netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jesper Dangaard Brouer <brouer@redhat.com>
To: Daniel Borkmann <daniel@iogearbox.net>
Cc: brouer@redhat.com, netdev@vger.kernel.org
Subject: Re: [PATCH bpf-next 05/11] bpf: avoid retpoline for lookup/update/delete calls on maps
Date: Tue, 29 May 2018 19:23:24 +0200	[thread overview]
Message-ID: <20180529192324.7a1db7ee@redhat.com> (raw)
In-Reply-To: <20180528004344.3606-6-daniel@iogearbox.net>

On Mon, 28 May 2018 02:43:38 +0200
Daniel Borkmann <daniel@iogearbox.net> wrote:

> While some of the BPF map lookup helpers provide a ->map_gen_lookup()
> callback for inlining the map lookup altogether it is not available
> for every map, so the remaining ones have to call bpf_map_lookup_elem()
> helper which does a dispatch to map->ops->map_lookup_elem(). In
> times of retpolines, this will control and trap speculative execution
> rather than letting it do its work for the indirect call and will
> therefore cause a slowdown. Likewise, bpf_map_update_elem() and
> bpf_map_delete_elem() do not have an inlined version and need to call
> into their map->ops->map_update_elem() resp. map->ops->map_delete_elem()
> handlers.
> 
> Before:
> 
>   # bpftool p d x i 1

I would really appreciate if we can use the long options in these kind
of examples.  It makes the command "self-documenting" and searchable by
google.

Here it would be:

 # bpftool prog dump xlated id 1

>     0: (bf) r2 = r10
>     1: (07) r2 += -8
>     2: (7a) *(u64 *)(r2 +0) = 0
>     3: (18) r1 = map[id:1]
>     5: (85) call __htab_map_lookup_elem#232656
>     6: (15) if r0 == 0x0 goto pc+4
>     7: (71) r1 = *(u8 *)(r0 +35)
>     8: (55) if r1 != 0x0 goto pc+1
>     9: (72) *(u8 *)(r0 +35) = 1
>    10: (07) r0 += 56
>    11: (15) if r0 == 0x0 goto pc+4
>    12: (bf) r2 = r0
>    13: (18) r1 = map[id:1]
>    15: (85) call bpf_map_delete_elem#215008  <-- indirect call via
>    16: (95) exit                                 helper
> 
> After:
> 
>   # bpftool p d x i 1

Same here

>     0: (bf) r2 = r10
>     1: (07) r2 += -8
>     2: (7a) *(u64 *)(r2 +0) = 0
>     3: (18) r1 = map[id:1]
>     5: (85) call __htab_map_lookup_elem#233328
>     6: (15) if r0 == 0x0 goto pc+4
>     7: (71) r1 = *(u8 *)(r0 +35)
>     8: (55) if r1 != 0x0 goto pc+1
>     9: (72) *(u8 *)(r0 +35) = 1
>    10: (07) r0 += 56
>    11: (15) if r0 == 0x0 goto pc+4
>    12: (bf) r2 = r0
>    13: (18) r1 = map[id:1]
>    15: (85) call htab_lru_map_delete_elem#238240  <-- direct call
>    16: (95) exit
> 


-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer

  reply	other threads:[~2018-05-29 17:23 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-28  0:43 [PATCH bpf-next 00/11] Misc BPF improvements Daniel Borkmann
2018-05-28  0:43 ` [PATCH bpf-next 01/11] bpf: test case for map pointer poison with calls/branches Daniel Borkmann
2018-05-29 18:01   ` Song Liu
2018-05-28  0:43 ` [PATCH bpf-next 02/11] bpf: add also cbpf long jump test cases with heavy expansion Daniel Borkmann
2018-05-29 18:09   ` Song Liu
2018-05-28  0:43 ` [PATCH bpf-next 03/11] bpf: fixup error message from gpl helpers on license mismatch Daniel Borkmann
2018-05-29 17:16   ` Jesper Dangaard Brouer
2018-05-29 18:10     ` Song Liu
2018-05-28  0:43 ` [PATCH bpf-next 04/11] bpf: show prog and map id in fdinfo Daniel Borkmann
2018-05-29 17:27   ` Jesper Dangaard Brouer
2018-05-29 19:55     ` Daniel Borkmann
2018-05-30 16:15       ` Song Liu
2018-05-30 17:15         ` Jesper Dangaard Brouer
2018-05-28  0:43 ` [PATCH bpf-next 05/11] bpf: avoid retpoline for lookup/update/delete calls on maps Daniel Borkmann
2018-05-29 17:23   ` Jesper Dangaard Brouer [this message]
2018-05-30 17:06   ` Song Liu
2018-05-28  0:43 ` [PATCH bpf-next 06/11] bpf: add bpf_skb_cgroup_id helper Daniel Borkmann
2018-05-29 12:15   ` Quentin Monnet
2018-05-29 15:43     ` Daniel Borkmann
2018-05-28  0:43 ` [PATCH bpf-next 07/11] bpf: make sure to clear unused fields in tunnel/xfrm state fetch Daniel Borkmann
2018-05-30 17:15   ` Song Liu
2018-05-28  0:43 ` [PATCH bpf-next 08/11] bpf: fix cbpf parser bug for octal numbers Daniel Borkmann
2018-05-30 17:16   ` Song Liu
2018-05-28  0:43 ` [PATCH bpf-next 09/11] bpf: fix context access in tracing progs on 32 bit archs Daniel Borkmann
2018-05-30 16:46   ` Song Liu
2018-05-28  0:43 ` [PATCH bpf-next 10/11] bpf: sync bpf uapi header with tools Daniel Borkmann
2018-05-30 16:10   ` Song Liu
2018-05-28  0:43 ` [PATCH bpf-next 11/11] bpf, doc: add missing patchwork url and libbpf to maintainers Daniel Borkmann
2018-05-30  0:16   ` Song Liu

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=20180529192324.7a1db7ee@redhat.com \
    --to=brouer@redhat.com \
    --cc=daniel@iogearbox.net \
    --cc=netdev@vger.kernel.org \
    /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;
as well as URLs for NNTP newsgroup(s).