + !d_backing_inode(root)->i_op->lookup || > > + !d_backing_inode(root)->i_op->mkdir || > > + !(d_backing_inode(root)->i_opflags & IOP_XATTR) || > > + !root->d_sb->s_op->statfs || > > + !root->d_sb->s_op->sync_fs || > > + root->d_sb->s_blocksize > PAGE_SIZE) > > + goto error_unsupported; > > + > > That's quite a collection of tests. > > Most are obvious, but some comments explaining the need for others would > not be a bad thing. In particular, why is s_blocksize > PAGE_SIZE > unsupported? It can't do page-sized DIO requests to a filesystem with a block size larger than a page. In the future I can work around that in conjunction with netfslib by expanding read and write sizes. > Also, should you vet whether the fs supports i_op->tmpfile here ? That's a good idea. David -- Linux-cachefs mailing list Linux-cachefs@xxxxxxxxxx https://listman.redhat.com/mailman/listinfo/linux-cachefs
