Re: [PATCH] mm/vmalloc: fix numa spreading for large hash tables

[Chen Wandun]

在 2021/9/29 6:33, Andrew Morton 写道:
> On Tue, 28 Sep 2021 20:10:40 +0800 Chen Wandun <chenwandun@xxxxxxxxxx> wrote:
>
> > Eric Dumazet reported a strange numa spreading info in [1], and found
> > commit 121e6f3258fe ("mm/vmalloc: hugepage vmalloc mappings") introduced
> > this issue [2].
> >
> > Dig into the difference before and after this patch, page allocation has
> > some difference:
> >
> > before:
> > alloc_large_system_hash
> >      __vmalloc
> >          __vmalloc_node(..., NUMA_NO_NODE, ...)
> >              __vmalloc_node_range
> >                  __vmalloc_area_node
> >                      alloc_page /* because NUMA_NO_NODE, so choose alloc_page branch */
> >                          alloc_pages_current
> >                              alloc_page_interleave /* can be proved by print policy mode */
> >
> > after:
> > alloc_large_system_hash
> >      __vmalloc
> >          __vmalloc_node(..., NUMA_NO_NODE, ...)
> >              __vmalloc_node_range
> >                  __vmalloc_area_node
> >                      alloc_pages_node /* choose nid by nuam_mem_id() */
> >                          __alloc_pages_node(nid, ....)
> >
> > So after commit 121e6f3258fe ("mm/vmalloc: hugepage vmalloc mappings"),
> > it will allocate memory in current node instead of interleaving allocate
> > memory.
> >
> > [1]
> > https://lore.kernel.org/linux-mm/CANn89iL6AAyWhfxdHO+jaT075iOa3XcYn9k6JJc7JR2XYn6k_Q@xxxxxxxxxxxxxx/
> >
> > [2]
> > https://lore.kernel.org/linux-mm/CANn89iLofTR=AK-QOZY87RdUZENCZUT4O6a0hvhu3_EwRMerOg@xxxxxxxxxxxxxx/
> >
> > Fixes: 121e6f3258fe ("mm/vmalloc: hugepage vmalloc mappings")
> > Reported-by: Eric Dumazet <edumazet@xxxxxxxxxx>
> > Signed-off-by: Chen Wandun <chenwandun@xxxxxxxxxx>
>
> This seems like it could cause significant performance regressions in
> some situations?
Yes，I indeed will cause some performance regressions, I will send a 
optimization patch based on this patch.
> If "yes" then wouldn't a cc:stable be appropriate?  And some (perhaps
> handwavy) quantification of the slowdown would help people understand
> why we're recommending a backport.
>
> If "no" then why the heck do we have that feature in there anyway ;)
>
> .