netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Alexei Starovoitov <ast@kernel.org>
To: <davem@davemloft.net>
Cc: <daniel@iogearbox.net>, <jakub.kicinski@netronome.com>,
	<jannh@google.com>, <netdev@vger.kernel.org>,
	<bpf@vger.kernel.org>, <kernel-team@fb.com>
Subject: [PATCH bpf-next 3/7] bpf: improve verification speed by not remarking live_read
Date: Fri, 29 Mar 2019 17:16:08 -0700	[thread overview]
Message-ID: <20190330001612.2354959-4-ast@kernel.org> (raw)
In-Reply-To: <20190330001612.2354959-1-ast@kernel.org>

With large verifier speed improvement brought by the previous patch
mark_reg_read() becomes the hottest function during verification.
On a typical program it consumes 40% of cpu.
mark_reg_read() walks parentage chain of registers to mark parents as LIVE_READ.
Once the register is marked there is no need to remark it again in the future.
Hence stop walking the chain once first LIVE_READ is seen.
This optimization drops mark_reg_read() time from 40% of cpu to <1%
and overall 2x improvement of verification speed.
For some programs the longest_mark_read_walk counter improves from ~500 to ~5

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index b18512ac205e..6dfd148b58f6 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1151,6 +1151,15 @@ static int mark_reg_read(struct bpf_verifier_env *env,
 				parent->var_off.value, parent->off);
 			return -EFAULT;
 		}
+		if (parent->live & REG_LIVE_READ)
+			/* The parentage chain never changes and
+			 * this parent was already marked as LIVE_READ.
+			 * There is no need to keep walking the chain again and
+			 * keep re-marking all parents as LIVE_READ.
+			 * This case happens when the same register is read
+			 * multiple times without writes into it in-between.
+			 */
+			break;
 		/* ... then we depend on parent's value */
 		parent->live |= REG_LIVE_READ;
 		state = parent;
-- 
2.20.0


  parent reply	other threads:[~2019-03-30  0:16 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-30  0:16 [PATCH bpf-next 0/7] bpf: improve verifier scalability Alexei Starovoitov
2019-03-30  0:16 ` [PATCH bpf-next 1/7] bpf: add verifier stats and log_level bit 2 Alexei Starovoitov
2019-03-30  0:16 ` [PATCH bpf-next 2/7] bpf: improve verification speed by droping states Alexei Starovoitov
2019-03-30  3:12   ` Jakub Kicinski
2019-03-30  3:29     ` Alexei Starovoitov
2019-03-30  0:16 ` Alexei Starovoitov [this message]
2019-03-30  3:12   ` [PATCH bpf-next 3/7] bpf: improve verification speed by not remarking live_read Jakub Kicinski
2019-04-01 15:38   ` Edward Cree
2019-03-30  0:16 ` [PATCH bpf-next 4/7] bpf: increase complexity limit and maximum program size Alexei Starovoitov
2019-03-30  0:16 ` [PATCH bpf-next 5/7] bpf: increase verifier log limit Alexei Starovoitov
2019-03-30  3:16   ` Jakub Kicinski
2019-03-30  0:16 ` [PATCH bpf-next 6/7] libbpf: teach libbpf about log_level bit 2 Alexei Starovoitov
2019-03-30  0:16 ` [PATCH bpf-next 7/7] selftests/bpf: add few verifier scale tests Alexei Starovoitov
2019-04-01 14:17   ` Daniel Borkmann
2019-04-01 16:35     ` Daniel Borkmann
2019-04-02  2:08       ` Alexei Starovoitov
2019-04-02 14:13         ` Daniel Borkmann
2019-04-02 18:12           ` Alexei Starovoitov
2019-03-30  3:18 ` [PATCH bpf-next 0/7] bpf: improve verifier scalability Jakub Kicinski
2019-03-30  3:35   ` Alexei Starovoitov

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=20190330001612.2354959-4-ast@kernel.org \
    --to=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=jakub.kicinski@netronome.com \
    --cc=jannh@google.com \
    --cc=kernel-team@fb.com \
    --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).