On Thu, 5 Nov 2015 19:25:14 +0300
Vladimir Davydov <vdavydov@xxxxxxxxxxxxx> wrote:
> On Thu, Nov 05, 2015 at 04:38:06PM +0100, Jesper Dangaard Brouer wrote:
> > Initial implementation missed support for kmem cgroup support
> > in kmem_cache_free_bulk() call, add this.
> >
> > If CONFIG_MEMCG_KMEM is not enabled, the compiler should
> > be smart enough to not add any asm code.
> >
> > Signed-off-by: Jesper Dangaard Brouer <brouer@xxxxxxxxxx>
> >
> > ---
> > V2: Fixes according to input from:
> > Vladimir Davydov <vdavydov@xxxxxxxxxxxxx>
> > and Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
> >
> > mm/slub.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/mm/slub.c b/mm/slub.c
> > index 8e9e9b2ee6f3..bc64514ad1bb 100644
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -2890,6 +2890,9 @@ void kmem_cache_free_bulk(struct kmem_cache *s, size_t size, void **p)
> > do {
> > struct detached_freelist df;
> >
> > + /* Support for memcg */
> > + s = cache_from_obj(s, p[size - 1]);
> > +
>
> AFAIU all objects in the array should come from the same cache (should
> they?), so it should be enough to call this only once:
Can we be sure all objects in the array come from same cache?
Imagine my use case:
1. application send packet alloc a SKB (from a slab)
2. packet TX to NIC via DMA
3. TX DMA completion cleans up 256 packets and kmem free SKBs
I don't know enough about mem cgroups... but I can imagine two
applications belonging to different mem-cgroups sending packet out same
NIC and later getting their SKB (pkt-metadata struct) free'ed during
the same TX completion (TX softirq) cycle, as a bulk free.
With my limited mem cgroups, it looks like memcg works on the slab-page
level? And what I'm doing in this code is to group object together
belonging to the same slab-page.
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
Author of http://www.iptv-analyzer.org
LinkedIn: http://www.linkedin.com/in/brouer
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>