From: "Samudrala, Sridhar" <sridhar.samudrala@intel.com>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: "Björn Töpel" <bjorn.topel@gmail.com>,
"Toke Høiland-Jørgensen" <toke@redhat.com>,
"Jakub Kicinski" <jakub.kicinski@netronome.com>,
Netdev <netdev@vger.kernel.org>,
intel-wired-lan <intel-wired-lan@lists.osuosl.org>,
"Herbert, Tom" <tom.herbert@intel.com>,
"Fijalkowski, Maciej" <maciej.fijalkowski@intel.com>,
"bpf@vger.kernel.org" <bpf@vger.kernel.org>,
"Björn Töpel" <bjorn.topel@intel.com>,
"Karlsson, Magnus" <magnus.karlsson@intel.com>
Subject: Re: [Intel-wired-lan] FW: [PATCH bpf-next 2/4] xsk: allow AF_XDP sockets to receive packets directly from a queue
Date: Tue, 22 Oct 2019 12:06:18 -0700 [thread overview]
Message-ID: <8956a643-0163-5345-34fa-3566762a2b7d@intel.com> (raw)
In-Reply-To: <CAADnVQ+jiEO+jnFR-G=xG=zz7UOSBieZbc1NN=sSnAwvPaJjUQ@mail.gmail.com>
On 10/21/2019 3:34 PM, Alexei Starovoitov wrote:
> On Mon, Oct 21, 2019 at 1:10 PM Samudrala, Sridhar
> <sridhar.samudrala@intel.com> wrote:
>>
>> On 10/20/2019 10:12 AM, Björn Töpel wrote:
>>> On Sun, 20 Oct 2019 at 12:15, Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>>>>
>>>> Alexei Starovoitov <alexei.starovoitov@gmail.com> writes:
>>>>
>>>>> On Fri, Oct 18, 2019 at 05:45:26PM -0700, Samudrala, Sridhar wrote:
>>>>>> On 10/18/2019 5:14 PM, Alexei Starovoitov wrote:
>>>>>>> On Fri, Oct 18, 2019 at 11:40:07AM -0700, Samudrala, Sridhar wrote:
>>>>>>>>
>>>>>>>> Perf report for "AF_XDP default rxdrop" with patched kernel - mitigations ON
>>>>>>>> ==========================================================================
>>>>>>>> Samples: 44K of event 'cycles', Event count (approx.): 38532389541
>>>>>>>> Overhead Command Shared Object Symbol
>>>>>>>> 15.31% ksoftirqd/28 [i40e] [k] i40e_clean_rx_irq_zc
>>>>>>>> 10.50% ksoftirqd/28 bpf_prog_80b55d8a76303785 [k] bpf_prog_80b55d8a76303785
>>>>>>>> 9.48% xdpsock [i40e] [k] i40e_clean_rx_irq_zc
>>>>>>>> 8.62% xdpsock xdpsock [.] main
>>>>>>>> 7.11% ksoftirqd/28 [kernel.vmlinux] [k] xsk_rcv
>>>>>>>> 5.81% ksoftirqd/28 [kernel.vmlinux] [k] xdp_do_redirect
>>>>>>>> 4.46% xdpsock bpf_prog_80b55d8a76303785 [k] bpf_prog_80b55d8a76303785
>>>>>>>> 3.83% xdpsock [kernel.vmlinux] [k] xsk_rcv
>>>>>>>
>>>>>>> why everything is duplicated?
>>>>>>> Same code runs in different tasks ?
>>>>>>
>>>>>> Yes. looks like these functions run from both the app(xdpsock) context and ksoftirqd context.
