Ocfs2 devel - Oct 2018 - [PATCH 10/25] vfs: create generic_remap_file_range_touch to update inode metadata

> +/* Update inode timestamps and remove security privileges when remapping.
*/
> +int generic_remap_file_range_touch(struct file *file, bool is_dedupe)
> +{
> +	int ret;
> +
> +	/* If can't alter the file contents, we're done. */
> +	if (is_dedupe)
> +		return 0;
> +
> +	/* Update the timestamps, since we can alter file contents. */
> +	if (!(file->f_mode & FMODE_NOCMTIME)) {
> +		ret = file_update_time(file);
> +		if (ret)
> +			return ret;
> +	}
> +
> +	/*
> +	 * Clear the security bits if the process is not being run by root.
> +	 * This keeps people from modifying setuid and setgid binaries.
> +	 */
> +	return file_remove_privs(file);
> +}
> +EXPORT_SYMBOL(generic_remap_file_range_touch);
The name seems a little out of touch with what it actually does.
Also why a bool argument instead of the more descriptive flags which
introduced a few patches ago?

On Sun, Oct 14, 2018 at 10:21:31AM -0700, Christoph Hellwig
wrote:
> > +/* Update inode timestamps and remove security privileges when
remapping. */
> > +int generic_remap_file_range_touch(struct file *file, bool is_dedupe)
> > +{
> > +	int ret;
> > +
> > +	/* If can't alter the file contents, we're done. */
> > +	if (is_dedupe)
> > +		return 0;
> > +
> > +	/* Update the timestamps, since we can alter file contents. */
> > +	if (!(file->f_mode & FMODE_NOCMTIME)) {
> > +		ret = file_update_time(file);
> > +		if (ret)
> > +			return ret;
> > +	}
> > +
> > +	/*
> > +	 * Clear the security bits if the process is not being run by root.
> > +	 * This keeps people from modifying setuid and setgid binaries.
> > +	 */
> > +	return file_remove_privs(file);
> > +}
> > +EXPORT_SYMBOL(generic_remap_file_range_touch);
> 
> The name seems a little out of touch with what it actually does.
I originally thought "touch" because it updates [cm]time. :)

Though looking at the final code, I think this can just be called from
the end of generic_remap_file_range_prep, so we can skip the export and
all that other stuff.
> Also why a bool argument instead of the more descriptive flags which
> introduced a few patches ago?
Hmm, yes, the remap_flags transition can move up to before this patch.

--D