ebied...@xmission.com (Eric W. Biederman) writes:
> Miklos Szeredi <mik...@szeredi.hu> writes:
>
> Minor nits below.
>
>> diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c
>> new file mode 100644
>> index 0000000..6b50823
>> --- /dev/null
>> +++ b/fs/overlayfs/dir.c
>> @@ -0,0 +1,598 @@
>> +/*
>> + *
>> + * Copyright (C) 2011 Novell Inc.
>> + *
>> + * This program is free software; you can redistribute it and/or modify it
>> + * under the terms of the GNU General Public License version 2 as published
>> by
>> + * the Free Software Foundation.
>> + */
>> +
>> +#include <linux/fs.h>
>> +#include <linux/namei.h>
>> +#include <linux/xattr.h>
>> +#include <linux/security.h>
>> +#include <linux/cred.h>
>> +#include "overlayfs.h"
>> +
>> +static const char *ovl_whiteout_symlink = "(overlay-whiteout)";
>> +
>> +static int ovl_whiteout(struct dentry *upperdir, struct dentry *dentry)
>> +{
>> + int err;
>> + struct dentry *newdentry;
>> + const struct cred *old_cred;
>> + struct cred *override_cred;
>> +
>> + /* FIXME: recheck lower dentry to see if whiteout is really
>> needed */
>
> Is that FIXME still valid?
It is, but it's not an important feature. Lacking this will mean once a
file/directory is marked whiteout or opaque on the upper filesystem it
will remain so forever even after the file/directory it is masking out
has been removed from the lower filesystem.
However this cannot be observed by looking at the overlay, only by
looking at the underlying filesystems.
>
>> + err = -ENOMEM;
>> + override_cred = prepare_creds();
>> + if (!override_cred)
>> + goto out;
>> +
>> + /*
>> + * CAP_SYS_ADMIN for setxattr
>> + * CAP_DAC_OVERRIDE for symlink creation
>> + * CAP_FOWNER for unlink in sticky directory
>> + */
>> + cap_raise(override_cred->cap_effective, CAP_SYS_ADMIN);
>> + cap_raise(override_cred->cap_effective, CAP_DAC_OVERRIDE);
>> + cap_raise(override_cred->cap_effective, CAP_FOWNER);
>> + override_cred->fsuid = 0;
>> + override_cred->fsgid = 0;
>
> Could you please make these GLOBAL_ROOT_UID and GLOBAL_ROOT_GID
> instead of 0? Otherwise this code won't compile with the usernamespace
> bits enabled.
Okay.
Thanks for the review.
Miklos
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