>>>>>>
>>>>>>>
>>>>>>>> 2.81% ksoftirqd/28 [kernel.vmlinux] [k] bpf_xdp_redirect_map
>>>>>>>> 2.78% ksoftirqd/28 [kernel.vmlinux] [k] xsk_map_lookup_elem
>>>>>>>> 2.44% xdpsock [kernel.vmlinux] [k] xdp_do_redirect
>>>>>>>> 2.19% ksoftirqd/28 [kernel.vmlinux] [k] __xsk_map_redirect
>>>>>>>> 1.62% ksoftirqd/28 [kernel.vmlinux] [k] xsk_umem_peek_addr
>>>>>>>> 1.57% xdpsock [kernel.vmlinux] [k] xsk_umem_peek_addr
>>>>>>>> 1.32% ksoftirqd/28 [kernel.vmlinux] [k] dma_direct_sync_single_for_cpu
>>>>>>>> 1.28% xdpsock [kernel.vmlinux] [k] bpf_xdp_redirect_map
>>>>>>>> 1.15% xdpsock [kernel.vmlinux] [k] dma_direct_sync_single_for_device
>>>>>>>> 1.12% xdpsock [kernel.vmlinux] [k] xsk_map_lookup_elem
>>>>>>>> 1.06% xdpsock [kernel.vmlinux] [k] __xsk_map_redirect
>>>>>>>> 0.94% ksoftirqd/28 [kernel.vmlinux] [k] dma_direct_sync_single_for_device
>>>>>>>> 0.75% ksoftirqd/28 [kernel.vmlinux] [k] __x86_indirect_thunk_rax
>>>>>>>> 0.66% ksoftirqd/28 [i40e] [k] i40e_clean_programming_status
>>>>>>>> 0.64% ksoftirqd/28 [kernel.vmlinux] [k] net_rx_action
>>>>>>>> 0.64% swapper [kernel.vmlinux] [k] intel_idle
>>>>>>>> 0.62% ksoftirqd/28 [i40e] [k] i40e_napi_poll
>>>>>>>> 0.57% xdpsock [kernel.vmlinux] [k] dma_direct_sync_single_for_cpu
>>>>>>>>
>>>>>>>> Perf report for "AF_XDP direct rxdrop" with patched kernel - mitigations ON
>>>>>>>> ==========================================================================
>>>>>>>> Samples: 46K of event 'cycles', Event count (approx.): 38387018585
>>>>>>>> Overhead Command Shared Object Symbol
>>>>>>>> 21.94% ksoftirqd/28 [i40e] [k] i40e_clean_rx_irq_zc
>>>>>>>> 14.36% xdpsock xdpsock [.] main
>>>>>>>> 11.53% ksoftirqd/28 [kernel.vmlinux] [k] xsk_rcv
>>>>>>>> 11.32% xdpsock [i40e] [k] i40e_clean_rx_irq_zc
>>>>>>>> 4.02% xdpsock [kernel.vmlinux] [k] xsk_rcv
>>>>>>>> 2.91% ksoftirqd/28 [kernel.vmlinux] [k] xdp_do_redirect
>>>>>>>> 2.45% ksoftirqd/28 [kernel.vmlinux] [k] xsk_umem_peek_addr
>>>>>>>> 2.19% xdpsock [kernel.vmlinux] [k] xsk_umem_peek_addr
>>>>>>>> 2.08% ksoftirqd/28 [kernel.vmlinux] [k] bpf_direct_xsk
>>>>>>>> 2.07% ksoftirqd/28 [kernel.vmlinux] [k] dma_direct_sync_single_for_cpu
>>>>>>>> 1.53% ksoftirqd/28 [kernel.vmlinux] [k] dma_direct_sync_single_for_device
>>>>>>>> 1.39% xdpsock [kernel.vmlinux] [k] dma_direct_sync_single_for_device
>>>>>>>> 1.22% ksoftirqd/28 [kernel.vmlinux] [k] xdp_get_xsk_from_qid
>>>>>>>> 1.12% ksoftirqd/28 [i40e] [k] i40e_clean_programming_status
>>>>>>>> 0.96% ksoftirqd/28 [i40e] [k] i40e_napi_poll
>>>>>>>> 0.95% ksoftirqd/28 [kernel.vmlinux] [k] net_rx_action
>>>>>>>> 0.89% xdpsock [kernel.vmlinux] [k] xdp_do_redirect
>>>>>>>> 0.83% swapper [i40e] [k] i40e_clean_rx_irq_zc
>>>>>>>> 0.70% swapper [kernel.vmlinux] [k] intel_idle
>>>>>>>> 0.66% xdpsock [kernel.vmlinux] [k] dma_direct_sync_single_for_cpu
>>>>>>>> 0.60% xdpsock [kernel.vmlinux] [k] bpf_direct_xsk
>>>>>>>> 0.50% ksoftirqd/28 [kernel.vmlinux] [k] xsk_umem_discard_addr
>>>>>>>>
>>>>>>>> Based on the perf reports comparing AF_XDP default and direct rxdrop, we can say that
>>>>>>>> AF_XDP direct rxdrop codepath is avoiding the overhead of going through these functions
>>>>>>>> bpf_prog_xxx
>>>>>>>> bpf_xdp_redirect_map
>>>>>>>> xsk_map_lookup_elem
>>>>>>>> __xsk_map_redirect
>>>>>>>> With AF_XDP direct, xsk_rcv() is directly called via bpf_direct_xsk() in xdp_do_redirect()
>>>>>>>
>>>>>>> I don't think you're identifying the overhead correctly.
