Re: [PATCH] mm/zblock: use vmalloc for page allocations

[Sergey Senozhatsky]

On (25/05/06 23:54), Christoph Hellwig wrote:
> On Wed, May 07, 2025 at 03:08:08PM +0900, Sergey Senozhatsky wrote:
> > > This sounds interesting.  We might get rid of lots of memcpy()
> > > in object read/write paths, and so on.  I don't know if 0-order
> > > chaining was the only option for zsmalloc, or just happened to
> > > be the first one.
> > 
> > I assume we might have problems with zspage release path.  vfree()
> > should break .swap_slot_free_notify, as far as I can see.
> > .swap_slot_free_notify is called under swap-cluster spin-lock,
> > so if we free the last object in the zspage we cannot immediately
> > free that zspage, because vfree() might_sleep().
> 
> Note that swap_slot_free_notify really needs to go away in favor
> of just sending a discard bio.  Having special block ops for a
> single user bypassing the proper block interface is not sustainable.

Oh, I didn't realize that zram was the only swap_slot_free_notify
user.  zram already handles REQ_OP_DISCARD/REQ_OP_WRITE_ZEROES so
I guess only swap-cluster needs some work.  Are there any
blockers/complications on the swap-cluster side?