Re: [RFC PATCH 1/3] staging, zsmalloc: introduce zs_mem_[read/write]

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

> > +EXPORT_SYMBOL_GPL(zs_mem_write);
> 
> Two API have same logic but just different memcpy argument order for input/output
> so you can factor out common logic.

Or have one function but just use an extra argument to define the direction of the
copy.

> 
> Patch looks good to me but I have a concern.
> I'd like to promote zram/zsmalloc as soon as possible.
> My last hurdle was LOCKDEP complaint so I decided to stop sending promoting patches
> until it was solved. At that same time, Nitin was sending some patches on zram meta
> diet and critical bug fix. Of course, they was conflict so we should line patches up
> following as
> 
> 1. Critical bug fix and merge <- merged two days ago.
> 2. Nitin diet patch merge <- pending
> 3. Minchan Lockdep patch merge <- pending
> 
> And then, my plan was trying to promote again.
> But unfortunately, I was not convinced of 2 at that time while we all agree on 3.
> So it takes some time to discuss 2 again and finally merge.
> So I would like to merge lockdep patch as top priority and then,
> Joonsoo/Nitin/Seth could try to send your patches to staging.
> (Seth already had a patch to solve lockdep problem simply in his zswap series
> but I don't like it although I didn't reply his patch.)
> 
> If anyone has objection, please raise your hand.
> I will do best effort to send lockdep patch until early next week.

No objections here.
> 
> -- 
> Kind regards,
> Minchan Kim