>>>>>>> xsk_map_lookup_elem is 1%
>>>>>>> but bpf_xdp_redirect_map() suppose to call __xsk_map_lookup_elem()
>>>>>>> which is a different function:
>>>>>>> ffffffff81493fe0 T __xsk_map_lookup_elem
>>>>>>> ffffffff81492e80 t xsk_map_lookup_elem
>>>>>>>
>>>>>>> 10% for bpf_prog_80b55d8a76303785 is huge.
>>>>>>> It's the actual code of the program _without_ any helpers.
>>>>>>> How does the program actually look?
>>>>>>
>>>>>> It is the xdp program that is loaded via xsk_load_xdp_prog() in tools/lib/bpf/xsk.c
>>>>>> https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git/tree/tools/lib/bpf/xsk.c#n268
>>>>>
>>>>> I see. Looks like map_gen_lookup was never implemented for xskmap.
>>>>> How about adding it first the way array_map_gen_lookup() is implemented?
>>>>> This will easily give 2x perf gain.
>>>>
>>>> I guess we should implement this for devmaps as well now that we allow
>>>> lookups into those.
>>>>
>>>> However, in this particular example, the lookup from BPF is not actually
>>>> needed, since bpf_redirect_map() will return a configurable error value
>>>> when the map lookup fails (for exactly this use case).
>>>>
>>>> So replacing:
>>>>
>>>> if (bpf_map_lookup_elem(&xsks_map, &index))
>>>> return bpf_redirect_map(&xsks_map, index, 0);
>>>>
>>>> with simply
>>>>
>>>> return bpf_redirect_map(&xsks_map, index, XDP_PASS);
>>>>
>>>> would save the call to xsk_map_lookup_elem().
>>>>
>>>
>>> Thanks for the reminder! I just submitted a patch. Still, doing the
>>> map_gen_lookup() for xsk/devmaps still makes sense!
>>>
>>
>> I tried Bjorn's patch that avoids the lookups in the BPF prog.
>> https://lore.kernel.org/netdev/20191021105938.11820-1-bjorn.topel@gmail.com/
>>
>> With this patch I am also seeing around 3-4% increase in xdpsock rxdrop performance and
>> the perf report looks like this.
>>
>> Samples: 44K of event 'cycles', Event count (approx.): 38749965204
>> Overhead Command Shared Object Symbol
>> 16.06% ksoftirqd/28 [i40e] [k] i40e_clean_rx_irq_zc
>> 10.18% ksoftirqd/28 bpf_prog_3c8251c7e0fef8db [k] bpf_prog_3c8251c7e0fef8db
>> 10.15% xdpsock [i40e] [k] i40e_clean_rx_irq_zc
>> 10.06% ksoftirqd/28 [kernel.vmlinux] [k] xsk_rcv
>> 7.45% xdpsock xdpsock [.] main
>> 5.76% ksoftirqd/28 [kernel.vmlinux] [k] xdp_do_redirect
>> 4.51% xdpsock bpf_prog_3c8251c7e0fef8db [k] bpf_prog_3c8251c7e0fef8db
>> 3.67% xdpsock [kernel.vmlinux] [k] xsk_rcv
>> 3.06% ksoftirqd/28 [kernel.vmlinux] [k] bpf_xdp_redirect_map
>> 2.34% ksoftirqd/28 [kernel.vmlinux] [k] __xsk_map_redirect
>> 2.33% xdpsock [kernel.vmlinux] [k] xdp_do_redirect
>> 1.69% ksoftirqd/28 [kernel.vmlinux] [k] xsk_umem_peek_addr
>> 1.69% xdpsock [kernel.vmlinux] [k] xsk_umem_peek_addr
>> 1.42% ksoftirqd/28 [kernel.vmlinux] [k] dma_direct_sync_single_for_cpu
>> 1.19% xdpsock [kernel.vmlinux] [k] bpf_xdp_redirect_map
