Messages in this thread | | | Date | Tue, 14 Jan 2014 09:38:54 +0900 | From | Minchan Kim <> | Subject | Re: [PATCH 4/7] zram: use atomic operation for stat |
| |
Hello Andrew,
On Mon, Jan 13, 2014 at 04:23:25PM -0800, Andrew Morton wrote: > On Tue, 14 Jan 2014 09:19:35 +0900 Minchan Kim <minchan@kernel.org> wrote: > > > On Mon, Jan 13, 2014 at 03:58:14PM -0800, Andrew Morton wrote: > > > On Mon, 13 Jan 2014 20:18:59 +0900 Minchan Kim <minchan@kernel.org> wrote: > > > > > > > Some of fields in zram->stats are protected by zram->lock which > > > > is rather coarse-grained so let's use atomic operation without > > > > explict locking. > > > > > > Confused. The patch didn't remove any locking, so it made the code > > > slower. > > > > True but it could make remove dependency of zram->lock for 32bit stat > > so further patches can remove messy code and enhance write performance. > > So, it's preparing patch for further step. > > Should I rewrite the description to explain this? > > That would be useful ;) I'd ask for performance testing results but I > expect they'll say "no difference". > > I grabbed patches 1-3 as they seems appropriate for 3.14.
I will resend it tomorrow with testing result. Thanks!
> > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/
-- Kind regards, Minchan Kim
| |