Re: Zram writeback feature unstable with heavy swap utilization - BUG: Bad page state in process...

[Minchan Kim <minchan@kernel.org>]

On Tue, Jul 24, 2018 at 07:55:25PM -0700, Matthew Wilcox wrote:
> On Wed, Jul 25, 2018 at 11:51:06AM +0900, Minchan Kim wrote:
> > On Tue, Jul 24, 2018 at 07:35:32PM -0700, Matthew Wilcox wrote:
> > > There is NOTHING in a union with _refcount.
> > 
> > Confusing. Matthew, what am I missing?
> > 
> > Before:
> > 
> > counters consumes 8 bytes
> > units and _refcount consumes each 4 bytes so memory space is not overlapped.
> 
> Correct.  _mapcount is at offset 24, as is units.  _refcount is at offset 28.
> 
> >         union {
> >                 unsigned long counters;
> >                 struct {
> >                         union {
> >                                 atomic_t _mapcount;
> >                                 int units;                      /* SLOB */
> >                         };
> >                         atomic_t _refcount;
> >                 };
> >         };
> > 
> > 
> > After:
> > 
> > Now, units is overlapped with _refcount and _mapcount.
> 
> No.  units is at offset 24, as is _mapcount.  But _refcount is still at
> offset 28.

Ah, got it what you mean. Sorry for the noise, Matthew.

> 
> >         union {
> >                 unsigned long counters;
> >                 int units;                      /* SLOB */
> >                 struct {                        /* Page cache */
> >                         atomic_t _mapcount;
> >                         atomic_t _refcount;
> >                 };
> >         };