>> 1.13% xdpsock [kernel.vmlinux] [k] dma_direct_sync_single_for_device
>> 0.95% ksoftirqd/28 [kernel.vmlinux] [k] dma_direct_sync_single_for_device
>> 0.92% swapper [kernel.vmlinux] [k] intel_idle
>> 0.92% xdpsock [kernel.vmlinux] [k] __xsk_map_redirect
>> 0.80% ksoftirqd/28 [kernel.vmlinux] [k] __x86_indirect_thunk_rax
>> 0.73% ksoftirqd/28 [i40e] [k] i40e_clean_programming_status
>> 0.71% ksoftirqd/28 [kernel.vmlinux] [k] __xsk_map_lookup_elem
>> 0.63% ksoftirqd/28 [kernel.vmlinux] [k] net_rx_action
>> 0.62% ksoftirqd/28 [i40e] [k] i40e_napi_poll
>> 0.58% xdpsock [kernel.vmlinux] [k] dma_direct_sync_single_for_cpu
>>
>> So with this patch applied, direct receive performance improvement comes down from 46% to 42%.
>> I think it is still substantial enough to provide an option to allow direct receive for
>> certain use cases. If it is OK, i can re-spin and submit the patches on top of the latest bpf-next
>
> I think it's too early to consider such drastic approach.
> The run-time performance of XDP program should be the same as C code.
> Something fishy in these numbers, since spending 10% cpu in few loads
> and single call to bpf_xdp_redirect_map() just not right.
OK. Here is another data point that shows the perf report with the same test but CPU mitigations
turned OFF. Here bpf_prog overhead goes down from almost (10.18 + 4.51)% to (3.23 + 1.44%).
21.40% ksoftirqd/28 [i40e] [k] i40e_clean_rx_irq_zc
14.13% xdpsock [i40e] [k] i40e_clean_rx_irq_zc
8.33% ksoftirqd/28 [kernel.vmlinux] [k] xsk_rcv
6.09% ksoftirqd/28 [kernel.vmlinux] [k] xdp_do_redirect
5.19% xdpsock xdpsock [.] main
3.48% ksoftirqd/28 [kernel.vmlinux] [k] bpf_xdp_redirect_map
3.23% ksoftirqd/28 bpf_prog_3c8251c7e0fef8db [k] bpf_prog_3c8251c7e0fef8db
3.06% ksoftirqd/28 [kernel.vmlinux] [k] __xsk_map_redirect
2.72% xdpsock [kernel.vmlinux] [k] xdp_do_redirect
2.27% xdpsock [kernel.vmlinux] [k] xsk_rcv
2.10% xdpsock [kernel.vmlinux] [k] xsk_umem_peek_addr
2.09% ksoftirqd/28 [kernel.vmlinux] [k] xsk_umem_peek_addr
1.89% ksoftirqd/28 [kernel.vmlinux] [k] dma_direct_sync_single_for_cpu
1.44% xdpsock bpf_prog_3c8251c7e0fef8db [k] bpf_prog_3c8251c7e0fef8db
1.36% xdpsock [kernel.vmlinux] [k] bpf_xdp_redirect_map
1.31% ksoftirqd/28 [kernel.vmlinux] [k] dma_direct_sync_single_for_device
1.30% xdpsock [kernel.vmlinux] [k] __xsk_map_redirect
1.23% xdpsock [kernel.vmlinux] [k] dma_direct_sync_single_for_device
0.97% ksoftirqd/28 [kernel.vmlinux] [k] __xsk_map_lookup_elem
0.90% ksoftirqd/28 [i40e] [k] i40e_clean_programming_status
0.81% xdpsock [kernel.vmlinux] [k] dma_direct_sync_single_for_cpu
0.76% swapper [i40e] [k] i40e_clean_rx_irq_zc
0.75% swapper [kernel.vmlinux] [k] intel_idle
0.59% ksoftirqd/28 [kernel.vmlinux] [k] net_rx_action
So a major component of the bpf_prog overhead seems to be due to the CPU vulnerability mitigations.
