On Tue, Jul 04, 2017 at 02:11:10PM +0300, Timofey Titovets wrote:
> >> @@ -0,0 +1,336 @@
> >> +enum compression_advice btrfs_compress_heuristic(struct inode *inode,
> >
> > This returns enum but the caller treats it as a bool, this should be
> > synced up.
> >
>
> I think for now, enum and values can be just dropped,
> because it's only a proof of concept that in future, heuristic
> can return more complex answer and then some of that values
> can be used as acceptable for zlib and unnacceptable for lzo & etc.
> Also that can be easy added again later.
Right.
> > As mentioned in the previous mail, we'll add the skeleton code only for
> > now, which means this loop, that simply calls find_get_page, kunmap and
> > put_page.
> >
>
> So, i will send a separate patch for skeleton code.
> And if i understood all correctly, later, we'll add complex code one by one.
Yes.
> >> +
> >> +enum compression_advice {
> >> + COMPRESS_NONE,
> >> + COMPRESS_COST_UNKNOWN,
> >> + COMPRESS_COST_EASY,
> >> + COMPRESS_COST_MEDIUM,
> >> + COMPRESS_COST_HARD
> >
> > I don't find the naming good, as the cost is not easy or hard, but
> > rather low/medium/high. Or you can rename the whole prefix if you find a
> > better naming scheme.
>
> (Explain above)
> So just drop that for now.
Ok.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
