linux-kernel - [PATCH 0/4] clock_gettime_ns and x86-64 optimizations

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <cover.1324831829.git.luto@amacapital.net>
Date:	Sun, 25 Dec 2011 08:50:59 -0800
From:	Andy Lutomirski <luto@...capital.net>
To:	linux-kernel@...r.kernel.org, Kumar Sundararajan <kumar@...com>,
	john stultz <johnstul@...ibm.com>, Arun Sharma <asharma@...com>
Cc:	Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...e.hu>,
	Thomas Gleixner <tglx@...utronix.de>,
	Richard Cochran <richardcochran@...il.com>,
	Andy Lutomirski <luto@...capital.net>
Subject: [PATCH 0/4] clock_gettime_ns and x86-64 optimizations

On x86-64, clock_gettime is so fast that the overhead converting to and
from nanoseconds is non-negligible.  clock_gettime_ns is a different and
faster interface.

Patch 1 adds the syscall and wires it up on x86-64.  Patch 2 implements
the corresponding vdso entry on x86-64.  Patch 3 optimizes the vdso
call, and patch 4 is a trivial change that speeds up the vdso
clock_gettime and clock_gettime_ns implementations.

The vdso timings are (on an 800MHz Sandy Bridge mobile):

Basic implementation:

realtime 77.4ns
monotonic 79.2ns
realtime_coarse 18.1ns
monotonic_coarse 22.0ns

realtime_ns 84.9ns
monotonic_ns 85.1ns
realtime_coarse_ns 19.49
monotonic_coarse_ns 27.32

Optimized implementation:

realtime 78.5ns
monotonic 77.4ns [a little faster -- maybe significant]
realtime_coarse 18.4ns
monotonic_coarse 19.4ns

realtime_ns 77.85ns [a nice improvement]
monotonic_ns 77.75ns [ditto]
realtime_coarse_ns 18.2ns
monotonic_coarse_ns 18.2ns [a lot faster]

Inlined (patch 4): [everything is improved]

realtime 73.4ns
monotonic 72.1ns
realtime_coarse 13.2ns
monotonic_coarse 15.8ns

realtime_ns 73.15ns
monotonic_ns 72.1ns
realtime_coarse_ns 14.1ns
monotonic_coarse_ns 15.6ns

This being the middle of the holidays, I reserve the right to have made
mistakes.

For the git-inclined, this series is at
https://git.kernel.org/?p=linux/kernel/git/luto/linux.git;a=shortlog;h=refs/heads/timing/clock_gettime_ns/patch_v1

Andy Lutomirski (4):
  Add clock_gettime_ns syscall
  x86-64: Add __vdso_clock_gettime_ns vsyscall
  x86-64: Optimize vdso clock_gettime
  x86-64: Inline vdso clock_gettime helpers

 arch/x86/include/asm/unistd_64.h |    2 +
 arch/x86/include/asm/vgtod.h     |   21 ++++--
 arch/x86/kernel/vsyscall_64.c    |   25 +++++++-
 arch/x86/vdso/vclock_gettime.c   |  136 ++++++++++++++++++++++++++------------
 arch/x86/vdso/vdso.lds.S         |    7 ++
 include/linux/syscalls.h         |    3 +
 include/linux/time.h             |    5 ++
 kernel/posix-timers.c            |   30 ++++++++
 8 files changed, 179 insertions(+), 50 deletions(-)

-- 
1.7.7.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/