The other component is the bpf_xdp_redirect_map() codepath.
Let me know if it helps to collect any other data that should further help with the perf analysis.
next prev parent reply other threads:[~2019-10-22 19:06 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-08 6:16 [PATCH bpf-next 0/4] Enable direct receive on AF_XDP sockets Sridhar Samudrala
2019-10-08 6:16 ` [PATCH bpf-next 1/4] bpf: introduce bpf_get_prog_id and bpf_set_prog_id helper functions Sridhar Samudrala
2019-10-08 6:16 ` [PATCH bpf-next 2/4] xsk: allow AF_XDP sockets to receive packets directly from a queue Sridhar Samudrala
2019-10-08 6:58 ` Toke Høiland-Jørgensen
2019-10-08 8:47 ` Björn Töpel
2019-10-08 8:48 ` Björn Töpel
2019-10-08 9:04 ` Toke Høiland-Jørgensen
2019-10-08 8:05 ` Björn Töpel
2019-10-09 16:32 ` Samudrala, Sridhar
2019-10-09 1:20 ` Alexei Starovoitov
[not found] ` <3ED8E928C4210A4289A677D2FEB48235140134CE@fmsmsx111.amr.corp.intel.com>
2019-10-09 16:53 ` FW: " Samudrala, Sridhar
2019-10-09 17:17 ` Alexei Starovoitov
2019-10-09 19:12 ` Samudrala, Sridhar
2019-10-10 1:06 ` Alexei Starovoitov
2019-10-18 18:40 ` Samudrala, Sridhar
2019-10-18 19:22 ` Toke Høiland-Jørgensen
2019-10-19 0:14 ` Alexei Starovoitov
2019-10-19 0:45 ` Samudrala, Sridhar
2019-10-19 2:25 ` Alexei Starovoitov
2019-10-20 10:14 ` Toke Høiland-Jørgensen
2019-10-20 17:12 ` [Intel-wired-lan] " Björn Töpel
2019-10-21 20:10 ` Samudrala, Sridhar
2019-10-21 22:34 ` Alexei Starovoitov
2019-10-22 19:06 ` Samudrala, Sridhar [this message]
2019-10-23 17:42 ` Alexei Starovoitov
2019-10-24 18:12 ` Samudrala, Sridhar
2019-10-25 7:42 ` Björn Töpel
2019-10-31 22:38 ` Samudrala, Sridhar
2019-10-31 23:15 ` Alexei Starovoitov
2019-11-01 0:21 ` Jakub Kicinski
2019-11-01 18:31 ` Samudrala, Sridhar
2019-11-04 2:08 ` dan
2019-10-25 9:07 ` Björn Töpel
2019-10-08 6:16 ` [PATCH bpf-next 3/4] libbpf: handle AF_XDP sockets created with XDP_DIRECT bind flag Sridhar Samudrala
2019-10-08 8:05 ` Björn Töpel
2019-10-08 6:16 ` [PATCH bpf-next 4/4] xdpsock: add an option to create AF_XDP sockets in XDP_DIRECT mode Sridhar Samudrala
2019-10-08 8:05 ` Björn Töpel
2019-10-08 8:05 ` [PATCH bpf-next 0/4] Enable direct receive on AF_XDP sockets Björn Töpel
2019-10-09 16:19 ` Samudrala, Sridhar
2019-10-09 0:49 ` Jakub Kicinski
2019-10-09 6:29 ` Samudrala, Sridhar
2019-10-09 16:53 ` Jakub Kicinski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=8956a643-0163-5345-34fa-3566762a2b7d@intel.com \
--to=sridhar.samudrala@intel.com \
--cc=alexei.starovoitov@gmail.com \
--cc=bjorn.topel@gmail.com \
--cc=bjorn.topel@intel.com \
--cc=bpf@vger.kernel.org \
--cc=intel-wired-lan@lists.osuosl.org \
--cc=jakub.kicinski@netronome.com \
--cc=maciej.fijalkowski@intel.com \
--cc=magnus.karlsson@intel.com \
--cc=netdev@vger.kernel.org \
--cc=toke@redhat.com \
--cc=tom.herbert@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox