lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
Open Source and information security mailing list archives
| ||
|
Message-ID: <ZGQJKRfuf4+av/MD@lore-desk>
Date: Wed, 17 May 2023 00:52:25 +0200
From: Lorenzo Bianconi <lorenzo@...nel.org>
To: Maciej Fijalkowski <maciej.fijalkowski@...el.com>
Cc: Lorenzo Bianconi <lorenzo.bianconi@...hat.com>,
Yunsheng Lin <linyunsheng@...wei.com>, netdev@...r.kernel.org,
bpf@...r.kernel.org, davem@...emloft.net, edumazet@...gle.com,
kuba@...nel.org, pabeni@...hat.com, ast@...nel.org,
daniel@...earbox.net, hawk@...nel.org, john.fastabend@...il.com
Subject: Re: [RFC net-next] net: veth: reduce page_pool memory footprint
using half page per-buffer
> On Mon, May 15, 2023 at 01:24:20PM +0200, Lorenzo Bianconi wrote:
> > > On 2023/5/12 21:08, Lorenzo Bianconi wrote:
> > > > In order to reduce page_pool memory footprint, rely on
> > > > page_pool_dev_alloc_frag routine and reduce buffer size
> > > > (VETH_PAGE_POOL_FRAG_SIZE) to PAGE_SIZE / 2 in order to consume one page
> > >
> > > Is there any performance improvement beside the memory saving? As it
> > > should reduce TLB miss, I wonder if the TLB miss reducing can even
> > > out the cost of the extra frag reference count handling for the
> > > frag support?
> >
> > reducing the requested headroom to 192 (from 256) we have a nice improvement in
> > the 1500B frame case while it is mostly the same in the case of paged skb
> > (e.g. MTU 8000B).
>
> Can you define 'nice improvement' ? ;)
> Show us numbers or improvement in %.
I am testing this RFC patch in the scenario reported below:
iperf tcp tx --> veth0 --> veth1 (xdp_pass) --> iperf tcp rx
- 6.4.0-rc1 net-next:
MTU 1500B: ~ 7.07 Gbps
MTU 8000B: ~ 14.7 Gbps
- 6.4.0-rc1 net-next + page_pool frag support in veth:
MTU 1500B: ~ 8.57 Gbps
MTU 8000B: ~ 14.5 Gbps
side note: it seems there is a regression between 6.2.15 and 6.4.0-rc1 net-next
(even without latest veth page_pool patches) in the throughput I can get in the
scenario above, but I have not looked into it yet.
- 6.2.15:
MTU 1500B: ~ 7.91 Gbps
MTU 8000B: ~ 14.1 Gbps
- 6.4.0-rc1 net-next w/o commits [0],[1],[2]
MTU 1500B: ~ 6.38 Gbps
MTU 8000B: ~ 13.2 Gbps
Regards,
Lorenzo
[0] 0ebab78cbcbf net: veth: add page_pool for page recycling
[1] 4fc418053ec7 net: veth: add page_pool stats
[2] 9d142ed484a3 net: veth: rely on napi_build_skb in veth_convert_skb_to_xdp_buff
>
> >
> > >
> > > > for two 1500B frames. Reduce VETH_XDP_PACKET_HEADROOM to 192 from 256
> > > > (XDP_PACKET_HEADROOM) to fit max_head_size in VETH_PAGE_POOL_FRAG_SIZE.
> > > > Please note, using default values (CONFIG_MAX_SKB_FRAGS=17), maximum
> > > > supported MTU is now reduced to 36350B.
> > >
> > > Maybe we don't need to limit the frag size to VETH_PAGE_POOL_FRAG_SIZE,
> > > and use different frag size depending on the mtu or packet size?
> > >
> > > Perhaps the page_pool_dev_alloc_frag() can be improved to return non-frag
> > > page if the requested frag size is larger than a specified size too.
> > > I will try to implement it if the above idea makes sense.
> > >
> >
> > since there are no significant differences between full page and fragmented page
> > implementation if the MTU is over the page boundary, does it worth to do so?
> > (at least for the veth use-case).
> >
> > Regards,
> > Lorenzo
> >
>
>
Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)
Powered by blists - more mailing lists