Re: [PATCH 1/2] mm: slub: allocate slab object extensions non-contiguously

[Uladzislau Rezki <urezki@gmail.com>]

On Tue, May 20, 2025 at 10:28:06AM -0400, Kent Overstreet wrote:
> On Tue, May 20, 2025 at 07:24:40AM -0700, Shakeel Butt wrote:
> > On Tue, May 20, 2025 at 10:01:27AM -0400, Kent Overstreet wrote:
> > > On Tue, May 20, 2025 at 02:46:14PM +0100, Usama Arif wrote:
> > > > 
> > > > 
> > > > On 20/05/2025 14:44, Kent Overstreet wrote:
> > > > > On Tue, May 20, 2025 at 01:25:46PM +0100, Usama Arif wrote:
> > > > >> When memory allocation profiling is running on memory bound services,
> > > > >> allocations greater than order 0 for slab object extensions can fail,
> > > > >> for e.g. zs_handle zswap slab which will be 512 objsperslab x 16 bytes
> > > > >> per slabobj_ext (order 1 allocation). Use kvcalloc to improve chances
> > > > >> of the allocation being successful.
> > > > >>
> > > > >> Signed-off-by: Usama Arif <usamaarif642@gmail.com>
> > > > >> Reported-by: Vlad Poenaru <vlad.wing@gmail.com>
> > > > >> Closes: https://lore.kernel.org/all/17fab2d6-5a74-4573-bcc3-b75951508f0a@gmail.com/
> > > > >> ---
> > > > >>  mm/slub.c | 2 +-
> > > > >>  1 file changed, 1 insertion(+), 1 deletion(-)
> > > > >>
> > > > >> diff --git a/mm/slub.c b/mm/slub.c
> > > > >> index dc9e729e1d26..bf43c403ead2 100644
> > > > >> --- a/mm/slub.c
> > > > >> +++ b/mm/slub.c
> > > > >> @@ -1989,7 +1989,7 @@ int alloc_slab_obj_exts(struct slab *slab, struct kmem_cache *s,
> > > > >>  	gfp &= ~OBJCGS_CLEAR_MASK;
> > > > >>  	/* Prevent recursive extension vector allocation */
> > > > >>  	gfp |= __GFP_NO_OBJ_EXT;
> > > > >> -	vec = kcalloc_node(objects, sizeof(struct slabobj_ext), gfp,
> > > > >> +	vec = kvcalloc_node(objects, sizeof(struct slabobj_ext), gfp,
> > > > >>  			   slab_nid(slab));
> > > > > 
> > > > > And what's the latency going to be on a vmalloc() allocation when we're
> > > > > low on memory?
> > > > 
> > > > Would it not be better to get the allocation slighly slower than to not get
> > > > it at all?
> > > 
> > > Our behaviour when thrashing sucks, we don't want to do anything to make
> > > that worse.
> > > 
> > > There's also the fact that vmalloc doesn't correctly respect gfp flags,
> > > so until that gets fixed this doesn't work at all.
> > 
> > Which gfp flags vmalloc is not respecting today?
> 
> GFP_NOWAIT.
> 
> As to why, you'd better ask Michal Hocko...
> 
It is mainly due to pte_alloc_one_kernel(), it uses the GFP_KERNEL

#define GFP_PGTABLE_KERNEL	(GFP_KERNEL | __GFP_ZERO)

to get a new pte entry.

I think we can fix it. For example if we populate some region and allocate
there for NOWAIT. But there are of course can be other hidden problems.

--
Uladzislau Rezki